ANALYSIS ON TWITTER DATA OF AUTOMOBILE DOMAIN USING ONTOLOGY

Corpus Data refers to as a collection of huge datasets. Sentiment analysis contributed to a popular research area for twitter. The sentiment analysis done without feature extraction fails to give the deep result about the users opinion but, features of the domain are extracted by building ontology which helps in getting the refined sentiment analysis. Ontology means a formal, explicit specification of a shared conceptualization. Conceptualization refers to an abstract model of some world phenomena. In this paper, we have used ontology to analyse the tweets to increase augmentation and efficiency of sentiments which is obtained using naïve Bayesian algorithm. The work is done in the following stages. In first stage, the tweets are extracted from Twitter4J and stored in a repository. Then sentences are extracted one by one. Sentences extracted are simplified by removing stop words and redundant words. In Second stage, the words left in the sentences are used for sense matching using WordNet-an online semantic dictionary. WordNet dictionary is used to extract features from tweets. In Third stage, Ontology is being generated by using java customized code. Crawler is being designed next to get the details about the automobile domain. The data is stored in text manner. In fourth stage, Mapping of data is done which includes mapping of ontology with the crawler data, together with ontology validation. In fifth stage, Analysis of tweets is done using ontology by applying naïve Bayesian algorithm and comparison of automobile is done which one is better and what all are the attributes that other automobile does not fall into this category.