While vision-language models (VLMs) have demonstrated remarkable performance across various tasks combining textual and visual information, they continue to struggle with fine-grained visual perception tasks that require detailed pixel-level analysis. Effectively eliciting comprehensive reasoning from VLMs on such intricate visual elements remains an open challenge. In this paper, we present VipAct, an agent framework that enhances VLMs by integrating multi-agent collaboration and vision expert models, enabling more precise visual understanding and comprehensive reasoning. VipAct consists of an orchestrator agent, which manages task requirement analysis, planning, and coordination, along with specialized agents that handle specific tasks such as image captioning and vision expert models that provide high-precision perceptual information. This multi-agent approach allows VLMs to better perform fine-grained visual perception tasks by synergizing planning, reasoning, and tool use. We evaluate VipAct on benchmarks featuring a diverse set of visual perception tasks, with experimental results demonstrating significant performance improvements over state-of-the-art baselines across all tasks. Furthermore, comprehensive ablation studies reveal the critical role of multi-agent collaboration in eliciting more detailed System-2 reasoning and highlight the importance of image input for task planning. Additionally, our error analysis identifies patterns of VLMs' inherent limitations in visual perception, providing insights into potential future improvements. VipAct offers a flexible and extensible framework, paving the way for more advanced visual perception systems across various real-world applications.

Large Language Models (LLMs) have been increasingly studied as neural knowledge bases for supporting knowledge-intensive applications such as question answering and fact checking. However, the structural organization of their knowledge remains unexplored. Inspired by cognitive neuroscience findings, such as semantic clustering and priming, where knowing one fact increases the likelihood of recalling related facts, we investigate an analogous knowledge homophily pattern in LLMs. To this end, we map LLM knowledge into a graph representation through knowledge checking at both the triplet and entity levels. After that, we analyze the knowledgeability relationship between an entity and its neighbors, discovering that LLMs tend to possess a similar level of knowledge about entities positioned closer in the graph. Motivated by this homophily principle, we propose a Graph Neural Network (GNN) regression model to estimate entity-level knowledgeability scores for triplets by leveraging their neighborhood scores. The predicted knowledgeability enables us to prioritize checking less well-known triplets, thereby maximizing knowledge coverage under the same labeling budget. This not only improves the efficiency of active labeling for fine-tuning to inject knowledge into LLMs but also enhances multi-hop path retrieval in reasoning-intensive question answering.

Manan Suri, Puneet Mathur, Nedim Lipka, Franck Dernoncourt, Ryan A. Rossi, Dinesh Manocha. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Yu Xia, Subhojyoti Mukherjee, Zhouhang Xie, Junda Wu, Xintong Li, Ryan Aponte, Hanjia Lyu, Joe Barrow, Hongjie Chen, Franck Dernoncourt, Branislav Kveton, Tong Yu, Ruiyi Zhang, Jiuxiang Gu, Nesreen K. Ahmed, Yu Wang, Xiang Chen, Hanieh Deilamsalehy, Sungchul Kim, Zhengmian Hu, Yue Zhao, Nedim Lipka, Seunghyun Yoon, Ting-Hao Kenneth Huang, Zichao Wang, Puneet Mathur, Soumyabrata Pal, Koyel Mukherjee, Zhehao Zhang, Namyong Park, Thien Huu Nguyen, Jiebo Luo, Ryan A. Rossi, Julian McAuley. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Flowcharts are a critical tool for visualizing decision-making processes.However, their non-linear structure and complex visual-textual relationships make it challenging to interpret them using LLMs, as vision-language models frequently hallucinate nonexistent connections and decision paths when analyzing these diagrams.This leads to compromised reliability for automated flowchart processing in critical domains such as logistics, health, and engineering.We introduce the task of Fine-grained Flowchart Attribution, which traces specific components grounding a flowchart referring LLM response.Flowchart Attribution ensures the verifiability of LLM predictions and improves explainability by linking generated responses to the flowchart's structure.We propose FlowPathAgent, a neurosymbolic agent that performs fine-grained post hoc attribution through graph-based reasoning.It first segments the flowchart, then converts it into a structured symbolic graph, and then employs an agentic approach to dynamically interact with the graph, to generate attribution paths.Additionally, we present FlowExplainBench, a novel benchmark for evaluating flowchart attributions across diverse styles, domains, and question types.Experimental results show that FlowPathAgent mitigates visual hallucinations in LLM answers over flowchart QA, outperforming strong baselines by 10-14% on our proposed FlowExplainBench dataset.

