Neural information retrieval: introduction to the special issue

Craswell, Nick; Croft, W. Bruce; Rijke, Maarten de; Guo, Jiafeng; Mitra, Bhaskar

doi:10.1007/s10791-017-9323-9

Cited by 9 publications

(7 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, the advances and the success of deep learning in many different tasks have promoted the diffusion of (deep) neural networks in IR, reviving the interest in semantic models in the research community. Since the very first approaches [116,210], neural IR has attracted a lot of attention: dedicated workshops were held at the ACM SIGIR Conference on Research & Development in Information Retrieval (SIGIR) in 2016 [49] and 2017 [47], while an in-depth monograph [165] and a special issue in the Information Retrieval Journal (IRJ) [48] were published in 2018. Also, SIGIR papers employing deep learning are increasing at a fast pace -i.e., from two articles published in 2014 to eleven articles published in 2017 [8] to more than fifty articles published in 2020.…”

Section: Knowledge-enhanced Neural Ir Modelsmentioning

confidence: 99%

Developing unsupervised knowledge-enhanced models to reduce the semantic gap in information retrieval

Marchesin

2021

SIGIR Forum

View full text Add to dashboard Cite

In this thesis we tackle the semantic gap, a long-standing problem in Information Retrieval (IR). The semantic gap can be described as the mismatch between users' queries and the way retrieval models answer to such queries. Two main lines of work have emerged over the years to bridge the semantic gap: (i) the use of external knowledge resources to enhance the bag-of-words representations used by lexical models, and (ii) the use of semantic models to perform matching between the latent representations of queries and documents. To deal with this issue, we first perform an in-depth evaluation of lexical and semantic models through different analyses [Marchesin et al., 2019]. The objective of this evaluation is to understand what features lexical and semantic models share, if their signals are complementary, and how they can be combined to effectively address the semantic gap. In particular, the evaluation focuses on (semantic) neural models and their critical aspects. Each analysis brings a different perspective in the understanding of semantic models and their relation with lexical models. The outcomes of this evaluation highlight the differences between lexical and semantic signals, and the need to combine them at the early stages of the IR pipeline to effectively address the semantic gap. Then, we build on the insights of this evaluation to develop lexical and semantic models addressing the semantic gap. Specifically, we develop unsupervised models that integrate knowledge from external resources, and we evaluate them for the medical domain - a domain with a high social value, where the semantic gap is prominent, and the large presence of authoritative knowledge resources allows us to explore effective ways to address it. For lexical models, we investigate how - and to what extent - concepts and relations stored within knowledge resources can be integrated in query representations to improve the effectiveness of lexical models. Thus, we propose and evaluate several knowledge-based query expansion and reduction techniques [Agosti et al., 2018, 2019; Di Nunzio et al., 2019]. These query reformulations are used to increase the probability of retrieving relevant documents by adding to or removing from the original query highly specific terms. The experimental analyses on different test collections for Precision Medicine - a particular use case of Clinical Decision Support (CDS) - show the effectiveness of the proposed query reformulations. In particular, a specific subset of query reformulations allow lexical models to achieve top performing results in all the considered collections. Regarding semantic models, we first analyze the limitations of the knowledge-enhanced neural models presented in the literature. Then, to overcome these limitations, we propose SAFIR [Agosti et al., 2020], an unsupervised knowledge-enhanced neural framework for IR. SAFIR integrates external knowledge in the learning process of neural IR models and it does not require labeled data for training. Thus, the representations learned within this framework are optimized for IR and encode linguistic features that are relevant to address the semantic gap. The evaluation on different test collections for CDS demonstrate the effectiveness of SAFIR when used to perform retrieval over the entire document collection or to retrieve documents for Pseudo Relevance Feedback (PRF) methods - that is, when it is used at the early stages of the IR pipeline. In particular, the quantitative and qualitative analyses highlight the ability of SAFIR to retrieve relevant documents affected by the semantic gap, as well as the effectiveness of combining lexical and semantic models at the early stages of the IR pipeline - where the complementary signals they provide can be used to obtain better answers to semantically hard queries.

show abstract

Section: Knowledge-enhanced Neural Ir Modelsmentioning

confidence: 99%

Developing unsupervised knowledge-enhanced models to reduce the semantic gap in information retrieval

Marchesin

2021

SIGIR Forum

View full text Add to dashboard Cite

show abstract

“…The purpose was to provide a forum for new and early work relating to deep learning and other neural approaches to IR, and discuss the main challenges facing this line of research. Since then, research publication in the area has been increasing (see Figure 8.1 and [466]), along with relevant workshops [467][468][469], tutorials [3][4][5][6]470], and plenary talks [471,472]. While there has been significant interest in deep learning for ad-hoc ranking [1], the work till recently has largely been done with small data, proprietary data or synthetic data.…”

