In benchmarking of Information Retrieval systems, the Wilcoxon signed-rank test is often treated as a safer alternative to the t-test. This belief is fueled by textbooks and recommendations that portray Wilcoxon as the proper non-parametric alternative because metric scores are not normally distributed. We argue that this narrative is misleading and harmful. A careful review of Statistics textbooks reveals inconsistencies and omissions in how the assumptions underlying these tests are presented, fostering confusion that has propagated into IR research. As a result, Wilcoxon has been routinely misapplied for decades, creating a false sense of safety against a threat that was never there to begin with, while introducing another one so severe that it virtually guarantees the test will break down and mislead researchers. Through a combination of systematic literature review, analysis and empirical demonstrations with TREC data, we show how and why the Wilcoxon test easily loses control of its Type I error rate in IR settings. We conclude that the continued use of Wilcoxon in IR evaluation is unjustified and that abandoning it would improve the methodological soundness of our field.

Rank-Biased Overlap (RBO) is a similarity measure for indefinite rankings: it is top-weighted, and can be computed when only a prefix of the rankings is known or when they have only some items in common. It is widely used for instance to analyze differences between search engines by comparing the rankings of documents they retrieve for the same queries. In these situations, though, it is very frequent to find tied documents that have the same score. Unfortunately, the treatment of ties in RBO remains superficial and incomplete, in the sense that it is not clear how to calculate it from the ranking prefixes only. In addition, the existing way of dealing with ties is very different from the one traditionally followed in the field of Statistics, most notably found in rank correlation coefficients such as Kendall's and Spearman's. In this paper we propose a generalized formulation for RBO to handle ties, thanks to which we complete the original definitions by showing how to perform prefix evaluation. We also use it to fully develop two variants that align with the ones found in the Statistics literature: one when there is a reference ranking to compare to, and one when there is not. Overall, these three variants provide researchers with flexibility when comparing rankings with RBO, by clearly determining what ties mean, and how they should be treated. Finally, using both synthetic and TREC data, we demonstrate the use of these new tie-aware RBO measures. We show that the scores may differ substantially from the original tie-unaware RBO measure, where ties had to be broken at random or by arbitrary criteria such as by document ID. Overall, these results evidence the need for a proper account of ties in rank similarity measures such as RBO.

Direct optimization of IR metrics has often been adopted as an approach to devise and develop ranking-based recommender systems. Most methods following this approach (e.g. TFMAP, CLiMF, Top-N-Rank) aim at optimizing the same metric being used for evaluation, under the assumption that this will lead to the best performance. A number of studies of this practice bring this assumption, however, into question. In this paper, we dig deeper into this issue in order to learn more about the effects of the choice of the metric to optimize on the performance of a ranking-based recommender system. We present an extensive experimental study conducted on different datasets in both pairwise and listwise learning-to-rank (LTR) scenarios, to compare the relative merit of four popular IR metrics, namely RR, AP, nDCG and RBP, when used for optimization and assessment of recommender systems in various combinations. For the first three, we follow the practice of loss function formulation available in literature. For the fourth one, we propose novel loss functions inspired by RBP for both the pairwise and listwise scenario. Our results confirm that the best performance is indeed not necessarily achieved when optimizing the same metric being used for evaluation. In fact, we find that RBP-inspired losses perform at least as well as other metrics in a consistent way, and offer clear benefits in several cases. Interesting to see is that RBP-inspired losses, while improving the recommendation performance for all uses, may lead to an individual performance gain that is correlated with the activity level of a user in interacting with items. The more active the users, the more they benefit. Overall, our results challenge the assumption behind the current research practice of optimizing and evaluating the same metric, and point to RBP-based optimization instead as a promising alternative when learning to rank in the recommendation context.