Nishant Balepur, Alexa Siu, Nedim Lipka, Franck Dernoncourt, Tong Sun, Jordan Lee Boyd-Graber, Puneet Mathur. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

The `pre-train, prompt, predict' paradigm of large language models (LLMs) has achieved remarkable success in open-domain question answering (OD-QA). However, few works explore this paradigm in multi-document question answering (MD-QA), a task demanding a thorough understanding of the logical associations among the contents and structures of documents. To fill this crucial gap, we propose a Knowledge Graph Prompting (KGP) method to formulate the right context in prompting LLMs for MD-QA, which consists of a graph construction module and a graph traversal module. For graph construction, we create a knowledge graph (KG) over multiple documents with nodes symbolizing passages or document structures (e.g., pages/tables), and edges denoting the semantic/lexical similarity between passages or document structural relations. For graph traversal, we design an LLM-based graph traversal agent that navigates across nodes and gathers supporting passages assisting LLMs in MD-QA. The constructed graph serves as the global ruler that regulates the transitional space among passages and reduces retrieval latency. Concurrently, the graph traversal agent acts as a local navigator that gathers pertinent context to progressively approach the question and guarantee retrieval quality. Extensive experiments underscore the efficacy of KGP for MD-QA, signifying the potential of leveraging graphs in enhancing the prompt design and retrieval augmented generation for LLMs. Our code: https://github.com/YuWVandy/KG-LLM-MDQA.

Knowledge workers often need to extract and analyze information from a collection of documents to solve complex information tasks in the workplace, e.g., hiring managers reviewing resumes or analysts assessing risk in contracts. However, foraging for relevant information can become tedious and repetitive over many documents and criteria of interest. We introduce Marco, a mixed-initiative workspace supporting sensemaking over diverse business document collections. Through collection-centric assistance, Marco reduces the cognitive costs of extracting and structuring information, allowing users to prioritize comparative synthesis and decision making processes. Users interactively communicate their information needs to an AI assistant using natural language and compose schemas that provide an overview of a document collection. Findings from a usability study (n=16) demonstrate that when using Marco, users complete sensemaking tasks 16% more quickly, with less effort, and without diminishing accuracy. A design probe with seven domain experts identifies how Marco can benefit various real-world workflows.

Aspect-based sentiment analysis (ABSA) delves into understanding sentiments specific to distinct elements within a user-generated review. It aims to analyze user-generated reviews to determine a) the target entity being reviewed, b) the high-level aspect to which it belongs, c) the sentiment words used to express the opinion, and d) the sentiment expressed toward the targets and the aspects. While various benchmark datasets have fostered advancements in ABSA, they often come with domain limitations and data granularity challenges. Addressing these, we introduce the OATS dataset, which encompasses three fresh domains and consists of 27,470 sentence-level quadruples and 17,092 review-level tuples. Our initiative seeks to bridge specific observed gaps in existing datasets: the recurrent focus on familiar domains like restaurants and laptops, limited data for intricate quadruple extraction tasks, and an occasional oversight of the synergy between sentence and review-level sentiments. Moreover, to elucidate OATS’s potential and shed light on various ABSA subtasks that OATS can solve, we conducted experiments, establishing initial baselines. We hope the OATS dataset augments current resources, paving the way for an encompassing exploration of ABSA (https://github.com/RiTUAL-UH/OATS-ABSA).

Most real-world networks are noisy and incomplete samples from an unknown target distribution. Refining them by correcting corruptions or inferring unobserved regions typically improves downstream performance. Inspired by the impressive generative capabilities that have been used to correct corruptions in images, and the similarities between "in-painting" and filling in missing nodes and edges conditioned on the observed graph, we propose a novel graph generative framework, SGDM, which is based on subgraph diffusion. Our framework not only improves the scalability and fidelity of graph diffusion models, but also leverages the reverse process to perform novel, conditional generation tasks. In particular, through extensive empirical analysis and a set of novel metrics, we demonstrate that our proposed model effectively supports the following refinement tasks for partially observable networks: (T1) denoising extraneous subgraphs, (T2) expanding existing subgraphs and (T3) performing ``style" transfer by regenerating a particular subgraph to match the characteristics of a different node or subgraph.

