MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123198 A) filed by SR University, Warangal, Telangana, on Dec. 6, 2025, for 'hybrid bi-gru and bi-lstm based ontology-driven natural language processing system for multi-tiered detection and classification of depression levels.'
Inventor(s) include P Lakshmi Priya; and Dr. R. Vijaya Prakash.
The application for the patent was published on Jan. 2, under issue no. 01/2026.
According to the abstract released by the Intellectual Property India: "With the rapid growth of social media, user-generated content has become a valuable source of information for understanding psychological health patterns. This invention discloses an ontology-driven Natural Language Processing (NLP) system integrated with a hybrid deep learning architecture combining Bidirectional Gated Recurrent Units (Bi-GRU) and Bidirectional Long Short-Term Memory networks (Bi-LSTM) for the multi-tiered detection and classification of depression levels. The system employs an ontology-based framework to enrich linguistic representation, providing semantic depth for recognizing depressive expressions in short and informal texts such as tweets. The Bi-GRU component captures sequential dependencies in both forward and backward directions, enabling the model to learn contextual features from social media posts. The Bi-LSTM, augmented with an attention mechanism, further extracts critical hidden features while emphasizing discriminative words and phrases that contribute significantly to depression identification. By integrating Bi-GRU with Bi-LSTM, the model leverages the complementary strengths of both architectures to achieve enhanced accuracy in multi-class classification of depression severity levels ranging from mild to severe conditions. Additionally, the invention incorporates Bag of Words (BoW) representations alongside ontology-based vocabulary for adolescent depression, enabling structured mapping between linguistic features and psychological states. This hybrid methodology allows the system to combine statistical text features with semantic reasoning, resulting in a highly accurate framework for automated depression assessment. Experimental evaluations demonstrate that the proposed system outperforms conventional machine learning and standalone deep learning models, offering a scalable and effective solution for early detection and classification of depressive disorders from large-scale social media data."
Disclaimer: Curated by HT Syndication.
