Copyright © 2026 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{179176,
        author = {prajwal p and sarath n and rohith b},
        title = {ATTACK ANALYSIS OVER WIDE NETWORK TRANSACTION SERVER},
        journal = {International Journal of Innovative Research in Technology},
        year = {2025},
        volume = {11},
        number = {12},
        pages = {6904-6913},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=179176},
        abstract = {Industrial digital transformation has intensified along with the dramatic growth of complex computer attacks throughout recent times. The current intrusion detection systems (IDS) find it challenging to detect contemporary advanced cyber-attack approaches. A Machine Learningbased Intrusion Detection System (IDS) that combines rule-based detection with deep learning techniques exists to accurately identify malicious network activities is proposed for addressing previous system limitations. Users can easily interact with the GUI-based system through Python's Tkinter framework because its design provides a complete interface which enables both novices and experts to work without extensive technical expertise. The system initiates with choosing the dataset before executing multiple preprocessing operations to handle missing data and eliminate unneeded records and extract apt features for model training purposes. The system reaches its critical point through its implementation of two classification techniques. A Decision Tree classifier based on the FURIA inspiration method enables identification of rules which lead to interpretable classifications. The system integrates a Functional Recurrent Neural Network (FRNN) which learns patterns across temporal sequences through its capability to detect advanced intrusion patterns. The joint usage of these models strengthens the detection capability of the system through extensive attack type recognition while increasing accuracy levels. Attack types receive clustering treatment in the solution while extracting statistical results from the dataset. The model performance gets measured through accuracy score with additional evaluations using confusion matrix and classification reports. The system implements Matplotlib and Seaborn visual analytics for generating user-friendly plots which present data from results and system performance. The proposed hybrid approach goes through comparison testing using SVM also known as support vector machines and conventional Decision Tree as baseline classifiers to show its superior characteristics. The solution presents a robust defensive system built with modular components which assures scalability for intrusion detection through data preparation and intelligent classification along with interactive capabilities},
        keywords = {},
        month = {May},
        }