• Aaquib Multani
• Atul Agrawal

Data mining techniques are used for analyzing the data and recovering the valuable information from the data. There two basic techniques of data analysis available namely supervised and unsupervised learning. The supervised algorithm are the accurate models by which the similar patterns are trained on the initial samples and accurate class labels can be identifiable for raw samples that not contains the class labels. In this presented work the supervised learning technique is used for classifying the unstructured twitter data for finding the emotional class labels. There are two class labels are available in data namely positive classes and negative classes. In order to perform classification the twits are first the pre-processed to make clean. In further the feature extraction technique is used by which the unstructured data is transformed into the structured format. For transformation of data NLP parser is used. The NLP parser is used to obtain the POS (part of speech) information from text and can be used for transforming the data into the 2D vector. This vector is used with the two different classification techniques namely C4.5 decision tree. First the decision tree model is developed using training data. And the developed decision tree is used for classify the test data. Further the data is again classified using the Bayesian classifier that replaces the misclassified data to improve the classification accuracy. The implemented technique is experimented a number of times and it is concluded the performance of the technique improves the classification ability by involving multiple classifiers.

Data Mining, Sentiment Analysis, Twitter, Social Media, Multi-class Classification, Text Mining, Naïve Bayes.

"A HYBRID TEXT CLASSIFIER FOR MINING SENTIMENTS ON TWITTER DATASET", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 5, page no.225-231, May-2019, Available :http://www.jetir.org/papers/JETIR1905436.pdf

"A HYBRID TEXT CLASSIFIER FOR MINING SENTIMENTS ON TWITTER DATASET", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 5, page no. pp225-231, May-2019, Available at : http://www.jetir.org/papers/JETIR1905436.pdf