Semantic Extraction of Named Entities from Bank Wire Text.

Online transactions have increased dramatically over the years due to rapid growth in digital innovation. These transactions are anonymous therefore user provide some details for identification. These comments contain information about entities involved and transfer details which are used for log analysis later. Log analysis can be used for fraud analytics and detect money laundering activities. In this paper, we discuss the challenges of entity extraction from such kind of data. We briefly explain what wired text is, what are the challenges and why semantic information is required for entity extraction. We explore why traditional IE approaches are in-su cient to solve the problem. We tested the approach with available open source tools for Entity extraction and describe how our approach is able to solve the problem of entity identification.