Potential Impact Factors of Generative-Based Chatbots versus Retrieval-Based Chatbots: A Study of Adoption in Indian Educational Institutes

In Indian educational institutions, generative-based chatbots have a number of advantages over retrieval-based chatbots. These AI-powered systems are very good at interpreting natural language, which enables them to interpret and react more accurately and sensitively to a variety of student inquiries. This feature not only improves the learning process by offering individualized and nuanced solutions, but it also potentially assists bilingual students and students from a variety of linguistic backgrounds. Although generative chatbots can produce absurd responses and need substantial training, they provide greater flexibility, creativity, and context comprehension. They struggle with scalability but excel at personalizing. Although retrieval-based chatbots are faster to set up and manage, they are less personalized and have trouble keeping up with lengthy context maintenance. As a part of methodology, an organized survey was given to administrators, professors, and students at different Indian educational institutions as part of the quantitative phase. On a scale of 1 to 5, participants evaluated their interactions with retrieval and generative-based chatbots. A score of 5 indicated the greatest level of satisfaction. Each type is appropriate for a distinct set of use scenarios because they are safer in terms of ethical considerations and response quality. In order to justify our study, we create and implement chatbots, one with multiple designing and in other hands with a retrieval and generative basis. The major difference of retrieval-based chatbots are limited to giving the responses to the in-puts that best fit the out-puts they are already familiar with, but generative-based chatbots are able to produce new content.