MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202641109603 A) filed by Ppg Institute Of Technology; Priyadarshini S; Dr. K. Chinnarasu; Thamaraikannan P; Sahana S; V. Vandhana; Mythilipriya K; S Shanthakumari; M. Navadharani; Manoj M; Dr. S. Nandhakumar; and Kaaviyakanth K on September 12, 2026, for An Ai-Powered Conversational Chatbot For College Information Retrieval Using Bert.
Inventors include Ppg Institute Of Technology; Priyadarshini S; Dr. K. Chinnarasu; Thamaraikannan P; Sahana S; V. Vandhana; Mythilipriya K; S Shanthakumari; M. Navadharani; Manoj M; Dr. S. Nandhakumar; and Kaaviyakanth K.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: With the increasing demand for automated assistance in educational institutions, the development of intelligent conversational agents has become an essential tool for improving communication and efficiency. Traditional query-handling methods often involve manual responses, which can be time-consuming and error-prone. In contrast, AI-powered chatbots can provide real-time, accurate answers to a wide range of queries, making them a valuable asset for educational institutions. This project aims to develop a College Chatbot using BERT (Bidirectional Encoder Representations from Transformers), a state-of-the-art pre-trained language model. BERT excels in understanding the context of text by processing sentences bidirectionally, making it highly effective for natural language understanding tasks, including question- answering. The College Chatbot is designed to provide students, faculty, and staff with an easy-to-use interface for obtaining instant information about various college- related topics, such as admissions, academic courses, faculty details, events, and more. The chatbot leverages BERT for natural language processing, enabling it to understand the intent behind user queries and extract relevant entities to generate appropriate responses. The chatbot is built upon the BERT model, which is fine-tuned using a custom dataset containing frequently asked questions (FAQs) and college-specific information. The system employs tokenization and text preprocessing techniques to transform user input into a format that can be understood by the model. It identifies the intent of the user's query and extracts key entities to generate a contextually relevant response. The chatbot's knowledge base is continuously updated by administrators, ensuring that it remains accurate and relevant. The chatbot's ability to understand complex queries and provide accurate answers positions it as an invaluable resource for automating information retrieval and improving communication within the college.
Disclaimer: Curated by HT Syndication.
