• Proceedings of the Korea Information Processing Society Conference
• /
• 2010.04a
• /
• pp.353-354
• /
• 2010

최근 사용자의 취향에 맞는 콘텐츠를 필터링하여 자동으로 추천하는 연구가 활발히 진행되고 있다. 참여형, 개방형, 공유형 서비스들의 증가와 함께 웹 3.0 시대에는 더욱 지능화되고 개인화된 서비스가 중요시되고, 이를 위한 맞춤형 정보 제공 연구가 필수적이다. 본 논문에서는 사용자 맞춤형 추천 방법의 대표적인 기술인 협업 필터링(Collaborative Filtering) 방안 분석에 대해 설명하고, 협업 필터링 방법의 적용 연구를 설명한다.

• Journal of Information Management
• /
• v.40 no.1
• /
• pp.199-215
• /
• 2009

The web 2.0 concept rapidly spreads to the various field which is based on an opening, the participation, and a share. And the research about the recommendation system, that is the personalize feature, and social network is very active. In the case of the recommendation system and social network, it had been developing in the respectively different area and the new research toward the service model of a form that it fuses these is insignificant. In this paper, I'm going to introduce efficient social network which is called the researcher connection network. It is possible to recommend the researcher intellectually who studies the similar field by analyzing the usage log and user profile. Through this study, we could solved the network expandability problem which is due to the user passive participation and the difficulty of the initial network construction that is the conventional social network problem.

• Information Systems Review
• /
• v.22 no.4
• /
• pp.75-92
• /
• 2020

As population and generation structures change, more and more customers tend to avoid facing relation due to the development of information technology and spread of smart phones. This phenomenon consists with efficiency and immediacy, which are the consumption patterns of modern customers who are used to information technology, so offline network-oriented distribution companies actively try to switch their sales and services to untact patterns. Recently, untact services are boosted in various fields, but beauty products are not easy to be recommended through untact services due to many options depending on skin types and conditions. There have been many studies on recommendations and development of recommendation systems in the online beauty field, but most of them are the ones that develop recommendation algorithm using survey or social data. In other words, there were not enough studies that classify segments based on user information such as skin types and product preference. Therefore, this study classifies customer segments using machine learning technique K-prototypesalgorithm based on customer information and search log data of mobile application, which is one of untact services in the beauty field, based on which, untact marketing strategy is suggested. This study expands the scope of the previous literature by classifying customer segments using the machine learning technique. This study is practically meaningful in that it classifies customer segments by reflecting new consumption trend of untact service, and based on this, it suggests a specific plan that can be used in untact services of the beauty field.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.673-676
• /
• 2021

본 연구에서는 Google Dialogflow 자연어 처리 엔진(NLP 엔진), 정보수집(크롤링), iOS 챗봇 애플리케이션을 통해 정부 및 지자체 혜택 및 정보 추천 서비스를 제공하는 챗봇 구현을 제안한다. 해당 챗봇은 디지털 기기 사용이 능숙하지 않은 중, 장년층 사용자가 쉽게 이용할 수 있도록 접근성을 높이고, 정부 및 지자체에서 제공하는 다양한 혜택 및 정보의 불균형과 격차의 해소를 목적으로, 사용자가 선택한 지역에 따른 혜택·복지·소식 정보를 제공 및 추천한다. 이 과정을 통해 사용자는 자신이 원하는 분야의 정부 및 지자체의 적절한 복지 정보를 추천 받을 수 있다.

• Science of Emotion and Sensibility
• /
• v.9 no.spc3
• /
• pp.179-185
• /
• 2006

This study examines and compares various recommendation techniques which have been used successfully in other fields and seeks for opportunity to improve design personalization service more effectively. Throughout the literature study, several major recommendation techniques were identified, namely 'contents-based filtering', 'collaborative filtering', and 'demographic filtering'. In order for finding out relative advantages and disadvantages, a case study was carried out by applying different techniques. The result showed that in general, demographic filtering was evaluated least efficient among the techniques. Content-based filtering showed the best efficiency among them. Another significant finding was that the collaborative filtering had a better efficiency as the number of test subjects is increased. In conclusion, we suggest that design recommendation services can be improved by applying contents-based or collaborative filtering for better efficiency of recommendation. And, if the number of test subjects is large enough, it may be possible to remarkably improve the efficiency of design recommendation services by using collaborative filtering.

• Journal of the Korean Society for Library and Information Science
• /
• v.54 no.3
• /
• pp.41-69
• /
• 2020

The purpose of this study is to design and develop a alternative material recommendation system using automatic classification, based on user preference. Details of usage data by users from DREAM was analysed in order to develop the way of a method on selecting proper alternative material, and then the data by user preference were allocated under each category of 10 KDC categories. The keyword, selected from the title of users' usage data from a certain period of time, were divided into 10 subject categories and ranked by the order of frequency of appearance. Books including high frequency of the keyword in title can be selected as a preferred target for producing alternative materials. Lastly, a dynamic linkage for sharing usage data among National Library for the Disabled and other libraries is proposed to produce more proper alternative materials, based on user preference.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.6
• /
• pp.79-94
• /
• 2018

The purpose of this study is to specify a probabilistic tracking mechanism for customer luxury purchase implemented by hidden Markov model, Bayesian inference, customer satisfaction and net promoter score. In this paper, we have designed a probabilistic model based on customer's actual data containing purchase or non-purchase states by tracking the SPC chain : customer satisfaction -> customer referral -> purchase/non-purchase. By applying hidden Markov model and Viterbi algorithm to marketing theory, we have developed the statistical model related to probability theories and have found the best purchase pattern scenario from customer's purchase records.

• 한국정보교육학회:학술대회논문집
• /
• 2010.08a
• /
• pp.281-287
• /
• 2010

Conventional paper-based assessment methods for Gifted students to be selected are changed into observed and recommended method, and this is focus on developing the potential of gifted students. Information is subject to the school curriculum does not address the important observation because of recommendations from teachers' attention away from ideas on how recommendations are being observed is very low. To solve this problem, existing information about gifted education gifted characteristics of the subjects studied by analyzing the proposal and recommended elements of a feasibility study is to describe. The results of this study recommended that teachers or teachers' information, observing the subjects in the field of gifted education is likely to take advantage of them when detail.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.3
• /
• pp.435-439
• /
• 2022

Many of the industry was changed because of the pandemic of covid 19. It combined with the tendency of modern people to pursue convenience. The industry of Cosmetics also changed business channel from offline to online. Before, people can not get suggestions after they complete the survey. This paper research how to suggest some cosmetics products with their skin type and skin data. We will develop Beauty Concierge system that can get suggestion after the survey. It's will make people attend activity and can make more benefit to the people.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.673-675
• /
• 2023

그래프 데이터는 데이터간의 관계를 효율적으로 분석할 수 있으며, 뛰어난 확장성, 다양한 종류의 데이터들을 쉽게 표현할 수 있어 화학, 의학, 추천시스템등 다양한 분야에 적용하려는 사례가 늘고 있다. 이러한 그래프 데이터를 머신러닝기법에 쉽게 사용할 수 있도록 적용된 것이 GNN모델이다. 그 중 Convolultion기법을 적용한 ConvGNNs 모델이 추천 시스템 등 다양한 분야에서 많이 연구 되고 있다. 본 논문은 실험을 통해 상이한 데이터셋 환경에서 Convolution 그래프 기반 모델들의 성능을 비교하였다.