In a collaborative-filtering recommendation scenario, biases in the data will likely propagate in the learned recommendations. In this paper we focus on the so-called mainstream bias: the tendency of a recommender system to provide better recommendations to users who have a mainstream taste, as opposed to non-mainstream users. We propose NAECF, a conceptually simple but effective idea to address this bias. The idea consists of adding an autoencoder (AE) layer when learning user and item representations with text-based Convolutional Neural Networks. The AEs, one for the users and one for the items, serve as adversaries to the process of minimizing the rating prediction error when learning how to recommend. They enforce that the specific unique properties of all users and items are sufficiently well incorporated and preserved in the learned representations. These representations, extracted as the bottlenecks of the corresponding AEs, are expected to be less biased towards mainstream users, and to provide more balanced recommendation utility across all users. Our experimental results confirm these expectations, significantly improving the recommendations for non-mainstream users while maintaining the recommendation quality for mainstream users. Our results emphasize the importance of deploying extensive content-based features, such as online reviews, in order to better represent users and items to maximize the de-biasing effect.

Inspired by the success of deploying deep learning in the fields of Computer Vision and Natural Language Processing, this learning paradigm has also found its way into the field of Music Information Retrieval. In order to benefit from deep learning in an effective, but also efficient manner, deep transfer learning has become a common approach. In this approach, it is possible to reuse the output of a pre-trained neural network as the basis for a new learning task. The underlying hypothesis is that if the initial and new learning tasks show commonalities and are applied to the same type of input data (e.g., music audio), the generated deep representation of the data is also informative for the new task. Since, however, most of the networks used to generate deep representations are trained using a single initial learning source, their representation is unlikely to be informative for all possible future tasks. In this paper, we present the results of our investigation of what are the most important factors to generate deep representations for the data and learning tasks in the music domain. We conducted this investigation via an extensive empirical study that involves multiple learning sources, as well as multiple deep learning architectures with varying levels of information sharing between sources, in order to learn music representations. We then validate these representations considering multiple target datasets for evaluation. The results of our experiments yield several insights into how to approach the design of methods for learning widely deployable deep data representations in the music domain.

Recently proposed methods allow the generation of simulated scores representing the values of an effectiveness metric, but they do not investigate the generation of the actual lists of retrieved documents. In this paper we address this limitation: we present an approach that exploits an evolutionary algorithm and, given a metric score, creates a simulated relevance profile (i.e., a ranked list of relevance values) that produces that score. We show how the simulated relevance profiles are realistic under various analyses.

Statistical significance testing is widely accepted as a means to assess how well a difference in effectiveness reflects an actual difference between systems, as opposed to random noise because of the selection of topics. According to recent surveys on SIGIR, CIKM, ECIR and TOIS papers, the t-test is the most popular choice among IR researchers. However, previous work has suggested computer intensive tests like the bootstrap or the permutation test, based mainly on theoretical arguments. On empirical grounds, others have suggested non-parametric alternatives such as the Wilcoxon test. Indeed, the question of which tests we should use has accompanied IR and related fields for decades now. Previous theoretical studies on this matter were limited in that we know that test assumptions are not met in IR experiments, and empirical studies were limited in that we do not have the necessary control over the null hypotheses to compute actual Type I and Type II error rates under realistic conditions. Therefore, not only is it unclear which test to use, but also how much trust we should put in them. In contrast to past studies, in this paper we employ a recent simulation methodology from TREC data to go around these limitations. Our study comprises over 500 million p-values computed for a range of tests, systems, effectiveness measures, topic set sizes and effect sizes, and for both the 2-tail and 1-tail cases. Having such a large supply of IR evaluation data with full knowledge of the null hypotheses, we are finally in a position to evaluate how well statistical significance tests really behave with IR data, and make sound recommendations for practitioners.

In test collection based evaluation of IR systems, score standardization has been proposed to compare systems across collections and minimize the effect of outlier runs on specific topics. The underlying idea is to account for the difficulty of topics, so that systems are scored relative to it. Webber et al. first proposed standardization through a non-linear transformation with the standard normal distribution, and recently Sakai proposed a simple linear transformation. In this paper, we show that both approaches are actually special cases of a simple standardization which assumes specific distributions for the per-topic scores. From this viewpoint, we argue that a transformation based on the empirical distribution is the most appropriate choice for this kind of standardization. Through a series of experiments on TREC data, we show the benefits of our proposal in terms of score stability and statistical test behavior.

