Neural Networks for Information Retrieval

Kenter, Tom; Borisov, Alexey; Gysel, C; Dehghani, Mostafa; Rijke, Maarten de; Mitra, Bhaskar

doi:10.1145/3077136.3082062

Cited by 17 publications

(9 citation statements)

References 53 publications

(61 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Beyond representation learning, there are more applications of neural models in IR [Craswell et al 2016, Kenter et al 2017, Onal et al 2018, Van Gysel et al 2017c]. In machine-learned ranking [Liu 2011], we have RankNet [Burges et al 2005].…”

Section: Neural Ir and Language Modelingmentioning

confidence: 99%

Neural Vector Spaces for Unsupervised Information Retrieval

Gysel

Rijke

Kanoulas

2018

ACM Trans. Inf. Syst.

Self Cite

View full text Add to dashboard Cite

We propose the Neural Vector Space Model (NVSM), a method that learns representations of documents in an unsupervised manner for news article retrieval. In the NVSM paradigm, we learn low-dimensional representations of words and documents from scratch using gradient descent and rank documents according to their similarity with query representations that are composed from word representations. We show that NVSM performs better at document ranking than existing latent semantic vector space methods. The addition of NVSM to a mixture of lexical language models and a state-of-the-art baseline vector space model yields a statistically significant increase in retrieval effectiveness. Consequently, NVSM adds a complementary relevance signal. Next to semantic matching, we find that NVSM performs well in cases where lexical matching is needed. NVSM learns a notion of term specificity directly from the document collection without feature engineering. We also show that NVSM learns regularities related to Luhn significance. Finally, we give advice on how to deploy NVSM in situations where model selection (e.g., cross-validation) is infeasible. We find that an unsupervised ensemble of multiple models trained with different hyperparameter values performs better than a single cross-validated model. Therefore, NVSM can safely be used for ranking documents without supervised relevance judgments.

show abstract

Section: Neural Ir and Language Modelingmentioning

confidence: 99%

Neural Vector Spaces for Unsupervised Information Retrieval

Gysel

Rijke

Kanoulas

2018

ACM Trans. Inf. Syst.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The idea of adversarial networks is initially presented by Goodfellow et al (Goodfellow et al, 2014) for image generation, besides, and has been applied broadly in many tasks of NLP field, such as information retrieval (Kenter et al, 2017), machine comprehension (Wang et al, 2017), dialog generation (Lu et al, 2017), and fake news detection (Wang et al, 2018). The goal of adversarial networks is to use a generative network G to generate a data distribution P G (x) that matches the real data distribution P data (x) as much as possible.…”

Section: Adversarial Networkmentioning

confidence: 99%

Discovering differential features: Adversarial learning for information credibility evaluation

Rao

Nazir

et al. 2020

Information Sciences

View full text Add to dashboard Cite

A series of deep learning approaches extract a large number of credibility features to detect fake news on the Internet. However, these extracted features still suffer from many irrelevant and noisy features that restrict severely the performance of the approaches. In this paper, we propose a novel model based on Adversarial Networks and inspirited by the Shared-Private model (ANSP), which aims at reducing common, irrelevant features from the extracted features for information credibility evaluation. Specifically, ANSP involves two tasks: one is to prevent the binary classification of true and false information for capturing common features relying on adversarial networks guided by reinforcement learning. Another extracts credibility features (henceforth, private features) from multiple types of credibility information and compares with the common features through two strategies, i.e., orthogonality constraints and KL-divergence for making the private features more differential. Experiments first on two six-label LIAR and Weibo datasets demonstrate that ANSP achieves the state-of-theart performance, boosting the accuracy by 2.1%, 3.1%, respectively and then on four-label Twit-ter16 validate the robustness of the model with 1.8% performance improvements.

show abstract

“…Other topics related to scoring and ranking are only partly covered, e.g., term dependency in [27,56] and semantic scoring in [15,27,56]. This is natural as some topics, such as word embeddings for ranking [2] and neural IR [48,58,60] have only been introduced recently. Given the above observations, we believe that today's IR textbooks have to be updated to include recent developments in scoring and ranking, especially in the fields of LTR and semantic learning to match (especially, neural IR).…”

Section: Offline Phasementioning

confidence: 99%

“…Ai et al [3] Unbiased learning to rank Arguello [5] Aggregated search Barocas and Hardt [8] Fairness in machine learning Bast et al [9] Semantic search, knowledge graphs Budylin et al [12,13] Online evaluation Burges [14] Learning to rank Cai and de Rijke [16] Query auto-completion Cambazoglu and Baeza-Yates [17] Infrastructure Chuklin et al [21,22,23,24] Click models Crestani et al [26] Mobile information retrieval Gao et al [30] Conversational search Glowacka [32] Bandit algorithms Grotov and de Rijke [33] Online learning to rank Hajian et al [35] Algorithmic bias Hofmann et al [40] Online evaluation Hui Yang and Zhang [41] Differential privacy in information retrieval Joachims and Swaminathan [44] Counterfactual evaluation and learning Jones [45] Mobile search Kanoulas [46] Online and offline evaluation Kelly [47] User studies Kenter et al [48] Neural methods in information retrieval Knijnenburg and Berkovsky [49] Privacy in recommender systems Lalmas [51] XML retrieval Lattimore and Szepesvári [52] Bandit algorithms Liu [55] Offline learning to rank Mehrotra et al [57] Task understanding Mitra and Craswell [58] Neural methods in information retrieval Onal et al [60] Neural methods in information retrieval Oosterhuis [61] Online evaluation and ranking Ren et al [66] E-commerce Sakai [67] Experimental design and methodology Santos et al [68] Diversification Silvestri…”

Section: Author(s) Topicmentioning

confidence: 99%

What Should We Teach in Information Retrieval?

Markov¹,

Rijke

2019

SIGIR Forum

Self Cite

View full text Add to dashboard Cite

Modern Information Retrieval (IR) systems, such as search engines, recommender systems, and conversational agents, are best thought of as interactive systems. And their development is best thought of as a two-stage development process: offline development followed by continued online adaptation and development based on interactions with users. In this opinion paper, we take a closer look at existing IR textbooks and teaching materials, and examine to which degree they cover the offline and online stages of the IR system development process. We notice that current teaching materials in IR focus mostly on search and on the offline development phase. Other scenarios of interacting with information are largely absent from current IR teaching materials, as is the (interactive) online development phase. We identify a list of scenarios and a list of topics that we believe are essential to any modern set of IR teaching materials that claims to fully cover IR system development. In particular, we argue for more attention, in basic IR teaching materials, to scenarios such as recommender systems, and to topics such as query and interaction mining and understanding, online evaluation, and online learning to rank.

show abstract

Neural Networks for Information Retrieval

Cited by 17 publications

References 53 publications

Neural Vector Spaces for Unsupervised Information Retrieval

Neural Vector Spaces for Unsupervised Information Retrieval

Discovering differential features: Adversarial learning for information credibility evaluation

What Should We Teach in Information Retrieval?

Contact Info

Product

Resources

About