Twitilyzer: Designing an approach for ad-hoc search engine

There are many micro blogging sites which provides rich source of information about personality, products, sports, politics, technology etc. As of February 2014, about 241 million tweets are being generated per day. In this paper Twitilyzer, Twitter, a micro blogging site has been used to gather the...

Full description

Saved in:
Bibliographic Details
Main Authors: Kanakia, Harshil T., Kalbande, Dhananjay R.
Format: Conference Proceeding
Language:English
Subjects:
Algorithm design and analysis
Classification algorithms
Conferences
Feature extraction
Machine learning algorithms
Naive Bayes classifier
Sentiment analysis
TextBlob
Top tweets classification
Twitter
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:There are many micro blogging sites which provides rich source of information about personality, products, sports, politics, technology etc. As of February 2014, about 241 million tweets are being generated per day. In this paper Twitilyzer, Twitter, a micro blogging site has been used to gather the information on different topics. The information will be in the form of tweets. These tweets are extracted from the Twitter through Twitter API. These tweets are then preprocessed to remove mistakes. The Naive Bayes machine learning algorithm is developed to classify the tweets into positive and negative sentiments. The Ranking algorithm is developed which will display top positive and top negative sentiments about the particular topic. The experimental result shows statistics of positive and negative sentiments. Thus it can contribute to organization for analysis of current market trends.
DOI:10.1109/ICCICT.2015.7045716