MORF: A Distributed Multimodal Information Filtering System

Abstract

The proliferation of objectionable information on the Internet has reached a level of serious concern. To empower end-users with the choice of blocking undesirable and offensive Web-sites, we propose a multimodal personalized information filter, named MORF. The design of MORF aims to meet three major performance goals: effeciency, accuracy, and personalization. To achieve these design goals, we have devised a multimodality classification algorithm and a personalization algorithm. Empirical study and initial statistics collected from the MORF filters deployed at sites in the U.S. and Asia show that MORF is both e.cient and effective, compared to the traditional URL- and text-based .ltering approaches.

In this paper, unless otherwise stated, objectionable information refers to undesirable and offensive information that can be personally defined; e.g., pornography, hate messages, etc.