MUMBAI, India, July 30 -- Intellectual Property India has published a patent application (202641088787 A) filed by Jawaharlal Nehru Technological University Hyderabad Jntuh on July 21, 2026, for A Novel Method For Multi Dimensional Temporal Association Rule Mining In Large-Scale Datasets.
Inventors include Mr B. B. L. V. Prasad; and Dr E. Hemalatha.
The application for the patent was published on July 24, 2026, under issue no. 30/2026.
Abstract: ABSTRACT The present invention relates to a computer-implemented method for multi-dimensional temporal association rule mining in large-scale datasets using a system (100) comprising a processor (109) and a memory (110). The system includes a data ingestion and preprocessing module (101), a temporal partitioning module (102), a multi-dimensional indexing module (103), a support-confidence computation module (104), an adaptive pruning engine (105), a rule drift detection module (106), an association rule generation module (107), and a rule repository module (108). The dataset is partitioned into temporal windows, and a hybrid temporal- attribute index structure is generated to efficiently compute support and confidence values. The adaptive pruning engine applies temporal decay-based pruning, while the rule drift detection module identifies emerging, stable, or declining association patterns across successive temporal windows. The generated association rules are stored for downstream applications, thereby reducing dataset scans, minimizing memory consumption, and improving computational efficiency. The figure associated with the abstract is Fig. 1.
Disclaimer: Curated by HT Syndication.
