Zero-shot document re-ranking with Large Language Models (LLMs) has evolved from Pointwise methods to Listwise and Setwise approaches that optimize computational efficiency. Despite their success, these methods predominantly rely on generative scoring or output logits, which face bottlenecks in inference latency and result consistency. In-Context Re-ranking (ICR) has recently been proposed as an O(1) alternative method. ICR extracts internal attention signals directly, avoiding the overhead of text generation. However, existing ICR methods simply aggregate signals across all layers; layer-wise contributions and their consistency across architectures have been left unexplored. Furthermore, no unified study has compared internal attention with traditional generative and likelihood-based mechanisms across diverse ranking frameworks under consistent conditions.

Human relevance assessment is time-consuming and cognitively intensive, limiting the scalability of Information Retrieval evaluation. This has led to growing interest in using large language models (LLMs) as proxies for human judges. However, it remains an open question whether LLM-based relevance judgments are reliable, stable, and rigorous enough to match humans for relevance assessment. In this work, we conduct a study of overrating behavior in LLM-based relevance judgments across model backbones, evaluation paradigms (pointwise and pairwise), and passage modification strategies. We show that models consistently assign inflated relevance scores-often with high confidence-to passages that do not genuinely satisfy the underlying information need, revealing a system-wide bias rather than random fluctuations in judgment. Furthermore, controlled experiments show that LLM-based relevance judgments can be highly sensitive to passage length and surface-level lexical cues. These results raise concerns about the usage of LLMs as drop-in replacements for human relevance assessors, and highlight the urgent need for careful diagnostic evaluation frameworks when applying LLMs for relevance assessments. Our code and results are publicly available. https://github.com/chutingyu/Exploring-Overrating

Document chunking is a critical preprocessing step in dense retrieval systems, yet the design space of chunking strategies remains poorly understood. Recent research has proposed several concurrent approaches, including LLM-guided methods (e.g., DenseX and LumberChunker) and contextualized strategies (e.g., Late Chunking), which generate embeddings before segmentation to preserve contextual information. However, these methods emerged independently and were evaluated on benchmarks with minimal overlap, making direct comparisons difficult.

Despite extensive research on a wide range of question answering (QA) systems, most existing work focuses on answer containment-i.e., assuming that answers can be directly extracted and/or generated from documents in the corpus. However, some questions require inference, i.e., deriving answers that are not explicitly stated but can be inferred from the available information. We introduce Inferential QA -- a new task that challenges models to infer answers from answer-supporting passages which provide only clues. To study this problem, we construct QUIT (QUestions requiring Inference from Texts) dataset, comprising 7,401 questions and 2.4M passages built from high-convergence human- and machine-authored hints, labeled across three relevance levels using LLM-based answerability and human verification. Through comprehensive evaluation of retrievers, rerankers, and LLM-based readers, we show that methods effective on traditional QA tasks struggle in inferential QA: retrievers underperform, rerankers offer limited gains, and fine-tuning provides inconsistent improvements. Even reasoning-oriented LLMs fail to outperform smaller general-purpose models. These findings reveal that current QA pipelines are not yet ready for inference-based reasoning. Inferential QA thus establishes a new class of QA tasks that move towards understanding and reasoning from indirect textual evidence.

Systematic reviews are comprehensive literature reviews that address highly focused research questions and represent the highest form of evidence in medicine. A critical step in this process is the development of complex Boolean queries to retrieve relevant literature. Given the difficulty of manually constructing these queries, recent efforts have explored Large Language Models (LLMs) to assist in their formulation. One of the first studies, Wang et al. [21], investigated ChatGPT for this task, followed by Staudinger et al. [14], which evaluated multiple LLMs in a reproducibility study. However, the latter overlooked several key aspects of the original work, including (i) validation of generated queries (ii) output formatting constraints, and (iii) selection of examples for chainof- thought (Guided) prompting. As a result, its findings diverged significantly from the original study. In this work, we systematically reproduce both studies while addressing these overlooked factors. Our results show that query effectiveness varies significantly across models and prompt designs, with guided query formulation benefiting from well-chosen seed studies. Overall, prompt design and model selection are key drivers of successful query formulation. Our findings provide a clearer understanding of LLMs' potential in Boolean query generation and highlight the importance of modeland prompt-specific optimisations. The complex nature of systematic reviews adds to challenges in both developing and reproducing methods but also highlights the importance of reproducibility studies in this domain.

2D Matryoshka Training is an advanced embedding representation training approach designed to train an encoder model simultaneously across various layer-dimension setups. This method has demonstrated higher effectiveness in Semantic Text Similarity (STS) tasks over traditional training approaches when using sub-layers for embeddings. Despite its success, discrepancies exist between two published implementations, leading to varied comparative results with baseline models. In this reproducibility study, we implement and evaluate both versions of 2D Matryoshka Training on STS tasks and extend our analysis to retrieval tasks. Our findings indicate that while both versions achieve higher effectiveness than traditional Matryoshka training on sub-dimensions, and traditional full-sized model training approaches, they do not outperform models trained separately on specific sub-layer and sub-dimension setups. Moreover, these results generalize well to retrieval tasks, both in supervised (MSMARCO) and zero-shot (BEIR) settings. Further explorations of different loss computations reveals more suitable implementations for retrieval tasks, such as incorporating full-dimension loss and training on a broader range of target dimensions. Conversely, some intuitive approaches, such as fixing document encoders to full model outputs, do not yield improvements. Our reproduction code is available at https://github.com/ielab/2DMSE-Reproduce.

Systematic reviews are fundamental to evidence-based medicine.Creating one is time-consuming and labour-intensive, mainly due to the need to screen, or assess, many studies for inclusion in the review.Existing tools help streamline this process, mostly using traditional machine learning.Large language models (LLMs) offer new opportunities to speed up screening, yet no tool currently enables users to directly apply LLMs or ensures systematic and transparent use of these methods.This paper presents (i) a flexible framework for using LLMs in systematic review tasks, especially title and abstract screening, and (ii) a web-based interface for LLMassisted screening.Together, they form AiReview-a novel platform that connects cutting-edge LLM-assisted screening methods with real-world systematic review practice.The live tool is available at https://aireview.ielab.io.We also release the code publicly at https://github.com/ielab/ai-review.

Large Language Models (LLMs) often exhibit hallucinations, which makes detecting and mitigating these errors a critical challenge. The Retrofit Attribution using Research and Revision (RARR) framework addresses this challenge by extracting key aspects of an LLM response, verifying them against retrieved evidence, and resolving errors through re-prompting. In this work, we critically examine RARR and adapt its framework to incorporate publicly available evidence retrieval systems and generative models, thereby operationalizing the approach. We focus on hallucination detection, analyzing how each pipeline component contributes to this task. We also conduct a sentence-level analysis of hallucinations to provide a more granular assessment of RARR's performance. A key finding is that while query generation and retrieval are effective, the agreement module emerges as the weakest link in the RARR pipeline. We offer deeper insights into RARR's strengths, limitations, and potential areas for improvement, thereby broadening our understanding of hallucination detection in LLMs.

This paper reports on findings from a comparative study on the effectiveness and efficiency of federated unlearning strategies within Federated Online Learning to Rank (FOLTR), with specific attention to systematically analysing the unlearning capabilities of methods in a verifiable manner.Federated approaches to ranking of search results have recently garnered attention to address users privacy concerns.In FOLTR, privacy is safeguarded by collaboratively training ranking models across decentralized data sources, preserving individual user data while optimizing search results based on implicit feedback, such as clicks.Recent legislation introduced across numerous countries is establishing the so called "the right to be forgotten", according to which services based on machine learning models like those in FOLTR should provide capabilities that allow users to remove their own data from those used to train models.This has sparked the development of unlearning methods, along with evaluation practices to measure whether unlearning of a user data successfully occurred.Current evaluation practices are however often controversial, necessitating the use of multiple metrics for a more comprehensive assessment -but previous proposals of unlearning methods only used single evaluation metrics.This paper addresses this limitation: our study rigorously assesses the effectiveness of unlearning strategies in managing both under-unlearning and over-unlearning scenarios using adapted, and newly proposed evaluation metrics.Thanks to our detailed

Dense retrievers utilize pre-trained backbone language models (e.g., BERT, LLaMA) that are fine-tuned via contrastive learning to perform the task of encoding text into sense representations that can be then compared via a shallow similarity operation, e.g. inner product.

