SOME DEFINITIONS
Object
An object o is a product, person, event, organization, or topic. o is represented as
a hierarchy of components, sub-components, and so on. Each node represents a component and is associated with a set of attributes of the component.
OPINION
An opinion is a quintuple (o, f, so, h, t), where
o
is a target object. f is a feature of the object o. so is the sentiment value of the opinion of the opinion holder h on feature f of object o at time t. h is an opinion holder. t is the time when the opinion is expressed
OPINION MINING AND SENTIMENT ANALYSIS
Given a set of evaluative text documents D that contain opinions (or sentiments) about an object, opinion mining aims to extract attributes and components of the object that have been commented on in each document d ∈ D and to determine whether the comments are positive, negative or neutral. Sentiment analysis is a synonymous term.
TRY THESE!!
“If you are reading this because it is your darling fragrance, please wear it at home exclusively, and tape the windows shut.” “Jane Austen‟s books madden me so that I can‟t conceal my frenzy from the reader. Every time I read „Pride and Prejudice‟ I want to dig her up and beat her over the skull with her own shin-bone.” - Mark Twain
SOME CONCEPTS
Sentiment polarity and degrees of positivity The binary classification task of labeling an opinionated document as expressing either an overall positive or an overall negative opinion Joint topic-sentiment analysis Term presence vs. frequency Presence is more important than frequency
SOME CONCEPTS
Parts of speech Adjectives, adverbs, nouns, verbs Negation “I like this book” and “I don‟t like this book” Similar but opposite! „No‟ does‟nt imply a negation. “No wonder it is one of the best”
A FEW APPROACHES TO THE PROBLEM
WORDNET
WORDNET BASED APPROACH
Determine sentiments of adjectives in WordNet by measuring relative distance of the term from exemplars, such as “good” and “bad”.
The polarity orientation of a term t is measured as follows O(t) = [d(t, good) − d(t, bad)] / d(good, bad) where d() is a WordNet based relatedness measure
GRAPH BASED METHODS
Mincuts
A mincut of a weighted graph G(V,E) is a partitioning the vertices V into V1 and V2 such that sum of the edge weights of all edges between V1 and V2 is minimal
GRAPH BASED METHODS
Classify datapoints by partitioning the similarity graph such that it minimizes the number of similar points being labeled differently
REFERENCES
Opinion mining and sentiment analysis , Bo Pang and Lillian Lee. Foundations and Trends in Information Retrieval Vol. 2, No 1-2 (2008) 1– 135 Ahmed Abbasi. Affect intensity analysis of dark web forums. In Proceedings of Intelligence and Security Informatics (ISI), pages 282–288, 2007. Lada A. Adamic and Natalie Glance. The political blogosphere and the 2004 U.S. election: Divided they blog. In Proceedings of LinkKDD, 2005. Alekh Agarwal and Pushpak Bhattacharyya. Sentiment analysis: A new approach for effective use of linguistic knowledge and exploiting similarities in a set of documents to be classified. In Proceedings of the International Conference on Natural Language Processing (ICON), 2005. Rakesh Agrawal, Sridhar Rajagopalan, Ramakrishnan Srikant, and Yirong Xu. Mining newsgroups using networks arising from social behavior. In Proceedings of WWW, pages 529–535, 2003. Edoardo M. Airoldi, Xue Bai, and Rema Padman. Markov blankets and metaheuristic search: Sentiment extraction from unstructured text. Lecture Notes in Computer Science, 3932 (Advances in Web Mining and Web Usage Analysis):167–187, 2006.
