Skip to main navigation Skip to search Skip to main content

A short texts matching method using shallow features and deep features

  • Harbin Institute of Technology Shenzhen
  • Zunyi Medical and Pharmaceutical College

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

Abstract

Semantic matching is widely used in many natural language processing tasks. In this paper, we focus on the semantic matching between short texts and design a model to generate deep features, which describe the semantic relevance between short “text object”. Furthermore, we design a method to combine shallow features of short texts (i.e., LSI, VSM and some other handcraft features) with deep features of short texts (i.e., word embedding matching of short text). Finally, a ranking model (i.e., RankSVM) is used to make the final judgment. In order to evaluate our method, we implement our method on the task of matching posts and responses. Results of experiments show that our method achieves the state-of-the-art performance by using shallow features and deep features.

Original languageEnglish
Title of host publicationNatural Language Processing and Chinese Computing - 3rd CCF Conference, NLPCC 2014, Proceedings
EditorsChengqing Zong, Jian-Yun Nie, Dongyan Zhao, Yansong Feng
PublisherSpringer Verlag
Pages150-159
Number of pages10
ISBN (Electronic)9783662459232
DOIs
StatePublished - 2014
Externally publishedYes
Event3rd CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2014 - Shenzhen, China
Duration: 5 Dec 20149 Dec 2014

Publication series

NameCommunications in Computer and Information Science
Volume496
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference3rd CCF Conference on Natural Language Processing and Chinese Computing, NLPCC 2014
Country/TerritoryChina
CityShenzhen
Period5/12/149/12/14

Keywords

  • Ranking Model
  • Semantic Matching
  • Short Text
  • Word Embedding

Fingerprint

Dive into the research topics of 'A short texts matching method using shallow features and deep features'. Together they form a unique fingerprint.

Cite this