|International Journal of Computer Applications
|Foundation of Computer Science (FCS), NY, USA
|Volume 155 - Number 6
|Year of Publication: 2016
|Authors: Padala S. Venkata Durga Gayatri, Archana Raghuvamshi
Padala S. Venkata Durga Gayatri, Archana Raghuvamshi . Analysis of Opinion Mining on Social Media Data Streams using Hadoop. International Journal of Computer Applications. 155, 6 ( Dec 2016), 45-49. DOI=10.5120/ijca2016912336
Twitter is a social networking site in which the data to be processed is in rich amounts and which can be structured, semi-structured and unstructured data streams. Opinion mining over the Twitter offers organizations a fast and effective way to monitor the feelings of public towards their services. It focuses on predicting the polarity of words and then classifies them into positive and negative feelings with the aim of identifying attitude and opinions that are expressed in any form or language. Bian et al.’s method (2012) annotated the twitter corpus which was focused on Adverse Drug Reaction (ADR) which includes the broad pharmacological coverage. Bingwei et al.’s method ( 2013) evaluates the scalability of Naive Bayes classifier (NBC) in large datasets instead of using the standard library. Skuza et al.’s method (2015) estimated the future stock prices by calculating in distributed environment according to Map Reduce programming model. Mohit et al.’s method, (2014) explains how the Map – Reduce paradigm can be applied to existing Naïve Bayes algorithm to handle a large number of tweets. All these approaches say about the real-world data sets at its accuracy level by using Hadoop File System. This paper analyses all the above methods comparatively.