• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.413-414
• /
• 2023

IT 기술의 발달로 E-Commerce 분야는 실시간으로 발생되는 데이터양이 증가하고 있으며, 발생된 데이터는 개인화 맞춤 서비스에 많이 활용되고 있다. 그러나 신생 E-commerce 기업은 신규 상품 및 기존 상품에 대한 정보와 고객 간의 상호 작용 데이터가 존재하지 않아 콜드 스타트 문제가 발생한다. 이에 본 논문에서는 E-commerce 환경에서 실시간 사용자 구매패턴 분석을 통한 사용자 상품 추천 시스템을 제안한다. 제안하는 시스템은 Kafka와 Spark를 사용해 실시간 스트림을 데이터를 처리한다. 주요 기능은 ALS 알고리즘과, FP-Growth 알고리즘을 적용해 콜트 스타트 문제를 해결하며, 사용자 구매 패턴 분석을 통한 분석 결과에 맞는 상품을 사용자에게 추천한다.

• International Journal of Computer Science & Network Security
• /
• v.24 no.9
• /
• pp.202-206
• /
• 2024

Robust system may not guaranty its applicability and adaptability. That is why research and development go together in the modern research concept. In this paper we are going to examine the applicability and adaptability of gait-based biometric identity verification system especially in the GCC (Gulf Cooperation Council). The system itself closely involved with human interaction where privacy and personality are in concern. As of 1^st phase of our research we will establish gait-based identity verification system and then we will explain them in and out of human interaction with the system. With involved interaction we will conduct an extensive survey to find out both applicability and adoptability of the system. To conduct our experiment, we will use UCMG databased [1] which is readily available for the research community with more than three thousand video sequences in different viewpoint collected in various walking pattern and clothing. For the survey we will prepare questioners which will cover approach of data collection, potential traits to collect and possible consequences. For analyzing gait biometric trait, we will apply multivariate statistical classifier through well-known machine learning algorithms in a ready platform. Similarly, for the survey data analysis we will use similar approach to co-relate the user view point for such system. It will also help us to find the perception of the user for the system.

• Journal of Information Technology and Architecture
• /
• v.11 no.4
• /
• pp.471-483
• /
• 2014

The purpose of this study was to look over the sharing and flow of information on tourism tourist destination made through social media by utilizing social network analysis. In this research for compared the exchange and diffusion relationships and made information sharing among users of the network group for the Seoul / Busan Tourism Twitter. The first part of analysis dealt with the network centralization and information dissemination patterns from a global perspective. The second part was an analysis of nodes with the highest degree centrality in order to identify the most influential user within the Twitter network and determine the information flow and user characteristics with respect to the information dissemination pattern in Seoul/Busan tourism Twitter. The results scientific evidence was presented that social media users' information search behavior information to understand the sharing and spreading patterns, social media as a marketing tool, simple products and services that will ultimately provide the desired information detainee information is not a means of promoting utilization was presented.

• Journal of the Ergonomics Society of Korea
• /
• v.26 no.2
• /
• pp.1-13
• /
• 2007

Ergonomic methods have been effectively applied to design and evaluation of functional clothing. The goals of the present study are to: (1) develop an ergonomic evaluation system for the systematic analysis of functional clothing and (2) examine the usefulness of the proposed evaluation system by applying to flame-proof clothing. Based on the survey of literature and the brainstorming of experts in clothing design and ergonomics, factors considered for clothing evaluation were selected, classified, and complemented, resulting in an ergonomic clothing evaluation system consisting of four factor categories (clothing construction, user, work and environment, and user response). Using the proposed system, a field survey and a laboratory experiment were conducted for flame-proof clothing to identify its design problems. The field survey to workers found a comprehensive set of problems on the flame-proof clothing design in terms of pattern, textile, and color. The laboratory experiment identified additional design problems using a questionnaire that was developed based on an analysis on the relationship between clothing design components and ergonomic evaluation measures. The present study showed the ergonomic evaluation system and the relationship analysis of clothing design components and ergonomic evaluation measures are of use to identify design problems of functional clothing in a comprehensive and analytic manner.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.5
• /
• pp.1001-1009
• /
• 2015

In Smart Grid, smart meter sends our electric signal data to the main server of power supply in real-time. However, the more efficient the management of power loads become, the more likely the user's pattern of usage leaks. This paper points out the threat of privacy and the need of security measures in smart device environment by showing that it's possible to identify the appliances and the specific usage patterns of users from the smart meter's data. Learning algorithm PCA is used to reduce the dimension of the feature space and k-NN Classifier to infer appliances and states of them. Accuracy is validated with 10-fold Cross Validation.

• The Journal of the Korea Contents Association
• /
• v.11 no.7
• /
• pp.497-507
• /
• 2011

This paper develops the steel-house design interface that helps make building design effectively done by constructing building materials as components and assembling them as patterns. It also aims to develop the Flexible Building Design System that supports a technology which makes change of constructing easy and reduces the cost effectively through a simulation of building design like design, analysis, change information, etc., by grafting the virtual building technology into the process of building. It especially is possible for a designer and user to change a building easily by using patterns and according to the change, pattern information of materials needed and the plans of the building are made automatically. Kin-search that can be happened through personal proficiency or knowledge visualization is also the reason why the technology should be embodied.

