A Feature Selection Algorithm based on Hoeffding Inequality and Mutual Information

Chunyong Yin,Lu Feng,Luyu Ma,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Signal Processing, Image Vol.8 No.11

With the rapid development of the Internet, the application of data mining in the Internet is becoming more and more extensive. However, the data source’s complex feature redundancy leads that data mining process becomes very inefficient and complex. So feature selection research is essential to make data mining more efficient and simple. In this paper, we propose a new way to measure the correlation degree of internal features of dataset which is a mutation of mutual information. Additionally we also introduce Hoeffding inequality as constraint of constructing algorithm. During the experiments, we use C4.5 classification algorithm as test algorithm and compare HSF with BIF(feature selection algorithm based on mutual information). Experiments results show that HSF performances better than BIF[1] in TP and FP rate, what’s more the feature subset obtained by HSF can significantly improve the TP, FP and memory usage of C4.5 classification algorithm.

Botnet Detection Based on Degree Distributions of Node Using Data Mining Scheme

Chunyong Yin,Lei Yang,Jin Wang 보안공학연구지원센터 2013 International Journal of Future Generation Communi Vol.6 No.6

Botnet Detection Based on Correlation of Malicious Behaviors

Chunyong Yin,Mian Zou,Darius Iko,Jin Wang 보안공학연구지원센터 2013 International Journal of Hybrid Information Techno Vol.6 No.6

A Feature Selection Algorithm towards Efficient Intrusion Detection

Chunyong Yin,Luyu Ma,Lu Feng,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.11

Feature selection algorithm plays a crucial role in intrusion detection, data mining and pattern recognition. According to some evaluation criteria, it gets optimal feature subset by deleting unrelated and redundant features of the original data set. Aiming at solving the problems about the low accuracy, the high false positive rate and the long detection time of the existing feature selection algorithm. In this paper, we come up with a feature selection algorithm towards efficient intrusion detection, this algorithm combines the correlation algorithm and redundancy algorithm to chooses the optimal feature subset. Experimental results show that the algorithm shows almost and even better than the traditional feature selection algorithm on the different classifiers.

A Mobile Recommendation Algorithm Based on Statistical Analysis of User Data

Chunyong Yin,Hui Zhang,Jun Xiang,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Hybrid Information Techno Vol.8 No.12

Recommendation technology is used to help people solve the problem of information overload. Recent years, it has been widely applied to the movie ratings, e-commerce and many other fields. Researchers have noticed its powerful application prospect. But with the exponential growth of information data, the recommendation systems also have to improve the ability of data processing and this leads to that the traditional collaborative filtering recommendation algorithms cannot meet the needs of the users. To solve the problem, we designed an algorithm based on the theory of statistical analysis. This algorithm classified the data simply firstly, and then system could give users the relatively satisfactory personalized recommendations by the statistical analysis of different attributes on the data sets.

An incremental approach to computing conditional complementary entropy for dynamic information systems with varying object set

Chunyong Wang,Bing Yang 원광대학교 기초자연과학연구소 2019 ANNALS OF FUZZY MATHEMATICS AND INFORMATICS Vol.18 No.3

Information entropies have been widely applied in constructing heuristic attribute reduction. However, little attention has been paid to the information entropies for dynamic information systems with varying object set. In this paper, we present an incremental approach to update conditional complementary entropy for dynamic information systems with varying objects. Based on the new incremental formulas, we develop an incremental attribute reduction algorithm for decision table with varying object set. By a numerical experiment, we express the efficiency of the new method.

Botnet Detection Based on Genetic Neural Network

Chunyong Yin,Ardalan Husin Awlla,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.11

Botnet have turned into the most serious security dangers on the present Internet framework. A botnet is most extensive and regularly happens in today's cyber-attacks, bringing about the serious risk of our system resources and association's properties. Botnets are accumulations of compromised computers (Bots) which are remotely regulated by its creator (BotMaster) under a typical Command-and-Control (C&C) framework. Botnets cannot just be implemented utilizing existing well-known applications and additionally developed by unknown or inventive applications. This makes the botnet detection a challenging issue. In this paper proposed an anomaly detection model based on genetic neural network system, which joined the significant global searching capability of genetic algorithm with the precise local searching element of back propagation feed forward neural networks to improve the initial weights of neural network.

