Message Classification for Breast Cancer Chatbot using Bidirectional LSTM
Pariwat Maktapwong, Pichathat Siriphornphokha, Supawadee Tubglam, Aurawan Imsombut · 2022
This study proposed a chatbot application for breast cancer patients in Thailand. This is to increase the communication channels and address the problem related to a shortage of medical staffs. The first step in the chatbot system is message classification. Natural language processing techniques such as word segmentation, stop words removal, word indexing, and word vectorization are used for the text pre-processing steps. After that, a deep learning technique with Bi-LSTM is used for text classification. There are 1,139 sample messages for learning which can be classified into 60 classes. The results from the experiment showed that the accuracy rate of the proposed technique is 0.869.