Examining the Runtime of NLTK and TensorFlow Algorithms for Chatbot Based on intents.json Length


Author(s): Rishi Hariharaprasad

This research paper aims to explain an empirical investigation concerning the correlation between the length of an intents.json file and the runtime of NLTK (Natural Language Toolkit) and TensorFlow algorithms in Python, tailored towards chatbot development. By examining the runtime efficiency based on various intents.json file sizes, the study aims to ascertain the impact of file length on algorithm performance through the utilization of a least squares regression line. Experimental analysis reveals a compelling linear relationship between intents.json length and training runtime, indicative of an O(N) runtime complexity. This research provides practical implications for developers seeking to enhance the runtime performance of chatbot systems by providing them with a baseline of runtime for dynamic chatbot creation.