An Effective Fuzzy Clustering of Crime Reports Embedded by a Universal Sentence Encoder Model

Crime reports clustering is crucial for identifying and preventing criminal activities that frequently happened in society. In the proposed work, named entities a report are recognized to extract crime-related phrases subsequently, preprocessed by applying stopword removal lemmatization operations. Next, module of universal encoder model, called transformer, applied get sentence embedding each associated sentence, aggregation which finally provides vector representation report. An innovative efficient graph-based algorithm consisting splitting merging operations has been cluster crime reports. The generates overlapping clusters, indicates existence multiple types. fuzzy theory used provide score expressing its membership into different accordingly, labelled categories. efficiency method assessed taking account datasets comparing them with other state-of-the-art approaches help various performance measure metrics.