As deep learning based models are increasingly being used for information retrieval (IR), a major challenge is to ensure the availability of test collections for measuring their quality. Test collections are generated based on pooling results of various retrieval systems, but until recently this did not include deep learning systems. This raises a major challenge for reusable evaluation: Since deep learning based models use external resources (e.g. word embeddings) and advanced representations as opposed to traditional methods that are mainly based on lexical similarity, they may return different types of relevant document that were not identified in the original pooling. If so, test collections constructed using traditional methods are likely to lead to biased and unfair evaluation results for deep learning (neural) systems. This paper uses simulated pooling to test the fairness and reusability of test collections, showing that pooling based on traditional systems only can lead to biased evaluation of deep learning systems.
University of Crete（克里特大学）；FORTH-ICS（希腊计算机科学研究所（基础研究与技术基金会））
Question Answering (QA) is a challenging topic since it requires tackling the various difficulties of natural language understanding. Since evaluation is important not only for identifying the strong and weak points of the various techniques for QA, but also for facilitating the inception of new methods and techniques, in this paper we present a collection for evaluating QA methods over free text that we have created. Although it is a small collection, it contains cases of increasing difficulty, therefore it has an educational value and it can be used for rapid evaluation of QA systems.
Johns Hopkins University（约翰斯·霍普金斯大学）；University of Maryland, College Park（马里兰大学帕克分校）
HC4 is a new suite of test collections for ad hoc Cross-Language Information Retrieval (CLIR), with Common Crawl News documents in Chinese, Persian, and Russian, topics in English and in the document languages, and graded relevance judgments. New test collections are needed because existing CLIR test collections built using pooling of traditional CLIR runs have systematic gaps in their relevance judgments when used to evaluate neural CLIR methods. The HC4 collections contain 60 topics and about half a million documents for each of Chinese and Persian, and 54 topics and five million documents f…
Query by Example is a well-known information retrieval task in which a document is chosen by the user as the search query and the goal is to retrieve relevant documents from a large collection. However, a document often covers multiple aspects of a topic. To address this scenario we introduce the task of faceted Query by Example in which users can also specify a finer grained aspect in addition to the input query document. We focus on the application of this task in scientific literature search. We envision models which are able to retrieve scientific papers analogous to a query scientific pap…
Information Sciences Institute（信息科学研究所）；University of Southern California（南加州大学）
Language models (LMs) show state of the art performance for common sense (CS) question answering, but whether this ability implies a human-level mastery of CS remains an open question. Understanding the limitations and strengths of LMs can help researchers improve these models, potentially by developing novel ways of integrating external CS knowledge. We devise a series of tests and measurements to systematically quantify their performance on different aspects of CS. We propose the use of cloze testing combined with word embeddings to measure the LM's robustness and confidence. Our results sho…
The University of Melbourne（墨尔本大学）；JD AI Research（京东AI研究院）
Conventional wisdom is that hand-crafted features are redundant for deep learning models, as they already learn adequate representations of text automatically from corpora. In this work, we test this claim by proposing a new method for exploiting handcrafted features as part of a novel hybrid learning approach, incorporating a feature auto-encoder loss component. We evaluate on the task of named entity recognition (NER), where we show that including manual features for part-of-speech, word shapes and gazetteers can improve the performance of a neural CRF model. We obtain a $F_1$ of 91.89 for t…
DFKI GmbH（德国人工智能研究中心）；Università di Bologna（博洛尼亚大学）；Qatar Computing Research Institute (QCRI-HBKU)（卡塔尔计算研究所（QCRI-HBKU））；Amazon（亚马逊）
We propose an automatic language-independent graph-based method to build à-la-carte article collections on user-defined domains from the Wikipedia. The core model is based on the exploration of the encyclopaedia's category graph and can produce both monolingual and multilingual comparable collections. We run thorough experiments to assess the quality of the obtained corpora in 10 languages and 743 domains. According to an extensive manual evaluation, our graph-based model outperforms a retrieval-based approach and reaches an average precision of 84% on in-domain articles. As manual evaluations…
While search efficacy has been evaluated traditionally on the basis of result relevance, fairness of search has attracted recent attention. In this work, we define a notion of distributional fairness and provide a conceptual framework for evaluating search results based on it. As part of this, we formulate a set of axioms which an ideal evaluation framework should satisfy for distributional fairness. We show how existing TREC test collections can be repurposed to study fairness, and we measure potential data bias to inform test collection design for fair search. A set of analyses show metric d…
