Sentiment Analysis on Twitter
The sentiment analysis on twitter messages are based on these four steps
ACCESS: Access the twitter API.Poling limit 100-150 in single fetch. Use the API Authentication. I used the PHP authentication scheme
COLLECT: Streaming the API, but the problem was duplication of data.Solution: With the use of Since ID we schedule the data and getting every minute unique result
ANALYZE: We analyze the data using the MySqlâ€™s Natural search.We clean the data and remove all symbols and character and make the data in Alpha numeric format.And then use Solr to analyze the count of positive and negative words on the basis of our algorithms.
STORE: Finally update the polarity of the data in the database where all the tweets are saved. And generate the graph