Intellectual Property India Publishes Patent Application for 'Next-Gen Content Suggestion: Hybrid Ensemble Machine Learning based Technique to Predict Content Creation Recommendations for social media platforms' Filed by Aafrin Akil Khan; and Advait Bajirao Mhalungekar

MUMBAI, India, March 14 -- Intellectual Property India has published a patent application (202421067614 A) filed by Aafrin Akil Khan; and Advait Bajirao Mhalungekar, Kolhapur, Maharashtra, on Sept. 6, 2024, for 'next-gen content suggestion: hybrid ensemble machine learning based technique to predict content creation recommendations for social media platforms.'

Inventor(s) include Aafrin Akil Khan; and Advait Bajirao Mhalungekar.

The application for the patent was published on March 13, under issue no. 11/2026.

According to the abstract released by the Intellectual Property India: "Disclosed herein is a framework for generating content creation recommendations on social media platforms through the application of machine learning and natural language processing techniques. User comments, such as those obtained from YouTube, are collected and preprocessed through text normalization, tokenization, stopword removal, stemming, and lemmatization. A HashingVectorizer is employed for scalable feature extraction. In one embodiment, the system utilizes incremental learning with classifiers such as Stochastic Gradient Descent and Multinomial Naive Bayes, updated through batch-wise training using the partial_fit method. The outputs of these classifiers are combined using a weighted ensemble mechanism to improve robustness and prediction accuracy. The trained ensemble model classifies comments into categories such as suggestive and non-suggestive. In another embodiment, the invention incorporates a summarization module powered by a large language model, such as LLaMA 3.1. The subset of comments identified as suggestive are aggregated and summarized to highlight audience-requested improvements, new content ideas, and recurring themes. The invention thus provides an integrated framework that combines incremental ensemble classification with advanced summarization, enabling scalable, real-time, and actionable content recommendations for creators across social media, online forums, and customer feedback systems."

Disclaimer: Curated by HT Syndication.