Study And Design Of Framework For Social Media Data Analytics Based On MultiCriteria Decision Making Mcdm Models

Abstract

In today s digital technology era, it is important for business enterprises to build a competitive benefit to customers through distinguishing services or responsiveness to dynamic business situation. Data or information such as consumer opinions, experiences and sentiments on various product and services produced from various Social Media platform or sites helps to improve value of business organizations. Gathering data from Social Media sites and evaluating that data to make business decisions is called Social Media Analytics (SMA). SMA has been getting importance in the recent times due to 3.5 billion active users presence in Facebook, Instagram, Twitter, LinkedIn etc. Social Media networks produce data through blog posts, likes, and comments that are collated and managed through a Big Data platform such as Apache Hadoop, IBM Cloudera. Business enterprises are expected to understand their customers, competitors or partners holistically with insight trends from both Social Media platforms and enterprise business data platforms such as Oracle Enterprise Resource Planning to make well informed business decisions. Previous research used Standard Centralities and conventional Network parameters to perform Social Media Analytics operations. Also, Not much research is found using Multi-Criteria Decision Making model to perform Social Media Analytics. newline newline newlineOur research work proposes a novel comprehensive framework based on Multi-Criteria Decision Making (MCDM) models to help business organizations effectively leverage Social Media insights along with Enterprise data. The proposed comprehensive framework solution is implemented in three stages. At first, Social Media datasets from Facebook and Twitter has been performed with Social Media Analytics operation using Multi-Criteria Decision Making (MCDM) models. Various MCDM based methods such as Pugh, SDI Matrix, Analytic Hierarchy Process (AHP), Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) has been executed and compared th