• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.19 no.1
• /
• pp.113-118
• /
• 2015

Recommendation system is filtering for users require appropriate information from increasing information. Recommendation system is provides the information based on user information or content that information entered in the original through process of filtering through the algorithm. Recommend system is problems with Cold-start, and Cold-start is not enough information in the occurrences for new users of recommend system in the new information to the user when recommend. Cold-start is should meet to resolve the user of information and item information. In this paper, Suggest for movie recommendation system on collaborative filtering techniques and content-based filtering techniques based to a hybrid of a hybrid filtering techniques to solve problems in cold-start.

• Journal of Digital Convergence
• /
• v.19 no.2
• /
• pp.299-305
• /
• 2021

This study examined to identify the factors influencing on inpatients' satisfaction with hospitalization and willingness to recommend hospital to others. Data from the 2018 National Patient Experience Survey were used for the analysis. Of the 593 patients experiencing inpatient services, multivariate linear regression analysis was conducted to explore the factors such as physician care, nursing care, facility and administrative support, and healthcare system on inpatients' satisfaction after controlling for their socio-demographic and health related factors. The study found that physician care, nursing care, administrative support and healthcare system were related to inpatients' satisfaction and willingness to recommend hospital. To improve inpatients' satisfaction, it is necessary to improve healthcare professionals' overall patient-centerdness attitude, user friendly hospital facilities and administrative support services and efforts to trust and satisfaction on healthcare system from the patients' perspectives.

• Journal of Society of Preventive Korean Medicine
• /
• v.26 no.3
• /
• pp.49-58
• /
• 2022

Objectives : The purpose of this study is to analyze the factors affecting the intention of traffic accident patients, who had visited Korean medicine clinics for the purpose of treating traffic accidents, on revisiting and recommending those clinics to others. Methods : This study conducted the frequency analysis, Rao-scott chi-square test, and logistic regression analysis on 389 people, who answered that they had once visited Korean medicine clinics for treatment in traffic accidents, using data from the 2020 Korean Medicine Utilization and Herbal Medicine Consumption Survey. Results : As a result of the analyses, it was revealed that the significant influencing factors entailed marital status, job status, the attitude of medical staff, and access to the Korean medicine clinics, while only access to the Korean medicine clinics was a significant influencing factor for the intention to recommend to others. Especially, the intention of to revisit and to recommend in case of satisfying access to the Korean medicine clinics were 8.476 times and 6.784 times higher than when it is not the case. Conclusions : The results of this study reflect the characteristics of automobile insurance, and indicate that both further study and policy establishment on the operation of the automobile insurance system are required to ensure sufficient treatment for traffic accident patients.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.7
• /
• pp.768-774
• /
• 2016

This paper deals with a method to recommend the combination of items as a group according to similarity to handle application area such as fashion and cooking, while the previous methods recommend single item such as a book, music or movie. Collaborative filtering is a method to recommend an item selected by users with similar tendency based on similarity between users. In this paper, the proposed method generates a set of frequent items based on collaborative filtering and association rules and recommends a group by similarity between groups. To show the validity of the proposed method, experiments are performed with purchase data collected from e-commerce for four months.

• Proceedings of the CALSEC Conference
• /
• 2005.11a
• /
• pp.147-153
• /
• 2005

When a customer wants to buy an item at the Internet shopping mall, one of the difficulties is to decide when to buy the item because its price changes over time. If the shopping mall can be able to recommend appropriate buying points, it will be greatly helpful for the customer. Therefore, in this presentation, we propose a method to recommend buying points based on the time series analysis using a database that contains past prices data of items. The procedure to provide buying points for an item is as follows. First, we search past time series patterns from the database using normalized similarity, which are similar to the current time series pattern of the item. Second, we analyze the retrieved past patterns and predict the future price pattern of the item. Third, using the future price pattern, we recommend when to buy the item.

