Description

World’s largest community Facebook’s ‘Like’ button pressed 2.7 billion times every day across the web revealing what people care about, such an impact of social media that internet user average almost spends 2.5 hours daily on liking, chatting, poking, tweeting on social media, which has become vast source of unstructured data. While dealing with big data it’s difficult for traditional databases and architecture to modify, grill and then structure this data, it can lead to many consumer insights which can help to create win-win situations. It has become necessary to find out value from large data sets to show relationships, dependencies as well as to perform predictions of outcomes and behaviors. Big Data has been characterized by 5 Vs – Volume, Velocity, Variety, Veracity and Value. This paper deals with all these 5Vs, features, challenges, future of Big Data in social media arena using data mining algorithms, tools and Hadoop framework for overcoming challenges of Big Data. Data mining detects useful knowledge from very large datasets like trends, patterns and rules. Big Data Analytics captures and analyzes Big Data for discovering interesting patterns and relationships in it. Data mining has predictive and descriptive tasks. Predictive tasks include Classification, Regression and Deviation Detection, Descriptive tasks includes Clustering, Summarization, Association Learning and Sequential Patterns. Various data mining techniques are applied to resolve the different social media issues like influence detection, community or graph detection, expert finding, link prediction, recommender systems, predicting trust and distrust among individuals, behavior and Mood Analysis, opinion mining, Data mining techniques are classified into Unsupervised, Semi-Supervised and Supervised. The overall process of discovering useful knowledge from data is referred as KDD (Knowledge Discovery in Databases) and data mining is one of the steps in KDD process which is used for extracting patterns, models from data with the help of specified algorithms. KDD process is nontrivial process of discovering valid, novel, potentially useful, and ultimately understandable patterns in data. Storing massive amount of data has a value only when useful knowledge is extracted to make decisions, target interesting events and trends on the basis of statistic AL analysis, utilizing the data for achieving business, operational or scientific objectives. Author Rahul Shoran Reno et al. has used knowledge discovery and data mining (KDD) algorithms through the Waikato Environment for Knowledge Analysis (WEKA) interface for automated analysis of historical work instructions. Analyzing large datasets of historical data was a challenging task while creating Decision support systems. Or social media analytics of customer opinions, text analytics and sentiment analytics are frequently adopted.