• International Journal of Computer Science & Network Security
• /
• v.21 no.7
• /
• pp.159-168
• /
• 2021

Detecting cyber-attacks using machine learning or deep learning is being studied and applied widely in network intrusion detection systems. We noticed that the application of deep learning algorithms yielded many good results. However, because each deep learning model has different architecture and characteristics with certain advantages and disadvantages, so those deep learning models are only suitable for specific datasets or features. In this paper, in order to optimize the process of detecting cyber-attacks, we propose the idea of building a new deep learning network model based on the association and combination of individual deep learning models. In particular, based on the architecture of 2 deep learning models: Convolutional Neural Network (CNN) and Long Short Term Memory (LSTM), we combine them into a combined deep learning network for detecting cyber-attacks based on network traffic. The experimental results in Section IV.D have demonstrated that our proposal using the CNN-LSTM deep learning model for detecting cyber-attacks based on network traffic is completely correct because the results of this model are much better than some individual deep learning models on all measures.

• Journal of Information Technology Services
• /
• v.11 no.3
• /
• pp.17-38
• /
• 2012

Social network service(SNS), provided by social network sites such as Facebook, Twitter and Cyworld is rapidly growing in online business. Furthermore, many companies have growing interests in finding effective ways to use SNSs for their innovations, marketing and advertisement. In fact, firms have recognized the utility value of the SNS for their business. In this aspect, this study attempts to identify key factors influencing the intention to continuous use of SNSs. Based on the UTAUT(the Unified Theory of Acceptance and Usage of Technology)model, this study proposes the research model, including the effects of social network service characteristics(social relationship support, information sharing, image expression) and individual characteristics(self-disclosure, extroversion, familiarity) on performance expectancy as well as the moderating effect of perceived information security among UTAUT variables. The 412T sets of data collected in a survey were tested against the modeling using SEM using SmartPLS. Results indicated that social network service and individual characteristics had significant effect on performance expectancy with exception of self-disclosure. In addition, the moderating effect of perceived information security had significant effect. The results had important implications for firms providing SNSs hoping to develop a successful business model.

• Journal of Digital Convergence
• /
• v.10 no.6
• /
• pp.33-39
• /
• 2012

The purpose of this study is to investigate the effect of SNS(Social Network Service) on individual learning performance. To do this, we distribute and collect data by using a survey method. Research results suggest that online social networking engagement and acculturation have an effect on interaction quality with professors. Interaction quality with professors influences individual learning performance as well as collaborative learning. The conclusion and implications are discussed.

• Journal of Wetlands Research
• /
• v.5 no.1
• /
• pp.15-27
• /
• 2003

The fractal study in river basin has been performed for the sinuosity of an individual stream and bifurcation of the stream network. The previous studies has suggested many methods or equations for the fractal dimension estimation in a river network. This study used those many equations for the estimation of fractal dimensions on the streams such as Bokha, Gonjiam, and Pocheon streams. The estimated dimensions are in the range of 1 to 1.359 for the individual stream and 1.634 to 2 for the stream network. The most of equations were suggested based on the assumption of self-similarity of a river basin for the individual stream and stream network. However, the real river basin could be characterized by self-affinity rather than self-similarity. Even though we estimate the dimensions by using many equations, we could not recommend which one is better equation for the estimation of fractal dimension. This might be from the self-similarity assumption of equations. Therefore, the assumption and research work of self-affinity will be needed for the appropriate estimation of fractal dimension in river basin.

• Journal of the Korean Society for information Management
• /
• v.6 no.1
• /
• pp.37-56
• /
• 1989

This study is a structural analysis of a multi-type and multi-level library network within the framework of a regional interlibrary loan (ILL) system. The study monitored to information network structure for resource sharing of academic and research library materials transmitted through the ILL. The local flow of academic and research information was measured by a survey of the filled ILL transactions by individual libraries in the Western 3Rs region. The major findings were as follows: 1) the regional ILL network showed less than half of participation of the total subject libraries, 2) existing structure surveyed was identified as a composite centralized network with three communication groups, 3) depending on the types of materials transacted, the structure were changed, 4) statewide and multi-state library cooperatives had direct interactions with some of the local libraries, 5) individual libraries participated in the ILL network more for periodicals than book materials, 6) academic libraries throughout the total six structure analyzed showed the highest percentage of participation.

