Desired Content Extraction and Filtering of Unwanted Messages

Abstract

Today, we mostly use online Social Networks (OSNs) to send messages to one another, but there are no restrictions on any sort of message flow. In this project, we will provide users the option to regulate the messages posted on their own private area in order to avoid the appearance of undesirable content. This will be accomplished using a flexible rule-based system that lets users to create the filtering criteria that will be applied to their wall, as well as a machine-based soft classifier that will automatically label messages in support of contend-based filtering. If this type of posting of undesirable messages on a user's wall occurs frequently, the system will automatically add that person to a blacklist. This is accomplished using a flexible rule-based framework that lets users to tailor the filtering criteria that are applied to their walls, as well as a Machine Learning-based soft classifier that labels messages automatically in support of content-based filtering. Each user is considered to act independently in content-based filtering. In this work, we suggest a system that, with the aid of information filtering, may allow OSN users to have direct control over posting or commenting on their walls. When a user submits a message, the filtered wall intercepts it and applies Filtering and Black List Rules to the message. If the message does not breach the filtering and black list criteria, it will be shown on user walls.