• 한국컴퓨터정보학회논문지
• /
• 제28권9호
• /
• pp.81-92
• /
• 2023

프라이버시 모델이란 프라이버시 공격을 통한 개인정보의 유출 가능성과 위험 정도를 정량적으로 제한하는 기법이다. 대표적인 모델로 k-익명성, l-다양성, t-근접성, 차분 프라이버시 등이 있다. 지금까지 많은 프라이버시 모델들이 연구되어 왔지만, 주어진 데이터에 대해 가장 적합한 모델을 선택하는 문제에 대한 연구는 미흡하다. 본 연구에서는 개인정보 유출 문제를 막기 위한 최적의 프라이버시 모델 추천 시스템을 개발한다. 본 논문에서는 프라이버시 모델 선택 시 고려해야 할 데이터 특성(예: 데이터 타입, 분포, 빈도, 범위 등)을 분석하고 데이터 특성과 모델 간의 연관관계정보를 포함하는 프라이버시 모델 배경지식에 기반한 최적 모델을 추천한다. 마지막으로 타당성과 유용성을 검증하기 위해 추천 프로토타입 시스템을 구현하였다.

• 디지털산업정보학회논문지
• /
• 제8권4호
• /
• pp.43-51
• /
• 2012

Web 3.0 is the next-generation of the World Wide Web and is included two main platforms, semantic technologies and social computing environment. The basic idea of web 3.0 is to define structure data and link them in order to more effective discovery, automation, integration, and reuse across various applications. The semantic technologies represent open standards that can be applied on the top of the web. The social computing environment allows human-machine co-operations and organizing a large number of the social web communities. In the recent years, recommender systems have been combined with ontologies to further improve the recommendation by adding semantics to the context on the web 3.0. In this paper, we study previous researches about recommendation method and propose a recommendation method based on web 3.0. Our method scores documents based on context tags and social network services. Our social scoring model is computed by both a tagging score of a document and a tagging score of a document that was tagged by a user's friends.

• 한국식품영양학회지
• /
• 제30권5호
• /
• pp.943-959
• /
• 2017

This study was conducted to survey the perception and preferences of customers that have dined at Korean restaurants in China and investigate the importance and performance level of quality attributes, customer satisfaction, revisit intention and recommendation intention. The survey was conducted January 31~March 1, 2016 in China. The 293 questionnaires (97.7%) were analyzed using SPSS(Ver. 23.0) and AMOS(Ver. 21.0). Results of this study are as follow: Customers that dined at a Korean restaurant in China were composed of 157 women and 136 men. Regarding the reason for preferring Korean cuisine, taste, hygiene and nutritional value of Korean food were the most significant quality factors. Regarding complaints about Korean food, Chinese people placed much emphasis on freshness of ingredients when dining out, based on the majority of complaints about ingredients that were not fresh. The main reason for leftover food were personal eating habits and that of customers revisiting food taste and nutrition. Path model among customer satisfaction, revisit intention and recommendation intention revealed the factor of menus and attributes of menu items regarding customer's age that had an impact on customers' satisfaction, and association with customers' satisfaction, revisit intention and recommendation intention as well.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권5호
• /
• pp.1708-1727
• /
• 2021

In the advent of the twenty-first century, human beings began to closely interact with technology. Today, technology is developing, and as a result, the world wide web (www) has a very important place on the Internet and the significant task is fulfilled by Web services. A lot of Web services are available on the Internet and, therefore, it is difficult to find matching Web services among the available Web services. The recommendation systems can help in fixing this problem. In this paper, our observation was based on the recommended method such as the collaborative filtering (CF) technique which faces some failure from the data sparsity and the cold-start problems. To overcome these problems, we first applied an ontology-based clustering and then the k-nearest neighbor (KNN) algorithm for each separate cluster group that effectively increased the data density using the past user interests. Then, user ratings were predicted based on the model-based approach, such as singular value decomposition (SVD) and the predictions used for the recommendation. The evaluation results showed that our proposed approach has a less prediction error rate with high accuracy after analyzing the existing recommendation methods.

• 아태비즈니스연구
• /
• 제13권2호
• /
• pp.169-182
• /
• 2022

Purpose - The purpose of this study was to investigate influence relationship among health lifestyle, customer satisfaction, repurchase intention and recommendation intention in the protein beverage market. Design/methodology/approach - This study collected 286 survey data from customers who had experience buying and drinking the protein beverage. The Exploratory Factor Analysis (EFA) and the multiple regression analysis were hired in order to analyze the data. Findings - First, four dimensions of health lifestyle("health confidence," "health sensitivity," "health intention," and "health eating habit") were found to be valid and reliable. Second, all four dimensions of health lifestyle had a positive effect on customer satisfaction. Third, customer satisfaction had a positive effect on repurchase intention. Lastly, customer satisfaction had a positive effect on recommendation intention. Research implications or Originality - This study provided research model among health lifestyle, customer satisfaction, repurchase intention and recommendation. Furthermore, the results of this study were useful for identifying the role of health lifestyle in estimating customer satisfaction and the strategies for strengthening customer satisfaction in the protein beverage market.