In this work, we analyze a pseudo-relevance retrieval method based on the results of web search engines. By enriching topics with text data from web search engine result pages and linked contents, we train topic-specific and cost-efficient classifiers that can be used to search test collections for relevant documents. Building upon attempts initially made at TREC Common Core 2018 by Grossman and Cormack, we address questions of system performance over time considering different search engines, queries, and test collections. Our experimental results show how and to which extent the considered c…
Citation recommendation is intended to assist researchers in the process of searching for relevant papers to cite by recommending appropriate citations for a given input text. Existing test collections for this task are noisy and unreliable since they are built automatically from parsed PDF papers. In this paper, we present our ongoing effort at creating a publicly available, manually annotated test collection for citation recommendation. We also conduct a series of experiments to evaluate the effectiveness of content-based baseline models on the test collection, providing results for future w…
Research community evaluations in information retrieval, such as NIST's Text REtrieval Conference (TREC), build reusable test collections by pooling document rankings submitted by many teams. Naturally, the quality of the resulting test collection thus greatly depends on the number of participating teams and the quality of their submitted runs. In this work, we investigate: i) how the number of participants, coupled with other factors, affects the quality of a test collection; and ii) whether the quality of a test collection can be inferred prior to collecting relevance judgments from human as…
Matching mentions of persons to the actual persons (the name disambiguation problem) is central for several digital library applications. Scientists have been working on algorithms to create this matching for decades without finding a universal solution. One problem is that test collections for this problem are often small and specific to a certain collection. In this work, we present an approach that can create large test collections from historical metadata with minimal extra cost. We apply this approach to the DBLP collection to generate two freely available test collections. One collection…
Retrieval systems for scholarly literature offer the ability for the scientific community to search, explore and download scholarly articles across various scientific disciplines. Mostly used by the experts in the particular field, these systems contain user community logs including information on user specific downloaded articles. In this paper we present a novel approach for automatically evaluating document similarity models in large collections of scholarly publications. Unlike typical evaluation settings that use test collections consisting of query documents and human annotated relevance…
Robust test collections are crucial for Information Retrieval research. Recently there is a growing interest in evaluating retrieval systems for domain-specific retrieval tasks, however these tasks often lack a reliable test collection with human-annotated relevance assessments following the Cranfield paradigm. In the medical domain, the TripClick collection was recently proposed, which contains click log data from the Trip search engine and includes two click-based test sets. However the clicks are biased to the retrieval model used, which remains unknown, and a previous study shows that the …
Extending TREC-style test collections by incorporating external resources is a time consuming and challenging task. Making use of freely available web data requires technical skills to work with APIs or to create a web scraping program specifically tailored to the task at hand. We present a light-weight alternative that employs the web data extraction language OXPath to harvest data to be added to an existing test collection from web resources. We demonstrate this by creating an extended version of GIRT4 called GIRT4-XT with additional metadata fields harvested via OXPath from the social scien…
We describe the development, characteristics and availability of a test collection for the task of Web table retrieval, which uses a large-scale Web Table Corpora extracted from the Common Crawl. Since a Web table usually has rich context information such as the page title and surrounding paragraphs, we not only provide relevance judgments of query-table pairs, but also the relevance judgments of query-table context pairs with respect to a query, which are ignored by previous test collections. To facilitate future research with this benchmark, we provide details about how the dataset is pre-pr…
TREC-COVID is a community evaluation designed to build a test collection that captures the information needs of biomedical researchers using the scientific literature during a pandemic. One of the key characteristics of pandemic search is the accelerated rate of change: the topics of interest evolve as the pandemic progresses and the scientific literature in the area explodes. The COVID-19 pandemic provides an opportunity to capture this progression as it happens. TREC-COVID, in creating a test collection around COVID-19 literature, is building infrastructure to support new research and techno…
Crowdsourcing offers an affordable and scalable means to collect relevance judgments for IR test collections. However, crowd assessors may show higher variance in judgment quality than trusted assessors. In this paper, we investigate how to effectively utilize both groups of assessors in partnership. We specifically investigate how agreement in judging is correlated with three factors: relevance category, document rankings, and topical variance. Based on this, we then propose two collaborative judging methods in which a portion of the document-topic pairs are assessed by in-house judges while …
