Abstract. In this paper we address the problem of acquiring knowledge by integration. Our aim is to construct an integrated knowledge base from several separate sources. The objective of integration is to construct one system that exploits all the knowledge that is available and has good performance. The aim of this paper is to discuss the methodology of knowledge integration and present some concrete results. In our experiments the performance of the integrated theory exceeded the performance of the individual theories by quite a significant amount. Also, the performance did not fluctuate much when the experiments were repeated. These results indicate knowledge integration can complement other existing ML methods.
5 500 words