GTAI: Geo-centric Technique for Advertisement and Improvisation using Sentiment Analysis

Abstract

The exponentially exceeding internet based user content has been taken as a call by researchers, to mine this gigantic form of user information. This kind of information has been utilized for new inventions. Especially, Social Media(SM) specific information has exploded enormously in the present era of internet. The user specific information in the form of structured and unstructured data has helped in getting familiarized with likes, dislikes of a particular user community. Moreover, it has also helped in providing knowledge regarding the prevalent trends and fashion in a particular area. Sentiment Analysis (SA) as a tool has been come to a rescue in identifying user sentiments. It has also assisted in outlying and analyzing sentiments on the basis of their polarity (i.e. on positive, negative and neutral scale) for any topic which come under experimentation. Many previous studies have been conducted analysing user reviews on the basis of polarity identification and feature extraction. However, only few studies surfaced which dealt with Sentiment Analysis from the point of view of Advertisements(Ads). Also, limited work has been done from the context of analysing sentiments on the basis of geolocations of particular user community. The objective of the present thesis is to target spatially distributed web users worldwide by means of Geocentric Technique for Advertisement and Improvisation (GTAI) .This approach is indented to be used for marketing and promoting various products companies. Furthermore, it can be used to recommend different advertisement agencies about different products which can be promoted or advertised in particular geographical region. It is an effective geo-centric strategy, which works in four phases: Extracting user opinions from social media based upon their geo-locations, Pre-processing of textual user information, analysing the user sentiments using machine learning algorithm, and taking decisions by the help of rule base. The strategy presented in the paper is further demonstrated by using Facebook page. In the initial phase of experimental setup of GTAI, User reviews are collected from Facebook page and after analysing sentiments and exploring there geographical-regions using Top-k rule and 50-50% rule , then Face book page is further used as a test bed for validating the proposed technique on 3 parameters i.e. Audience Building, Consistency, Maintaining Goodwill.