Boosting the Accuracy of Optimisation Chatbot by Random Forest With Halving Grid Search Hyperparameter Tuning

Computer science, engineering and technologies are witnessing a vital role in providing challenging demands of users. Artificial intelligence, machine learning and robotic process automation strive to improve the intelligent behavior of computers. Fast human like responses of text chatbot can perform better if and only if it is optimized. Hyper parameter optimization methods are popular for successfully boosting up the overall performance of model. In this paper we focus on creating chatbot using random forest and optimizing its performance by hyper parameter tuning halving grid search. We propose chatbot model 1 without optimization, chatbot model 2 with optimization and chatbot model3 with optimization and best values of key performance indicators. Computations are performed before optimization and after optimization for measurement factors including accuracy, precision, recall and f1-scores. Three different models proposed, and performance are compared for each model with respect to precision, recall, f1-scores and accuracy.