Application of NLP on Big Data Using Hadoop: Case Study Using Trouble Tickets

Telecommunication company trouble tickets system contains reported incidents tickets that related to networks service interruption or problems. Trouble tickets system is the example of the application that deals with a large amount of textual data. In order for the team that is handling the trouble tickets system, they spend a lot of time to analyze the data manually. In this paper, Natural Language Processing (NLP) approach has been introduced to solve the problem. By applying this technique, manual of activities can be automated and reduce the time and effort to find the classification of the closing resolution code. It also helps the trouble tickets system to collect incidents trending and resource utilization. New data processing method with NLP and sublanguage introduced via big data platform to deliver faster classification computation. The outcome of this study is transformation method of the original data set into the analytics series, and identification the characteristics of the trouble tickets data set to enable the classification of the resolution code. The data processing workflow shows that the linguistics of the trouble tickets fit the sublanguage theoretical framework thus enabling to tap into the unrealized value inside it. The data processing and data transformation workflow describe the linguistics of the trouble tickets fit the sublanguage theoretical framework, therefore, supporting the research to tap into the unexposed content inside it.