Recent advancements in dense retrieval have introduced vision-language model (VLM)-based retrievers, such as DSE and ColPali, which leverage document screenshots embedded as vectors to enable effective search and offer a simplified pipeline over traditional text-only methods. In this study, we propose three pixel poisoning attack methods designed to compromise VLM-based retrievers and evaluate their effectiveness under various attack settings and parameter configurations. Our empirical results demonstrate that injecting even a single adversarial screenshot into the retrieval corpus can significantly disrupt search results, poisoning the top-10 retrieved documents for 41.9% of queries in the case of DSE and 26.4% for ColPali. These vulnerability rates notably exceed those observed with equivalent attacks on text-only retrievers. Moreover, when targeting a small set of known queries, the attack success rate raises, achieving complete success in certain cases. By exposing the vulnerabilities inherent in vision-language models, this work highlights the potential risks associated with their deployment.

Managing the review process for large-scale academic conferences poses significant challenges in effectively matching papers to the right reviewers, detecting conflicts of interest, ensuring review quality, and addressing potential ethical issues such as dual submissions. In this demonstration paper, we introduce ReviewHQ, an API-based system designed to streamline conference management. For matching reviewers to publications, ReviewHQ constructs expertise profiles for potential reviewers by mining their publication records and representing both papers and reviewers as dense vectors. It then formulates paper assignment as a constrained optimization problem, leveraging dense vector similarity scores and other reviewer's and submission's features to produce high-quality assignments. Beyond its core functionality, ReviewHQ identifies conflicts of interest by analyzing co-authorship histories and flags suspected dual submissions by comparing manuscripts against submission information from other conferences. Further, the system employs methods to detect low-quality or automated (AI-generated) reviews and pinpoint discrepancies between reviewers' recommendations and final acceptance decisions. ReviewHQ has been used since 2024 across the SIGIR and SIGIR-AP conferences.

Generative Large Language Models (LLMs) like GPT, Gemini, and Llama are transforming Information Retrieval, enabling new and more effective approaches to document retrieval and ranking. The switch from the previous generation pre-trained language models backbones (e.g., BERT, T5) to the new generative LLMs backbones has required the field to adapt training processes; it also has provided unprecedented capabilities and opportunities, stimulating research into zero-shot approaches, reasoning approaches, reinforcement learning based training, and multilingual and multimodal applications. This tutorial will provide a structured overview of LLM-based retrievers and rankers, covering fundamental architectures, training paradigms, real-world deployment considerations, and open challenges and research directions.

Recent advances in large language models have enabled the development of viable generative retrieval systems. Instead of a traditional document ranking, generative retrieval systems often directly return a grounded generated text as a response to a query. Quantifying the utility of the textual responses is essential for appropriately evaluating such generative ad hoc retrieval. Yet, the established evaluation methodology for ranking-based ad hoc retrieval is not suited for the reliable and reproducible evaluation of generated responses. To lay a foundation for developing new evaluation methods for generative retrieval systems, we survey the relevant literature from the fields of information retrieval and natural language processing, identify search tasks and system architectures in generative retrieval, develop a new user model, and study its operationalization.

In this demo we present a web-based application for selecting an effective pre-trained dense retriever to use on a private collection. Our system, DenseQuest, provides unsupervised selection and ranking capabilities to predict the best dense retriever among a pool of available dense retrievers, tailored to an uploaded target collection. DenseQuest implements a number of existing approaches, including a recent, highly effective method powered by Large Language Models (LLMs), which requires neither queries nor relevance judgments. The system is designed to be intuitive and easy to use for those information retrieval engineers and researchers who need to identify a general-purpose dense retrieval model to encode or search a new private target collection. Our demonstration illustrates conceptual architecture and the different use case scenarios of the system implemented on the cloud, enabling universal access and use. DenseQuest is available at https://densequest.ielab.io.

In this paper we present Large Language Model Assisted Retrieval Model Ranking (LARMOR), an effective unsupervised approach that leverages LLMs for selecting which dense retriever to use on a test corpus (target). Dense retriever selection is crucial for many IR applications that rely on using dense retrievers trained on public corpora to encode or search a new, private target corpus. This is because when confronted with domain shift, where the downstream corpora, domains, or tasks of the target corpus differ from the domain/task the dense retriever was trained on, its performance often drops. Furthermore, when the target corpus is unlabeled, e.g., in a zero-shot scenario, the direct evaluation of the model on the target corpus becomes unfeasible. Unsupervised selection of the most effective pre-trained dense retriever becomes then a crucial challenge. Current methods for dense retriever selection are insufficient in handling scenarios with domain shift.

Document expansion is a technique that aims to reduce the likelihood of term mismatch by augmenting documents with related terms or queries. Doc2Query minus minus (Doc2Query-) represents an extension to the expansion process that uses a neural model to identify and remove expansions that may not be relevant to the given document, thereby increasing the quality of the ranking while simultaneously reducing the amount of augmented data. In this work, we conduct a detailed reproducibility study of Doc2Query- to better understand the trade-offs inherent to document expansion and filtering mechanisms. After successfully reproducing the best-performing method from the Doc2Query- family, we show that filtering actually harms recall-based metrics on various test collections. Next, we explore whether the two-stage "generate-then-filter" process can be replaced with a single generation phase via reinforcement learning. Finally, we extend our experimentation to learned sparse retrieval models and demonstrate that filtering is not helpful when term weights can be learned. Overall, our work provides a deeper understanding of the behaviour and characteristics of common document expansion mechanisms, and paves the way for developing more efficient yet effective augmentation models.

The goal of screening prioritisation in systematic reviews is to identify relevant documents with high recall and rank them in early positions for review. This saves reviewing effort if paired with a stopping criterion, and speeds up review completion if performed alongside downstream tasks. Recent studies have shown that neural models have good potential on this task, but their time-consuming fine-tuning and inference discourage their widespread use for screening prioritisation. In this paper, we propose an alternative approach that still relies on neural models, but leverages dense representations and relevance feedback to enhance screening prioritisation, without the need for costly model fine-tuning and inference. This method exploits continuous relevance feedback from reviewers during document screening to efficiently update the dense query representation, which is then applied to rank the remaining documents to be screened. We evaluate this approach across the CLEF TAR datasets for this task. Results suggest that the investigated dense query-driven approach is more efficient than directly using neural models and shows promising effectiveness compared to previous methods developed on the considered datasets. Our code is available at https://github.com/ielab/dense-screening-feedback.

Federated search systems aggregate results from multiple search engines, selecting appropriate sources to enhance result quality and align with user intent. With the increasing uptake of Retrieval-Augmented Generation (RAG) pipelines, federated search can play a pivotal role in sourcing relevant information across heterogeneous data sources to generate informed responses. However, existing datasets, such as those developed in the past TREC FedWeb tracks, predate the RAG paradigm shift and lack representation of modern information retrieval challenges.

Text stemming is a natural language processing technique that is used to reduce words to their base form, also known as the root form. In Information Retrieval (IR), stemming is used in keyword-based matching pipelines to normalise text before indexing and query processing to improve subsequent matching between document and query keywords. The use of stemming has been shown to often improve the effectiveness of keyword-matching models such as BM25. However, traditional stemming methods, focusing solely on individual terms, overlook the richness of contextual information.

We propose a novel zero-shot document ranking approach based on Large Language Models (LLMs): the Setwise prompting approach.Our approach complements existing prompting approaches for LLM-based zero-shot ranking: Pointwise, Pairwise, and Listwise.Through the first-of-its-kind comparative evaluation within a consistent experimental framework and considering factors like model size, token consumption, latency, among others, we show that existing approaches are inherently characterised by trade-offs between effectiveness and efficiency.We find that while Pointwise approaches score high on efficiency, they suffer from poor effectiveness.Conversely, Pairwise approaches demonstrate superior effectiveness but incur high computational overhead.Our Setwise approach, instead, reduces the number of LLM inferences and the amount of prompt token consumption during the ranking procedure, compared to previous methods.This significantly improves the efficiency of LLM-based zero-shot ranking, while also retaining high zero-shot ranking effectiveness.We make our code and results publicly available at https://github.com/ielab/llm-rankers.