In this position paper we argue that certain aspects of relevance assessment in the evaluation of IR systems are oversimplified and that human assessments represented by qrels should be augmented to take account of contextual factors and the subjectivity of the task at hand. We propose enhancing test collections used in evaluation with information related to human assessors and their interpretation of the task. Such augmented collections would provide a more realistic and user-focused evaluation, enabling us to better understand the evaluation process, the performance of systems and user inter…
To create a new IR test collection at low cost, it is valuable to carefully select which documents merit human relevance judgments. Shared task campaigns such as NIST TREC pool document rankings from many participating systems (and often interactive runs as well) in order to identify the most likely relevant documents for human judging. However, if one's primary goal is merely to build a test collection, it would be useful to be able to do so without needing to run an entire shared task. Toward this end, we investigate multiple active learning strategies which, without reliance on system ranki…
Improvements of entity-relationship (E-R) search techniques have been hampered by a lack of test collections, particularly for complex queries involving multiple entities and relationships. In this paper we describe a method for generating E-R test queries to support comprehensive E-R search experiments. Queries and relevance judgments are created from content that exists in a tabular form where columns represent entity types and the table structure implies one or more relationships among the entities. Editorial work involves creating natural language queries based on relationships represented…
Urdu is a widely spoken language with 163 million speakers worldwide across the globe. Information Retrieval (IR) for Urdu entails special consideration of research community due to its rich morphological features and a large number of speakers. In general, IR evaluation task is not extensively explored for Urdu. The most important missing element is the availability of a standardized evaluation corpus specific to Urdu. In this research work, we propose and construct a standard test collection of Urdu documents for IR evaluation and named it Collection for Urdu Retrieval Evaluation (CURE). We …
This article introduces a new language-independent approach for creating a large-scale high-quality test collection of tweets that supports multiple information retrieval (IR) tasks without running a shared-task campaign. The adopted approach (demonstrated over Arabic tweets) designs the collection around significant (i.e., popular) events, which enables the development of topics that represent frequent information needs of Twitter users for which rich content exists. That inherently facilitates the support of multiple tasks that generally revolve around events, namely event detection, ad-hoc …
Neural retrieval models are generally regarded as fundamentally different from the retrieval techniques used in the late 1990's when the TREC ad hoc test collections were constructed. They thus provide the opportunity to empirically test the claim that pooling-built test collections can reliably evaluate retrieval systems that did not contribute to the construction of the collection (in other words, that such collections can be reusable). To test the reusability claim, we asked TREC assessors to judge new pools created from new search results for the TREC-8 ad hoc collection. These new search …
This paper presents a test collection for contextual point of interest (POI) recommendation in a narrative-driven scenario. There, user history is not available, instead, user requests are described in natural language. The requests in our collection are manually collected from social sharing websites, and are annotated with various types of metadata, including location, categories, constraints, and example POIs. These requests are to be resolved from a dataset of POIs, which are collected from a popular online directory, and are further linked to a geographical knowledge base and enriched wit…
The scalability, as well as the effectiveness, of the different Content-based Image Retrieval (CBIR) approaches proposed in literature, is today an important research issue. Given the wealth of images on the Web, CBIR systems must in fact leap towards Web-scale datasets. In this paper, we report on our experience in building a test collection of 100 million images, with the corresponding descriptive features, to be used in experimenting new scalable techniques for similarity searching, and comparing their results. In the context of the SAPIR (Search on Audio-visual content using Peer-to-peer I…
Microsoft（微软）；University College London（伦敦大学学院）；University of Illinois, Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；National Institute of Standards and Technology（美国国家标准与技术研究院）
The TREC Deep Learning (DL) Track studies ad hoc search in the large data regime, meaning that a large set of human-labeled training data is available. Results so far indicate that the best models with large data may be deep neural networks. This paper supports the reuse of the TREC DL test collections in three ways. First we describe the data sets in detail, documenting clearly and in one place some details that are otherwise scattered in track guidelines, overview papers and in our associated MS MARCO leaderboard pages. We intend this description to make it easy for newcomers to use the TREC…
The combination of informetric analysis and information retrieval allows a twofold application. (1) While in-formetrics analysis is primarily used to gain insights into a scientific domain, it can be used to build recommen-dation or alternative ranking services. They are usually based on methods like co-occurrence or citation analyses. (2) Information retrieval and its decades-long tradition of rigorous evaluation using standard document corpora, predefined topics and relevance judgements can be used as a test bed for informetric analyses. We show a preliminary experiment on how both domains c…