• 한국컴퓨터정보학회논문지
• /
• 제26권6호
• /
• pp.19-28
• /
• 2021

코로나로 인해 건강관리에 대한 관심이 증가하고 있는 요즘, 여러 사람이 함께 이용하는 헬스장이나 공용시설을 이용하는데 어려움이 늘어남에 따라 홈 트레이닝을 하는 이들이 늘어나고 있다. 이에 본 연구에서는 홈 트레이닝 사용자들에게 좀 더 정확하고 의미 있는 운동 추천을 제공하기 위해 개인 성향 정보를 활용한 개인화된 운동 추천 알고리즘을 제안한다. 이를 위해 식습관 정보, 육체적 조건 등 개인을 나타낼 수 있는 개인 성향 정보를 사용해 k-최근접 이웃 알고리즘으로 데이터를 비만의 기준에 따라 분류하였다. 또한, 운동 데이터 셋을 운동의 레벨에 따라 등급을 구별하였으며 각 데이터 셋의 이웃 정보를 바탕으로 모델 기반 협업 필터링 방법 중 차원 축소모델인 특이값 분해 알고리즘(SVD)을 통해 사용자들에게 개인화된 운동 추천을 제공한다. 따라서 메모리 기반 협업 필터링 추천 기법의 데이터 희소성과 확장성의 문제를 해결할 수 있고, 실험을 통해 본 연구에서 제안하는 알고리즘의 정확도와 성능을 검증한다.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 추계학술발표대회
• /
• pp.377-379
• /
• 2022

POI refers to the point of Interest in Location-Based Social Networks (LBSNs). With the rapid development of mobile devices, GPS, and the Web (web2.0 and 3.0), LBSNs have attracted many users to share their information, physical location (real-time location), and interesting places. The tremendous demand of the user in LBSNs leads the recommendation systems (RSs) to become more widespread attention. Recommendation systems assist users in discovering interesting local attractions or facilities and help social network service (SNS) providers based on user locations. Therefore, it plays a vital role in LBSNs, namely POI recommendation system. In the machine learning model, most of the training data are stored in the centralized data storage, so information that belongs to the user will store in the centralized storage, and users may face privacy issues. Moreover, sharing the information may have safety concerns because of uploading or sharing their real-time location with others through social network media. According to the privacy concern issue, the paper proposes a recommendation model to prevent user privacy and eliminate traditional RS problems such as cold-start and data sparsity.

• 경영정보학연구
• /
• 제23권4호
• /
• pp.69-86
• /
• 2021

인공지능 기술이 발달함에 따라 빅데이터를 활용한 개인화 추천시스템에 대한 관심이 증가하고 있다. 특히 뷰티 제품의 경우 개인의 취향과 더불어 피부 특성 및 민감도에 따라 제품 선호도가 명확히 구분되므로 축적된 고객 데이터를 활용하여 고객 맞춤형 추천서비스를 제공하는 것이 필요하다. 따라서 본 연구에서는 딥러닝 기법을 활용하여 제품 검색 기록과 개인 사용자의 피부 타입과 고민 등의 콘텍스트 정보를 함께 반영한 심층 신경망 기반의 추천시스템 모델을 제시하고자 한다. 본 연구에서는 실제 화장품 검색 플렛폼의 데이터를 활용하여 성능 평가를 실시하였다. 본 연구의 실험 결과, 고객의 콘텍스트 정보를 포함한 모델이 제품 검색 기록만을 활용한 기존의 협업 필터링 모델들 보다 우수한 성능을 보임을 확인하였다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권7호
• /
• pp.1569-1584
• /
• 2013

Ubiquitous interaction in a pervasive environment is the main attribute of smart spaces. Pervasive systems are weaving themselves in our daily life, making it possible to collect user information invisibly, in an unobtrusive manner by known and even unknown parties. Huge number of interactions between users and pervasive devices necessitate a comprehensive trust model which unifies different trust factors like context, recommendation, and history to calculate the trust level of each party precisely. Trusted computing enables effective solutions to verify the trustworthiness of computing platforms. In this paper, we elaborate Unified Trust Model (UTM) which calculates entity's trustworthiness based on history, recommendation, context and platform integrity measurement, and formally use these factors in trustworthiness calculation. We evaluate UTM behaviour by simulating in different scenario experiments using a Trust and Reputation Models Simulator for Wireless Sensor Networks. We show that UTM offers responsive behaviour and can be used effectively in the low interaction environments.

• International Journal of Contents
• /
• 제5권2호
• /
• pp.20-25
• /
• 2009

The purpose of this study is to seek new u-business services in restaurant management. Using the concept of business model methodology in family restaurant management domain, this study identifies customers' needs in services at the stage of management of purchase of materials, the production management, and the sales management. In addition, this study suggests two killer applications of a family restaurant management linking with the latest ubiquitous computing technologies: the service of the customer-oriented menu recommendation and the service of the inventory-oriented menu recommendation. These findings may offer practical insights in the context of ubiquitous service model of restaurant management.