• 한국HCI학회:학술대회논문집
• /
• 2007.02b
• /
• pp.693-704
• /
• 2007

지금까지 사이버문화에 대한 연구는 해당 문화의 가장 큰 특성 중인 익명성과 그것이 개별 이용자들의 정체성에 미치는 영향에 대한 많은 관심이 있어왔다. 최근 주목을 받고 있는 싸이월드나 블로그 등 CMC환경에서 매체가 일반적으로 여성의 라이프스타일 및 여성성과 연관되어 정체성의 구성에 영향을 미치게 되며, 이러한 이용 형태를 마케팅과 접목하여 높은 여성 참여와 구매력을 이끌어내고 있다. 반면 온라인게임은 남성 중심의 매체로써의 특징을 가지고 있으며, 상대적으로 여성의 온라인게임 이용 행태 분석 및 마케팅 전략 연구가 미흡하였다. 온라인게임에서 여성 이용자의 참여나 구매력을 이끌어 내기 위해서는 우선적으로 남성 중심적 매체에서 여성 온라인게임 유저의 이용 행태 분석 이 선행되어야 하며, 이를 통하여 여성의 온라인게임 소비 패턴과 특성을 파악하고, 나아가 고객 충성도, 구매력과의 관련성 속에서 마케팅적 시사점을 도출하여야 할 것이다. 본 연구는 온라인게임 여성 유저의 심층 인터뷰를 통하여 우선적으로 연령별 게임 이용 패턴 - 게임이용 장소, 초기 진입 방식, 커뮤니티 활동, 아이템 구매 형태 등 - 에 대해 알아보고, 다음으로 여성 유저의 게임 플레이 방식으로 게임 내 캐릭터의 전유 방식(몰입)과 게임 플레이 스타일을 고찰하고자 한다. 이때 여성의 온라인게임 이용 방식이 남성적 매체라는 구조적 공간 내에서 여성의 성적 특성이 온라인게임 소비 행태에 어떠한 영향을 미치며, 또한 여성 유저의 이용 방식에 따라 고객 충성도에 미치는 효과에 대해 알아보도록 한다.

• Journal of the Korean Institute of Educational Facilities
• /
• v.23 no.6
• /
• pp.19-26
• /
• 2016

In this study, moving awareness of students having different physical conditions in homebase and attitudes of students to spacial use in user's aspect were analyzed. As shown in the results, the dissatisfaction degree of the second grade using homebase type was higher than the first and the third grades using exclusive space for a class, homeroom type. The necessity of exclusive space for a class was highly shown in the second grade students. Students of the first and the third grades showed similar pattern in using frequency of homeroom. The results showed that behavior pattern of students was different with their physical conditions. Especially, factors showing obvious differences between the two groups were possession awareness of their belongings, tendency to route choice of high frequency, stay awareness of main stay in time to rest and point of time using restroom. These results will be useful to design various models in variation type of schools.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.9
• /
• pp.4560-4575
• /
• 2018

Since smartphone contains various personal data, security is one of the important aspects in smartphone technologies. Up to now, various authentication techniques have been proposed to protect smartphones. The pattern lock on the Android system is one of the most widely used authentication methods for low-cost devices but it is known to be vulnerable to smudge attack or shoulder surfing attack. LG's smartphone uses its own technique, which is called "Knock Code." The knock code completes the authentication by touching the user defined area in turn on the screen. In this paper, we propose the new, enhanced version of knock code by adding the sliding operation and by using flexible area recognition. We conducted security analysis, which shows that under the same password size, the search space is overwhelmingly larger than the original algorithm. Also, by using the sliding operation, the proposed scheme shows resilience against smudge attacks. We implemented the prototype of our scheme. Experimental results show that compared with the original Knock Code and Android pattern lock, our scheme is more convenient while providing better security.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.42 no.4
• /
• pp.153-164
• /
• 2019

Since the appearance of smartphones, the smartphone market has been in fierce completion by new technologies and marketing trends. The smartphone market is now somewhat saturated, and the manufacturers are trying to improve their position in the market through the repurchase of existing customers and the influx of competitors. At the same time, customers have their own purchasing criteria for smartphones. Therefore, manufacturers need to determine new technology and marketing trends based on customer purchasing trends and usage characteristics. The aim of this study is to analyze the quality attributes of smartphones. We conducted a survey on 220 respondents, and divided the respondents into several groups by purchasing trends and usage characteristic through cluster analysis. The groups are analyzed and compared based on the Kano model for the quality attributes of smartphone. The analysis result are as follows. Firstly, purchasing trends divide responders into groups that prefers high-end premium smartphones and those that take into account practicality in terms of purchasing trends. Secondly, usage characteristic divide responders into three groups: those with clear usage pattern, those who prefer ease of use, and the rest, and we find out that those with clear usage pattern are important customer in viral marketing. Lastly, Kano analysis is revealed the 'Slow/hi-speed camera', 'Private mode', 'Widget', 'Health care' are attractive quality attributes.