The origin and primary areas of application of natural language processing

• Main Authors: Krutilla, Zsolt, Kovari, Attila
• Format: Conference Proceeding
• Language: English
• Subjects: Artificial Intelligence; Banking; Computational modeling; Computers; Decision making; Deep learning; Insurance; Law; Natural Language Processing
• ISSN: 2471-9269
• DOI: 10.1109/CINTI-MACRo57952.2022.10029432

Natural language processing with computers, i.e. NLP, is increasingly gaining ground in today's IT. The purpose of this study is to analyze the articles published in the NLP topic and to summarize it from several aspects, especially with regards to the development and practical application of the technology. The idea of NLP as a technology was not formulated in the 21st century, but already in the 1950s, relevant research was carried out and theories were created regarding how a machine can become capable of interpreting human speech. Technology has come a long way from simple algorithms to Deep Learning, but it has not yet solved the problem of machine processing of natural language. The research has confirmed the idea that mathematical modelling and models based on statistical probability dominated the field in the early days of the discipline and before the advent of computers. However, despite technological progress, no solution has yet been found to the comprehensive and all-encompassing problem of natural language processing with computers. To determine the future research area of NLP, we need to take into account the text files on which the technology is based. From this perspective, we can focus on the insurance and banking sectors, where a series of administrative processes take place where centralized paper processing is the incoming data set of the process and a database going back several years may be available.