• 디지털산업정보학회논문지
• /
• 제20권1호
• /
• pp.47-58
• /
• 2024

The serving robot is defined as a robot that carries cooked food to a customer table or collects and carries bowls after eating. This study presented a research model to find out the causal relationship between customer satisfaction and continuous use intention through the three factors of serving robot service quality. In addition, in order to empirically verify the above research model, a survey was conducted on customers using serving robots within the last 5 months. 300 copies were analyzed using SPSS 20 as the final analysis, excluding unfaithful responses. The main findings of this study are as follows. First, serving robot service quality (typical) found to have significant effect on customer satisfaction and continuous use intention. Second, it was found that serving robot service quality (reliability) had a significant effect on customer satisfaction and continuous use intention. Third, it was found that serving robot service quality (ease) did not significantly affect customer satisfaction and continuous use intention.

• 인터넷정보학회논문지
• /
• 제23권3호
• /
• pp.97-110
• /
• 2022

연합학습은 학습데이터의 전송없이 모델의 학습을 수행할 수 있는 학습방법이다. IoT 혹은 헬스케어 분야는 사용자의 개인정보를 다루는 만큼 정보유출에 민감하여 시스템 디자인에 많은 주의를 기울여야 하지만 연합학습을 사용하는 경우 데이터가 수집되는 디바이스에서 데이터가 이동하지 않기 때문에 개인정보 유출에 자유로운 학습방법으로 각광받고 있다. 이에 따라 많은 연합학습 구현체가 개발되었으나 연합학습을 사용하는 시스템의 개발과 운영을 위한 시스템 설계에 관한 구체적인 연구가 부족하다. 본 연구에서는 연합학습을 실제 프로젝트에 적용하여 IoT 디바이스에 배포하고자 할 때 연합학습의 수명주기, 코드 버전 관리, model serving, 디바이스 모니터링에 대한 대책이 필요함을 보이고 이러한 점을 보완해주는 개발환경에 대한 설계를 제안하고자 한다. 본 논문에서 제안하는 시스템은 중단 없는 model-serving을 고려하였고 소스코드 및 모델 버전 관리와 디바이스 상태 모니터링, 서버-클라이언트 학습 스케쥴 관리기능을 포함한다.

• 한국재난정보학회 논문집
• /
• 제19권2호
• /
• pp.432-438
• /
• 2023

연구목적: 생활안전 예방서비스 앱에서 신고되는 이미지를 AI를 사용하여 실시간으로 위험 카테고리를 자동으로 분류하여 사용자에게 편리한 위험신고를 가능하게 하는 것을 목적으로 한다. 연구방법: 인터넷으로 상호연결되는 생활안전 예방서비스 플랫폼, 생활안전 예방서비스 앱, AI 모델 서빙 서버와 sftp 서버로 구성되는 시스템을 통하여 신고된 생활안전 이미지를 실시간으로 자동분류하며, 이때 사용되는 AI모델 생성을 위한 AI 학습 알고리즘도 개발하였다. 연구결과: 이미지를 실시간으로 AI 처리하여 자동으로 분류할 수 있게 되어, 신고자가 생활안전 관련 사항을 보다 편리하게 신고할 수 있게 되었다. 결론: 본 논문에서 제시하는 AI 이미지 자동분류 시스템은 90% 이상의 분류 정확도로 신고 이미지를 실시간으로 자동분류하여 신고자가 간편하게 생활안전 관련 이미지를 신고할 수 있게 되었으며 향후 생활안전 예방서비스 앱의 사용자의 증가에 따라 더욱 빠르고 정확한 AI 모델 개발 및 시스템 처리용량 향상이 필요하다.

• International Journal of Internet, Broadcasting and Communication
• /
• 제13권4호
• /
• pp.135-142
• /
• 2021

Estimating the locations of workers in a shipyard is beneficial for a variety of applications such as selecting potential forwarders for transferring data in IoT services and quickly rescuing workers in the event of industrial disasters or accidents. In this work, we propose a human movement stream processing system for estimating worker locations in shipyards based on Apache Spark and TensorFlow serving. First, we use Apache Spark to process location data streams. Then, we design a worker location prediction model to estimate the locations of workers. TensorFlow serving manages and executes the worker location prediction model. When there are requirements from clients, Apache Spark extracts input data from the processed data for the prediction model and then sends it to TensorFlow serving for estimating workers' locations. The worker movement data is needed to evaluate the proposed system but there are no available worker movement traces in shipyards. Therefore, we also develop a mobility model for generating the workers' movements in shipyards. Based on synthetic data, the proposed system is evaluated. It obtains a high performance and could be used for a variety of tasksin shipyards.

• 한국식생활문화학회지
• /
• 제33권5호
• /
• pp.426-436
• /
• 2018

This study assessed effective strategies to reduce the sodium intake among consumers using pork cutlet sauce as a model food system. Original pork cutlet sauce and sodium-reduced sauce (29% reduced by a salt substitute) were analyzed to characterize the sensory properties using descriptive analysis. The effects of sodium-reduction of the sauce, consumer type (nutrition teachers vs. general consumers), information related to the sodium content, serving method, and consumer's health, taste and sodium-related attitudes on the consumer's preference, perception, and intake of the sauce were analyzed using a consumer test. In descriptive analysis, the original and sodium-reduced sauce showed similar sensory characteristics but did not differ in saltiness. In the consumer test, there were no significant differences in the overall preference levels between the two sauces. On the other hand, there were significant differences in preference and perception between nutrition teachers and general consumer groups, which were due largely to their age as well as the health and sodium-related attitudes and nutritional knowledge differences. Sodium-reduced information decreased the perceived saltiness intensity. In addition, reducing sodium intake by serving pork cutlet sauce in a bottle can be an effective strategy because this serving method increased the acceptance and induced the smaller intake of sauce.

