Performance Analysis and Development of QnA Chatbot Model Using LSTM in Answering Questions

This research aims to evaluate the performance of a Long Short-Term Memory (LSTM) based chatbot in answering questions (QnA). LSTM is type Recurrent Neural Network (RNN) architecture specifically designed overcome vanishing gradient problems and can store long-term information. The method used 5-fold cross-validation train model with 15 epochs at each fold using dataset provided. results showed variations fold. At 5th fold, there was decrease 84.63% accuracy, 96.36% precision, 64.9% recall, 69.84% loss value. finding shows that variability QnA In conclusion, provide good answers high accuracy precision. Still, need be considered use this chatbot.