Systematic reviews are comprehensive literature reviews for a highly focused research question. These reviews are considered the highest form of evidence in medicine. Complex Boolean queries are developed as part of the systematic review creation process to retrieve literature, as they permit reproducibility and understandability. However, it is difficult and time-consuming to develop high-quality Boolean queries, often requiring the expertise of expert searchers like librarians. Recent advances in transformer-based generative models have shown their ability to effectively follow user instructions and generate answers based on these instructions. In this paper, we investigate ChatGPT as a means for automatically formulating and refining complex Boolean queries for systematic review literature search. Overall, our research finds that ChatGPT has the potential to generate effective Boolean queries. The ability of ChatGPT to follow complex instructions and generate highly precise queries makes it a tool of potential value for researchers conducting systematic reviews, particularly for rapid reviews where time is a constraint and where one can trade off higher precision for lower recall. We also identify several caveats in using ChatGPT for this task, highlighting that this technology needs further validation before it is suitable for widespread uptake.

Knowledge distillation plays a key role in boosting the effectiveness of rankers based on pre-trained language models (PLMs); this is achieved using an effective but inefficient large model to teach a more efficient student model. In the context of knowledge distillation for a student dense passage retriever, the balanced topic-aware sampling method has been shown to provide state-of-the-art effectiveness. This method intervenes in the creation of the training batches by creating batches that contain positive-negative pairs of passages from the same topic, and balancing the pairwise margins of the positive and negative passages.

Effective cross-lingual dense retrieval methods that rely on multilingual pre-trained language models (PLMs) need to be trained to encompass both the relevance matching task and the cross-language alignment task. However, cross-lingual data for training is often scarcely available. In this paper, rather than using more cross-lingual data for training, we propose to use cross-lingual query generation to augment passage representations with queries in languages other than the original passage language. These augmented representations are used at inference time so that the representation can encode more information across the different target languages. Training of a cross-lingual query generator does not require additional training data to that used for the dense retriever. The query generator training is also effective because the pre-training task for the generator (T5 text-to-text training) is very similar to the fine-tuning task (generation of a query). The use of the generator does not increase query latency at inference and can be combined with any cross-lingual dense retrieval method. Results from experiments on a benchmark cross-lingual information retrieval dataset show that our approach can improve the effectiveness of existing cross-lingual dense retrieval methods. Implementation of our methods, along with all generated query files are made publicly available at https://github.com/ielab/xQG4xDR.

While large amounts of potentially useful agricultural resources (journal articles, manuals, reports) are available, their value cannot be realised if they cannot be easily searched and presented to the agriculture users in a digestible form.AgAsk is a conversational search system for the agricultural domain, providing tailored answers to growers questions. AgAsk is underpinned by an efficient and effective neural passage ranking model fine-tuned on real world growers' questions. An adaptable, messaging-style user interface is deployed via the Telegram messaging platform, allowing users to ask natural language questions via text or voice, and receive short natural language answers as replies.

Entity Alignment (EA), which aims to detect entity mappings (i.e. equivalent entity pairs) in different Knowledge Graphs (KGs), is critical for KG fusion. Neural EA methods dominate current EA research but still suffer from their reliance on labelled mappings. To solve this problem, a few works have explored boosting the training of EA models with self-training, which adds confidently predicted mappings into the training data iteratively. Though the effectiveness of self-training can be glimpsed in some specific settings, we still have very limited knowledge about it. One reason is the existing works concentrate on devising EA models and only treat self-training as an auxiliary tool. To fill this knowledge gap, we change the perspective to self-training to shed light on it. In addition, the existing self-training strategies have limited impact because they introduce either much False Positive noise or a low quantity of True Positive pseudo mappings. To improve self-training for EA, we propose exploiting the dependencies between entities, a particularity of EA, to suppress the noise without hurting the recall of True Positive mappings. Through extensive experiments, we show that the introduction of dependency makes the self-training strategy for EA reach a new level. The value of self-training in alleviating the reliance on annotation is actually much higher than what has been realised. Furthermore, we suggest future study on smart data annotation to break the ceiling of EA performance.

Boolean query construction is often critical for medical systematic review literature search. To create an effective Boolean query, systematic review researchers typically spend weeks coming up with effective query terms and combinations. One challenge to creating an effective systematic review Boolean query is the selection of effective MeSH Terms to include in the query. In our previous work, we created neural MeSH term suggestion methods and compared them to state-of-the-art MeSH term suggestion methods. We found neural MeSH term suggestion methods to be highly effective.

Pseudo Relevance Feedback (PRF) is known to improve the effectiveness of bag-of-words retrievers. At the same time, deep language models have been shown to outperform traditional bag-of-words rerankers. However, it is unclear how to integrate PRF directly with emergent deep language models. This article addresses this gap by investigating methods for integrating PRF signals with rerankers and dense retrievers based on deep language models. We consider text-based, vector-based and hybrid PRF approaches and investigate different ways of combining and scoring relevance signals. An extensive empirical evaluation was conducted across four different datasets and two task settings (retrieval and ranking). Text-based PRF results show that the use of PRF had a mixed effect on deep rerankers across different datasets. We found that the best effectiveness was achieved when (i) directly concatenating each PRF passage with the query, searching with the new set of queries, and then aggregating the scores; (ii) using Borda to aggregate scores from PRF runs. Vector-based PRF results show that the use of PRF enhanced the effectiveness of deep rerankers and dense retrievers over several evaluation metrics. We found that higher effectiveness was achieved when (i) the query retains either the majority or the same weight within the PRF mechanism, and (ii) a shallower PRF signal (i.e., a smaller number of top-ranked passages) was employed, rather than a deeper signal. Our vector-based PRF method is computationally efficient; thus, this represents a general PRF method others can use with deep rerankers and dense retrievers.

Entity Alignment (EA) aims to match equivalent entities that refer to the same real-world objects and is a key step for Knowledge Graph (KG) fusion. Most neural EA models cannot be applied to large-scale real-life KGs due to their excessive consumption of GPU memory and time. One promising solution is to divide a large EA task into several subtasks such that each subtask only needs to match two small subgraphs of the original KGs. However, it is challenging to divide the EA task without losing effectiveness. Existing methods display low coverage of potential mappings, insufficient evidence in context graphs, and largely differing subtask sizes.

We present SCC, a test collection for evaluating search in chat conversations. Chat applications such as Slack, WhatsApp and Wechat have become popular communication methods. Typical search requirements in these applications revolve around the task of known item retrieval, i.e. find information that the user has previously experienced in their chats. However, the search capabilities of these chat applications are often very basic. Our collection aims to support new research into building effective methods for chat conversations search. We do so by building a collection with 114 known item retrieval topics for searching over 437,893 Slack chat messages. An important aspect when searching through conversations is the unit of indexing (indexing granularity), e.g., it being a single message vs. an entire conversation. To support researchers to investigate this aspect and its influence on retrieval effectiveness, the collection has been processed with conversation disentanglement methods: these mark cohesive segments in which each conversation consists of messages whose senders interact with each other regarding a specific event or topic. This results in a total of 38,955 multi-participant conversations being contained in the collection. Finally, we also provide a set of baselines with related empirical evaluation, including traditional bag-of-words methods and zero-shot neural methods, at both indexing granularity levels.

Pseudo-Relevance Feedback (PRF) assumes that the top results retrieved by a first-stage ranker are relevant to the original query and uses them to improve the query representation for a second round of retrieval. This assumption however is often not correct: some or even all of the feedback documents may be irrelevant. Indeed, the effectiveness of PRF methods may well depend on the quality of the feedback signal and thus on the effectiveness of the first-stage ranker. This aspect however has received little attention before.

Current pre-trained language model approaches to information retrieval can be broadly divided into two categories: sparse retrievers (to which belong also non-neural approaches such as bag-of-words methods, e.g., BM25) and dense retrievers. Each of these categories appears to capture different characteristics of relevance. Previous work has investigated how relevance signals from sparse retrievers could be combined with those from dense retrievers via interpolation. Such interpolation would generally lead to higher retrieval effectiveness.

Recent advances in Information Retrieval utilise energy-intensive hardware to produce state-of-the-art results. In areas of research highly related to Information Retrieval, such as Natural Language Processing and Machine Learning, there have been efforts to quantify and reduce the power and emissions produced by methods that depend on such hardware. Research that is conscious of the environmental impacts of its experimentation and takes steps to mitigate some of these impacts is considered 'Green'. Given the continuous demand for more data and power-hungry techniques, Green research is likely to become more important within the broader research community. Therefore, within the Information Retrieval community, the consequences of non-Green (in other words, Red) research should at least be considered and acknowledged. As such, the aims of this perspective paper are fourfold: (1) to review the Green literature not only for Information Retrieval but also for related domains in order to identify transferable Green techniques; (2) to provide measures for quantifying the power usage and emissions of Information Retrieval research; (3) to report the power usage and emission impacts for various current IR methods; and (4) to provide a framework to guide Green Information Retrieval research, taking inspiration from 'reduce, reuse, recycle' waste management campaigns, including salient examples from the literature that implement these concepts.

