Adding Domain Knowledge In Terms Of Pair-wise Instance Level Constraints, In Clustering, To Refine Road Map
Prarthana A. Deshkar · IOSR Journal of Engineering · 2012
Clustering algorithms, which automatically divide a data set into meaningful sub-groups, have been especially successful in both areas, in the corporate world to extract useful information from large database, as well as in academic research.In some cases information about the problem domain is available in addition to the data instances themselves.This information can be added in the clustering algorithm to make the algorithm domain specific.Domain knowledge should be added in algorithm because, any clustering algorithm can detect general trends and patterns in data, but they cannot make use of additional knowledge specific to the problem at hand, as a human expert can.In this paper we show how instance level constraints are added to make general K-Means algorithm more intelligent.For implementation purpose we have taken the GPS traces, which after implementation can represent road lanes.