Published Paper Details:

Online social networks have become a popular means of communication for internet users, and as a result, harmful assaults that target these platforms are also on the rise. In this case, our goal is to build a hybrid specialized system that uses artificial intelligence and a fuzzy system technique to identify malware. Malicious assaults can cause harm to people and businesses by allowing crucial data to be completely altered or misrepresented while being transferred or stored on social media platforms. The hybrid model was put through malware detection tests that were made accessible in several public datasets in order to execute the fuzzy rules extraction and to confirm the effectiveness of the hybrid technique. It was contrasted in binary classification tests with artificial neural network models and hybrid models of fuzzy neural networks. The simulation findings show that the fuzzy neural network method to treating malware detection is workable and that it permits the construction of fuzzy rules that can help in the development of specialized systems. Next, utilizing the IPS-MD5 algorithm, we enhance it with the Text Mining and Op code-based learning method to stop the propagation of harmful software in social applications. As a result, the frequency of malicious assaults will decline.