Hybrid Software Classifier of Internet Applications

Abstract

Traffic classification refers to the processes used to categorize traffic based on features in the traffic and in accordance with specific classification objectives. In general, there are numerous forests, fields, and aims in the classification of internet traffic. Internet service providers must be aware of the types of traffic that are sent across their network Classifying Internet applications that are carried by computer networks is the specific goal, create a hybrid software classifier that can be applied to the classification of Internet traffic.   Internet service providers (ISPs) attempt to boost bandwidth as a result of rising Internet traffic, but at the same time, more bandwidth is needed for Internet applications. Problems arise from the exponential growth of new internet applications using unregistered ports. The new programs may also contain a lot of viruses and dangerous code. In recent years, traffic classification has drawn more and more attention. With the help of direct and passive observation of the individual packets or stream of packets moving over the network, it seeks to provide the capability of automatically identifying the program that created a specific stream of packets. Data mining (DM) is a method for sifting through enormous databases in search of fresh, obscured, and practical information patterns. The knowledge discovery process includes the DM idea.   This study employs a variety of techniques and functions as a hybrid classification. The foundation of many essential network monitoring and controlling jobs, such as billing, quality of service, security, and trend analyzers, is the classification of networks flows by their application type. Identification of Internet traffic is a crucial tool for network management. It enables operators to more accurately forecast upcoming traffic patterns and demand, and it enables security staff to spot unusual conduct.