The emergence of contract specific programming languages has struggled to translate into widespread adoption of computable contracts due largely to high conversion costs. In this work, we present the first system for converting natural language contracts into code through the extraction of key entities, relationships, and formulas into a graph representation called the Obligation Logic Graph (OLG). This approach allows the semantic meaning of contract obligations, including dependencies between obligations, to be captured through the OLG and mapped to code downstream. We also introduce OLG extraction as a new joint entity and relation prediction task for legal contracts, and present the Contract-OLG dataset, consisting of 1,876 contract provisions, 18,597 entities and 18,170 relationships. We perform detailed experiments to understand the capabilities of state-of-the-art Transformer and graph-based models at completing these tasks, and identify where there is currently a significant gap between human expert and machine performance, particularly for relation extraction.

Siva Uday Sampreeth Chebolu, Franck Dernoncourt, Nedim Lipka, Thamar Solorio. Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Graph models based on factorization of the adjacency matrix often fail to capture network structures related to links between dissimilar nodes (heterophily). We introduce a novel graph factorization model that leverages two nonnegative vectors per node to interpretably account for links between both similar and dissimilar nodes. We prove that our model can exactly represent any graph with low *arboricity*, a property that many real-world networks satisfy; our proof also applies to related models but has much greater scope than the closest prior bound, which is based on low *max degree*. Our factorization also has compelling properties besides expressiveness: due to its symmetric structure and nonnegativity, fitting the model inherently finds node communities, and the model's link predictions can be interpreted in terms of these communities. In experiments on real-world networks, we demonstrate our factorization's effectiveness on a variety of tasks, including community detection and link prediction.

Story ideation is a critical part of the story-writing process. It is challenging to support computationally due to its exploratory and subjective nature. Tropes, which are recurring narrative elements across stories, are essential in stories as they shape the structure of narratives and our understanding of them. In this paper, we propose to use tropes as an intermediate representation of stories to approach story ideation. We present TaleStream, a canvas system that uses tropes as building blocks of stories while providing steerable suggestions of story ideas in the form of tropes. Our trope suggestion methods leverage data from the tvtropes.org wiki. We find that 97% of the time, trope suggestions generated by our methods provide better story ideation materials than random tropes. Our system evaluation suggests that TaleStream can support writers’ creative flow and greatly facilitates story development. Tropes, as a rich lexicon of narratives with available examples, play a key role in TaleStream and hold promise for story-creation support systems.

In this paper, we initiate the study of one-pass algorithms for solving the maximum-a-posteriori (MAP) inference problem for Non-symmetric Determinantal Point Processes (NDPPs). In particular, we formulate streaming and online versions of the problem and provide one-pass algorithms for solving these problems. In our streaming setting, data points arrive in an arbitrary order and the algorithms are constrained to use a single-pass over the data as well as sub-linear memory, and only need to output a valid solution at the end of the stream. Our online setting has an additional requirement of maintaining a valid solution at any point in time. We design new one-pass algorithms for these problems and show that they perform comparably to (or even better than) the offline greedy algorithm while using substantially lower memory.

Unlike single images, capturing bursts enables many possible downstream tasks (e.g. superresolution, HDR enhancement) due to the rich information preserved in the consecutive frames. Efficient compression of these bursts is therefore essential given the additional frames to store. In this paper, we propose a novel near-lossless compression method that can preserve the most relevant information in the burst to enable multiple downstream image enhancement tasks, while at the same time reducing the file size. Specifically, we propose a two-bitstream near-lossless compression pipeline that controls the image-space distortion at frame level, and introduce the Lipschitz condition to bound the task-space distortion at burst level. Experiments conducted on a real-world burst dataset confirm the benefit of the proposed solution in terms of rate-distortion both in the burst frame space and the superresolution task space, a popular downstream task in burst processing.

Graph Neural Networks (GNNs) have proven to be useful for many different practical applications. However, many existing GNN models have implicitly assumed homophily among the nodes connected in the graph, and therefore have largely overlooked the important setting of heterophily, where most connected nodes are from different classes. In this work, we propose a novel framework called CPGNN that generalizes GNNs for graphs with either homophily or heterophily. The proposed framework incorporates an interpretable compatibility matrix for modeling the heterophily or homophily level in the graph, which can be learned in an end-to-end fashion, enabling it to go beyond the assumption of strong homophily. Theoretically, we show that replacing the compatibility matrix in our framework with the identity (which represents pure homophily) reduces to GCN. Our extensive experiments demonstrate the effectiveness of our approach in more realistic and challenging experimental settings with significantly less training data compared to previous works: CPGNN variants achieve state-of-the-art results in heterophily settings with or without contextual node features, while maintaining comparable performance in homophily settings.

