Approach for Improving Similarity Measure

This book proposes a new method to implement a similarity measures on fuzzylogic for IR. The proposed similarity measures are based on different features of IR system which improves the performance of IR systems. The main strength is to extract most of the features of documents. Fuzzy rules are also designed to make this approach most effective, which transforms the domain knowledge into fuzzy sets. The CACM benchmark dataset is used to validate our proposed similarity measure. The drawback of this approach is that, if more features are used, than the approach will be more complex. This book has significant effort to improve the performance of IR systems.