Skip to main navigation Skip to search Skip to main content

Combine non-text features with deep learning structures based on attention-LSTM for answer selection

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Because of the lexical gap between questions and answer candidates, methods with only word features cannot solve Answer Selection (AS) problem well. In this paper, we apply a LSTMs with Attention model to extract the latent semantic information of sentences and propose a method to learning non-text features. Besides, we propose an index to evaluate the sorting ability of models with the same accuracy value. Our model achieved the best accuracy and F1 performance than other known models, and the ranking index results, including MAP, AvgRec and MRR index’s result, are after only KeLP system and Beihang MSRA system in SemEval-2017 Task 3 Subtask A.

Original languageEnglish
Title of host publicationInformation Retrieval - 23rd China conference, CCIR 2017, Proceedings
EditorsJianyun Nie, Tong Ruan, Tieyun Qian, Jirong Wen, Yiqun Liu
PublisherSpringer Verlag
Pages261-271
Number of pages11
ISBN (Print)9783319686981
DOIs
StatePublished - 2017
Event23rd China conference on Information Retrieval, CCIR 2017 - Shanghai, China
Duration: 13 Jul 201714 Jul 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10390 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference23rd China conference on Information Retrieval, CCIR 2017
Country/TerritoryChina
CityShanghai
Period13/07/1714/07/17

Keywords

  • Answer selection
  • Attention
  • LSTMs
  • Non-text features

Fingerprint

Dive into the research topics of 'Combine non-text features with deep learning structures based on attention-LSTM for answer selection'. Together they form a unique fingerprint.

Cite this