Section: Chaptermentioning

confidence: 99%

Neural methods for effective, efficient, and exposure-aware information retrieval

Mitra

2021

SIGIR Forum

Self Cite

View full text Add to dashboard Cite

Neural networks with deep architectures have demonstrated significant performance improvements in computer vision, speech recognition, and natural language processing. The challenges in information retrieval (IR), however, are different from these other application areas. A common form of IR involves ranking of documents---or short passages---in response to keyword-based queries. Effective IR systems must deal with query-document vocabulary mismatch problem, by modeling relationships between different query and document terms and how they indicate relevance. Models should also consider lexical matches when the query contains rare terms---such as a person's name or a product model number---not seen during training, and to avoid retrieving semantically related but irrelevant results. In many real-life IR tasks, the retrieval involves extremely large collections---such as the document index of a commercial Web search engine---containing billions of documents. Efficient IR methods should take advantage of specialized IR data structures, such as inverted index, to efficiently retrieve from large collections. Given an information need, the IR system also mediates how much exposure an information artifact receives by deciding whether it should be displayed, and where it should be positioned, among other results. Exposure-aware IR systems may optimize for additional objectives, besides relevance, such as parity of exposure for retrieved items and content publishers. In this thesis, we present novel neural architectures and methods motivated by the specific needs and challenges of IR tasks. We ground our contributions with a detailed survey of the growing body of neural IR literature [Mitra and Craswell, 2018]. Our key contribution towards improving the effectiveness of deep ranking models is developing the Duet principle [Mitra et al., 2017] which emphasizes the importance of incorporating evidence based on both patterns of exact term matches and similarities between learned latent representations of query and document. To efficiently retrieve from large collections, we develop a framework to incorporate query term independence [Mitra et al., 2019] into any arbitrary deep model that enables large-scale precomputation and the use of inverted index for fast retrieval. In the context of stochastic ranking, we further develop optimization strategies for exposure-based objectives [Diaz et al., 2020]. Finally, this dissertation also summarizes our contributions towards benchmarking neural IR models in the presence of large training datasets [Craswell et al., 2019] and explores the application of neural methods to other IR tasks, such as query auto-completion.

show abstract

“…ML classification is the way to categorize documents; therefore, it is little surprising that classification in IR can be found dating back to the seventies. Based on the given set of topic, there are a number of publications addressing classification with IR [11]- [16]. Apart from IR, classification is a topic of computer vision where classification is based on contextual information in the images.…”

Section: Literature Surveymentioning

confidence: 99%

“…l:e l >0 e l (16) Quantum SDT in IR was introduced in [42] to re-weight the query terms and re-rank the retrieved documents. Consider the vector |y , which is the input query of an IR system providing a ranked list of documents, each document being represented by a vector |x .…”

Section: A Quantum Sdtmentioning

confidence: 99%

Towards a Quantum-Inspired Binary Classifier

Tiwari

Melucci

2019

IEEE Access

View full text Add to dashboard Cite

Machine Learning classification models learn the relation between input as features and output as a class in order to predict the class for the new given input. Several research works have demonstrated the effectiveness of machine learning algorithms but the state-of-the-art algorithms are based on the classical theories of probability and logic. Quantum Mechanics (QM) has already shown its effectiveness in many fields and researchers have proposed several interesting results which cannot be obtained through classical theory. In recent years, researchers have been trying to investigate whether the QM can help to improve the classical machine learning algorithms. It is believed that the theory of QM may also inspire an effective algorithm if it is implemented properly. From this inspiration, we propose the quantum-inspired binary classifier, which is based on quantum detection theory. We used text corpora and image corpora to explore the effect of our proposed model. Our proposed model outperforms the state-of-the-art models in terms of precision, recall, and F-measure for several topics (categories) in the 20 newsgroup text corpora. Our proposed model outperformed all the baselines in terms of recall when the MNIST handwritten image dataset was used; F-measure is also higher for most of the categories and precision is also higher for some categories. Our proposed model suggests that binary classification effectiveness can be achieved by using quantum detection theory. In particular, we found that our Quantum-Inspired Binary Classifier can increase the precision, recall, and F-measure of classification where the state-of-the-art methods cannot.

show abstract

Neural information retrieval: introduction to the special issue

Cited by 9 publications

References 36 publications

Developing unsupervised knowledge-enhanced models to reduce the semantic gap in information retrieval

Developing unsupervised knowledge-enhanced models to reduce the semantic gap in information retrieval

Neural methods for effective, efficient, and exposure-aware information retrieval

Towards a Quantum-Inspired Binary Classifier

Contact Info

Product

Resources

About