Part of Information Retrieval evaluation research is limited by the fact that we do not know the distributions of system effectiveness over the populations of topics and, by extension, their true mean scores. The workaround usually consists in resampling topics from an existing collection and approximating the statistics of interest with the observations made between random subsamples, as if one represented the population and the other a random sample. However, this methodology is clearly limited by the availability of data, the impossibility to control the properties of these data, and the fact that we do not really measure what we intend to. To overcome these limitations, we propose a method based on vine copulas for stochastic simulation of evaluation results where the true system distributions are known upfront. In the basic use case, it takes the scores from an existing collection to build a semi-parametric model representing the set of systems and the population of topics, which can then be used to make realistic simulations of the scores by the same systems but on random new topics. Our ability to simulate this kind of data not only eliminates the current limitations, but also offers new opportunities for research. As an example, we show the benefits of this approach in two sample applications replicating typical experiments found in the literature. We provide a full R package to simulate new data following the proposed method, which can also be used to fully reproduce the results in this paper.

Abstract Named Entity Recognition is a basic task in Information Extraction that aims at identifying entities of interest within full text documents. The patterns used to recognize entities can be rule based, as in the popular JAPE system. However, hand-crafting effective patterns is often difficult, and yet there is little research devoted to methods capable of learning human-readable patterns, possibly with arbitrary sets of features. In this paper, we present a semi-automatic method to generate both regular expressions and a subset of the JAPE language. It does not need a corpus annotated beforehand. Instead, it employs active learning and combines clustering with an algorithm that finds alignments between symbols present in the entities discovered during the learning process. The method currently supports a fixed set of character features and an arbitrary set of token features, but it can incorporate other kinds of features as well. Through several experiments with an English corpus, we show the ability of the method to generate effective patterns at a low annotation cost, and how it can successfully help in the annotation of brand new corpora.

The Kendall ? and AP rank correlation coefficients have become mainstream in Information Retrieval research for comparing the rankings of systems produced by two different evaluation conditions, such as different effectiveness measures or pool depths. However, in this paper we focus on the expected rank correlation between the mean scores observed with a test collection and the true, unobservable means under the same conditions. In particular, we propose statistical estimators of ? and AP correlations following both parametric and non-parametric approaches, and with special emphasis on small topic sets. Through large scale simulation with TREC data, we study the error and bias of the estimators. In general, such estimates of expected correlation with the true ranking may accompany the results reported from an evaluation experiment, as an easy to understand figure of reliability. All the results in this paper are fully reproducible with data and code available online

Microtexts are a valuable, albeit noisy, source to infer collaborative information. As music plays an important role in many human lives, microblogs on music-related activities are available in abundance. This paper investigates different strategies to estimate music similarity from these data sources. In particular, we first present a framework to extract co-occurrence scores between music artists from microblogs and then investigate 12 similarity estimation functions to subsequently derive resemblance scores. We evaluate the approaches on a collection of microblogs crawled from Twitter over a period of 10 months and compare them to standard tf-idf approaches. As evaluation criteria we use precision and recall in an artist retrieval task as well as rank proximity. We show that collaborative chatter on music can be effectively used to develop music artist similarity measures, which are a core part of every music retrieval and recommendation system. Furthermore, we analyze the effects of the “long tail” on retrieval results and investigate whether results are consistent over time, using a second dataset.

SUMMARY Software process improvement in software development organizations is a complex task that can be solved using knowledge management strategies. The definition and use of process patterns is a proven approach to apply knowledge management strategies in software engineering organizations. One of the main problems for the effective application of process patterns is the difficulty of formalizing the knowledge about the development process using these approaches. In this paper, a framework to manage software project patterns is presented. This framework is composed of a model called reusable project pattern (RPP) and a tool called sdpReuser. It is able to formalize the knowledge on software development projects from software development methodologies, reference frameworks, best practices and also tacit knowledge, useful to software engineers. The authors carried out a case study at the University Carlos III of Madrid, where junior software engineers used RPP. This study shows that the quality of pattern formalization correlates with experience in specific software engineering methods, the quality of the references in software engineering practices and the collaboration among software engineers to develop an RPP. It also analyzed the usefulness of RPP information elements to help software engineers to identify the most appropriate RPP and how to adapt it to project restrictions. Copyright © 2013 John Wiley & Sons, Ltd.

