Implementation of Inception V3-Based CNN Classification for Malware Detection

This is the authors' abstract. We don't add key points for this paper.

A common issue with distributed networks, steal data and damage or destroy computer system software is refer to as malware. Finding different malware strains is very helpful for early detection. This study proposes an inception v3-based CNN classification for malware detection. Python is the programming language that is used to accomplish the proposed approach. To facilitate the process, the input dataset is pre-processed. Data collection and data enhancement are two categories of data pre-processing. The dataset is delivered into the data imagining block following pre-processing. These data representations aid in determining the structure of the data and the presence or absence of irregularities. The data visualization method comes after the classification procedure. Inception V3 is a CNN-based categorization network, is highly competitive in making accurate predictions, making it a valuable tool in various simulations. This proposed work generated a confusion matrix in order to determine the replica's performance. The outcome of the study demonstrate the effectiveness of the system that was created and point to its potential for improving malware detection's precision and effectiveness. On comparing the proposed accuracy of the controller has 90.6% and specificity has 92.6%.