Imparting Data Knowledge in discrete data volumes using crowded agent approach for multi-perspective and visualized big data

Abstract

This Master’s dissertation is description of methodology implemented for Imparting Data Knowledge in discrete data volumes using crowded agent approach for multi-perspective and visualized big data. The project has been simulated and developed on the business as well as technical needs acquired from the business sector. The end product of the project development is a simulation and reinforced learning for general purpose artificially intelligent software. The algorithm is adaptive and learns over time with different cases and scenarios presented forward to it. The system was developed and computed in three major phases mainly composed of requirement elicitation & understanding, knowledge representation and knowledge casting and finally agent based application development. The modern world is focused on two areas at the moment i.e. the amount of data that is incurred from time phases and also the relevancy or usage of information for benefits. The benefits than further classify into human, corporate, social benefits etc. the modern world is faced with the issues and concerns of business intelligence. Methodologies and techniques have been developed to facilitate the process of business analysis and comprehension. One such scientific field is focused on achieving the intelligent data before it can be utilized for intelligent analysis. The current size of information is huge and the tasks aimed out of analysis present a complex situation. These perceptions can be handled by using the right and optimal techniques from artificial intelligence. This dissertation is focused on achieving multi-agent perspective architecture for using data rawness and discrepancies to turn them into data intelligence and opportunities. The MAS technique has been used to generate faster data processing and for imparting data with knowledge of its own. The three different agents function independently on here different data sets using reinforcement learning and selective reinforcement to attain high levels of self-actualization in the system. There were three domain datasets that were attained to accomplish the tasks for reinforcement. The first dataset was classified for gaming and associated features. The second dataset was from MIT’s Sloan’s publicly available dataset. The third dataset constituted from the AWS publicly available accounting data set. The first agent performs approximately to 91% accuracy as viii compared to the traditional models of classification. The second agent functions on the selective reinforcement and produces accuracy of over 90% for unlabeled and unstructured data. The application has been developed by using Microsoft Visual Studio 2013. The programming language used is c#. The development was focused on developing an application that is upward compatible. Results and analysis have been submitted as a part of project. The framework used is .NET 4.