Discovery of data models using genetic programming

Abstract

A central challenge and a common goal of AI and machine learning is to get a computer to solve a problem without explicitly programming. The field is known as automatic programming. This paper proposes a solution to address the difficulty of automatically discovering data models in any given data set using Genetic Programming which is a form of automatic programming. The proposed system will lead us to discover previously unknown data models in a data set and help to readily predict the outcomes of a given input data. The concept has been demonstrated by hypothetical data for the agricultural domain.