• The Journal of Society for e-Business Studies
• /
• v.21 no.3
• /
• pp.29-50
• /
• 2016

We explore the impact of individuals'network position toward knowledge sharing intention. In order to identify network positions, we utilize three centrality measures (degree/closeness/betweenness) of individual network participants. The research findings show that the individual network positions significantly affect knowledge sharing intentions. Since an individual with high degree centrality might be the leader or the hub, one makes considerable effort to maintain the network position by actively participating in intra-team and inter-team knowledge sharing, A participant who can quickly interact with many other participants within a team (high closeness centrality) is more interested in intra-team knowledge sharing than inter-team knowledge sharing. Unlike degree centrality and closeness centrality, the betweenness centrality provides a participant with diverse resources located in multiple sub-groups. Although an individual with high betweenness centrality is not at the center of the networks, one plays a crucial role in disseminating and regulating information. Therefore, the individual is likely to have more positive intention toward inter-team knowledge sharing than intra-team knowledge sharing.

• Journal of Korea Multimedia Society
• /
• v.20 no.2
• /
• pp.216-224
• /
• 2017

Abnormal situation caused by aggressive behavior of pigs adversely affects the growth of pigs, and comes with an economic loss in intensive pigsties. Therefore, IT-based video surveillance system is needed to monitor the abnormal situations in pigsty continuously in order to minimize the economic demage. Recently, some advances have been made in pig monitoring; however, detecting each pig is still challenging problem. In this paper, we propose a new color image-based monitoring system for the detection of the individual pig using a fast region-based convolution neural network with consideration of detecting touching pigs in a crowed pigsty. The experimental results with the color images obtained from a pig farm located in Sejong city illustrate the efficiency of the proposed method.

• The Journal of the Korea Contents Association
• /
• v.15 no.11
• /
• pp.616-625
• /
• 2015

The purpose of the study is to structure the friendship network by the social network analysis and investigate the effects of social network centrality and learners' performances in college students. Both the in-degree centrality of 1st grade class study-network(t=2.722, P<.005) and the in-degree centrality of and $2^{nd}$ grade class study-network(t=2.708, P<.005)are predicted the individual student's learning performances. But there is no correlation between the in-degree centrality of $1^{st}$ and $2^{nd}$ grade class entertainment-network and the individual student's learning performances. Results of the study suggested the significant effect of social network analysis measures on learners' performance in the friendship networks. Based on the results, implication to the teaching strategy and future research direction were discussed.

• The Journal of the Korea Contents Association
• /
• v.12 no.5
• /
• pp.393-403
• /
• 2012

This study is to investigate social network of sports for all's instructor and occupational achievement. To accomplish this goal, this study sampled 240 instructors of sports facilities of Seoul utilizing purposive sampling method in 2011. But this study finally used 203 samples in data analysis. Validity and reliability of instruments tested by expert meetings and reliability analysis. Chronbach's ${\alpha}$ is over .674. Data analysis is logistic regression analysis and multiple regression analysis using SPSSWIN 18.0. Conclusions are the followings. First, gender, number of employment influence on employment throughout individual network. Second, social network influences on occupational achievement. Namely, individual network influences on wage satisfaction, social network influences fitness of academic career.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.2
• /
• pp.171-177
• /
• 2002

In this paper, we propose a new data fusion method to improve the performance of individual prediction models for time series data. Individual models used are ARIMA and neural network and their results are combined based on the weight reflecting the inverse of EWMA of squared prediction error of each individual model. Monte Carlo simulation is used to identify the situation where the proposed approach can take a vintage point over typical fusion methods which utilize MSE for weight. Study results indicate the following: EWMA performs better than MSE fusion when the data size is large with a relatively big amplitude, which is often observed in intra-cranial pressure data. Additionally, EWMA turns out to be a best choice among MSE fusion and the two individual prediction models when the data size is large with relatively small random noises, often appearing in tax revenue data.