Medical systematic review query formulation is a highly complex task done by trained information specialists. Complexity comes from the reliance on lengthy Boolean queries, which express a detailed research question. To aid query formulation, information specialists use a set of exemplar documents, called 'seed studies', prior to query formulation. Seed studies help verify the effectiveness of a query prior to the full assessment of retrieved studies. Beyond this use of seeds, specific IR methods can exploit seed studies for guiding both automatic query formulation and new retrieval models. One major limitation of work to date is that these methods exploit 'pseudo seed studies' through retrospective use of included studies (i.e., relevance assessments). However, we show pseudo seed studies are not representative of real seed studies used by information specialists. Hence, we provide a test collection with real world seed studies used to assist with the formulation of queries. To support our collection, we provide an analysis, previously not possible, on how seed studies impact retrieval and perform several experiments using seed study based methods to compare the effectiveness of using seed studies versus pseudo seed studies. We make our test collection and the results of all of our experiments and analysis available at http://github.com/ielab/sysrev-seed-collection.

In this perspective paper we study the effect of non independent and identically distributed (non-IID) data on federated online learning to rank (FOLTR) and chart directions for future work in this new and largely unexplored research area of Information Retrieval. In the FOLTR process, clients participate in a federation to jointly create an effective ranker from the implicit click signal originating in each client, without the need to share data (documents, queries, clicks). A well-known factor that affects the performance of federated learning systems, and that poses serious challenges to these approaches, is that there may be some type of bias in the way data is distributed across clients. While FOLTR systems are on their own rights a type of federated learning system, the presence and effect of non-IID data in FOLTR has not been studied. To this aim, we first enumerate possible data distribution settings that may showcase data bias across clients and thus give rise to the non-IID problem. Then, we study the impact of each setting on the performance of the current state-of-the-art FOLTR approach, the Federated Pairwise Differentiable Gradient Descent (FPDGD), and we highlight which data distributions may pose a problem for FOLTR methods. We also explore how common approaches proposed in the federated learning literature address non-IID issues in FOLTR. This allows us to unveil new research gaps that, we argue, future research in FOLTR should consider.

In this paper we study how to effectively exploit implicit feedback in Dense Retrievers (DRs). We consider the specific case in which click data from a historic click log is available as implicit feedback. We then exploit such historic implicit interactions to improve the effectiveness of a DR. A key challenge that we study is the effect that biases in the click signal, such as position bias, have on the DRs. To overcome the problems associated with the presence of such bias, we propose the Counterfactual Rocchio (CoRocchio) algorithm for exploiting implicit feedback in Dense Retrievers. We demonstrate both theoretically and empirically that dense query representations learnt with CoRocchio are unbiased with respect to position bias and lead to higher retrieval effectiveness. We make available the implementations of the proposed methods and the experimental framework, along with all results at https://github.com/ielab/Counterfactual-DR.

Current dense retrievers are not robust to out-of-domain and outlier queries, i.e. their effectiveness on these queries is much poorer than what one would expect. In this paper, we consider a specific instance of such queries: queries that contain typos. We show that a small character level perturbation in queries (as caused by typos) highly impacts the effectiveness of dense retrievers. We then demonstrate that the root cause of this resides in the input tokenization strategy employed by BERT. In BERT, tokenization is performed using the BERT's WordPiece tokenizer and we show that a token with a typo will significantly change the token distributions obtained after tokenization. This distribution change translates to changes in the input embeddings passed to the BERT-based query encoder of dense retrievers. We then turn our attention to devising dense retriever methods that are robust to such queries with typos, while still being as performant as previous methods on queries without typos. For this, we use CharacterBERT as the backbone encoder and an efficient yet effective training method, called Self-Teaching (ST), that distills knowledge from queries without typos into the queries with typos. Experimental results show that CharacterBERT in combination with ST achieves significantly higher effectiveness on queries with typos compared to previous methods. Along with these results and the open-sourced implementation of the methods, we also provide a new passage retrieval dataset consisting of real-world queries with typos and associated relevance assessments on the MS MARCO corpus, thus supporting the research community in the investigation of effective and robust dense retrievers. Code, experimental results and dataset are made available at https://github.com/ielab/CharacterBERT-DR.

The process of model checkpoint validation refers to the evaluation of the performance of a model checkpoint executed on a held-out portion of the training data while learning the hyperparameters of the model. This model checkpoint validation process is used to avoid over-fitting and determine when the model has converged so as to stop training. A simple and efficient strategy to validate deep learning checkpoints is the addition of validation loops to execute during training. However, the validation of dense retrievers (DR) checkpoints is not as trivial -- and the addition of validation loops is not efficient. This is because, in order to accurately evaluate the performance of a DR checkpoint, the whole document corpus needs to be encoded into vectors using the current checkpoint before any actual retrieval operation for checkpoint validation can be performed. This corpus encoding process can be very time-consuming if the document corpus contains millions of documents (e.g., 8.8M for MS MARCO v1 and 21M for Natural Questions). Thus, a naïve use of validation loops during training will significantly increase training time. To address this issue, we propose Asyncval: a Python-based toolkit for efficiently validating DR checkpoints during training. Instead of pausing the training loop for validating DR checkpoints, Asyncval decouples the validation loop from the training loop, uses another GPU to automatically validate new DR checkpoints and thus permits to perform validation asynchronously from training. Asyncval also implements a range of different corpus subset sampling strategies for validating DR checkpoints; these strategies allow to further speed up the validation process. We provide an investigation of these methods in terms of their impact on validation time and validation fidelity. Asyncval is made available as an open-source project at https://github.com/ielab/asyncval.

Abstract Online learning to rank (OLTR) aims to learn a ranker directly from implicit feedback derived from users’ interactions, such as clicks. Clicks however are a biased signal: specifically, top-ranked documents are likely to attract more clicks than documents down the ranking (position bias). In this paper, we propose a novel learning algorithm for OLTR that uses reinforcement learning to optimize rankers: Reinforcement Online Learning to Rank (ROLTR). In ROLTR, the gradients of the ranker are estimated based on the rewards assigned to clicked and unclicked documents. In order to de-bias the users’ position bias contained in the reward signals, we introduce unbiased reward shaping functions that exploit inverse propensity scoring for clicked and unclicked documents. The fact that our method can also model unclicked documents provides a further advantage in that less users interactions are required to effectively train a ranker, thus providing gains in efficiency. Empirical evaluation on standard OLTR datasets shows that ROLTR achieves state-of-the-art performance, and provides significantly better user experience than other OLTR approaches. To facilitate the reproducibility of our experiments, we make all experiment code available at https://github.com/ielab/OLTR .

Evaluation is central to Information Retrieval, and is how we compare the quality of systems. One important principle of evaluation is that the measured score should reflect the user's experience with the system. Hence, there should be direct connection between how users interact with the system and the characteristics of the metric. In this tutorial we introduce the C/W/L approach to user modeling and show how different user models lead to different metrics. We then describe the recent innovations and approaches to evaluation that it has facilitated. The tutorial is presented as a mix of on-line synchronous lecture, pre-recorded in-depth videos, and hands-on activities using the C/W/L toolkit for participants' own evaluation tasks. A followup consultation session is also provided, to allow extended questions and individual discussion with the four presenters.

We present a search engine aimed to help clinicians find targeted treatments for children with cancer. Childhood cancer is a leading cause of death and clinicians increasingly seek treatments that are tailored to an individual patient, particularly their tumour genetics. Finding treatments that are specific to paediatrics and match individual genetics is a real challenge amongst the vast and growing body of medical literature and clinical trials. We aim to help clinicians through a search system tailored to this problem.

Often, existing chat services that organisations and individuals use today provide a way to search through previously sent messages. However, many of these chat services provide far-limited search functionalities, typically exact matching on individual messages. In this paper, we introduce a new task for addressing this problem, called searching for conversations, whereby the aim is to retrieve and rank groups of related messages given a search query. We promote this task by providing a platform for research and development called PECAN. Our platform provides all the necessary functionality researchers need to conduct experiments on searching for conversations. Our system is also generic so as to support organisations and individuals who wish to search through their chat message archives.

