An intelligent decision support system for machine learning algorithms recommendation

Abstract

Machine learning is a very central topic in Artificial Intelligence and even computer science in general. Nowadays, its use in Big Data problems is quite well known. However, while the big data, and machine learning problems in general, are quite varied and in needing of different kinds of solutions, there are as well many different methods in machine learning that can be used. In this work, we propose an application that might help deciding on which machine learning methods a user needs for a specified problem. The application is an Intelligent Decision Support System for Machine Learning Algorithm Recommendation for which we present the design, which is centered around the combined use of the Case-Based Reasoning and RuleBased Reasoning, for the recommending process, while also trying to make the system easy to use and manage. We present a prototype of such a system, and the implementation details of the two recommender algorithms. The preliminary testing of the prototype shows it to be a promising tool.