Improving service quality through classifying chatbot messages based on natural language processing: A bidirectional long short-term memory network model

Improving service quality through classifying chatbot messages based on natural language processing: A bidirectional long short-term memory network model

The rapid development of the times and the influence of globalization have enormously changed human life. One of the affected fields is service, and machines are gradually replacing services. Not without reason, the number of bad assessments of services is one of the factors why machines begin to replace the role of humans. The existence of machines also makes it easier for companies to provide services and help cut costs for labor. The machine used in this research was a chatbot. Chatbot is a computer program designed to simulate conversations between humans. The long short- term memory network (LSTM) algorithm was implemented on the chatbot with a natural language processing (NLP) approach in this research. Our experiment was carried out using the NLP approach, where the results were used in the data training process using the bidirectional LSTM algorithm to produce a chatbot model. Next, after evaluating the model, our proposed method outperformed other models in the experiment. Bidirectional LSTM had 98.09% accuracy, 98.23% precision, 98.29% recall, and 98.25% f1 score.