Fine-grained logging of interactions in user studies is important for studying user behaviour, among other reasons. However, in many research scenarios, the way interactions are logged is usually tied to a monolithic system. We present a generic, application-independent service for logging interactions in web-pages, specifically targetting user studies. Our service, Big Brother, can be dropped-in to existing user interfaces with almost no configuration required by researchers. Big Brother has already been used in several user studies to record interactions in a number of user study research scenarios, such as lab-based and crowdsourcing environments. We further demonstrate the ability for Big Brother to scale to very large user studies through benchmarking experiments. Big Brother also provides a number of additional tools for visualising and analysing interactions. Big Brother significantly lowers the barrier to entry for logging user interactions by providing a minimal but powerful, no configuration necessary, service for researchers and practitioners of user studies that can scale to thousands of concurrent sessions. We have made the source code and releases for Big Brother available for download at https://github.com/hscells/bigbro.

Online learning to rank (OLTR) uses interaction data, such as clicks, to dynamically update rankers. OLTR has been thought to capture user intent change overtime - a task that is impossible for rankers trained on statistic datasets such as in offline and counterfactual learning to rank. However, this feature has never been demonstrated and empirically studied, as previous work only considered simulated online data with single user intent or real online data with no explicit notion of intents and how they change over interactions. In this paper, we address this gap by study the capability of OLTR algorithms to adapt to user intent change. Our empirical experiments show that the adaptation to intent change does vary across OLTR methods, and is also dependent on the amount of noise in the implicit feedback signal. This is an important result, as it highlights that intent change adaptation should be studied alongside online and offline performance. Investigating how OLTR algorithms adapt to intent change is challenging as current LTR datasets do not explicitly contain the required intent data. Along with the main findings reported in this paper related to intent change, we also contribute a methodology to investigate this aspect of OLTR methods. Specifically, we create a collection for OLTR with explicit intent change by adapting an existing TREC collection to this task. We further introduce methods to model and simulate click behaviour related to intent change. We further propose novel evaluation metrics tailored to study different aspects of how OLTR methods adapt to intent change.

Deep language models (deep LMs) are increasingly being used for full text retrieval or within cascade retrieval pipelines as later-stage re-rankers. A problem with using deep LMs is that, at query time, a slow inference step needs to be performed -- this hinders the practical adoption of these powerful retrieval models, or limits sensibly how many documents can be considered for re-ranking.

Increasingly, people go online to seek health advice. They commonly use the symptoms they are experiencing to identify the health conditions they may have (self-diagnosis task) as well as to determine an appropriate action to take (triaging task); e.g., should they seek emergent medical attention or attempt to treat themselves at home? This paper investigates the effectiveness of two of the most common methods people use for self-diagnosis and triaging: online symptom checkers and traditional web search engines. To this end, we conducted a user study with 64 real-world users performing 8 simulated self-diagnosis tasks. Participants were exposed to both a representative symptom checker and a search engine. The results of our study provides empirical evidence for whether using a search engine for health information improves people’s understanding of their health condition and their ability to act on them, compared to interacting with a symptom checker, which bases its interaction model on a question-answering process. Additionally, recorded answers to qualitative questionnaires from study participants provide insights into which style of interaction and system they prefer to use for obtaining medical information, and how helpful they thought each system was. These findings can help inform the development of better search engines and symptom checkers that support people seeking health advice online.

Abstract Previous research has found that improved search engine effectiveness—evaluated using a batch‐style approach—does not always translate to significant improvements in user task performance; however, these prior studies focused on simple recall and precision‐based search tasks. We investigated the same relationship, but for realistic, complex search tasks required in clinical decision making. One hundred and nine clinicians and final year medical students answered 16 clinical questions. Although the search engine did improve answer accuracy by 20 percentage points, there was no significant difference when participants used a more effective, state‐of‐the‐art search engine. We also found that the search engine effectiveness difference, identified in the lab, was diminished by around 70% when the search engines were used with real users. Despite the aid of the search engine, half of the clinical questions were answered incorrectly. We further identified the relative contribution of search engine effectiveness to the overall end task success. We found that the ability to interpret documents correctly was a much more important factor impacting task success. If these findings are representative, information retrieval research may need to reorient its emphasis towards helping users to better understand information, rather than just finding it for them.

The empirical nature of Information Retrieval (IR) mandates strong experimental practices. A keystone of such experimental practices is the Cranfield evaluation paradigm. Within this paradigm, the collection of relevance judgments has been the subject of intense scientific investigation. This is because, on one hand, consistent, precise, and numerous judgements are keys to reducing evaluation uncertainty and test collection bias; on the other hand, however, relevance judgements are costly to collect. The selection of which documents to judge for relevance, known as pooling method, has therefore a great impact on IR evaluation. In this paper we focus on the bias introduced by the pooling method, known as pool bias, which affects the reusability of test collections, in particular when building test collections with a limited budget. In this paper we formalize and evaluate a set of 22 pooling strategies based on: traditional strategies, voting systems, retrieval fusion methods, evaluation measures, and multi-armed bandit models. To do this we run a large-scale evaluation by considering a set of 9 standard TREC test collections, in which we show that the choice of the pooling strategy has significant effects on the cost needed to obtain an unbiased test collection. We also identify the least biased pooling strategy in terms of pool bias according to three IR evaluation measures: AP, NDCG, and P@10.

Systematic reviews are used widely in the biomedical and healthcare domains. Systematic reviews aim to provide a complete and exhaustive overview of the medical literature for a specific research question. Core to the construction of a systematic review is the search strategy. The main component of a search strategy is a complex Boolean query, typically developed by information specialists (e.g., librarians). The aim of the search strategy is to retrieve relevant studies that will contribute to the outcomes of the systematic review. One barrier information specialists face when developing a search strategy is the enormous amount of medical literature that exists in databases. This vast amount of literature means that search strategies often suffer from biases (e.g., lack of expertise, overconfidence, limited knowledge of the domain) and are incomplete, or retrieve far too many studies (possibly as a result of the biases, but also due to the tools used to develop search strategies). Retrieving too many studies impacts the time and financial costs of the review, and retrieving too few studies may impact the outcomes of the review. Therefore, it is vital to support expert searchers develop effective search strategies. In this paper, we present a novel end-to-end set of advanced tools for information specialists. These tools are tightly integrated into an existing Open Source search strategy refining package (searchrefiner). These tools aim to address the problems associated with search strategy development by providing a complete framework from query development, to refinement, to documentation. The implementation of these tools also offers a glimpse at the ease at which related tools may be implemented within the searchrefiner ecosystem. More information about the tools including installation, documentation, and screenshots is made available on the searchrefiner website: https://ielab.io/searchrefiner.

Formulating Boolean queries for systematic review literature search is a challenging task. Commonly, queries are formulated by information specialists using the protocol specified in the review and interactions with the research team. Information specialists have in-depth experience on how to formulate queries in this domain, but may not have in-depth knowledge about the reviews’ topics. Query formulation requires a significant amount of time and effort, and is performed interactively; specialists repeatedly formulate queries, attempt to validate their results, and reformulate specific Boolean clauses. In this paper, we investigate the possibility of automatically formulating a Boolean query from the systematic review protocol. We propose a novel five-step approach to automatic query formulation, specific to Boolean queries in this domain, which approximates the process by which information specialists formulate queries. In this process, we use syntax parsing to derive the logical structure of high-level concepts in a query, automatically extract and map concepts to entities in order to perform entity expansion, and finally apply post-processing operations (such as stemming and search filters).

Searching medical literature for synthesis in a systematic review is a complex and labour intensive task. In this context, expert searchers construct lengthy Boolean queries. The universe of possible query variations can be massive: a single query can be composed of hundreds of field-restricted search terms/phrases or ontological concepts, each grouped by a logical operator nested to depths of sometimes five or more levels deep. With the many choices about how to construct a query, it is difficult to both formulate and recognise effective queries. To address this challenge, automatic methods have recently been explored for generating and selecting effective Boolean query variations for systematic reviews. The limiting factor of these methods is that it is computationally infeasible to process all query variations for training the methods. To overcome this, we propose novel query variation sampling methods for training Learning to Rank models to rank queries. Our results show that query sampling methods do directly impact the ability of a Learning to Rank model to effectively identify good query variations. Thus, selecting appropriate query sampling methods is a key problem for the automatic reformulation of effective Boolean queries for systematic review literature search. We find that the best sampling strategies are those which balance the diversity of queries with the quantity of queries.

