• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.1059-1062
• /
• 2021

투자에 대한 관심 증가에 따라 적은 비용과 시간으로 객관적인 정보 제공의 필요성 증가와 함께 인공지능 기술을 활용한 로보어드바이저 서비스가 확대되었다. 또한, 최근 ETF 를 통한 안정적인 투자에 대한 선호도가 증가함에 따라 ETF 중심의 AI 로보어드바이저 추천 서비스가 필요할 것으로 보인다. 하지만, 기존의 투자 어플리케이션에서는 뉴스 기반의 감성적인 요인이 반영되지 않은 추천 방식으로 주가에 영향을 미치는 다양한 요인들을 고려하지 못하는 문제점이 있다. 이에 본 연구에서는 뉴스의 감성분석을 통한 감성지수를 기반으로 새로운 주가 예측 모델을 제안하고, 사용자의 투자 성향 분석을 통한 맞춤 추천 서비스를 통해 개인화된 ETF 서비스를 제공한다.

• Information Systems Review
• /
• v.23 no.3
• /
• pp.51-75
• /
• 2021

Global e-commerce websites offer personalized recommendation services to gain sustainable competitiveness. Existing studies have offered personalized recommendation services using quantitative preferences such as ratings. However, offering personalized recommendation services using only quantitative data has raised the problem of decreasing recommendation performance. For example, a user gave a five-star rating but wrote a review that the user was unsatisfied with hotel service and cleanliness. In such cases, has problems where quantitative and qualitative preferences are inconsistent. Recently, a growing number of studies have considered review data simultaneously to improve the limitations of existing personalized recommendation service studies. Therefore, in this study, we identify review and rating mismatches and build a new user profile to offer personalized recommendation services. To this end, we use deep learning algorithms such as CNN, LSTM, CNN + LSTM, which have been widely used in sentiment analysis studies. And extract sentiment features from reviews and compare with quantitative preferences. To evaluate the performance of the proposed methodology in this study, we collect user preference information using real-world hotel data from the world's largest travel platform TripAdvisor. Experiments show that the proposed methodology in this study outperforms the existing other methodologies, using only existing quantitative preferences.

• Journal of Internet Computing and Services
• /
• v.8 no.5
• /
• pp.81-89
• /
• 2007

In this thesis, we propose ontology based context-aware recommendation system using concept hierarchy(OCARCH), Context-aware recommendation services are useful to provide an user with relevant information and/or services bared on his current context, However several approaches to context-aware recommendation system have been already proposed, each of them provide information without considering level of information concept bared on his current context, For this reason, we propose OCARCH as system capable of helping people to find their way quickly and easily through large amounts of information by determining level of information concept based on his current context, We are also using prefetching algorithm to store recommendation information that the user is likely to need in the near future based on current predictions, Therefore the OCARCH enables users to obtain relevant information efficiently, Several experiments are performed and the experimental results show that the proposed system provides more effective than conventional context-aware recommendation system.

• Smart Media Journal
• /
• v.13 no.1
• /
• pp.67-75
• /
• 2024

The purpose of this research is to investigate consumer responses to production recommendations by AI. In order to test hypotheses of this study, we conducted experimental study that was a 2(Brand reputation: high vs. low) X 2(Social comparison: high vs. low). The results of this study showed the interaction effects of brand reputation and social comparison on brand attitude. Based on the results, we provide theoretical implications to extent the existing research regarding product recommendations. Moreover, the results of this study provide some practical implications and a new aspect about AI recommendations.

• The Journal of the Korea Contents Association
• /
• v.17 no.11
• /
• pp.183-191
• /
• 2017

In order to secure the convenience of information retrieval by users of scientific information service platforms and to reduce the time required to acquire the proper information, this study proposes an optimized content recommendation algorithm among the algorithms that currently provide service menus and content information for each service, and conducts comparative evaluation on the results. To enhance the recommendation accuracy, users' major items were added to the original algorithm, and performance evaluations on the recommendation results from the original and optimized algorithms were performed. As a result of this evaluation, we found that the relevance of the content provided to the users through the optimized algorithm was increased by 21.2%. This study proposes a method to shorten the information acquisition time and extend the life cycle of the results as valuable information by automatically computing and providing content suitable for users in the system for each service menu.

• Science of Emotion and Sensibility
• /
• v.25 no.2
• /
• pp.71-78
• /
• 2022

This study investigates the recommendation for tax accounting services used in many companies. In particular, it aims to create guidelines for small businesses with fewer than 100 employees, which are relatively difficult to manage in terms of cost or time. We surveyed 100 corporate officials on basic business information, such as the number of employees, job titles, and business type, as well as the type of tax accounting service, the recommended score for the service, the reason for the score, and other opinions related to tax accounting services. In particular, the recommendation score seeks to obtain more effective results by using the Net Promoter Score method, which is known to be more effective in understanding customer opinions than general customer satisfaction surveys. The survey revealed a Net Promoter Score for a recommendation of -33 points, lower than the general Net Promoter Score. It also indicated that tax accounting services need improvement. Specifically, the opinions of the respondents who gave a non-recommendation score were as follows: "Not inconvenient or comfortable," "It was just okay," "I don't know if it would be helpful," and "There is no differentiation and there are no special alternatives." We concluded that an improved service for raising recommendation scores was necessary. This survey focused on recommendations for companies with fewer than 100 employees; future studies should incorporate larger companies and more variables.

• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.15-28
• /
• 2015

As personal devices and pervasive technologies for interacting with networked objects continue to proliferate, there is an unprecedented world of scattered pieces of contextualized information available. However, the explosive growth and variety of information ironically lead users and service providers to make poor decision. In this situation, recommender systems may be a valuable alternative for dealing with these information overload. But they failed to utilize various types of contextual information. In this study, we suggest a methodology for context-aware recommender systems based on the concept of contextual boundary. First, as we suggest contextual boundary-based profiling which reflects contextual data with proper interpretation and structure, we attempt to solve complexity problem in context-aware recommender systems. Second, in neighbor formation with contextual information, our methodology can be expected to solve sparsity and cold-start problem in traditional recommender systems. Finally, we suggest a methodology about context support score-based recommendation generation. Consequently, our methodology can be first step for expanding application of researches on recommender systems. Moreover, as we suggest a flexible model with consideration of new technological development, it will show high performance regardless of their domains. Therefore, we expect that marketers or service providers can easily adopt according to their technical support.

• 한국IT서비스학회:학술대회논문집
• /
• 2009.05a
• /
• pp.345-348
• /
• 2009

추천시스템에서 널리 사용되고 있는 협력적 여과기법은 이웃의 정보를 추천대상 고객에게 적용하여 추천에 사용한다. 이 방법을 이용한 추천은 인터넷 사용자에게 알맞은 정보를 제공하여 보다 편리하게 자신이 원하는 정보에 접근하도록 한다. 따라서 추천시스템의 성능향상에 대한 연구가 활발히 진행되고 있으며, 본 논문은 추천시스템의 기능에 대한 정확성을 향상시키기 위한 것이다. 본 논문에서는 먼저, 협력적 여과기법에서 사용되는 고객의 선호도 평가 값에 대한 평균값을 조사하고, 이웃들이 평가한 선호도 평가 값을 분석하였다. 그리고 협력적 여과기법에 두 개의 분석 값을 변수로 적용하여 추천시스템의 예측 정확도를 계산하였다. 본 논문이 제안한 방법과 기존의 알고리즘을 비교한 결과 추천시스템의 성능이 향상됨을 알 수 있다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2016.01a
• /
• pp.81-82
• /
• 2016

최근 사용자들의 추천 서비스를 위해 다른 사용자들의 평가값을 이용하여 특정 사용자에게 서비스를 추천해주는 추천 시스템은 협력 필터링 방법을 널리 사용되고 있다. 하지만 이러한 추천 시스템은 클러스터링 과정에서 이미 분류된 그룹에 특정 사용자가 분류되어 정확히 분류되지 못하고, 사용자들의 평가값 오차가 클 경우 정확하지 못한 결과를 추천하는 문제점이 있다. 본 논문에서는 이러한 문제점을 해결하기 위하여 협력 필터링 알고리즘을 클러스터링 기반으로 분산 환경에서 구현하여, 추천의 효과를 최적화 하는 기법을 제안하며 하둡 및 스파크 기반으로 시스템을 구성하여 협력 필터링 추천 알고리즘을 비교 하였다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.05a
• /
• pp.258-265
• /
• 2005

전자상거래의 확산에 따라 인터넷 쇼핑몰에서의 구매활동은 일반적인 현상이 되었다. 그 결과, 유사한 업종이나 업태의 인터넷 쇼핑몰이 범람하게 되었고 업체들 간의 경쟁도 심화되어 차별화된 서비스를 제공하지 않는 업체는 도태되기 쉬운 상황이다. 본 연구에서는 치열한 경쟁환경 하에서 인터넷 쇼핑몰의 차별화된 마케팅 서비스의 수단으로써 이용되고 있는 상품추천시스템의 개선된 모형을 제시하고자 한다. 본 연구에서 제안하는 모형은 전역 최적화 기법 중의 하나인 유전자 알고리즘을 데이터 마이닝의 도구로 활용한 인터넷 쇼핑몰에서의 개인화된 상품추천시스템 모형이다. 유전자 알고리즘은 추출하기가 어려운 소비자의 성향을 데이터를 통해 추출하고 이에 맞는 상품군을 선택할 수 있도록 해주는 최적화 기법으로 상품추천시스템의 추천엔진으로써 유용할 것으로 기대된다. 본 연구에서는 제안한 유전자 알고리즘에 기반한 추천 규칙들이 장착된 웹 기반의 개인화된 상품추천시스템의 프로토타입을 개발하고 이에 대한 실제 사용자들의 이용 만족도를 확인함으로써 본 연구에서 제안한 방법론의 유용성을 확인하고자 한다.