Copyright © 2025 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{186040,
        author = {Priyanga.P and Dr.K.Saraswathi},
        title = {Attention-Enhanced CNN–RNN Framework with Adaptive Coyote Optimization–Based Feature Selection for High-Dimensional Data Mining and Anomaly Detection},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {12},
        number = {6},
        pages = {10-21},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=186040},
        abstract = {Data mining (DM) is fundamental for extracting meaningful knowledge from large-scale, high-dimensional datasets, were redundant attributes and noisy features often hinder classification and anomaly detection. In order to produce a smaller but more informative feature set, the suggested ACOA technique uses a process of eliminating features that are superfluous or redundant. This method produces small, extremely informative subsets. The spatial and sequential dependencies in data can be effectively captured by the application of an attention based- hybrid CNN–RNN (A-CNN-RNN). For anomaly detection study, and DM analysis, a widely accessible dataset named KDD Cup 1999 dataset is utilized to assess suggested method. When compared this suggested model with some standard deep learning (DL) models (CNN-only, RNN-only, CNN–RNN) and traditional machine learning (ML) techniques (Random Forest (RF), XGBoost), this suggested model executes well than other methods, and it was demonstrated by the outcomes of simulation. High scores in prediction accuracy (ACC), precision (P), recall (R), F1-score, and AUC-ROC were attained using ACOA. Additionally, feature selection (FS) was used to minimise the input feature space by more than 50%. By effectively finding pertinent features, metaheuristic-guided feature selection (FS) improves classification and anomaly detection in high-dimensional data. It facilitates Strong pattern recognition and extracting valuable insights over a wide range of DM backgrounds},
        keywords = {Data Mining; Anomaly Detection; High-Dimensional Data; CNN–RNN; Attention Mechanism; Adaptive Coyote Optimization (ACOA); Feature Selection; KDD Cup 1999},
        month = {October},
        }