This paper investigates methods to rank health cards, a domain-specific type of entity cards, for consumer health search (CHS) queries. A key challenge in this context is which card(s) should be presented to the user. In particular, little evidence exists to determine the effectiveness of retrieval and ranking methods for health cards in CHS. CHS is a challenging domain, where users lack domain expertise and thus are often unable to formulate effective queries, and to interpret the retrieved results. In addition, unlike in other contexts, CHS presents the opportunity to exploit a number of domain specific characteristics and features. In this paper, we focus on difficult queries with self-diagnosis intents. Our study makes the following contributions: (1) it assembles and releases the first test collection of health cards for research purposes, and (2) it empirically evaluates a large range of entity retrieval methods adapted to health cards retrieval, including features specific to health cards for learning to rank. This is the first study that thoroughly investigates methods to rank health cards.

Economics provides an intuitive and natural way to formally represent the costs and benefits of interacting with applications, interfaces and devices. By using economic models it is possible to reason about interaction, make predictions about how changes to the system will affect behavior, and measure the performance of people's interactions with the system. In this tutorial, we first provide an overview of relevant economic theories, before showing how they can be applied to formulate different ranking principles to provide the optimal ranking to users. This is followed by a session showing how economics can be used to model how people interact with search systems, and how to use these models to generate hypotheses about user behavior. The third session focuses on how economics has been used to underpin the measurement of information retrieval systems and applications using the CWL framework (which reports the expected utility, expected total utility, expected total cost, and so on) -- and how different models of user interaction lead to different metrics. We then show how information foraging theory can be used to measure the performance of an information retrieval system -- connecting the theory of how people search with how we measure it. The final session of the day will be spent building economic models and measures of search. Here sample problems will be provided to challenge participants, or participants can bring their own.

This paper investigates the impact of health cards in consumer health search (CHS) - people seeking health advice online. Health cards are a concise presentations of a health concept shown along side search results to specific health queries; they have the potential to convey health information in easily digestible form for the general public. However, little evidence exists on how effective health cards actually are for users when searching health advice online, and whether their effectiveness is limited to specific health search intents. To understand the impact of health cards on CHS, we conducted a laboratory study to observe users completing CHS tasks using two search interface variants: one just with result snippets and one containing both result snippets and health cards. Our study makes the following contributions: (1) it reveals how and when health cards are beneficial to users in completing consumer health search tasks, and (2) it identifies the features of health cards that helped users in completing their tasks. This is the first study that thoroughly investigates the effectiveness of health cards in supporting consumer health search.

This paper introduces TrecTools, a Python library for assisting Information Retrieval (IR) practitioners with TREC-like campaigns. IR practitioners tasked with activities like building test collections, evaluating systems, or analysing results from empirical experiments commonly have to resort to use a number of different software tools and scripts that each perform an individual functionality - and at times they even have to implement ad-hoc scripts of their own. TrecTools aims to provide a unified environment for performing these common activities.

The HS2019 tutorial will cover topics from an area of information retrieval (IR) with significant societal impact --- health search. Whether it is searching patient records, helping medical professionals find best-practice evidence, or helping the public locate reliable and readable health information online, health search is a challenging area for IR research with an actively growing community and many open problems. This tutorial will provide attendees with a full stack of knowledge on health search, from understanding users and their problems to practical, hands-on sessions on current tools and techniques, current campaigns and evaluation resources, as well as important open questions and future directions.

In the medical domain, systematic reviews are a highly trustworthy evidence source used to inform clinical diagnosis and treatment, and governmental policy making. Systematic reviews must be complete in that all relevant literature for the research question of the review must be synthesised in order to produce a recommendation. To identify the literature to screen for inclusion in systematic reviews, information specialists construct complex Boolean queries that capture the information needs defined by the research questions of the systemic review. However, in the quest for total recall, these Boolean queries return many non relevant results.

In this paper, we proposed a framework to evaluate information retrieval systems in presence of multidimensional relevance. This is an important problem in tasks such as consumer health search, where the understandability and trustworthiness of information greatly influence people's decisions based on the search engine results, but common topicality-only evaluation measures ignore these aspects. We used synthetic and real data to compare our proposed framework, named MM, to the understandability-biased information evaluation (UBIRE), an existing framework used in the context of consumer health search. We showed how the proposed approach diverges from the UBIRE framework, and how MM can be used to better understand the trade-offs between topical relevance and the other relevance dimensions.

We present an open source tool, searchrefiner, for researchers that conduct medical systematic reviews to assist in formulating, visualising, and understanding Boolean queries. The searchrefiner web interface allows researchers to explore how Boolean queries retrieve citations in existing, popular query syntaxes used in systematic review literature search. The web interface allows researchers to perform tasks such as using validation citations to ensure queries are retrieving a minimum set of known relevant citations, and editing Boolean queries by dragging and dropping clauses in a structured editor. In addition, the tools provided by the searchrefiner interface allow researchers to visualise why the queries they formulate retrieve citations, and ways to understand how to refine queries into more effective ones. searchrefiner is targeted at both experts and novices, as a tool for query formulation and refinement, and as a tool for training users to search for literature to compile systematic reviews.

We studied the volatility of commercial search engines and reflected on its impact on research that uses them as basis of algorithmical techniques or for user studies. Search engine volatility refers to the fact that a query posed to a search engine at two different points in time returns different documents. By comparing search results retrieved every 2 days over a period of 64 days, we found that the considered commercial search engine API consistently presented volatile search results: it both retrieved new documents, and it ranked documents previously retrieved at different ranks throughout time. Moreover, not only results are volatile: we also found that the effectiveness of the search engine in answering a query is volatile. Our findings reaffirmed that results from commercial search engines are volatile and that care should be taken when using these as basis for researching new information retrieval techniques or performing user studies.

Test collection based evaluation represents the standard of evalua- tion for information retrieval systems. Legal IR, more speci cally case law retrieval, has no such standard test collection for evalua- tion. In this paper, we present a test collection for use in evaluating case law search, being the retrieval of judicial decisions relevant to a particular legal question. The collection is made available at ielab.io/caselaw.

When conducting systematic reviews, medical researchers heavily deliberate over the final query to pose to the information retrieval system. Given the possible query variations that they could construct, selecting the best performing query is difficult. This motivates a new type of query performance prediction (QPP) task where the challenge is to estimate the performance of a set of query variations given a particular topic. Query variations are the reductions, expansions and modifications of a given seed query under the hypothesis that there exists some variations (either generated from permutations or hand crafted) which will improve retrieval effectiveness over the original query. We use the CLEF 2017 TAR Collection, to evaluate sixteen pre and post retrieval predictors for the task of Query Variation Performance Prediction (QVPP). Our findings show the IDF based QPPs exhibits the strongest correlations with performance. However, when using QPPs to select the best query, little improvement over the original query can be obtained, despite the fact that there are query variations which perform significantly better. Our findings highlight the difficulty in identifying effective queries within the context of this new task, and motivates further research to develop more accurate methods to help systematic review researchers in the query selection process.

We present a framework for constructing and executing information retrieval experiment pipelines. The framework as a whole is built primarily for domain specific applications such as medical literature search for systematic reviews, or finding factually or legally applicable case law in the legal domain; however it can also be used for more general tasks. There are a number of pre-implemented components that enable common information retrieval experiments such as ad-hoc retrieval or query analysis through query performance predictors. In addition, this collection of tools seeks to be user friendly, well documented, and easily extendible. Finally, the entire pipeline can be distributed as a single binary with no dependencies, ready to use with a simple domain specific language (DSL) for constructing pipelines.

Systematic reviews form the cornerstone of evidence based medicine, aiming to answer complex medical questions based on all evidence currently available. Key to the effectiveness of a systematic review is an (often large) Boolean query used to search large publication repositories. These Boolean queries are carefully crafted by researchers and information specialists, and often reviewed by a panel of experts. However, little is known about the effectiveness of the Boolean queries at the time of formulation. In this paper we investigate whether a better Boolean query than that defined in the protocol of a systematic review, can be created, and we develop methods for the transformation of a given Boolean query into a more effective one. Our approach involves defining possible transformations of Boolean queries and their clauses. It also involves casting the problem of identifying a transformed query that is better than the original into: (i) a classification problem; and (ii) a learning to rank problem. Empirical experiments are conducted on a real set of systematic reviews. Analysis of results shows that query transformations that are better than the original queries do exist, and that our approaches are able to select more effective queries from the set of possible transformed queries so as to maximise different target effectiveness measures.