• The Journal of the Korea Contents Association
• /
• v.7 no.12
• /
• pp.191-198
• /
• 2007

The main purpose of the study is to figure out visitors' intentions for the Suanbo Spa Festival after they participated the events, programs offered from the Suanbo Festival organization. The visitors were also asked to evaluate contents, images, and convenience facilities of the festival. The study found that only visitors over 60s of age positively evaluate the festival, therefore, products such as traditional tea-making, herb-massage will attract old aged group. The findings suggests that event image and convenient facilities have impact on tourists' responses and intentions to their neighbors. The result of the study also recommend traditional participations such as horseback-riding, GaMa-riding, Korean traditional archery shooting to be added to the participating program.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.798-802
• /
• 2010

The integration of GIS and fuzzy MCDA(Multi-Criteria Decision Analysis) allows the engineer to determine the preferred alternative for each spatial location in the study area. The next step is to recommend to the final decision makers a single flood management alternative for the entire region. Note that if the study area is large, it might be possible to use the kind of information to recommend different alternatives for different portions of the region. However, for this study it is assumed that only a single alternative will be used. In this study, a "cost of uniformity" metric is proposed that allows decision makers to compute the impact of selecting a single alternative for the entire floodplain. This metric represents the increase in the average distance metric value as compared to the spatially diverse solution from the MCDA and GIS analysis. The results could be applied to any region of the floodplain as desired. Whether the decision makers decide to apply these calculations to the entire floodplain or to specific important regions within the floodplain, an analysis of the increases in the cost of uniformity provides an integrated way for the decision maker to rank the alternatives. This should provide an improvement in their engineering analysis.

• The Journal of the Korea Contents Association
• /
• v.21 no.6
• /
• pp.81-93
• /
• 2021

This study examined the effects of cosmetic selection attributes on consumers' purchase propensity and recommend intention. In addition, the moderating effect according to the consumer's deal proneness was verified. Finally, a multi-group analysis was conducted to verify the difference in the research model path according to the information source. Through this study, the selection attributes of consumers who purchase cosmetics were clarified. This study aims to meet the needs and demands of the related industry for more detailed and effective strategic insights by clarifying the structure of the influence of these selection attributes on purchase intention and recommendation intention according to the discount purchase intention. In order to achieve the research objectives due to this necessity, a questionnaire of 258 Korean female consumers was collected and used for research. The analysis results showed that product selection attributes, purchase intention, and recommendation intention all had a positive influence. As a result of analyzing the moderation effect according to the consumer's Deal Proneness, the results showed a moderating effect between the selection attribute and the purchase propensity, the selection attribute and the recommend intention, and the purchase propensity and the recommend intention. Finally, it was partially adopted as a result of conducting a multi-group analysis to verify whether individual paths of the model differ according to information sources.

• The KIPS Transactions:PartB
• /
• v.15B no.4
• /
• pp.347-354
• /
• 2008

In this paper, we propose a method for personalizing and intelligence of speech Web. The proposed system records information that was demanded in the past as a transaction, explores association rules from those transactions, and discovers itemsets from frequent requests. This method is to recommend relevant information, based on frequent itemsets, to users who have similar inclinations to previous users. As a result of experimenting and implementation of proposed system for verification, we confirmed that the proposed system can recommend previously frequently requested information as relevant information.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2006.05a
• /
• pp.109-112
• /
• 2006

최근 대용량 추천시스템에 대한 필요성이 증가하고 있고, 특히 대규모 인터넷 쇼핑몰을 위한 개인화 추천 시스템 구조에 대한 관심이 높아지고 있다. 본 논문에서는 k-means 클러스터링과 순차 패턴 기법을 이용한 인터넷 쇼핑몰 상품 추천 시스템을 설계 및 구현한다. 사용자 정보의 일괄처리와 카테고리의 계층적 특성을 반영하면서 데이터 마이닝 기법을 활용하여 개인화된 추천 엔진을 대형 시스템에서 동작하도록 설계 하였다. 설계 구현한 시스템의 평가를 위해, 대형 쇼핑몰의 데이터를 이용하여 추천 예측 정확율(PRP: Predictive Recommend Precision), 추천 예측 재현율(PRR: Predictive Recommend Recall), 정확도 인수(PF1 : Predictive Factor One-measure)를 구하였다.