Joe Barrow, Rajiv Jain, Nedim Lipka, Franck Dernoncourt, Vlad Morariu, Varun Manjunatha, Douglas Oard, Philip Resnik, Henning Wachsmuth. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Search engines for domain-specific media collections often rely on rich metadata being available for the content items. The annotations may not be complete or rich enough to support an adequate retrieval effectiveness. As a result, some search queries receive only a small result set (low recall) and others might suffer from reduced relevance (low precision). To alleviate this, we present a framework that exploits external knowledge to provide entity-oriented reformulation suggestions for queries that contain entities. We propose that queries be added as surrogate nodes to an external Knowledge Graph (KG) via the use of state-of-the-art entity linking algorithms. Embedding methods are invoked on the augmented graph, which contains additional edges between surrogate nodes and KG entities. We introduce a new evaluation setting to evaluate the quality of these embeddings. Experimental results on seven datasets confirm the effectiveness of the approach.

Conversations aimed at determining good recommendations are iterative in nature.People often express their preferences in terms of a critique of the current recommendation (e.g., "It doesn't look good for a date"), requiring some degree of common sense for a preference to be inferred.In this work, we present a method for transforming a user critique into a positive preference (e.g., "I prefer more romantic") in order to retrieve reviews pertaining to potentially better recommendations (e.g., "Perfect for a romantic dinner").We leverage a large neural language model (LM) in a fewshot setting to perform critique-to-preference transformation, and we test two methods for retrieving recommendations: one that matches embeddings, and another that fine-tunes an LM for the task.We instantiate this approach in the restaurant domain and evaluate it using a new dataset of restaurant critiques.In an ablation study, we show that utilizing critiqueto-preference transformation improves recommendations, and that there are at least three general cases that explain this improved performance.

Sangwoo Cho, Franck Dernoncourt, Tim Ganter, Trung Bui, Nedim Lipka, Walter Chang, Hailin Jin, Jonathan Brandt, Hassan Foroosh, Fei Liu. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.

Accurately discovering user intents from their written or spoken language plays a critical role in natural language understanding and automated dialog response. Most existing research models this as a classification task with a single intent label per utterance. Going beyond this formulation, we define and investigate a new problem of open intent discovery. It involves discovering one or more generic intent types from text utterances, that may not have been encountered during training. We propose a novel, domain-agnostic approach, OPINE, which formulates the problem as a sequence tagging task in an open-world setting. It employs a CRF on top of a bidirectional LSTM to extract intents in a consistent format, subject to constraints among intent tag labels. We apply multi-headed self-attention and adversarial training to effectively learn dependencies between distant words, and robustly adapt our model across varying domains. We also curate and release an intent-annotated dataset of 25K real-life utterances spanning diverse domains. Extensive experiments show that OPINE outperforms state-of-art baselines by 5-15% F1 score.

Saed Rezayi, Handong Zhao, Sungchul Kim, Ryan Rossi, Nedim Lipka, Sheng Li. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

In this paper, we aim to learn associations between visual attributes of fonts and the verbal context of the texts they are typically applied to. Compared to related work leveraging the surrounding visual context, we choose to focus only on the input text as this can enable new applications for which the text is the only visual element in the document. We introduce a new dataset, containing examples of different topics in social media posts and ads, labeled through crowd-sourcing. Due to the subjective nature of the task, multiple fonts might be perceived as acceptable for an input text, which makes this problem challenging. To this end, we investigate different end-to-end models to learn label distributions on crowd-sourced data and capture inter-subjectivity across all annotations.

Accurately discovering user intents from their written or spoken language plays a critical role in natural language understanding and automated dialog response. Most existing research models this as a classification task with a single intent label per utterance, grouping user utterances into a single intent type from a set of categories known beforehand. Going beyond this formulation, we define and investigate a new problem of open intent discovery. It involves discovering one or more generic intent types from text utterances, that may not have been encountered during training. We propose a novel domain-agnostic approach, OPINE, which formulates the problem as a sequence tagging task under an open-world setting. It employs a CRF on top of a bidirectional LSTM to extract intents in a consistent format, subject to constraints among intent tag labels. We apply a multi-head self-attention mechanism to effectively learn dependencies between distant words. We further use adversarial training to improve performance and robustly adapt our model across varying domains. Finally, we curate and plan to release an open intent annotated dataset of 25K real-life utterances spanning diverse domains. Extensive experiments show that our approach outperforms state-of-the-art baselines by 5-15% F1 score points. We also demonstrate the efficacy of OPINE in recognizing multiple, diverse domain intents with limited (can also be zero) training examples per unique domain.