The HS2018 tutorial will cover topics from an area of information retrieval (IR) with significant societal impact --- health search. Whether it is searching patient records, helping medical professionals find best-practice evidence, or helping the public locate reliable and readable health information online, health search is a challenging area for IR research with an actively growing community and many open problems. This tutorial will provide attendees with a full stack of knowledge on health search, from understanding users and their problems to practical, hands-on sessions on current tools and techniques, current campaigns and evaluation resources, as well as important open questions and future directions.

The PICO process is a technique used in evidence based practice to frame and answer clinical questions. It involves structuring the question around four types of clinical information: population, intervention, control or comparison and outcome. The PICO framework is used extensively in the compilation of systematic reviews as the means of framing research questions. However, when a search strategy (comprising of a large Boolean query) is formulated to retrieve studies for inclusion in the review, PICO is often ignored. This paper evaluates how PICO annotations can be applied and integrated into retrieval to improve the screening of studies for inclusion in systematic reviews. The task is to increase precision while maintaining the high level of recall essential to ensure systematic reviews are representative and unbiased. Our results show that restricting the search strategies to match studies using PICO annotations improves precision, however recall is slightly reduced, when compared to the non-PICO baseline. This can lead to both time and cost savings when compiling systematic reviews.

As an empirical discipline, information access and retrieval research requires substantial software infrastructure to index and search large collections. This workshop is motivated by the desire to better align information retrieval research with the practice of building search applications from the perspective of open-source information retrieval systems. Our goal is to promote the use of Lucene for information access and retrieval research.

This paper investigates how automated query generation methods can be used to derive effective ad-hoc queries from verbose patient narratives. In a clinical setting, automatic query generation provides a means of retrieving information relevant to a clinician, based on a patient record, but without the need for the clinician to manually author a query. Given verbose patient narratives, we evaluated a number of query reduction methods, both generic and domain specific. Comparison was made against human generated queries, both in terms of retrieval effectiveness and characteristics of human queries. Query reduction was an effective means of generating ad-hoc queries from narratives. However, human generated queries were still significantly more effective than automatically generated queries. Further improvements were possible if parameters of the query reduction methods were set on a per-query basis and a means of predicting this was developed. Under ideal conditions, automated methods can exceed humans. Effective human queries were found to contain many novel keywords not found in the narrative. Automated reduction methods may be handicapped in that they only use terms from narrative. Future work, therefore, may be directed toward better understanding effective human queries and automated query rewriting methods that attempt to model the inference of novel terms by exploiting semantic inference processes.

Evidence-based medicine (EBM) is the practice of making clinical decisions based on rigorous scientific evidence. EBM relies on effective access to peer-reviewed literature - a task hampered by both the exponential growth of medical literature and a lack of efficient and effective means of searching and presenting this literature. This paper describes a search engine specifically designed for searching medical literature for the purpose of EBM and in a clinical decision support setting.

This paper introduces a test collection for evaluating the effectiveness of different methods used to retrieve research studies for inclusion in systematic reviews. Systematic reviews appraise and synthesise studies that meet specific inclusion criteria. Systematic reviews intended for a biomedical science audience use boolean queries with many, often complex, search clauses to retrieve studies; these are then manually screened to determine eligibility for inclusion in the review. This process is expensive and time consuming. The development of systems that improve retrieval effectiveness will have an immediate impact by reducing the complexity and resources required for this process. Our test collection consists of approximately 26 million research studies extracted from the freely available MEDLINE database, 94 review (query) topics extracted from Cochrane systematic reviews, and corresponding relevance assessments. Tasks for which the collection can be used for information retrieval system evaluation are described and the use of the collection to evaluate common baselines within one such task is demonstrated. The test collection is available at https://github.com/ielab/SIGIR2017-PICO-Collection.

The HS2017 tutorial will cover topics from an area of information retrieval (IR) with significant societal impact - health search. Whether it is searching patient records, helping medical professionals find best-practice evidence, or helping the public locate reliable and readable health information online, health search is a challenging area for IR research with an actively growing community and many open problems. This tutorial will provide attendees with a full stack of knowledge on health search, from understanding users and their problems to practical, hands-on sessions on current tools and techniques, current campaigns and evaluation resources, as well as important open questions and future directions.

This article demonstrates the benefits of using sequence representations based on word embeddings to inform the seed selection and sample selection processes in an active learning pipeline for clinical information extraction. Seed selection refers to choosing an initial sample set to label to form an initial learning model. Sample selection refers to selecting informative samples to update the model at each iteration of the active learning process. Compared to supervised machine learning approaches, active learning offers the opportunity to build statistical classifiers with a reduced amount of training samples that require manual annotation. Reducing the manual annotation effort can support automating the clinical information extraction process. This is particularly beneficial in the clinical domain, where manual annotation is a time‐consuming and costly task, as it requires extensive labor from clinical experts. Our empirical findings demonstrate that (a) using sequence representations along with the length of sequence for seed selection shows potential towards more effective initial models, and (b) using sequence representations for sample selection leads to significantly lower manual annotation efforts, with up to 3% and 6% fewer tokens and concepts requiring annotation, respectively, compared to state‐of‐the‐art query strategies.

In this paper, we perform an in‐depth study into how clinicians represent their information needs and the influence this has on information retrieval (IR) effectiveness. While much research in IR has considered the effectiveness of IR systems, there is still a significant gap in the understanding of how users contribute to the effectiveness of these systems. The paper aims to contribute to this by studying how clinicians search for information. Multiple representations of an information need—from verbose patient case descriptions to ad‐hoc queries—were considered in order to understand their effect on retrieval. Four clinicians provided queries and performed relevance assessment to form a test collection used in this study. The different query formulation strategies of each clinician, and their effectiveness, were investigated. The results show that query formulation had more impact on retrieval effectiveness than the particular retrieval systems used. The most effective queries were short, ad‐hoc keyword queries. Different clinicians were observed to consistently adopt specific query formulation strategies. The most effective queriers were those who, given their information need, inferred novel keywords most likely to appear in relevant documents. This study reveals aspects of how people search within the clinical domain. This can help inform the development of new models and methods that specifically focus on the query formulation process to improve retrieval effectiveness.

A recurring question in information retrieval is whether term associations can be properly integrated in traditional information retrieval models while preserving their robustness and effectiveness. In this paper, we revisit a wide spectrum of existing models (Pivoted Document Normalization, BM25, BM25 Verboseness Aware, Multi-Aspect TF, and Language Modelling) by introducing a generalisation of the idea of the translation model. This generalisation is a de facto transformation of the translation models from Language Modelling to the probabilistic models. In doing so, we observe a potential limitation of these generalised translation models: they only affect the term frequency based components of all the models, ignoring changes in document and collection statistics. We correct this limitation by extending the translation models with the 15 statistics of term associations and provide extensive experimental results to demonstrate the benefit of the newly proposed methods. Additionally, we compare the translation models with query expansion methods based on the same term association resources, as well as based on Pseudo-Relevance Feedback (PRF). We observe that translation models always outperform the first, but provide complementary information with the second, such that by using PRF and our translation models together we observe results better than the current state of the art.

We explore the implications of using query variations for evaluating information retrieval systems and how these variations should be exploited to compare system effectiveness. Current evaluation approaches consider the availability of a set of topics (information needs), and only one expression of each topic in the form of a query is used for evaluation and system comparison. While there is strong evidence that considering query variations better models the usage of retrieval systems and accounts for the important user aspect of user variability, it is unclear how to best exploit query variations for evaluating and comparing information retrieval systems.

We present a test collection to study the use of search engines for matching eligible patients (the query) to clinical trials (the document). Clinical trials are experiments conducted in the development of new medical treatments, drugs or devices. Recruiting candidates for a trial is often a time-consuming and resource intensive effort, and imposes delays or even the cancellation of trials.

We propose a method that integrates relevance and understandability to rank health web documents. We use a learning to rank approach with standard retrieval features to determine topical relevance and additional features based on readability measures and medical lexical aspects to determine understandability. Our experiments measured the effectiveness of the learning to rank approach integrating understandability on a consumer health benchmark. The findings suggest that this approach promotes documents that are at the same time topically relevant and understandable.

