• 융합신호처리학회논문지
• /
• 제20권1호
• /
• pp.31-36
• /
• 2019

사용자로부터 입력되는 비정형 데이터를 대화 형태로 이해하여 사용자가 원하는 정보에 대한 맞춤 서비스를 제공하는 챗봇은 모바일 서비스의 중요한 분야로서 주목받고 있다. 그러나 사용자의 자연 언어 형태의 질의 대화를 완전하게 이해하여 서비스할 수 있는 방법은 아직 미흡한 실정이다. 따라서 본 논문에서는 사용자가 음식점 추천을 위하여 입력하는 대화 문장으로부터 지역, 음식분류, 음식점명 등의 의미 단어를 추출하고, 추출된 단어를 SNS의 음식점 추천 관련 해시태그를 기반으로 구축된 지식 데이터베이스의 내용과 대조하여 통계적으로 단어 유사성이 가장 큰 사용자 목적 정보를 제공한다. 본 논문에서 구현한 음식점 추천 챗봇 시스템의 성능 평가를 위해서 웹 기반의 모바일 환경을 구축하여 다양한 사용자 질의 정보에 대한 접근 편의성을 측정한 결과, 기존 유사 서비스와 비교하여 터치 횟수와 화면 전환 횟수에서 각각 37.2%와 73.3%의 감소함을 보였다.

• 한국컴퓨터정보학회논문지
• /
• 제22권3호
• /
• pp.35-43
• /
• 2017

In IoT(Internet of Things) environment, users want to receive customized service by users' personal device such as smart watch and pendant. To fulfill this requirement, the mobile device should support a lot of functions. However, the miniaturization of mobile devices is another requirement and has limitation such as tiny display. limited I/O, and less powerful processors. To solve this limitation problem and provide customized service to users, this paper proposes a collaboration system for sharing various computing resources. The paper also proposes the method for reasoning and recommending suitable resources to compose the user-requested service in small device with limited power on expected time. For this goal, our system adopts MBTI(Myers-Briggs Type Indicator) to analyzes user's behavior pattern and recommends personalized resources based on the result of the analyzation. The evaluation in this paper shows that our approach not only reduces recommendation time but also increases user satisfaction with the result of recommendation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제4권2호
• /
• pp.174-190
• /
• 2010

In ubiquitous computing, invisible devices and software are connected to one another to provide convenient services to users [1][2]. Users hope to obtain a personalized service which is composed of customized devices among sharable devices in a ubiquitous smart space (which is called USS in this paper). However, the situations of each user are different and user preferences also are various. Although users request the same service in the same USS, the most suitable devices for composing the service are different for each user. For these user requirements, this paper proposes a device recommender system which infers and recommends customized devices for composing a user required service. The objective of this paper is the development of the systems for recommending devices through context-aware inference in peer-to-peer environments. For this goal, this paper considers the context and user preference. Also I implement a prototype system and test performance on the real ubiquitous mobile object (UMO).

• 국제학술발표논문집
• /
• The 3th International Conference on Construction Engineering and Project Management
• /
• pp.935-941
• /
• 2009

In competitive society, performance management is an essential element of business success. Despite the importance of performance management, it has not been widely implemented in construction companies. In the recent years, construction companies have become more aware of its need to identify, implement, and sustain performance improvements more systematically. Thus, the objective of this paper is to develop performance management system (PMS) for construction companies. The roles of performance management system is not only measuring performance but also offering guidance to improve performance. Therefore, PMS needs benchmarking process that provides Best Practice and new knowledge. This paper investigates the case of PMSs in UK, USA, Brazil, and Chile and discusses the lessons learned. To overcome the limitations of existing PMSs, new performance measurement framework, in form of 'Construction' BSC, is proposed. Based on the 'Construction' BSC, key performance indicators are derived and methodology of performance management is suggested. This paper concludes by developing PMS for benchmarking in construction companies and recommending some further directions on this research topic.

• 대한안전경영과학회지
• /
• 제10권4호
• /
• pp.309-314
• /
• 2008

Many studies on predicting and recommending information and products have been studying to meet customers' preference. Unnecessary information should be removed to satisfy customers' needs in massive information. The some information filtering methods to remove unnecessary information have been suggested but these methods have scarcity and scalability problems. Therefore, this paper explores a personalized recommendation system based on artificial neural network (ANN) to solve these problems. The insurance product recommendation is adapted as an example to demonstrate the proposed method. The proposed recommendation system is expected to recommended a suitable and personalized insurance products for customers' satisfaction.

