For Project Enquiry +91 9962 067 067

ACTIVE LEARNING OF CONSTRAINTS FOR

SEMI-SUPERVISED CLUSTERING

ABSTRACT

For clustering biomedical documents, we can consider three different types of information: the local-content (LC) information from documents, the global-content (GC) information from the whole document collections, and the medical subject heading (MeSH)-semantic (MS) information. Previous methods for clustering biomedical documents are not necessarily effective for integrating different types of information, by which only one or two types of information have been used. Recently, the performance of document clustering has been enhanced by linearly combining both the LC and MS information. However, the simple linear combination could be ineffective because of the limitation of the representation space for combining different types of information (similarities) with different reliability. Semi supervised clustering to select a pair-wise must link and cannot link constraints. We consider active learning as an iteration process which means, an each iteration queries are selected based on the current clustering solution and existing constraint set. We apply a general framework that builds on the concept of neighborhood, where neighborhoods contain “labeled examples” of different clusters according to the pair-wise constraints. Here the active learning method expands the neighborhoods by selecting informative points and querying their relationship with neighborhoods. To resolve uncertainty problem and to select queries that have a highest information rate. We evaluate the proposed method on the benchmark data sets and the results demonstrate consistent and substantial improvements over the current state of the art.