Analysis of Socio-Commercial-Daily Living Activities Big Data Using Artificial Intelligence to Design Prediction Models for Smart Living

Smart Living means more than creating intelligent homes and social environment, where people can use devices and information to make life comfortable and convenient. In this competitive world, life is so busy that it is difficult to pay personal attention to the old age parents and children in a home. Life can be made convenient if daily activities of the old age parents and children can be monitored remotely. It can be achieved by automatizing the alerts through efficient analysis of Big Data generated from the Internet of Things (IoT) devices used in the smart homes. Besides, it is highly essential to monitor the social behavior of the relatives, friends and own children to avoid any unwanted situation or complicated relationship, which can improve the social aspects of smart living. Analysis of social media Big Data and mining of social behaviors can help to predict the psychological, personal and unusual behaviors of a person such as suicidal tendency, confronting attitude etc. Besides, smart living of an individual will be incomplete, if someone cannot invest and utilize the hard earned money in buying or selling properties smartly of own interest. However, it is highly difficult, even if impossible to study the huge volume of commercial data available on the Web and to correlate them manually for a judicious decision. Moreover, the data are so diversified and confusing that Artificial Intelligence (AI) is necessary to classify, extract, correlate and predict the most valuable information for the benefits of the people and to improve quality of life. Hence, in these THREE years of the proposal, we would like to analyze the smart home, social and commercial Big Data using different algorithms of Artificial Intelligence and Machine Learning for designing most accurate prediction models. In the 1st Year of the project, real time Big Data generated by IoT devices fitted in smart old age home will be analyzed using AI to recognize the activities of the daily livings. Besides, prediction models for daily activity recognition will be developed to deal with smart communities Big Data. In the 2nd Year, AI will be used to analyze the huge volume of unstructured social media Big Data for predicting psychological behaviors of a person. Each individual’s shared social media Big Data will be used to train the machine for personality analysis and behavioral anomaly recognition. In the 3rd Year, commercial Big Data belonging to Real Estate and Feature based Car pricing will be analyzed for predictive recommendation and predictive analysis, respectively. Besides, customer profiling and sentiment analysis on commercial products will also be done to generate interest-specific recommender systems. The proposed project will be implemented using open source Machine Learning tools such as Sci-Kit Learn and Massive Online Analysis (MOA) using real open data. Processing and analysis of huge volume of data will also be experimented in Apache Hadoop and SPARK platform for batch and streaming data, respectively.

Project ID:PB10708-2348
External Project ID:MOST107-2221-E182-073