As 98 percent of shoppers do not make a purchase on the first visit, we study the problem of predicting whether they would come back for a purchase later (i.e., conversion prediction). This problem is important for strategizing “retargeting”, for example, by sending coupons for customers who are likely to convert. For this goal, we study the following two problems, prediction of market and predictability of customer. First, prediction of market aims at identifying a conversion rate for a given product and its customer behavior modeling, which is an important analytics metric for retargeting process. Compared to existing approaches using either of customer or product-level conversion pattern, we propose a joint modeling of both patterns based on the well-studied buying decision process. Second, we can observe customer-specific behaviors after showing retargeting ads, to predict whether this specific customer follows the market model (high predictability) or not (low predictability). For the former, we apply the market model, and for the latter, we propose a new customer-specific prediction based on dynamic ad behavior features. To evaluate the effectiveness of our methods, we perform extensive experiments on the simulated dataset generated based on a set of real-world web logs and retargeting campaign logs. The evaluation results show that conversion predictions and predictability by our approach are consistently more accurate and robust than those by existing baselines in dynamic market environment.

Amirreza Shirani, Franck Dernoncourt, Paul Asente, Nedim Lipka, Seokhwan Kim, Jose Echevarria, Thamar Solorio. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. 2019.

With most summarization research focused on the news domain and scientific papers, little is known about the capabilities of the state of the art at summarizing more informal text.Today, the vast majority of text on the web is informally written on social media, and staying on top of the fast-paced stream of posts originating from one's subscriptions and followees is a burden to many.The TL;DR challenge marks a first step towards developing new summarization technology for social media, focusing on abstractive summarization.This paper reports the results of the challenge and describes our manual evaluation of the submissions.Finally, we discuss the expected properties of a good summary after analyzing the comments provided by human annotators.

Popular e-commerce websites such as Amazon offer community question answering systems for users to pose product-related questions and experienced customers may provide answers voluntarily. In this paper, we show that the large volume of existing community question answering data can be beneficial when building a system for answering questions related to product facts and specifications. Our experimental results demonstrate that the performance of a model for answering questions related to products listed in the Home Depot website can be improved by a large margin via a simple transfer learning technique from an existing large-scale Amazon community question answering dataset. Transfer learning can result in an increase of about 10% in accuracy in the experimental setting where we restrict the size of the data of the target task used for training. As an application of this work, we integrate the best performing model trained in this work into a mobile-based shopping assistant and show its usefulness.

The TL;DR challenge fosters research in abstractive summarization of informal text, the largest and fastest-growing source of textual data on the web, which has been overlooked by summarization research so far.The challenge owes its name to the frequent practice of social media users to supplement long posts with a "TL;DR"for "too long; didn't read"-followed by a short summary as a courtesy to those who would otherwise reply with the exact same abbreviation to indicate they did not care to read a post for its apparent length.Posts featuring TL;DR summaries form an excellent ground truth for summarization, and by tapping into this resource for the first time, we have mined millions of training examples from social media, opening the door to all kinds of generative models.

Generally 2% of shoppers make a purchase on the first visit to an online store while the other 98% enjoys only window-shopping. To bring people back to the store and close the deal, "retargeting" has been a vital online advertising strategy that leads to "conversion" of window-shoppers into buyers. As such retargeting is more effective as a focused tool, in this paper, we study the problem of identifying a conversion rate for a given product and its current customers, which is an important analytics metric for retargeting process. Compared to existing approaches using either of customer- or product-level conversion pattern, we propose a joint modeling of both level patterns based on the well-studied buying decision process. To evaluate the effectiveness of our method, we perform extensive experiments on the simulated dataset generated based on a set of real-world web logs. The evaluation results show that conversion predictions by our approach are consistently more accurate and robust than those by existing baselines in dynamic market environment.

We demonstrate an exploration tool that organizes social media content under diverse aspects enabling comprehensive explorations. Unlike existing approaches that group content by trending topics, we present a holistic view of diverse and relevant content with respect to a given query.

