DEVELOPMENT AND DEPLOYMENT OF A TEXT-BASED INTERACTIVE NATURAL LANGUAGE

ABSTRACT

The aim of this project report is to explore the implementation of a machine learning model using serverless computing architecture. As a new paradigm, serverless computing enables developers to concentrate entirely on the functionality of their code without having to worry about maintaining the supporting infrastructure. In this project, A methodical strategy was used to develop and deploy a machine learning model using serverless computing platforms like AWS Lambda, Render Cloud, and Azure functions. The procedure involved picking an appropriate machine learning method, preparing the data, and identifying an adequate dataset. The model was integrated into a serverless function, allowing for smooth and automatic scaling in response to demand, and the model's efficacy was evaluated using the performance metrics. The results suggest that serverless computing offers a scalable and effective method for implementing machine learning models. The model needed the least amount of operational management while achieving accuracy and inference speed comparable to existing deployment approaches. The model's integration into the serverless function ensured greater functions and improved capabilities. The project's findings aim to help researchers and practitioners in this field gain a clearer grasp of the possible advantages and difficulties of using serverless computing for machine learning activities.