An aggregation approach to the classification problem using multiple prediction experts

摘要

A general problem of combining the classifications or predictions of n local information systems S1, S2, …, Sn into a single system S is considered in this paper. We introduce a new approach for solving an aggregation problem where a decision maker needs to classify an observation based on group-membership predictions coming from multiple experts. We empirically test our approach along with two alternative approaches found in the literature using real-world data and four popular classification techniques. The results show that our approach produced the best performance according to certain performance measures. Researchers may use this method to create a knowledge base of information for a distributed expert system or one that acquires prediction information from the Internet.