Pengembangan Pendekatan Matriks Hubungan untuk Pengukuran Similaritas

Abstract

The similarity calculation is a basic function which is used by a wide range of infotn1ation processing applications such as searching for clustering and classification. There are different types of relationships can be used to reinforce the value of the similarity of data objects in pairs and to help improve the quality of the calculation of similarity between data objects. It is helpful to keep track of how relationships between data objects used to measure the similarity of data objects. Results of this study can showed that the Unified Relationship matrix (URM) can be used to provide heterogeneous data objects and the linkages between each object data in an integrated manner.