Automated text classification is one of the most important learning technologies to fight information overload. However, the information society is not only confronted with an information flood but also with an increase in "information volatility", by which we understand the fact that kind and distribution of a data source's emissions can significantly vary. In this paper we show how to estimate the expected effectiveness of a classification solution when the underlying data source undergoes a shift in the distribution of its subclasses (modes). Subclass distribution shifts are observed among others in online media such as tweets, blogs, or news articles, where document emissions follow topic popularity. To estimate the expected effectiveness of a classification solution we partition a test sample by means of clustering. Then, using repetitive resampling with different margin distributions over the clustering, the effectiveness characteristics is studied. We show that the effectiveness is normally distributed and introduce a probabilistic lower bound that is used for model selection. We analyze the relation between our notion of expected effectiveness and the mean effectiveness over the clustering both theoretically and on standard text corpora. An important result is a heuristic for expected effectiveness estimation that is solely based on the initial test sample and that can be computed without resampling.

The detection and improvement of low-quality information is a key concern in Web applications that are based on user-generated content; a popular example is the online encyclopedia Wikipedia. Existing research on quality assessment of user-generated content deals with the classification as to whether the content is high-quality or low-quality. This paper goes one step further: it targets the prediction of quality flaws, this way providing specific indications in which respects low-quality content needs improvement. The prediction is based on user-defined cleanup tags, which are commonly used in many Web applications to tag content that has some shortcomings. We apply this approach to the English Wikipedia, which is the largest and most popular user-generated knowledge source on the Web. We present an automatic mining approach to identify the existing cleanup tags, which provides us with a training corpus of labeled Wikipedia articles. We argue that common binary or multiclass classification approaches are ineffective for the prediction of quality flaws and hence cast quality flaw prediction as a one-class classification problem. We develop a quality flaw model and employ a dedicated machine learning approach to predict Wikipedia's most important quality flaws. Since in the Wikipedia setting the acquisition of significant test data is intricate, we analyze the effects of a biased sample selection. In this regard we illustrate the classifier effectiveness as a function of the flaw distribution in order to cope with the unknown (real-world) flaw-specific class imbalances. The flaw prediction performance is evaluated with 10,000 Wikipedia articles that have been tagged with the ten most frequent quality flaws: provided test data with little noise, four flaws can be detected with a precision close to 1.

A number of relevant information retrieval classification problems are one-class classification problems at heart. I.e., labeled data is only available for one class, the so-called target class, and common discrimination-based classification approaches, be them binary or multiclass, are not applicable. Achieving a high effectiveness when solving one-class problems is difficult anyway and it becomes even more challenging when the target class data is multimodal, which is often the case. To address these concerns we propose a cluster-based one-class ensemble that consists of four steps: (1) applying a clustering algorithm to the target class data, (2) training an individual one-class classifier for each of the identified clusters, (3) aggregating the decisions of the individual classifiers, and (4) selecting the best fitting clustering model. We evaluate our approach with four datasets: an artificially generated dataset, a dataset compiled from a known multiclass text corpus, and two datasets related to one-class problems that received much attention recently, namely authorship verification and quality flaw prediction. Our approach outperforms a one-class SVM on all four datasets.

For Web applications that are based on user generated content the detection of text quality flaws is a key concern. Our research contributes to automatic quality flaw detection. In particular, we propose to cast the detection of text quality flaws as a one-class classification problem: we are given only positive examples (= texts containing a particular quality flaw) and decide whether or not an unseen text suffers from this flaw. We argue that common binary or multiclass classification approaches are ineffective in here, and we underpin our approach by a real-world application: we employ a dedicated one-class learning approach to determine whether a given Wikipedia article suffers from certain quality flaws. Since in the Wikipedia setting the acquisition of sensible test data is quite intricate, we analyze the effects of a biased sample selection. In addition, we illustrate the classifier effectiveness as a function of the flaw distribution in order to cope with the unknown (real-world) flaw-specific class imbalances. Altogether, provided test data with little noise, four from ten important quality flaws in Wikipedia can be detected with a precision close to 1.

Wikipedia provides an information quality assessment model with criteria for human peer reviewers to identify featured articles. For this classification task "Is an article featured or not?" we present a machine learning approach that exploits an article's character trigram distribution. Our approach differs from existing research in that it aims to writing style rather than evaluating meta features like the edit history. The approach is robust, straightforward to implement, and outperforms existing solutions. We underpin these claims by an experiment design where, among others, the domain transferability is analyzed. The achieved performances in terms of the F-measure for featured articles are 0.964 within a single Wikipedia domain and 0.880 in a domain transfer situation.