Learning higher-level features with convolutional restricted Boltzmann machines for sentiment analysis

Trung Huynh, Yulan He, Stefan Rüger

Research output: Chapter in Book/Published conference outputConference publication


In recent years, learning word vector representations has attracted much interest in Natural Language Processing. Word representations or embeddings learned using unsupervised methods help addressing the problem of traditional bag-of-word approaches which fail to capture contextual semantics. In this paper we go beyond the vector representations at the word level and propose a novel framework that learns higher-level feature representations of n-grams, phrases and sentences using a deep neural network built from stacked Convolutional Restricted Boltzmann Machines (CRBMs). These representations have been shown to map syntactically and semantically related n-grams to closeby locations in the hidden feature space. We have experimented to additionally incorporate these higher-level features into supervised classifier training for two sentiment analysis tasks: subjectivity classification and sentiment classification. Our results have demonstrated the success of our proposed framework with 4% improvement in accuracy observed for subjectivity classification and improved the results achieved for sentiment classification over models trained without our higher level features.

Original languageEnglish
Title of host publicationAdvances in information retrieval
Subtitle of host publication37th European conference on IR research, ECIR 2015, Vienna, Austria, March 29 - April 2, 2015. Proceedings
EditorsAllan Hanbury, Gabriella Kazai, Andreas Rauber, Norbert Fuhr
Place of PublicationCham (CH)
Number of pages6
ISBN (Electronic)978-3-319-16354-3
ISBN (Print)978-3-319-16353-6
Publication statusPublished - 2015
Event37th European Conference on Information Retrieval Research - Vienna, Austria
Duration: 29 Mar 20152 Apr 2015

Publication series

NameLecture notes in computer science
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


Conference37th European Conference on Information Retrieval Research
Abbreviated titleECIR 2015


  • convolutional restricted Boltzmann machines
  • sentiment analysis
  • stacked restricted Boltzmann Machine
  • word embedding


Dive into the research topics of 'Learning higher-level features with convolutional restricted Boltzmann machines for sentiment analysis'. Together they form a unique fingerprint.

Cite this