Real Time Analysis of Twitter Data

Abstract

In last two decades, big data has emerged as one of the most trending fields of research. In today’s world, every minute, huge amount of structured and unstructured data is produced via various platforms. This data includes texts, images, and forums gathered from social media platforms. This increasing data implies a fundamental change in the way we analyze data since it is unable to analyze data using conventional methodologies and therefore, we need more advanced technologies and methodologies. There is increasing interest in business society to draw conclusions about the diverse type of data available to it from consumers. By proper analysis of users’ opinions business professionals can identify a trend, a problem or an opportunity, microsecond before anyone else, which can give them an edge over others in today’s competitive environment. One of the best methods to get valuable information is to extract sentiments and opinions from the text. Sentiment analytics aim at analyzing sentiments, emotions, attitudes, opinions etc., towards topics, products, people, organizations, services, etc. Nowadays data is evolving at a rapid pace, thus, there is a growing demand for analysis of big data in real-time. This paper aims at real-time sentiment analytics of twitter stream. More than 340 millions of tweets are generated daily, and this number is increasing rapidly, thus, there arises a demand for a much robust system to process these tweets in real-time. Here apache storm is being used for real-time sentiment analytics. The tweets are first parsed and then classified, followed by grouping.

Keywords: Big data, sentiment, analytics, real-time, apache storm, parsing, spout, bolt, topology

Cite this Article

Rishu Chaudhary, Anurag Singhbaghel. Real Time Analysis of Twitter Data. Current Trends in Information Technology. 2016; 6(1): 6–14p.