Jointly modeling aspects, ratings and sentiments for movie recommendation (JMARS)

By Qiming Diao, Minghui Qiu, Chao-Yuan Wu, Alex Smola, Jing Jiang, Chong Wang. 2014

461 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Information Systems, Management Science and Operations Research, Sentiment & Opinion Analysis, Recommender Systems

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Xiangnan He
2. Yongfeng Zhang
3. Tat-seng Chua
4. Xu Chen
5. Bing Liu
6. Jianfeng Gao
7. Xiang Wang
8. Philip S. Yu

In Information Systems

1. Yongfeng Zhang
2. Xiangnan He
3. Xu Chen
4. Tat-seng Chua
5. Xiang Wang
6. Xing Xie
7. Liqiang Nie
8. Fuli Feng

In Management Science and Operations Research

1. Xiangnan He
2. Xiang Wang
3. Tat-seng Chua
4. Yongfeng Zhang
5. Xu Chen
6. Fuli Feng
7. Le Wu
8. Yi Tay

In Sentiment & Opinion Analysis

1. Bing Liu
2. Shuai Wang
3. Bing Qin
4. Duyu Tang
5. Ting-Wei Liu
6. Hongwei Wang
7. B. Liu
8. Zhiyuan Chen

In Recommender Systems

1. Xiangnan He
2. Xing Xie
3. Yongfeng Zhang
4. Xiang Wang
5. Xu Chen
6. Tat-seng Chua
7. Hongwei Wang
8. Yi Tay

Citations in the corpus

• Knowledge Graph Convolutional Networks for Recommender Systems (2019, background): “With the advance of Internet technology, people can access a vast amount of online content, such as news [25], movies [5], and commodities [26].”
• DKN: Deep Knowledge-Aware Network for News Recommendation (2018, background): “First, unlike other items such as movies [9] Figure 1: Illustration of two pieces of news connected through knowledge entities. and restaurants [12], news articles are highly time-sensitive and their relevance expires quickly within a short period (see Section 5.1).”
• Joint Topic-Semantic-aware Social Recommendation for Online Voting (2017, neutral): “LDA is widely used in sentiment analysis [20], aspects and opinions mining [33], and recommendation [5].”
• Document Modeling with Gated Recurrent Neural Network for Sentiment Classification (2015, methodology): “Specifically, we use one movie review dataset from IMDB (Diao et al., 2014) and three restaurant review datasets from Yelp Dataset Challenge in 2013, 2014 and 2015.”
• Is Attention Interpretable? (2019, methodology): “We note that our IMDB data and Yelp data are different sets of reviews from those used by Yang et al. (2016), so our reported performances are not directly comparable to theirs.”
• Mining Aspect-Specific Opinion using a Holistic Lifelong Topic Model (2016, background): “There are also many topic models that have been used for the task of aspect extraction and categorization in, for example, [4, 11, 24, 28, 49, 27, 32, 33, 43, 10, 5, 14, 22, 23, 30, 45, 46].”
• Latent Relational Metric Learning via Memory-based Attention for Collaborative Ranking (2017, unknown): “IMDb that was introduced in [4].”
• Regularizing Neural Networks via Adversarial Model Perturbation (2020, background): “The IMDB dataset (Diao et al. 2014) contains movie reviews labelled by user ratings (from 1 to 10), which has 67426 sentences for training and 9112 sentences for testing.”
• Toward Controlled Generation of Text (2017, methodology): “We use a large IMDB text corpus (Diao et al., 2014) for training the generative models.”
• A Capsule Network for Recommendation and Explaining What You Like and Dislike (2019, methodology): “matically infer explainable aspects from reviews by devising an internal component. JMARS learns the aspect representations of user and item by using topic modeling to conduct collaborative filtering [8]. AFLM proposes an aspect-aware topic model over the reviews to learn diferent aspects of user an”
• Explainable Recommendation via Multi-Task Learning in Opinionated Text Data (2018, methodology): “A follow-up work [7] introduces aspect-level topic modeling to capture users’ finer-grained sentiment on different aspects of an item, so that aspect-level explanations become possible.”
• Hierarchical Attention Networks for Document Classification (2016, unknown): “IMDB reviews are obtained from (Diao et al., 2014).”
• Aspect-Aware Latent Factor Model: Rating Prediction with Ratings and Reviews (2018, methodology): “These works analyze user’s sentiments on items in reviews to boost the rating prediction performance, such as [12, 27, 28, 42].”
• A Survey on Text Classification: From Shallow to Deep Learning (2020, methodology): “The IMDB review [183] is developed for binary sentiment classification of film reviews with the same amount in each class.”
• Personalized Reason Generation for Explainable Song Recommendation (2019, unknown): “They provide recommendations by matching user preference with the available item content features [9, 10, 27, 34, 45–48].”
• Joint Representation Learning for Top-N Recommendation with Heterogeneous Information Sources (2017, methodology): “Furthermore, [13, 39] proposed leveraging probabilistic graphical models to include more ƒexible prior knowledge for review modeling.”
• Why I like it: multi-task learning for recommendation and explanation (2018, unknown): “(iii) CTR : Collaborative Topic Regression (CTR) [43] learns inter-pretable latent structure from user generated content so as to integrate probabilistic topic modeling into CF. (iv) JMARS : Jointly Modeling Aspects, Ratings, and Sentiments (JMARS) [7] is a state-of-the-art recommendation model that”