PSYCHOLOGICAL HEALTH Relation between Perception, Appearance Satisfaction, and Happiness of Women Participating in Jewelry Therapy

Chunyong Lee,이재범,Eungyeol Na J-INSTITUTE 2021 Protection Convergence Vol.6 No.1

Purpose: This study was conducted in order to examine the relations among the recognition, happiness and appearance satisfaction of jewelry therapy participant women and find out which structural effect appearance satisfaction as mediation effect has on feeling of happiness. Method: Study participants are 152 women in their 30s~ 60s purchased color jewelry at the jewelry shopping mall in Jongrogu, Republic of Korea and they were selected in a way of purposive sampling as the ones subject to the analysis of this study. Second, in order to analyze relationship which recognition and appearance satisfaction influence on the happiness, Multiple Regression Analysis is conducted. First, Structural Equation Modeling(SEM) is set in order to verify mediating effect of structural relations of each variables and appearance satisfaction variables and the assessment of model fitness and path coefficient between variables is conducted. Fifth, positive statistical analysis verifies at the p<.05 level. Results: First, in examining the result of multiple regression analysis in terms of the effect of participants’ recognition and appearance satisfaction to happiness, it is discovered that, to the happiness, physical recognition, social recognition, educational recognition and appearance satisfaction are statistically significant but psychological recognition is not. Second, it is discovered that appearance satisfaction works as significant med-iation effect between the recognition and happiness. In other words, it is discovered that the recognition, an external variable, has a significant effect on appearance satisfaction, an internal variable. Conclusion: Through this study, we can see that the recognition of women participating in jewelry therapy has positive effect not only on appearance satisfaction but also happiness with appearance satisfaction as mediation variable. That result of this study is significant in that it provides basic information of psychological health relations with which emotional response can be scientifically analyzed and utilized in the reality which lacks objective evaluation standard for jewelry therapy.

An Algorithm of Clustering by Density Peaks Using in Anomaly Detection

Chunyong Yin,Sun Zhang,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Security and Its Applicat Vol.9 No.12

With the development of the networks, the security of computer networks is becoming more and more serious. The information openness, sharing and interconnection are three important characteristics of computer networks. However, the amounts of intruders and attackers have been grows with the popularization of computers. Therefore, the focus of network security is preventing systems from being invaded effectively. Intrusion detection as a key technology of network security active defense system is designed to distinguish normal behaviors and attack behaviors. Intrusion detection is divided into misuse detection and anomaly detection, and using clustering algorithm is one of the most effective methods for anomaly detection. In this paper, a clustering algorithm based on fast search and find of density peaks is used to distinguish the normal and abnormal network connections to achieve the purpose of anomaly detection. The performance of the algorithm is tested by a data set selected from KDD CUP99. Experiment results show that this algorithm is more suitable than the traditional K-means in data sets containing a large amount of data and uneven density distribution.

Short Text Classification Algorithm Based on Semi-Supervised Learning and SVM

Chunyong Yin,Jun Xiang,Hui Zhang,Zhichao Yin,Jin Wang 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.12

Short text is a popular text form, which is widely used in real-time network news, short commentary, micro-blog and many other fields. With the development of the application such as QQ, mobile phone text messages and movie websites, the size of data is also becoming larger and larger. Most data is useless for us while other data is significant for us. Therefore, it is necessary for us to extract the useful short text from the big data. However, there are many problems with the short text classification, such as fewer features, irregularity and so on. To solve these problems, we should pretreat the short text set first, and then choose the significant features. This paper use semi-supervised learning method and SVM classifier to improve the traditional methods and it can classify a large number of short texts to mining the useful massage from the short text. The experimental results in this paper also show a good promotion.