MUMBAI, India, Jan. 2 -- Intellectual Property India has published a patent application (202541123099 A) filed by Malla Reddy (MR) Deemed to be University; Malla Reddy University; Malla Reddy Engineering College For Women; Malla Reddy College Of Engineering And Technology; and Malla Reddy Vishwavidyapeeth, Medchal-Malkajgiri, Telangana, on Dec. 6, 2025, for 'dynamic customer sentiment intelligence engine for real-time decision making.'
Inventor(s) include Dr. K. Maddileti; Dr. Vijayakumar Sajjan; Mrs. Gazala Akhtar; L Abdul Saleem; Mrs. Suneeta Netala; and Dr. Halesh Koti.
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: "The current innovation reveals a Dynamic Customer Sentiment Intelligence Engine (DCSIE), a new computational system that is capable of capturing, synthesizing, and exploiting real-time customer sentiment on heterogeneous, digital mediums in the automation of immediate and context-driven business decisions. Traditional customer relationship management (CRM) and sentiment analysis solutions are reactive, meaning that they analyze information based on historical data or on a long time scale (daily/weekly reports). Such inherent latency results in missed prospects of intervention, e.g. responding to a negative service experience, or taking advantage of a positive buying cue, which results in customer churn, negative reputation, or lost revenue. The DCSIE surpasses this drawback by creating a closed-loop feedback mechanism with low latency. The main component of the DCSIE is the Multi-Channel Affective Synthesis (MCAS) Module. This module takes in high velocity, multi-modern data streams such as text (social media, chat logs), voice (call transcripts, voice tone) as well as explicit behavioral cues (abandoned carts, frequent clicking of navigation). The MCAS Module uses a special Temporal Graph Network (TGN) to combine these two different signals and predicts the temporal dynamics of emotion intensity and will. TGN output is a discrete, running Customer Sentiment Score (CSS), which has been enhanced with a Contextual Action Vector (CAV) that determines the particular product, service or point in the customer journey that the current emotion is being caused by. The DCSIE incorporates such realtime CSS and CAV in the Adaptive Decision Controller (ADC). The ADC, in contrast to straightforward rule-based system, employs a reinforcement learning (RL) policy that is fit to optimize good business (e.g. maximize probability of conversion or reduce churn rate). As an illustration, when the CSS suggests high frustration (negative valence) and the CAV refers to the product page, the ADC immediately provides a custom intervention, including making an proactive chat offer with a discounted special code, or redirecting the customer to a higher level human agent. This is an intelligence-led dynamic system, which makes sure that the response of the business is not generic, but is precise in timing and is specifically specific to the needs of the customer in both emotional and contextual sense. The DCSIE radically enhances customer experience, resource allocation and the value of conversion and retention of customer interactions through active, real-time intervention of passive sentiment analysis. The capacity to deal with semantic ambiguity and irony is also an important facet of the DCSIE. The MCAS Module uses deep contextual embeddings, which are trained on domain-specific corpora, and thus it can identify ambiguous or figurative language (e.g., sarcasm or hidden discontent). Such a level of sophisticated natural language processing will guarantee that the final Customer Sentiment Score (CSS) will be a genuine representation of the intent of the user to avoid false-positive interventions and deliver on the overall accuracy and effectiveness of the real-time decision-making process."
Disclaimer: Curated by HT Syndication.
