Performance Evaluation of a Natural Language Processing approach applied in White Collar crime investigation

摘要

In today world we are confronted with increasing amounts of information everyday coming from a large variety of sources. People and co-operations areproducing data on a large scale, and since the rise of the internet, e-mail andsocial media the amount of produced data has grown exponentially. From a lawenforcement perspective we have to deal with these huge amounts of data when acriminal investigation is launched against an individual or company. Relevantquestions need to be answered like who committed the crime, who were involved,what happened and on what time, who were communicating and about what? Not onlythe amount of available data to investigate has increased enormously, but alsothe complexity of this data has increased. When these communication patternsneed to be combined with for instance a seized financial administration orcorporate document shares a complex investigation problem arises. Recently,criminal investigators face a huge challenge when evidence of a crime needs tobe found in the Big Data environment where they have to deal with large andcomplex datasets especially in financial and fraud investigations. To tacklethis problem, a financial and fraud investigation unit of a European countryhas developed a new tool named LES that uses Natural Language Processing (NLP)techniques to help criminal investigators handle large amounts of textualinformation in a more efficient and faster way. In this paper, we presentbriefly this tool and we focus on the evaluation its performance in terms ofthe requirements of forensic investigation: speed, smarter and easier forinvestigators. In order to evaluate this LES tool, we use different performancemetrics. We also show experimental results of our evaluation with large andcomplex datasets from real-world application.