Twitter Analytics: Project Overview

Home

Project Overview

Project Management

Documentation

Findings

About Me

Project Motivation Twitter, an online social network that allows users to post short text messages (tweets) of up to 140 characters have been gaining popularity in the recent years. The 140 characters restriction has encourages users to construct a focused updated regularly. As of 2013, there are more than 500 million Tweets a day on average. With this huge amount of data, stakeholders will be able to get insight into their area of interest. One of the features in Twitter is the hashtag capability to group relevant information together. Each hashtag in Twitter consists of topics made up of several keywords. By understanding the “buzz” in Twitter and their forecasted movement over time, stakeholders will be able to make better decisions based on the topics discussed. Social Media Analytics Framework Project Objective The social media analysis will be used as Proof of Concept (POC) to further enhance the analytics capabilities of the client’s company. This project plans to explore the various analytics capability that can be done   The solution shall be an independent system and support the following key capabilities: Data Gathering: Building of data collection point by scanning and retrieving data from Twitter Analytics: Analysis of various key components and other relevant information using various customized code Visualization: Displaying key objectives results in various form for better users’ understanding Possible Stakeholders Defence Ministry: Identify and monitor possible terrorism activities in the country Marketing: Improve Branding and Marketing, Tap into New Markets, Gain a competitive advantage Public Figures: Popularity over time Policy Makers: Monitor the public response regarding new policy or possible unhappiness can be minimized through corrective measures Primary Project Scope Secondary Project scope Identifying hot keywords that will be trending in the next course of time. Various intensive researches have been done to analyze the problem. One of the major key findings is done by MIT Researchers, which is able to predict the keyword trend. However, one major constraint is the data collection, as training data is needed to build the model. 500 keywords that became trend and 500 keywords that fail to trends are used in the model. The researches are able to get all the data necessary build the model as they are part of Twitter institute. In order to achieve similar things, the difficulty of getting data from Twitter needs to be overcome. However, since the data collection should not be part of this Analytics Practicum, this feature will be treated as secondary scope after the primary scope is done. References Twitter blog Streamcrab TMCnet Analytics Training MIT Quora Fract AnalyticsHcil2 IBM