• 한국문헌정보학회지
• /
• 제22권
• /
• pp.397-441
• /
• 1992

This study aims to present an improved model of the organizational structure for national university libraries in Korea. The results of this study are as follows : 1. A small scale library serving to a population of less than 15,000 should be organized with a staff of 50-60 persons, 3 departments and 10-13 sections (3 departments and 9-12 sections of central library + 1 section of branch library) 2. A medium scale library serving to 15,000-25,000 population should be organized with a staff of 80-120 persons, an associate librarian, 5-6 departments and 18-21 sections (4-5 departments and 15-18 sections of central library + 1 department and 3 sections of branch library) 3. A large scale library serving to a population of over 25,000 should be organized with a staff of 140-170 persons, an associate librarian, 7-8 departments and 25-28 sections (5-6 department and 20-21 sections of central library + 2 departments and 6 sections of branch libraries)

• 식품보건융합연구
• /
• 제8권1호
• /
• pp.9-16
• /
• 2022

Recently, the use of serving robots has been increasing due to the increase in preference for non-face-to-face services and the rise in the minimum wage due to the coronavirus. When analyzing previous studies related to serving robots, it was confirmed that most of the studies on the functions and technologies of serving robots were conducted. Therefore, this study analyzed the factors affecting the attitude and customer satisfaction of restaurant consumers toward serving robots by adding performance expectations, effort expectations, and speed factors among the UTAUT2 models. The survey period was conducted from July 28, 2021 to September 9, 2021, and 306 out of a total of 310 surveys were used for analysis, excluding 4 unfaithful surveys. For the analysis, exploratory factor analysis, reliability analysis, confirmatory factor analysis, and hypothesis test were performed using SPSS 20.0 and AMOS 20.0, and the research results are as follows. First, it was found that performance expectation, effort expectation, and speed had a significant positive (+) effect on attitudes. Second, it was found that attitude had a significant positive (+) effect on customer satisfaction. This study researched customer selection attributes of robot service restaurants using the UTAUT2 model, and also provided academic and practical implications.

• 한국도서관정보학회지
• /
• 제5권
• /
• pp.33-65
• /
• 1978

This study is an attempt to present a model of the organization structure for public libraries based on Standards for Public Libraries in Korea. The result of this study is as follows : (1) A library serving $50,000{\sim}80,000$ population should be organized with a staff of 4 persons and a part-time typist without actual departments. (2) A library serving $100,000{\sim}150,000$ population should be organized with a staff of 10 persons without actual departments. (3) A library serving $400,000{\sim}500,000$ population should be organized with a staff of 33 persons. $4{\sim}5$ sections and 2 branch libraries leveled with a section. (4) A library serving $900,000{\sim}1,000,000$ population should be organized with a staff of $62{\sim}66$ persons, an associate librarian, $4{\sim}5$ departments, $9{\sim}10$ sections and $4{\sim}5$ branch libraries leveled with a section. (5) A library serving $1,400,000{\sim}1,500,000$ population should be organized with a staff of 95 persons, an associate librarian, $5{\sim}6$ departments, $13{\sim}16$ sections, $6{\sim}8$ branch libraries leveled with a section and a full-time person taking charge of publicity. (6) A library serving $1,900,000{\sim}2,000,000$ population should be organized with a staff of 124 persons, 2 associate librarians, $7{\sim}8$ departments, 14 sections, 4 separate rooms without actual departments and $9{\sim}11$ branch libraries leveled with a section.

• 한국사회복지학
• /
• 제29권
• /
• pp.68-104
• /
• 1996

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권5호
• /
• pp.2122-2140
• /
• 2020

Proper and precise embedding of commercial ads within Webpages requires Ad-hoc analysis and understanding of their content. By the successful implementation of this step, both publishers and advertisers gain mutual benefits through increasing their revenues on the one hand, and improving user experience on the other. In this research work, we propose a novel multi-level context-based ads serving approach through which ads will be served at generic publisher websites based on their contextual relevance. In the proposed approach, knowledge encoded in domain-specific and generic semantic repositories is exploited in order to analyze and segment Webpages into sets of contextually-relevant segments. Semantically-enhanced indexes are also constructed to index ads based on their textual descriptions provided by advertisers. A modified cosine similarity matching algorithm is employed to embed each ad from the Ads repository into one or more contextually-relevant segments. In order to validate our proposal, we have implemented a prototype of an ad serving system with two datasets that consist of (11429 ads and 93 documents) and (11000 documents and 15 ads), respectively. To demonstrate the effectiveness of the proposed techniques, we experimentally tested the proposed method and compared the produced results against five baseline metrics that can be used in the context of ad serving systems. In addition, we compared the results produced by our system with other state-of-the-art models. Findings demonstrate that the accuracy of conventional ad matching techniques has improved by exploiting the proposed semantically-enhanced context-based ad serving model.