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{199131,
        author = {Akshada Solase and Shraddha Solase and Atik Pathan and Sangeeta Mohapatra},
        title = {Designing Anti hacking Software for securing social media platforms like What’sApp},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {14226-14234},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=199131},
        abstract = {Phishing attacks have evolved into one of the most persistent and dangerous cybersecurity threats in today's digital landscape. Such attacks usually target users via deceptive emails, harmful URLs, counterfeit login pages, and manipulative communication tactics. As attack techniques evolve, cybercriminals are increasingly employing image-based phishing tactics like fake screenshots, malicious embedded visuals, and cloned interfaces, thereby making detection significantly more difficult. Conventional phishing detection systems, which depend mainly on rule-based filtering or text analysis, frequently struggle to effectively identify such sophisticated threats.
This research proposes an intelligent phishing detection system based on artificial intelligence. The system combines text-based analysis with image-based classification powered by Convolutional Neural Networks (CNNs). The architecture comprises two primary modules. The initial module is dedicated to analyzing textual data, including URLs and message content, by extracting pertinent features. The second module employs deep learning techniques to analyze images and detect visual patterns indicative of phishing attacks.
Combining these two approaches allows the system to attain greater detection accuracy while substantially lowering false positive rates. The model is trained and evaluated on well-labeled datasets containing both phishing and legitimate samples. To ensure reliability, performance evaluation is conducted using metrics such as accuracy, precision, recall, and F1-score.
The findings demonstrate that the hybrid approach outperforms conventional single-method detection systems. The proposed system strengthens cybersecurity by offering an automated, intelligent, and real-time phishing detection solution that supports multiple data formats. This study underscores the efficacy of integrating deep learning and machine learning methods to counteract emerging cyber threats and enhance the security of online communication platforms.},
        keywords = {},
        month = {April},
        }