MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641071842 A) filed by Sona College Of Technology on June 10, 2026, for Deep Learning Based Interview Feedback And Question Organization Sharing System To Facilitate Campus-Based Preparation.
Inventors include Shanmugaraja P; Sowmya A; Prathima S; and Swathi R.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: ABSTRACT The present invention relates to an artificial intelligence–based interview preparation and question organization system for automatically organizing interview questions and feedback using deep learning and semantic analysis techniques. The system receives interview questions and company feedback submitted by authenticated users and preprocesses the textual data through normalization, tokenization, and lemmatization. Semantic embeddings are generated using a fine-tuned transformer-based language model including SciBERT with LoRA adaptation. The generated embeddings are utilized for semantic similarity–based topic clustering and automated difficulty classification of interview questions into predefined levels. The invention further performs sentiment analysis on company reviews using a RoBERTa-based model. Classified interview questions and metadata are stored in a centralized hybrid database architecture and retrieved through topic-wise, company- wise, and difficulty-wise filtering interfaces. The invention provides scalable, context-aware, and automated organization of interview preparation content, thereby improving retrieval efficiency and structured campus-based interview preparation.
Disclaimer: Curated by HT Syndication.
