Mohamed Hammami

WebGuard: a Web filtering engine combining textual, structural, and visual content-based analysis

Along with the ever-growing Web comes the proliferation of objectionable content, such as sex, violence, racism, etc. We need efficient tools for classifying and filtering undesirable Web content. In this paper, we investigate this problem and describe WebGuard, an automatic machine learning-based pornographic Web site classification and filtering system. Unlike most commercial filtering products, which are mainly based on textual content-based analysis such as indicative keywords detection or manually collected black list checking, WebGuard relies on several major data mining techniques associated with textual, structural content-based analysis, and skin color related visual content-based analysis as well. Experiments conducted on a testbed of 400 Web sites including 200 adult sites and 200 nonpornographic ones showed WebGuards filtering effectiveness, reaching a 97.4 percent classification accuracy rate when textual and structural content-based analysis was combined with visual content-based analysis. Further experiments on a black list of 12,311 adult Web sites manually collected and classified by the French Ministry of Education showed that WebGuard scored a 95.62 percent classification accuracy rate. The basic framework of WebGuard can apply to other categorization problems of Web sites which combine, as most of them do today, textual and visual content.

Face detection in video using combined data-mining and histogram based skin-color model

An improved technique of skin-color modelling for face detection in video is proposed. The skin-color is modelled as a combination of histogram based and data-mining based approaches. To solve a skin-color modelling problem in the presence of lighting variation, unconstrained environment, variation in races, resulting skin-color model is a product that minimizes registration errors and increases robustness and accuracy of the face detection in video system. The method improves applications performance on 7%, which is a significant advantage.

Adult content Web filtering and face detection using data-mining based kin-color model

The paper presents a novel approach for robust skin-color detection using data-mining techniques. The goal of skin-color detection is to select the appropriate color model that allows pixels to be verified under different lighting conditions and other variations. When the appropriate color model is selected, it is implied that we have good skin-color classifier properties for skin detection. This model has been successfully applied to face detection and Web based adult content filtering issues.

Violent web images classification based on MPEG7 color descriptors

In this article, we present a contribution to the violent Web images classification. This subject is deeply important as it has a potential use for many applications such as violent Web sites filtering. We propose to combine the techniques of image analysis and data-mining to relate low level characteristics extracted from the images colors to a higher characteristic of violence which could be contained in the image. We present a comparative study of different data mining techniques to classify violent Web images. Also, we discuss how the combination learning based methods can improve accuracy rate. Our results show that our approach can detect violent content effectively.

Adapted pruning scheme for the framework of imbalanced data-sets

Abstract Learning from imbalanced data is attracting an increasing interest by the machine learning community. This is mainly due to the high number of real applications that are affected by this situation. The adaptation of the standard decision trees to deal with imbalanced data represents one of the important number of approaches that have been developed to address this problem. This adaptation has been proposed under three different perspectives: splitting criterion, assignment rule and pruning. In this paper, we focus our attention to the pruning of decision trees. We propose an adaptation of the standard pruning algorithm MCCP to address the skewed-data problem. Our contribution affects two levels: adaption of the metric used in selecting nodes to be firstly pruned and change of the evaluation measure used in selecting the best decision-tree through the pruning set. Our goal is to show that, contrary to the popular belief in the literature enquiring into the uselessness of decision tree pruning, an adaptive pruning technique for imbalanced situations is more efficient and more accurate towards the minority class. A total of twelve binary class data-sets having different imbalance ratio are used to test the performance of the proposed method. Experimental results show that the proposed post-pruning approach can increase the performance of imbalanced decision trees in terms of evaluation measures that are recent and appropriate for the context of imbalanced classification.

Classification of violent web images using context based analysis

The development of the Web has been paralleled by the proliferation of a harmful content on its pages. Using Violent Web images as a case study, we tend to present a novel approach to their classification. This subject is of high importance as it has a potential use in many applications such as violent Web sites filtering. We, therefore, focus our attention on the extraction of contextual image features from the Web page. Also, we present a comparative study of different data mining techniques to classify violent Web images. The results we achieved show that our approach can detect violent content effectively.

Automatic violent content web filtering approach based on the KDD process

Purpose – The growth of the web and the increasing number of documents electronically available has been paralleled by the emergence of harmful web pages content such as pornography, violence, racism, etc. This emergence involved the necessity of providing filtering systems designed to secure the internet access. Most of them process mainly the adult content and focus on blocking pornography, marginalizing violence. The purpose of this paper is to propose a violent web content detection and filtering system, which uses textual and structural content‐based analysis.Design/methodology/approach – The violent web content detection and filtering system uses textual and structural content‐based analysis based on a violent keyword dictionary. The paper focuses on the keyword dictionary preparation, and presents a comparative study of different data mining techniques to block violent content web pages.Findings – The solution presented in this paper showed its effectiveness by scoring a 89 per cent classification ...

WebGuard: An Adult Content Detection and Filtering System

This paper describes a Web filtering system “WebGuard,” which aims to automatically detect and filter adult content on the Web. WebGuard uses data mining techniques to classify URLs into two classes: suspect URLs and normal URLs. The suspect URLs are stored in a database, which is constantly and automatically updated in order to reflect the highly dynamic evolution of the Web. When working, WebGuard simply captures a user’s URL, matches it with the suspect URLs stored in the database and takes an appropriate action — filtering or blocking — according to the result of the analysis. We started out with a study of most existing software so as to get to know the possibilities and functionalities available on the market at the moment. This phase enabled us to better evaluate the performances of our product as it was being developed. Thus, the second phase of our work was devoted to research into the usual algorithms regarding their advantages and drawbacks. Having gathered this knowledge, we are currently implementing a system that will combine several algorithms in order to increase the software’s performance. Our preliminary results show that it can detect and filter adult content effectively.

Data-Mining Based Skin-Color Modeling Using the ECL Skin-Color Images Database

Many human image processing techniques use skin detection as a first stage in subsequent feature extraction. In this paper we describe methods of skin detection using a data-mining technique. We also show the importance of the choice of the base simple to the performance of our skin analysis techniques. We present the details and the process of construction of our database which we have called “the ECL Skin-color Images Database from video ”. We will show that the use of a database derived from live video gives better results than one derived from internet images for face detection in video application.

AECID: Asymmetric entropy for classifying imbalanced data

Abstract In class imbalance problems, it is often more important and expensive to recognize examples from the minority class than from the majority. Standard entropies are known to exhibit poor performance towards the rare class since they take their maximal value for the uniform distribution. To deal with this issue, the present paper introduced a novel adaption of the decision-tree algorithm to imbalanced data situations. We focused, more specifically, on how to let the split criterion discriminate the minority-class examples on a binary-classification problem. Our algorithm uses a new asymmetric entropy measure, termed AECID, which adjusts the most uncertain class distribution to the prior class distribution and includes it in the evaluation of a node impurity. Unlike most competitive split, which include only the prior imbalanced class distribution in their formula, the proposed entropy is customizable with an adjustable concavity to take into account the specificities of each data-set and to better comply with the users’ requirements. Extensive experiments were conducted on thirty-six real life imbalanced data-sets to apprise the effectiveness of the proposed approach. Furthermore, the comparative results prove that the new proposal outperforms various algorithmic, data level and ensemble approaches that have been already proposed for imbalanced learning.