MUMBAI, India, Aug. 24 -- Intellectual Property India has published a patent application (202641095582 A) filed by Bekim Fetaji on August 07, 2026, for Net-Sentinel Chrome Browser Plugin-Anomaly Detection System Using Machine Learning To Identify Unusu.
Inventors include Bekim Fetaji; Suplar Kanti Podder; Gurunatii R; and Irsa.
The application for the patent was published on August 21, 2026, under issue no. 34/2026.
Abstract: Anomaly Detector from the Browser. Knows quietly what normal is and will let you know when it's not. NetSentinel constantly surveils everything you do on your network, and uses machine learning algorithms on your devices and within your browser to detect unusual, suspicious connections. It is usable and not annoying due to two design decisions. First, a baseline is adaptive: It gives greater weight to recent behavior and allows for changing habits, which means legitimate change does not appear to be an attack, but also it has a cap on how quickly it will accept a change so that an attacker can't slowly "normalize" bad behavior. Second, all alerts are explainable - the model provides a report on which features contributed to the score, making the opaque number something a person can now act on. It is the same anomaly-detection idea as in enterprise network defense, but applied to the one network surface you have the most direct control over: The browser. The principle behind the operation is quite simple: first, specify what the normal distribution is, and then compute the distance between new observations and the normal distribution. This focus change is effective in capturing more anomalous states, rather than focusing on categorical threat identification. The system can be used for identifying unusual network behavior at the real time detection and protect the system like firewall. The present invention may ensure the maximum protection from the unusual threats by measuring the patterns of the threats. The system may facilitate for future threat prediction and guide for necessary precautions from unusual behavior.
Disclaimer: Curated by HT Syndication.