The reliability of a test collection is proportional to the number of queries it contains. But building a collection with many queries is expensive, so researchers have to find a balance between reliability and cost. Previous work on the measurement of test collection reliability relied on data-based approaches that contemplated random what if scenarios, and provided indicators such as swap rates and Kendall tau correlations. Generalizability Theory was proposed as an alternative founded on analysis of variance that provides reliability indicators based on statistical theory. However, these reliability indicators are hard to interpret in practice, because they do not correspond to well known indicators like Kendall tau correlation. We empirically established these relationships based on data from over 40 TREC collections, thus filling the gap in the practical interpretation of Generalizability Theory. We also review the computation of these indicators, and show that they are extremely dependent on the sample of systems and queries used, so much that the required number of queries to achieve a certain level of reliability can vary in orders of magnitude. We discuss the computation of confidence intervals for these statistics, providing a much more reliable tool to measure test collection reliability. Reflecting upon all these results, we review a wealth of TREC test collections, arguing that they are possibly not as reliable as generally accepted and that the common choice of 50 queries is insufficient even for stable rankings.

Previous research has suggested the permutation test as the theoretically optimal statistical significance test for IR evaluation, and advocated for the discontinuation of the Wilcoxon and sign tests. We present a large-scale study comprising nearly 60 million system comparisons showing that in practice the bootstrap, t-test and Wilcoxon test outperform the permutation test under different optimality criteria. We also show that actual error rates seem to be lower than the theoretically expected 5%, further confirming that we may actually be underestimating significance.

Personalization and context-awareness are highly important topics in research on Intelligent Information Systems. In the fields of Music Information Retrieval (MIR) and Music Recommendation in particular, user-centric algorithms should ideally provide music that perfectly fits each individual listener in each imaginable situation and for each of her information or entertainment needs. Even though preliminary steps towards such systems have recently been presented at the “International Society for Music Information Retrieval Conference” (ISMIR) and at similar venues, this vision is still far away from becoming a reality. In this article, we investigate and discuss literature on the topic of user-centric music retrieval and reflect on why the breakthrough in this field has not been achieved yet. Given the different expertises of the authors, we shed light on why this topic is a particularly challenging one, taking computer science and psychology points of view. Whereas the computer science aspect centers on the problems of user modeling, machine learning, and evaluation, the psychological discussion is mainly concerned with proper experimental design and interpretation of the results of an experiment. We further present our ideas on aspects crucial to consider when elaborating user-aware music retrieval systems.

We describe a pilot experiment to update the program of an Information Retrieval course for Computer Science undergraduates. We have engaged the students in the development of a search engine from scratch, and they have been involved in the elaboration, also from scratch, of a complete test collection to evaluate their systems. With this methodology they get a whole vision of the Information Retrieval process as they would find it in a real-world setting, and their direct involvement in the evaluation makes them realize the importance of these laboratory experiments in Computer Science. We show that this methodology is indeed reliable and feasible, and so we plan on improving and keep using it in the next years, leading to a public repository of resources for Information Retrieval courses.

Structured Information Retrieval is gaining a lot of interest in recent years, as this kind of information is becoming an invaluable asset for professional communities such as Software Engineering. Most of the research has focused on XML documents, with initiatives like INEX to bring together and evaluate new techniques focused on structured information. Despite the use of XML documents is the immediate choice, the Web is filled with several other types of structured information, which account for millions of other documents. These documents may be collected directly using standard Web search engines like Google and Yahoo, or following specific search patterns in online repositories like SourceForge. This demo describes a distributed and focused web crawler for any kind of structured documents, and we show with it how to exploit general-purpose resources to gather large amounts of real-world structured documents off the Web. This kind of tool could help building large test collections of other types of documents, such as Java source code for software-oriented search engines or RDF for semantic searching.