Implementation of Natural Language Processing in Customer Service

The use of artificial intelligence and natural language processing (NLP) in customer service is growing quickly. Technology is being used to interact with users and answer their questions. Using NLP in customer services is in the form of artificial intelligence applications that allow users to communicate with models using different languages through text or speech, and the model will provide answers to the users. The main goal of this systematic review is to locate and analyze the existing articles and studies on the use of NLP technology in customer service in terms of research domain, applications, datasets used, and evaluation methods. To create the final review article, relevant papers were sorted and filtered based on inclusion exclusion standards and quality assessment. In addition, discovered that Twitter dataset was the second dataset in terms of the most often used datasets. For the evaluation, most of the researchers used Accuracy, Precision, Recall, and F1 as the methods to evaluate the performance. However, the most important limitation was the dataset because it can be associated to the volume, diversity and quality of the dataset, thus the dataset may have a huge impact on the outcomes.