Sentiment Analysis and Opinion Mining

1 Sentiment Analysis and Opinion Mining April 22, 2012 Bing Liu Draft: Due to copyediting, the published version is slightly different Bing Liu. Sentiment Analysis and Opinion Mining , Morgan & Claypool Publishers, May 2012. Sentiment Analysis and Opinion Mining 2 Table of Contents Preface ..5 Sentiment Analysis : A Fascinating Problem ..7 Sentiment Analysis Applications ..8 Sentiment Analysis Research ..10 Different Levels of Analysis .. 10 Sentiment Lexicon and Its Issues .. 12 Natural Language Processing Issues .. 13 Opinion Spam Detection.

2 14 What s Ahead ..14 The Problem of Sentiment Analysis ..16 Problem Definitions ..17 Opinion Defintion .. 17 Sentiment Analysis Tasks .. 21 Opinion Summarization ..24 Different Types of Opinions ..25 Regular and Comparative Opinions .. 25 Explicit and Implicit Opinions .. 26 Subjectivity and Emotion ..27 Author and Reader Standing Point ..29 Summary ..29 Document Sentiment Classification ..30 Sentiment Classification Using Supervised Learning ..31 Sentiment Classification Using Unsupervised Learning ..34 Sentiment Rating Prediction ..36 Cross-Domain Sentiment Classification ..38 Cross-Language Sentiment Classification ..41 Summary ..43 Sentence Subjectivity and Sentiment Classification ..44 Sentiment Analysis and Opinion Mining 3 Subectivity Classification.

3 45 Sentence Sentiment Classification ..49 Dealing with Conditional Sentences ..51 Dealing with Sarcastic Sentences ..52 Cross-language Subjectivity and Sentiment Classification .53 Using Discourse Information for Sentiment Classification 55 Summary ..56 Aspect-based Sentiment Analysis ..58 Aspect Sentiment Classification ..59 Basic Rules of Opinions and Compositional Semantics ..62 Aspect Extraction ..67 Finding Frequent Nouns and Noun 68 Using Opinion and Target Relations .. 71 Using Supervised 71 Using Topic Models .. 73 Mapping Implicit Aspects .. 77 Identifying Resource Usage Aspect ..78 Simutaneous Opinion Lexicon Expansion and Aspect Extraction ..79 Grouping Aspects into Categories ..81 Entity, Opinion Holder and Time Extraction.

4 84 Coreference Resolution and Word Sense Disambiguation .86 Summary ..88 Sentiment Lexicon Generation ..90 Dictionary-based Approach ..91 Corpus-based Approach ..95 Desirable and Undesirable Facts ..99 Summary ..100 Opinion Summarization ..102 Aspect-based Opinion Summarization ..102 Improvements to Aspect-based Opinion Summarization ..105 Contrastive View Summarization ..107 Traditional Summarization ..108 Summary ..108 Sentiment Analysis and Opinion Mining 4 Analysis of Comparative Opinions ..110 Problem Definitions ..110 Identify Comparative Sentences ..113 Identifying Preferred Entities ..115 Summary ..117 Opinion Search and Retrieval ..118 Web Search vs. Opinion Search ..118 Existing Opinion Retrieval Techniques.

5 119 Summary ..122 Opinion Spam Detection ..123 Types of Spam and Spamming ..124 Harmful Fake Reviews .. 125 Individual and Group Spamming .. 125 Types of Data, Features and Detection .. 126 Supervised Spam Detection ..127 Unsupervised Spam Detection ..130 Spam Detection based on Atypical Behaviors .. 130 Spam Detection Using Review Graph .. 133 Group Spam Detection ..134 Summary ..135 Quality of Reviews ..136 Quality as Regression Problem ..136 Other Methods ..138 Summary ..140 Concluding Remarks ..141 Bibliography ..143 Sentiment Analysis and Opinion Mining 5 Preface Opinions are central to almost all human activities and are key influencers of our behaviors. Our beliefs and perceptions of reality, and the choices we make, are, to a considerable degree, conditioned upon how others see and evaluate the world.

6 For this reason, when we need to make a decision we often seek out the opinions of others. This is not only true for individuals but also true for organizations. Opinions and its related concepts such as sentiments, evaluations, attitudes, and emotions are the subjects of study of Sentiment Analysis and Opinion Mining . The inception and rapid growth of the field coincide with those of the social media on the Web, , reviews, forum discussions, blogs, micro-blogs, Twitter, and social networks, because for the first time in human history, we have a huge volume of opinionated data recorded in digital forms. Since early 2000, Sentiment Analysis has grown to be one of the most active research areas in natural language processing.

7 It is also widely studied in data Mining , Web Mining , and text Mining . In fact, it has spread from computer science to management sciences and social sciences due to its importance to business and society as a whole. In recent years, industrial activities surrounding Sentiment Analysis have also thrived. Numerous startups have emerged. Many large corporations have built their own in-house capabilities. Sentiment Analysis systems have found their applications in almost every business and social domain. The goal of this book is to give an in-depth introduction to this fascinating problem and to present a comprehensive survey of all important research topics and the latest developments in the field.

8 As evidence of that, this book covers more than 400 references from all major conferences and journals. Although the field deals with the natural language text, which is often considered the unstructured data, this book takes a structured approach in introducing the problem with the aim of bridging the unstructured and structured worlds and facilitating qualitative and quantitative Analysis of opinions. This is crucial for practical applications. In this book, I first define the problem in order to provide an abstraction or structure to the problem. From the abstraction, we will naturally see its key sub-problems. The subsequent chapters discuss the existing techniques for solving these sub-problems.

9 This book is suitable for students, researchers, and practitioners who are interested in social media Analysis in general and Sentiment Analysis in particular. Lecturers can readily use it in class for courses on natural Sentiment Analysis and Opinion Mining 6 language processing, social media Analysis , text Mining , and data Mining . Lecture slides are also available online. Acknowledgements I would like to thank my former and current students Zhiyuan Chen, Xiaowen Ding, Geli Fei, Murthy Ganapathibhotla, Minqing Hu, Nitin Jindal, Huayi Li, Arjun Mukherjee, Guang Qiu (visiting student from Zhejiang University), William Underwood, Andrea Vaccari, Zhongwu Zhai (visiting student from Tsinghua University), and Lei Zhang for contributing numerous research ideas over the years.

10 Discussions with many researchers also helped shape the book: Malu G. Castellanos, Dennis Chong, Umesh Dayal, Eduard Dragut, Riddhiman Ghosh, Natalie Glance, Meichun Hsu, Jing Jiang, Birgit K nig, Xiaoli Li, Tieyun Qian, Gang Xu, Philip S. Yu, Clement Yu, and ChengXiang Zhai. I am also very grateful to two anonymous reviewers. Despite their busy schedules, they read the book very carefully and gave me many excellent suggestions. I have taken each and every one of them into consideration while improving this book. On the publication side, I thank the Editor, Dr. Graeme Hirst, and the President and CEO of Morgan & Claypool Publishers, Mr. Michael Morgan, who have managed to get everything done on time and provided me with many pieces of valuable advice.