• International Journal of Computer Science & Network Security
• /
• v.21 no.6
• /
• pp.200-206
• /
• 2021

Fraud in e-commerce transaction increased in the last decade especially with the increasing number of online stores and the lockdown that forced more people to pay for services and groceries online using their credit card. Several machine learning methods were proposed to detect fraudulent transaction. Neural networks showed promising results, but it has some few drawbacks that can be overcome using optimization methods. There are two categories of learning optimization methods, first-order methods which utilizes gradient information to construct the next training iteration whereas, and second-order methods which derivatives use Hessian to calculate the iteration based on the optimization trajectory. There also some training refinements procedures that aims to potentially enhance the original accuracy while possibly reduce the model size. This paper investigate the performance of several NN models in detecting fraud in e-commerce transaction. The backpropagation model which is classified as first learning algorithm achieved the best accuracy 96% among all the models.

• International Journal of Computer Science & Network Security
• /
• v.21 no.4
• /
• pp.214-222
• /
• 2021

Recently, the growth of e-commerce in Saudi Arabia has been exponential, bringing new remarkable challenges. A naive approach for product matching and categorization is needed to help consumers choose the right store to purchase a product. This paper presents a machine learning approach for product matching that combines deep learning techniques with standard artificial neural networks (ANNs). Existing methods focused on product matching, whereas our model compares products based on unstructured descriptions. We evaluated our electronics dataset model from three business-to-consumer (B2C) online stores by putting the match products collectively in one dataset. The performance evaluation based on k-mean classifier prediction from three real-world online stores demonstrates that the proposed algorithm outperforms the benchmarked approach by 80% on average F1-measure.

• The Journal of the Korea Contents Association
• /
• v.20 no.4
• /
• pp.1-13
• /
• 2020

In order to understand the recommendation algorithm applied to the online video platform, this study examines the relationship between the content characteristics of K-pop music videos and related videos recommended for playback on YouTube, and analyses which videos are recommended as related videos through network analysis. As a result, the more liked videos, the higher recommendation ranking and most of the videos belonging to the same channel or produced by the same agency were recommended as related videos. As a result of the network analysis of the related video, the network of K-pop music video is strongly formed, and the BTS music video is highly centralized in the network analysis of the related video. These results suggest that the network between K-pops is strong, so when you enter K-pop as a search query and watch videos, you can enjoy K-pop continuously. But when watching other genres of video, K-pop may not be recommended as a related video.

• Journal of Korean Academy of Nursing
• /
• v.47 no.6
• /
• pp.806-816
• /
• 2017

Purpose: As comprehensive nursing care service has gradually expanded, it has become necessary to explore the various opinions about it. The purpose of this study is to explore the large amount of text data regarding comprehensive nursing care service extracted from online news and social media by applying a semantic network analysis. Methods: The web pages of the Korean Nurses Association (KNA) News, major daily newspapers, and Twitter were crawled by searching the keyword 'comprehensive nursing care service' using Python. A morphological analysis was performed using KoNLPy. Nodes on a 'comprehensive nursing care service' cluster were selected, and frequency, edge weight, and degree centrality were calculated and visualized with Gephi for the semantic network. Results: A total of 536 news pages and 464 tweets were analyzed. In the KNA News and major daily newspapers, 'nursing workforce' and 'nursing service' were highly rated in frequency, edge weight, and degree centrality. On Twitter, the most frequent nodes were 'National Health Insurance Service' and 'comprehensive nursing care service hospital.' The nodes with the highest edge weight were 'national health insurance,' 'wards without caregiver presence,' and 'caregiving costs.' 'National Health Insurance Service' was highest in degree centrality. Conclusion: This study provides an example of how to use atypical big data for a nursing issue through semantic network analysis to explore diverse perspectives surrounding the nursing community through various media sources. Applying semantic network analysis to online big data to gather information regarding various nursing issues would help to explore opinions for formulating and implementing nursing policies.

• Agribusiness and Information Management
• /
• v.1 no.2
• /
• pp.81-94
• /
• 2009

Recently, social network service is developing rapidly as technology changes with new mobile dimensions and features creating positive opportunities and benefits to all users and companies. Social network services are allowing companies to expand their businesses and brands by utilizing it as a marketing tool to reach customers. This research is intended to identify major online social network services and their trends while enhancing the understanding of food service business expansion through social networking.

