On the Benefit of Incorporating External Features in a Neural Architecture for Answer Sentence Selection

Chen, Ruey-Cheng; Yulianti, Evi; Sanderson, Mark; Croft, W. Bruce

doi:10.1145/3077136.3080705

Cited by 10 publications

(9 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model is simple, fast and well studied. It has also been reproduced in other work Chen et al, 2017;Sequiera et al, 2017).…”

Section: Answer Sentence Selection Networksupporting

confidence: 65%

“…We remove from the dev. and test sets questions without answers, and questions with System MAP MRR Santos et al 201675.30 85.11 75.88 82.19 Severyn and Moschitti (2016) 76.54 81.86 Wang et al (2016b) 77.14 84.47 Rao et al (2016) 80.10 87.70 80.20 87.50 Shen et al (2017) 82 Yin et al (2016) 69.21 71.08 Severyn and Moschitti (2016) 69.51 71.07 Chen et al (2017) 70.10 71.80 Rao et al (2016) 70.90 72.30 Tymoshenko et al (2016) 71.25 72.30 Guo et al (2017) 71.71 73.36 71.80 73.10 Shen et al 201773.30 75.00 Wang et al (2016a) 73.41 74.18 Wang and Jiang (2017) 74 training instances with difficult negative examples. Our system beats several others that use word alignments and attention mechanisms.…”

Section: Trecqamentioning

confidence: 99%

“…We remove from the dev. and test sets questions without answers, and questions with Yin et al (2016) 69.21 71.08 Severyn and Moschitti (2016) 69.51 71.07 Chen et al (2017) 70.10 71.80 Rao et al (2016) 70.90 72.30 Tymoshenko et al (2016) 71.25 72.30 Guo et al (2017) 71.71 73.36 71.80 73. 10 Shen et al (2017) 73.30 75.00 Wang et al (2016a) 73.41 74.18 Wang and Jiang (2017) 74 training instances with difficult negative examples.…”

Section: Trecqamentioning

confidence: 99%

See 2 more Smart Citations

Semantic Linking in Convolutional Neural Networks for Answer Sentence Selection

Nicosia

Moschitti²

2018

Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

State-of-the-art networks that model relations between two pieces of text often use complex architectures and attention. In this paper, instead of focusing on architecture engineering, we take advantage of small amounts of labelled data that model semantic phenomena in text to encode matching features directly in the word representations. This greatly boosts the accuracy of our reference network, while keeping the model simple and fast to train. Our approach also beats a tree kernel model that uses similar input encodings, and neural models which use advanced attention and compare-aggregate mechanisms. * Now at Google † This work was partially carried out when the author was at the University of Trento 1 http://dawn.cs.stanford.edu/ benchmark/

show abstract

“…This model is simple, fast and well studied. It has also been reproduced in other work Chen et al, 2017;Sequiera et al, 2017).…”

Section: Answer Sentence Selection Networksupporting

confidence: 65%

Section: Trecqamentioning

confidence: 99%

Section: Trecqamentioning

confidence: 99%

See 1 more Smart Citation

Semantic Linking in Convolutional Neural Networks for Answer Sentence Selection

Nicosia

Moschitti²

2018

Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing

View full text Add to dashboard Cite

show abstract

“…That is, passages are ranked in response to a query using passage-relevance estimates. The merits of integrating the estimates using learning-to-rank (LTR) approaches were also demonstrated [5,9,10,39,56,58].…”

Section: Introductionmentioning

confidence: 99%

A passage-based approach to learning to rank documents

Sheetrit

Shtok²,

Kurland

2020

Inf Retrieval J

View full text Add to dashboard Cite

According to common relevance-judgments regimes, such as TREC's, a document can be deemed relevant to a query even if it contains a very short passage of text with pertinent information. This fact has motivated work on passage-based document retrieval: document ranking methods that induce information from the document's passages. However, the main source of passage-based information utilized was passage-query similarities. We address the challenge of utilizing richer sources of passage-based information to improve document retrieval effectiveness. Specifically, we devise a suite of learning-to-rank-based document retrieval methods that utilize an effective ranking of passages produced in response to the query; the passage ranking is also induced using a learning-to-rank approach. Some of the methods quantify the ranking of the passages of a document. Others utilize the feature-based representation of passages used for learning a passage ranker. Empirical evaluation attests to the clear merits of our methods with respect to highly effective baselines. Our best performing method is based on learning a document ranking function using document-query features and passage-query features of the document's passage most highly ranked.

show abstract

“…With the small size of the dataset, traditional features demonstrate good performance in comparison with the neural network models. This, along with more evidence on the usefulness of combining traditional features in deep learning architecture (R.-C. Chen, Yulianti, Sanderson, & Bruce Croo, 2017;Sequiera et al, 2017), encouraged us to build a hybrid model. An essential dilemma for building the feature-assisted model is how to incorporate engineered features to sentence embeddings inputs.…”

Section: A Feature-assisted Neural Network Architecture Modelmentioning

confidence: 99%

Text Mining for Precision Medicine

Abdelrahman¹

View full text Add to dashboard Cite

On the Benefit of Incorporating External Features in a Neural Architecture for Answer Sentence Selection

Cited by 10 publications

References 15 publications

Semantic Linking in Convolutional Neural Networks for Answer Sentence Selection

Semantic Linking in Convolutional Neural Networks for Answer Sentence Selection

A passage-based approach to learning to rank documents

Text Mining for Precision Medicine

Contact Info

Product

Resources

About