• Journal of Information Science Theory and Practice
• /
• 제10권1호
• /
• pp.45-58
• /
• 2022

Community enterprise plays an important role for developing local business. Products from local communities apply local specialties such as high-quality materials and inherited wisdom. This work aims to support merchandises from local community enterprises by bringing out their specialties related to local wisdom and intangible cultural aspects. An ontology is applied to demonstrate the innate information regarding the implicit values of the products and is used as a core for a semantic search system. Details of the products are gathered from their respective community using an interview method and are extracted to align with the developed ontological schema. The semantic search system thus is implemented with a recommendation process for online accessibility for providing the organised information. From evaluation, the developed ontology and its instances are rated highly for their consistency, conciseness, and completeness. In usage, accuracy of the query and recommendation results are evaluated at 97.38% searching accuracy and 85.03% for recommending interesting products.

• 응용통계연구
• /
• 제22권6호
• /
• pp.1229-1237
• /
• 2009

노인장기요양보험에서 가장 중요한 이슈는 급여대상자의 희망, 건강 및 기능상태에 따라 어떤 급여를 제공할 것인가 이다. 이를 해결하고자 노인장기요양보험의 보험자인 국민건강보험 공단은 급여대상자에게 '표준장기요양이용계획서'를 제공하고 있다. 본 연구에서는 표준장기요양이용계획 작성의 효율화 방안을 마련하고자 노인장기요양보험 3차 시범사업 표준이용계획 자료를 활용하여 노인장기요양급여 권고모형을 개발하였다. 모형개발에는 데이터마이닝의 의사결정나무모형, 로지스틱회귀모형, 앙상블 모형의 배깅과 부스팅 기법을 사용하였고, 이 중 실무자가 이해하기 쉬운 의사결정나무를 채택하여 권고모형을 설명 하였다. 본 연구는 노인장기요양보험 제도의 이용계획 수립의 객관성 및 과학성을 확보하고 이용계획 업무를 효율화하는 데에 기여할 것으로 기대된다.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2004년도 추계학술대회
• /
• pp.203-210
• /
• 2004

Due to the explosion of e-commerce, recommender systems are rapidly becoming a core tool to accelerate cross-selling and strengthen customer loyalty. There are two prevalent approaches for building recommender systems - content-based recommending and collaborative filtering. Collaborative filtering recommender systems have been very successful in both information filtering domains and e-commerce domains, and many researchers have presented variations of collaborative filtering to increase its performance. However, the current research on recommendation has paid little attention to the use of time related data in the recommendation process. Up to now there has not been any study on collaborative filtering to reflect changes in user interest. This paper proposes dynamic fuzzy clustering algorithm and apply it to collaborative filtering algorithm for dynamic recommendations. The proposed methodology detects changes in customer behavior using the customer data at different periods of time and improves the performance of recommendations using information on changes. The results of the evaluation experiment show the proposed model's improvement in making recommendations.

• Journal of information and communication convergence engineering
• /
• 제17권1호
• /
• pp.41-48
• /
• 2019

In recent years, sentiment analysis research has become popular. The research results of sentiment analysis have achieved remarkable results in practical applications, such as in Amazon's book recommendation system and the North American movie box office evaluation system. Analyzing big data based on user preferences and evaluations and recommending hot-selling books and hot-rated movies to users in a targeted manner greatly improve book sales and attendance rate in movies [1, 2]. However, traditional machine learning-based sentiment analysis methods such as the Classification and Regression Tree (CART), Support Vector Machine (SVM), and k-nearest neighbor classification (kNN) had performed poorly in accuracy. In this paper, an improved kNN classification method is proposed. Through the improved method and normalizing of data, the purpose of improving accuracy is achieved. Subsequently, the three classification algorithms and the improved algorithm were compared based on experimental data. Experiments show that the improved method performs best in the kNN classification method, with an accuracy rate of 11.5% and a precision rate of 20.3%.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2022년도 춘계학술대회
• /
• pp.655-656
• /
• 2022

추천 시스템을 위한 분석방법들 가운데 협업 필터링은 데이터 분석에 기반한 추천 시스템에서 주요 대표적 방법이다. 일반적 사용 방법은 다양한 아이템에 대해서 사용자들의 평가 데이터를 활용하여 공통적인 패턴을 찾으며, 특정 사용자에 대한 선호 아이템을 추천하는 기법이다. 이에 본 논문에서는 여러가지 알고리즘을 사용하여 지표 측정에 활용하였으며, 사용자 선호에 대한 예측에 적합한 알고리즘을 찾아서 제시하였다.