• Journal of KIISE:Information Networking
• /
• v.37 no.2
• /
• pp.94-98
• /
• 2010

In this paper, we formalize a new adaptive online price control scheme based on the Stackelberg game model. By using the hierarchical interaction strategy, control decisions in each mechanism act cooperatively and collaborate with each other to satisfy conflicting performance criteria. In addition, our dynamic online approach is practical for real network implementation. With a simulation study, the proposed scheme can adaptively adjust the network price to approximate an optimized solution under widely diverse network situations.

• Journal of Advanced Navigation Technology
• /
• v.11 no.2
• /
• pp.217-222
• /
• 2007

Recently PC games are trending toward online. But, mobile games are not using network actively. Because there are two problems that low speed of mobile network and low specification of mobile device in wireless internet environment. But, solving problems by these started HSDPA services and increased performance of devices. In this paper, we implement mobile on-line real time game for develop mobile network in wireless internet environment.

• Journal of the Korean Society of Radiology
• /
• v.10 no.8
• /
• pp.637-643
• /
• 2016

Due to the popularity of artificial intelligent, medical image processing using artificial neural network is increasingly attracting the attention of academic and industry researches. Deep learning with a convolutional neural network has been proved to very effective representation of images. However, the training process requires high performance H/W platform. Thus, the realtime learning of a large number of high dimensional samples within low-power devices is a challenging problem. In this paper, we attempt to establish this possibility by presenting a realtime neural network method on Raspberry pi using online sequential extreme learning machine. Our experiments on high-dimensional dataset show that the proposed method records an almost real-time execution.

• The Journal of the Korea Contents Association
• /
• v.17 no.2
• /
• pp.404-414
• /
• 2017

With the development of smartphones and online social networks, users produce a lot of contents and share them with each other. Therefore, users spend time by viewing or receiving the contents they do not want. In order to solve such problems, schemes for recommending useful contents have been actively studied. In this paper, we propose a contents recommendation scheme using collaborative filtering for users on online social networks. The proposed scheme consider a user trust in order to remove user data that lower the accuracy of recommendation. The user trust is derived by analyzing the user activity of online social network. For evaluating the user trust from various points of view, we collect user activities that have not been used in conventional techniques. It is shown through performance evaluation that the proposed scheme outperforms the existing scheme.

• The Korean Journal of Food And Nutrition
• /
• v.29 no.6
• /
• pp.1050-1057
• /
• 2016

SNS (Social network service) characteristics are perceived to simplify use. We carried out empirical studies on these parameters to observe the impact on the image of catering SNS online by word-of-mouth. The subjects of the study were as follows: 32.3% (392 persons) 19 years old, 67.7% (821 persons) over 19 years, 51.0% (619 persons) in their 20s, 22.1% (268 persons) in their 30s, 17.6% in their 50s, and 9.3% (112 persons) over 50 years. After verifying the hypothesis proposed that SNS characteristics perceived the ease of use, the significant factor identified in usability were connectivity Speed (${\beta}=0.213$), playfulness (${\beta}=0.246$), information (${\beta}=0.115$), and reciprocity (${\beta}=0.357$). Dual reciprocity had the most impact. It was observed that a longer impact of these significant factors improved the feel and fun of use. If SNS companies cater to, quick and easy, diverse, reliable and latest information, they can increase the ease of use, and availability, depending on the goals. Also, significant factors in the SNS features and online word of mouth was playfulness (${\beta}=0.312$), information (${\beta}=0.207$), reciprocity (${\beta}=0.066$) and perceived ease of use, and usefulness (${\beta}=0.293$), double playfulness had the maximum impact. These features provided more fun, reliable information, and could quickly deliver the latest information. The more the perceived usefulness, and ease of use, higher was the online word-of-mouth effect. SNS usage characteristics of connectivity Speed did not show any statistical significance in online word-of-mouth. Thus, catering businesses need to find ways to increase the ease of use, make the usefulness multifaceted, constantly checking the catering information on the SNS and ensuring to get the latest information is from diverse and reliable sources. This would increase the fun for the customer making the SNS to actively be utilized as a marketing tool.