Zhongfu Wu

Userrank for item-based collaborative filtering recommendation

With the recent explosive growth of the Web, recommendation systems have been widely accepted by users. Item-based Collaborative Filtering (CF) is one of the most popular approaches for determining recommendations. A common problem of current item-based CF approaches is that all users have the same weight when computing the item relationships. To improve the quality of recommendations, we incorporate the weight of a user, userrank, into the computation of item similarities and differentials. In this paper, a data model for userrank calculations, a PageRank-based user ranking approach, and a userrank-based item similarities/differentials computing approach are proposed. Finally, the userrank-based approaches improve the recommendation results of the typical Adjusted Cosine and Slope One item-based CF approaches.

An unsupervised anomaly intrusion detection algorithm based on swarm intelligence

An approach to network intrusion detection is investigated, based on swarm intelligence. The basic idea of the method is to produce the cluster by swarm intelligence-based clustering. With the classified data instances, anomaly data clusters can be easily identified by normal cluster ratio. And then the identified cluster can be used in real data detection. In the traditional clustering-based intrusion detection algorithms, clustering using a simple distance-based metric and detection based on the centers of clusters, which generally degrade detection accuracy and efficiency. Our approach based on swarm intelligence can settle these problems effectively. The experiment result shows that our approach can detect unknown intrusions efficiently in the real network connections.

Efficient feature selection for high-dimensional data using two-level filter

Feature selection is a key problem to pattern recognition and machine learning, and it is difficult to get the optimal feature subset for its NP-hard. Currently, the dimensionality of feature set or instance set is very high in many applications, such as information retrieval, so the feature selection from high-dimensional data is also an urgent task for researchers. This paper presents a new approach, which is a two-level filter model system integrating the relief and a newly developed algorithm of feature cluster, to reduce the dimensionality of large-scale feature set via the feature correlation (relevance) including the feature-feature correlation and feature-class correlation. Our major contributions are: (1) to present a system to perform feature selection from high-dimensional data; (2) to analyze the change of system architecture according to the time cost of the parts in the system; (3) to summarize and comment on the calculations of feature correlation; (4) to perform experiments to show the effective of the proposed approach, which has shown that the system can efficiently get a better compromise between dimensionality reduction and accuracy rate of classification than just part of the system. In many cases, it can improve the accuracy rate and dimensionality reduction.

Breaking Monocultures in P2P Networks for Worm Prevention

Peer-to-peer (P2P) worm has become an interesting research topic during recent years. It is widely believed that the monoculture of P2P systems results in fast worm propagation in P2P networks and software diversity can decrease the virulence of worms and the effectiveness of repeated applications of single attacks. In order to break monocultures in P2P networks, we propose a P2P-based software diversity model and conduct both theoretical analysis and extensive simulations to test the performance of software diversity in containing P2P worm propagation. We find that our proposed software diversity model can efficiently improve the performance of worm prevention in P2P networks

An enhanced swarm intelligence clustering-based RBFNN classifier and its application in deep Web sources classification

The central problem in training a radial basis function neural network (RBFNN) is the selection of hidden layer neurons, which includes the selection of the center and width of those neurons. In this paper, we propose an enhanced swarm intelligence clustering (ESIC) method to select hidden layer neurons, and then, train a cosine RBFNN based on the gradient descent learning process. Also, we apply this new method for classification of deep Web sources. Experimental results show that the average Precision, Recall and F of our ESIC-based RBFNN classifier achieve higher performance than BP, Support Vector Machines (SVM) and OLS RBF for our deep Web sources classification problems.

An Enhanced Swarm Intelligence Clustering-Based RBF Neural Network Detection Classifier

The central problem in training a radial basis function neural network (RBFNN) is the selection of hidden layer neurons, which includes the selection of the center and width of those neurons. In this paper, we propose an enhanced swarm intelligence clustering (ESIC) method to select hidden layer neurons, and then, training a cosine RBFNN base on gradient descent learning process. Also, the new method is applied for intrusion detection. Experimental results show that the average DR and FPR of our ESIC-based RBFNN detection classifier maintained a better performance than BP, SVM and OLS RBF.

Clustering based on Self-Organizing Ant Colony Networks with Application to Intrusion Detection

Due to the fact that it is more and more improbable to a system administrator to recognize and manually intervene to stop an attack, there is an increasing recognition that ID systems should have a lot to earn on following its basic principles on the behavior of complex natural systems, namely in what refers to self-organization, allowing for a real distributed and collective perception of this phenomena. A clustering model based on self-organizing ant colony networks (CSOACN) is systematically proposed for intrusion detection system. Instead of using the linear segmentation function of the CSI model, here we propose to use a nonlinear probability conversion function and can help to solve linearly inseparable problems. Using a set of benchmark data from 1998 DARPA, we demonstrate that the efficiency and accuracy of CSOACN

The fuzzy similarity measures for content-based image retrieval

Many fuzzy methods have been applied to the content-based image retrieval (CBIR) to retrieve the similar images according to the similarity of fuzzy sets. In this paper, firstly the principle of fuzzy similarity measure for CBIR is deeply inspected, then the properties and the classes about fuzzy similarity measures are introduced and remarked, lastly we present a faster algorithm on similarity measure using center of gravity of fuzzy sets in CBIR. The discussion nearly covers all the fuzzy similarity measures, which may be greatly helpful to both the development and application of fuzzy set theory for CBIR.

A data-type-based approach for identifying corresponding attributes in heterogeneous databases

In order to support interoperability among databases, the probability and efficiency of using metadata and data contents to automate some aspect of semantic integration become a key challenge. An important task of semantic integration in heterogeneous database is to determine which fields refer to the same data. Although many methods can be employed to automate the process, we argue that the existing methods of identifying corresponding attributes are time-consuming. In this paper we contribute to lower time complexity without reducing the precision ratio and recall ratio by providing a more efficient method, a data-type-based approach, to identify corresponding attributes in heterogeneous databases. The experimental results showed our method is effective.

A Novel Authentication Mechanism Based on CGA for BU Message Disposal in Mobile IPv6

In mobile IPv6, the location of a mobile node (MN) is obtained by implementing a binding update (BU) process. During the process, the unauthenticated BU message may give rise to many security risks, such as denial-of-service (DoS) attack. Thus, authentication of the BU message is most important for the security of mobile IPv6. The purpose of cryptographically generated addresses (CGA) is to prevent stealing and spoofing of existent IP addresses. In this paper, a novel authentication mechanism based on CGA is proposed for BU message in mobile IPv6. With the implementation of the novel authentication mechanism, care-of address of MN will be automatically generated, and this will be related to MN private attribute, hence, it would not be possible for a vicious node to steal it. With this method, the BU message can be authenticated.