Opinion Recommendation using Neural Memory Model

We present opinion recommendation, a novel task of jointly predicting a custom review with a rating score that a certain user would give to a certain product or service, given existing reviews and rating scores to the product or service by other users, and the reviews that the user has given to other products and services. A characteristic of opinion recommendation is the reliance of multiple data sources for multi-task joint learning, which is the strength of neural models. We use a single neural network to model users and products, capturing their correlation and generating customised product representations using a deep memory network, from which customised ratings and reviews are constructed jointly. Results show that our opinion recommendation system gives ratings that are closer to real user ratings on Yelp.com data compared with Yelp's own ratings, and our methods give better results compared to several pipelines baselines using state-of-the-art sentiment rating and summarization systems.

• Publication year

  2017

• Venue

  arXiv.org

• Publication date

  2017-02-06

• Fields of study

  Computer Science

• Identifiers
  arXiv 1702.01517
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• The World Wide Web Conference

  2019cited by this paper
• Opinion Mining and Sentiment Analysis

  2018cited by this paper
• Dynamic Memory Networks for Visual and Textual Question Answering

  2016cited by this paper
• Memory-enhanced Decoder for Neural Machine Translation

  2016cited by this paper
• Neural Sentiment Classification with User and Product Attention

  2016cited by this paper
• Neural Network-Based Abstract Generation for Opinions and Arguments

  2016influential reference
• Neural Network for Heterogeneous Annotations

  2016cited by this paper
• Recurrent Memory Networks for Language Modeling

  2016cited by this paper
• Context-Sensitive Lexicon Features for Neural Sentiment Analysis

  2016cited by this paper
• Stack-propagation: Improved Representation Learning for Syntax

  2016cited by this paper
• Fast Matrix Factorization for Online Recommendation with Implicit Feedback

  2016cited by this paper
• Hierarchical Attention Networks for Document Classification

  2016cited by this paper
• FLAME: A Probabilistic Model Combining Aspect Based Opinion Mining and Collaborative Filtering

  2015cited by this paper
• A Neural Attention Model for Abstractive Sentence Summarization

  2015cited by this paper
• Improved Semantic Representations From Tree-Structured Long Short-Term Memory Networks

  2015cited by this paper
• End-To-End Memory Networks

  2015cited by this paper
• Fine-grained Opinion Mining with Recurrent Neural Networks and Word Embeddings

  2015cited by this paper
• User Modeling with Neural Network for Review Rating Prediction

  2015cited by this paper
• Ask Me Anything: Dynamic Memory Networks for Natural Language Processing

  2015cited by this paper
• Recommender Systems Handbook

  2015cited by this paper
• Neural Machine Translation by Jointly Learning to Align and Translate

  2014cited by this paper
• Explicit factor models for explainable recommendation based on phrase-level sentiment analysis

  2014cited by this paper
• A survey of collaborative filtering based social recommender systems

  2014cited by this paper
• Sequence to Sequence Learning with Neural Networks

  2014cited by this paper
• Convolutional Neural Networks for Sentence Classification

  2014cited by this paper
• Neural Turing Machines

  2014cited by this paper
• Opinion Mining with Deep Recurrent Neural Networks

  2014cited by this paper
• SUIT: A Supervised User-Item Based Topic Model for Sentiment Analysis

  2014cited by this paper
• Joint Inference for Fine-grained Opinion Extraction

  2013cited by this paper
• Recursive Deep Models for Semantic Compositionality Over a Sentiment Treebank

  2013cited by this paper
• Distributed Representations of Words and Phrases and their Compositionality

  2013cited by this paper
• Co-Regression for Cross-Language Review Rating Prediction

  2013cited by this paper
• Multilingual Joint Parsing of Syntactic and Semantic Dependencies with a Latent Variable Model

  2013cited by this paper
• Modeling User Leniency and Product Popularity for Sentiment Classification

  2013cited by this paper
• Improving neural networks by preventing co-adaptation of feature detectors

  2012influential reference
• Association for Computational Linguistics: Human Language Technologies

  2011cited by this paper
• Natural Language Processing (Almost) from Scratch

  2011cited by this paper
• Combining Collaborative Filtering and Sentiment Classification for Improved Movie Recommendations

  2011cited by this paper
• A Comprehensive Survey of Neighborhood-based Recommendation Methods

  2011cited by this paper
• Adaptive Subgradient Methods for Online Learning and Stochastic Optimization

  2011cited by this paper
• Opinion Word Expansion and Target Extraction through Double Propagation

  2011cited by this paper
• Incorporating Reviewer and Product Information for Review Rating Prediction

  2011cited by this paper
• A Pilot Study of Opinion Summarization in Conversations

  2011cited by this paper
• Opinosis: A Graph Based Approach to Abstractive Summarization of Highly Redundant Opinions

  2010cited by this paper
• Extracting Opinion Targets in a Single and Cross-Domain Setting with Conditional Random Fields

  2010cited by this paper
• The Bag-of-Opinions Method for Review Rating Prediction from Sparse Text Patterns

  2010cited by this paper
• Optimizing Informativeness and Readability for Sentiment Summarization

  2010cited by this paper
• A Non-negative Matrix Tri-factorization Approach to Sentiment Classification with Lexical Prior Knowledge

  2009cited by this paper
• A Survey of Collaborative Filtering Techniques

  2009cited by this paper
• Joint sentiment/topic model for sentiment analysis

  2009cited by this paper
• Factorization meets the neighborhood: a multifaceted collaborative filtering model

  2008cited by this paper
• Orthogonal nonnegative matrix t-factorizations for clustering

  2006influential reference
• Integrating Collaborative Filtering and Sentiment Analysis: A Rating Inference Approach

  2006cited by this paper
• Seeing Stars: Exploiting Class Relationships for Sentiment Categorization with Respect to Rating Scales

  2005cited by this paper
• Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions

  2005cited by this paper
• Mining and summarizing customer reviews

  2004cited by this paper
• ROUGE: A Package for Automatic Evaluation of Summaries

  2004influential reference
• Item-based collaborative filtering recommendation algorithms

  2001cited by this paper
• Long Short-Term Memory

  1997influential reference

• ExpScore: Learning Metrics for Recommendation Explanation

  2022cites this paper
• Memory Reorganization: A Symmetric Memory Network for Reorganizing Neighbors and Topics to Complete Rating Prediction

  2020cites this paper
• Generate Neural Template Explanations for Recommendation

  2020influential citation
• Generation for Recommender Systems

  2019cites this paper
• Multimodal Review Generation for Recommender Systems

  2019cites this paper
• Long-term Product Rating Prediction Based on Users' Short-term Multiple Ratings

  2019cites this paper
• Recommendation for Multi-stakeholders and through Neural Review Mining

  2019cites this paper
• Sentiment analysis using deep learning approaches: an overview

  2019cites this paper
• Short Review of Sentiment-Based Recommender Systems

  2018cites this paper
• Multi-Modal Multi-Task Learning for Automatic Dietary Assessment

  2018cites this paper
• Statutes Recommendation Using Classification and Co-occurrence Between Statutes

  2018cites this paper
• Encoding User as More Than the Sum of Their Parts: Recurrent Neural Networks and Word Embedding for People-to-people Recommendation

  2017cites this paper