This paper presents a new active learning query strategy for information extraction, called Domain Knowledge Informativeness (DKI). Active learning is often used to reduce the amount of annotation effort required to obtain training data for machine learning algorithms. A key component of an active learning approach is the query strategy, which is used to iteratively select samples for annotation. Knowledge resources have been used in information extraction as a means to derive additional features for sample representation. DKI is, however, the first query strategy that exploits such resources to inform sample selection. To evaluate the merits of DKI, in particular with respect to the reduction in annotation effort that the new query strategy allows to achieve, we conduct a comprehensive empirical comparison of active learning query strategies for information extraction within the clinical domain. The clinical domain was chosen for this work because of the availability of extensive structured knowledge resources which have often been exploited for feature generation. In addition, the clinical domain offers a compelling use case for active learning because of the necessary high costs and hurdles associated with obtaining annotations in this domain. Our experimental findings demonstrated that (1) amongst existing query strategies, the ones based on the classification model's confidence are a better choice for clinical data as they perform equally well with a much lighter computational load, and (2) significant reductions in annotation effort are achievable by exploiting knowledge resources within active learning query strategies, with up to 14% less tokens and concepts to manually annotate than with state-of-the-art query strategies.

This paper investigates the effect that text pre-processing approaches have on the estimation of the readability of web pages. Readability has been highlighted as an important aspect of web search result personalisation in previous work. The most widely used text readability measures rely on surface level characteristics of text, such as the length of words and sentences. We demonstrate that different tools for extracting text from web pages lead to very different estimations of readability. This has an important implication for search engines because search result personalisation strategies that consider users reading ability may fail if incorrect text readability estimations are computed.

Understanding how people interact with information systems when searching is central to the study of Interactive Information Retrieval (IIR). While much of the prior work in this area has either been conceptual, observational or empirical, recently there has been renewed interest in developing mathematical models of information seeking and search. This is because such models can provide a concise and compact representation of search behaviours and naturally generate testable hypotheses about search behaviour. This full day tutorial focuses on explaining and building formal models of Information Seeking and Retrieval. The tutorial is structured into four sessions. In the first session we will discuss the rationale of modelling and examine a number of early formal models of search (including early cost models and the Probability Ranking Principle). Then we will examine more contemporary formal models (including Information Foraging Theory, the Interactive Probability Ranking Principle, and Search Economic Theory). The focus will be on the insights and intuitions that we can glean from the math behind these models. The latter sessions will be dedicated to building models that optimise particular objectives which drive how users make decisions, along with a how-to guide on model building, where we will describe different techniques (including analytical, graphical and computational) that can be used to generate hypotheses from such models. In the final session, participants will be challenged to develop a simple model of interaction applying the techniques learnt during the day, before concluding with an overview of challenges and future directions.

Advances in neural network language models have demonstrated that these models can effectively learn representations of words meaning. In this paper, we explore a variation of neural language models that can learn on concepts taken from structured ontologies and extracted from free-text, rather than directly from terms in free-text.

We present a study to understand the effect that negated terms (e.g., "no fever") and family history (e.g., "family his- tory of diabetes") have on searching clinical records. Our analysis is aimed at devising the most effective means of handling negation and family history. In doing so, we explicitly represent a clinical record according to its different content types: negated, family history and normal content; the retrieval model weights each of these separately. Empirical evaluation shows that overall the presence of negation harms retrieval effectiveness while family history has little effect. We show negation is best handled by weighting negated content (rather than the common practise of re- moving or replacing it). However, we also show that many queries benefit from the inclusion of negated content and that negation is optimally handled on a per-query basis. Additional evaluation shows that adaptive handing of negated and family history content can have significant benefits.

Relevation! is a system for performing relevance judgements for information retrieval evaluation. Relevation! is web-based, fully configurable and expandable; it allows researchers to effectively collect assessments and additional qualitative data. The system is easily deployed allowing assessors to smoothly perform their relevance judging tasks, even remotely. Relevation! is available as an open source project at: http://ielab.github.io/relevation.

A user's query is considered to be an imprecise description of their information need. Automatic query expansion is the process of reformulating the original query with the goal of improving retrieval effectiveness. Many successful query expansion techniques model syntagmatic associations that infer two terms co‐occur more often than by chance in natural language. However, structural linguistics relies on both syntagmatic and paradigmatic associations to deduce the meaning of a word. Given the success of dependency‐based approaches to query expansion and the reliance on word meanings in the query formulation process, we argue that modeling both syntagmatic and paradigmatic information in the query expansion process improves retrieval effectiveness. This article develops and evaluates a new query expansion technique that is based on a formal, corpus‐based model of word meaning that models syntagmatic and paradigmatic associations. We demonstrate that when sufficient statistical information exists, as in the case of longer queries, including paradigmatic information alone provides significant improvements in retrieval effectiveness across a wide variety of data sets. More generally, when our new query expansion approach is applied to large‐scale web retrieval it demonstrates significant improvements in retrieval effectiveness over a strong baseline system, based on a commercial search engine.

Many successful query expansion techniques ignore information about the term dependencies that exist within natural language. However, researchers have recently demonstrated that consistent and significant improvements in retrieval effectiveness can be achieved by explicitly modelling term dependencies within the query expansion process. This has created an increased interest in dependency-based models.

Measures of semantic similarity between medical concepts are central to a number of techniques in medical informatics, including query expansion in medical information retrieval. Previous work has mainly considered thesaurus-based path measures of semantic similarity and has not compared different corpus-driven approaches in depth. We evaluate the effectiveness of eight common corpus-driven measures in capturing semantic relatedness and compare these against human judged concept pairs assessed by medical professionals. Our results show that certain corpus-driven measures correlate strongly (approx 0.8) with human judgements. An important finding is that performance was significantly affected by the choice of corpus used in priming the measure, i.e., used as evidence from which corpus-driven similarities are drawn. This paper provides guidelines for the implementation of semantic similarity measures for medical informatics and concludes with implications for medical information retrieval.

In the TREC Web Diversity track, novelty-biased cumulative gain (α-NDCG) is one of the official measures to assess retrieval performance of IR systems. The measure is characterised by a parameter, α, the effect of which has not been thoroughly investigated. We find that common settings of α, i.e. α=0.5, may prevent the measure from behaving as desired when evaluating result diversification. This is because it excessively penalises systems that cover many intents while it rewards those that redundantly cover only few intents. This issue is crucial since it highly influences systems at top ranks. We revisit our previously proposed threshold, suggesting α be set on a query-basis. The intuitiveness of the measure is then studied by examining actual rankings from TREC 09-10 Web track submissions. By varying α according to our query-based threshold, the discriminative power of α-NDCG is not harmed and in fact, our approach improves α-NDCG's robustness. Experimental results show that the threshold for α can turn the measure to be more intuitive than using its common settings.

The assumptions underlying the Probability Ranking Principle (PRP) have led to a number of alternative approaches that cater or compensate for the PRP's limitations. In this poster we focus on the Interactive PRP (iPRP), which rejects the assumption of independence between documents made by the PRP. Although the theoretical framework of the iPRP is appealing, no instantiation has been proposed and investigated. In this poster, we propose a possible instantiation of the principle, performing the first empirical comparison of the iPRP against the PRP. For document diversification, our results show that the iPRP is significantly better than the PRP, and comparable to or better than other methods such as Modern Portfolio Theory.

The Quantum Probability Ranking Principle (QPRP) has been recently proposed, and accounts for interdependent document relevance when ranking. However, to be instantiated, the QPRP requires a method to approximate the "interference" between two documents. In this poster, we empirically evaluate a number of different methods of approximation on two TREC test collections for subtopic retrieval. It is shown that these approximations can lead to significantly better retrieval performance over the state of the art.

Recently, Portfolio Theory (PT) has been proposed for Information Retrieval. However, under non-trivial conditions PT violates the original Probability Ranking Principle (PRP). In this poster, we shall explore whether PT upholds a different ranking principle based on Quantum Theory, i.e. the Quantum Probability Ranking Principle (QPRP), and examine the relationship between this new model and the new ranking principle. We make a significant contribution to the theoretical development of PT and show that under certain circumstances PT upholds the QPRP, and thus guarantees an optimal ranking according to the QPRP. A practical implication of this finding is that the parameters of PT can be automatically estimated via the QPRP, instead of resorting to extensive parameter tuning.

Retrieval with Logical Imaging is derived from belief revision and provides a novel mechanism for estimating the relevance of a document through logical implication (i.e. P(q->d). In this poster, we perform the first comprehensive evaluation of Logical Imaging (LI) in Information Retrieval (IR) across several TREC test Collections. When compared against standard baseline models, we show that LI fails to improve performance. This failure can be attributed to a nuance within the model that means non-relevant documents are promoted in the ranking, while relevant documents are demoted. This is an important contribution because it not only contextualizes the effectiveness of LI, but crucially explains why it fails. By addressing this nuance, future LI models could be significantly improved.