Sentiment Analysis
- 3. Sentiment Analysis (SA) is the computational study of people’s opinions, attitudes and emotions toward an entity. The entity can represent individuals, events or topics. These topics are most likely to be covered by reviews.
- 4. 3 main classification levels in SA Document-level Sentence-level Aspect-level
- 5. Ok!! but what kind of data will we use for it?
- 6. Product reviews Stock markets News articles Political debates
- 7. How to do it?
- 8. FEATURE SELECTION Terms presence and frequency individual words or word n-grams and their frequency counts. Parts of speech (POS) finding adjectives, as they are important indicators of opinions. Opinion words and phrases words commonly used to express opinions including good or bad, like or hate. Negations the appearance of negative words may change the opinion orientation like not good is equivalent to bad.
- 9. FEATURE SELECTION METHODS Point-wise Mutual Information (PMI): This measure was derived from the information theory . The point-wise mutual information (PMI) Mi(w) between the word w and the class i is defined on the basis of the level of co-occurrence between the class i and word w. Chi-square (χ2): A chi square (X2) statistic is used to investigate whether distributions of categorical variables differ from one another. Latent Semantic Indexing (LSI): Latent semantic indexing adds an important step to the document indexing process. The method examines the document collection as a whole, to see which other documents contain some of those same words. LSI considers documents that have many words in common to be semantically close, and ones with few words in common to be semantically distant.
- 10. SENTIMENT CLASSIFICATION TECHNIQUES Machine learning approach Supervised leaning Unsupervised learning Lexicon based approach Dictionary based approach Corpus based approach
- 11. MACHINE LEARNING Machine learning approach relies on the famous ML algorithms to solve the SA as a regular text classification problem that makes use of syntactic and/or linguistic features. Supervised learning methods depend on the existence of labeled training documents. Probabilistic classifiers Linear classifiers Decision tree classifier Unsupervised learning is a type of machine learning algorithm used to draw inferences from datasets consisting of input data without labeled responses. Example cluster analysis, which is used for exploratory data analysis to find hidden patterns or grouping in data.
- 12. LEXICON-BASED APPROACH Positive opinion words express desired states, while negative opinion words express undesired states. Dictionary-based approach: A small set of opinion words is collected manually with known orientations. Then, this set is grown by searching in the well known corpora WordNet or thesaurus for their synonyms and antonyms. The newly found words are added to the seed list then the next iteration starts. The iterative process stops when no new words are found. Corpus-based approach: The Corpus-based approach helps to solve the problem of finding opinion words with context specific orientations. Its methods depend on syntactic patterns or patterns that occur together along with a seed list of opinion words to find other opinion words in a large corpus.