Frame Work to Classify Data in Interactive System to Enhance Decision Making

Abstract

Decision-making is a process of choosing among alternative courses of action for the purpose of attaining a goal or goals. The ultimate objective of data analytics is to ease the decision making process but this has lot of challenges and proper planning is the only way to overcome. The idea behind this research work is to propose a novel framework for data analytics to make effective decisions in an organization by aiding the various stages of the decision making mechanism. According to the design science methodology, the research has been formulated and used in the frame work design process. A novel framework was proposed that combines different aspects of data analytics, needed architectures and tools are incorporated in the various stages of decision making process. Based on the Simons the decision-making process, a new framework was designed with 4 phases namely Data, analytical, model deployment and visualization. The decisive objective of the proposed framework is to ease the process of decision making and also to take effective decision.

In the process of future planning by the organization, it needs simple accurate estimation techniques for predictions to make effective decisions. Predictions always deal with the future events based on past incidents or records. Different kinds of predictions have been done regularly in many fields for the benefit of an individual, a group of people, an organization or a country. Support Vector Machines can be used to create a powerful prediction model because of its capability in classification and regression. The purpose of this research work is to develop a decision support system model was developed using novel algorithm. The newly developed framework has been proposed for the purpose of data analytics and for prediction.  In this work, the machine learning algorithms Support Vector Machines (SVM), Random Forest, Decision Tree, Naïve Bayes and the newly proposed algorithm has been analyzed and results are compared. The outcome of this research work proves that the proposed framework model provides better result than other model.

The objective of designing and developing the proposed framework is to ease the process of decision making in the scenario of interactive system. A student performance assessment is used to evaluate the proposed framework using real data. To test the correctness of proposed framework, an experiment was done with the student data to predict student performance using newly proposed machine learning framework. The result justifies the proposed framework for the decision making process, gives added value.