• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.6 no.8
• /
• pp.15-23
• /
• 2016

Recently the interest in secondary use of medical information has emerged. But the domestic legislation or guidelines, such as being able to say that already specialize in healthcare information, can be seen a 'national medical privacy guidelines'. However the guidelines have suggested that only a violation of privacy laws in the medical information, it does not defined clearly with respect to protected health information(PHI) for secondary use. In this paper, we learn the HIPAA(Health Insurance Portability and Accountability Act) Privacy Rule of the US legislation which provides a non-identifiable screen instructions for secondary utilization of medical information, domestic guidelines and other country's guidelines. comparing with the HIPAA, national medical privacy guidelines and the domestic studies, we propose a new domestic target non-identifying information suitable for the domestic field and present future research direction.

• Journal of Convergence for Information Technology
• /
• v.8 no.1
• /
• pp.179-185
• /
• 2018

Big Data is strictly positioning one of method to deal with problems faced with mankind, not an icon of revolution in future anymore. Application of Big Data and protection of personal information have contradictoriness. When we weight more to usage of Big Data, someone's privacy is necessarily invaded. otherwise, we care more about keeping safe of individual information, only low-level of research using Big Data can be used to accomplish public purpose. In this study, we propose a method to anonymize Big Data collected in order to investigate the problems of personal information infringement and utilize Big Data and protect personal. This will solve the problem of personal information infringement as well as utilizing Big Data.

• Convergence Security Journal
• /
• v.22 no.5
• /
• pp.11-20
• /
• 2022

As the Internet and web technology develop around mobile devices, image data contains various types of sensitive information such as people, text, and space. In addition to these characteristics, as the use of SNS increases, the amount of damage caused by exposure and abuse of personal information online is increasing. However, research on de-identification technology based on multi-type object detection for personal information protection is insufficient. Therefore, this paper proposes an artificial intelligence model that detects and de-identifies multiple types of objects using existing single-type object detection models in parallel. Through cutmix, an image in which person and text objects exist together are created and composed of training data, and detection and de-identification of objects with different characteristics of person and text was performed. The proposed model achieves a precision of 0.724 and mAP@.5 of 0.745 when two objects are present at the same time. In addition, after de-identification, mAP@.5 was 0.224 for all objects, showing a decrease of 0.4 or more.

• The KIPS Transactions:PartD
• /
• v.14D no.4 s.114
• /
• pp.427-434
• /
• 2007

Although wireless devices are the most suitable device for ubiquitous environment, they have restrictive capacities when using internet services than desktop environments. Therefore this research proposes a wireless internet service method that uses contents-based personalization. The existing websites can easily and promptly access desired news articles and other data through RSS-linked web contents and by the personalization method. The proposed method will make using wireless internet easier while lowering contents production costs. Moreover, personalized mobile web news contents that satisfy the preferences of users can be serviced.

• 한국사회정책
• /
• v.25 no.4
• /
• pp.105-124
• /
• 2018

This article analyzes individualization of family policies and gender division of unpaid work in Germany, Netherlands and South Korea. Measured in terms of women's employment promotion, the individualization at the policy level was greater in Korea than both Germany and Netherlands. However, married women's share of unpaid work was substantially larger in Korea than Germany and Netherlands. The strong unequal divison of unpaid work was also the case among working married women. Regression analysis showed that share of unpaid work in Korea was related to relative income in all three countries. But while gender ideology was positively related to married women's share of unpaid work in Germany and Netherlands, the relationship was not found in Korea. Gender of respondent was also an explanatory factor for all three countries. The influence of married women's relative income and gender ideology on share of unpaid work was similar to or larger than the effect of gender of respondents in Germany and Netherlands. By contrast, it was gender of respondent that mattered most for married women's share of unpaid work in Korea. Among working married women, we found that both relative income and gender ideology were related to differences in share of unpaid work in Germany and Netherlands which was either similar to or larger than the effect of gender of respondent. However, we found that gender of respondent mattered most in Korea.

• Journal of Advanced Navigation Technology
• /
• v.13 no.1
• /
• pp.151-155
• /
• 2009

We proposes a new method using the non-negative matrix factorization (NMF) and clustering method to extract the sentences for personalized document summarization. The proposed method uses clustering method for retrieving documents to extract sentences which are well reflected topics and sub-topics in document. Beside it can extract sentences with respect to query which are well reflected user interesting by using the inherent semantic features in document by NMF. The experimental results shows that the proposed method achieves better performance than other methods use the similarity and the NMF.

• Proceedings of the Korean Information Science Society Conference
• /
• 1999.10b
• /
• pp.539-541
• /
• 1999

기존의 문장종속형 화자인식 기법에서는 음성 신호의 각 분석 프레임이 같은 기여도를 갖는 것으로 간주한다. 화자인식 시스템의 성능향상을 위해서는 음운정보보다는 인식의 단서가 되는 화자의 개인성 정보가 잘 반영되도록 하는 것이 중요하다. 본 논문에서는 HMM (hidden Markov model)을 기반으로 한 문장종속형 화자확인 시스템의 성능향상을 위해 프레임별로 인식의 단서가 되는 개인성 정보의 양을 측정하는 방법과, 이를 화자확인 시스템에 적용하는 기법을 제안한다. 제안한 방법을 적용한 결과, 기존의 우도비(likelihood ratio) 정규화 점수를 사용하는 방법에 비해 동일오류율(EER; equal error rate)을 평균 34% 감소시켜 인식율 향상을 얻을 수 있다.

• Korean Journal of Social Welfare Studies
• /
• v.42 no.4
• /
• pp.187-216
• /
• 2011

This study examined the effects of elderly care facility social workers' work environment (role conflict, role ambiguity, role overload, and adequacy of the reward system) and personal tendency (emotional empathy, cognitive empathy, external attribution, and internal attribution) on their burnout (emotional burnout, low personal achievement, and depersonalization about clients) and suggested strategies for preventing burnout in social workers at elderly care facilities. For this purpose, we conducted a questionnaire survey of social workers working at institutions designated as elderly long-term care facilities and collected a total of 312 questionnaires. According to the results of analyzing the data using a structural equation model, among the sub-factors of work environment, role ambiguity had a significant positive effect on low personal achievement and depersonalization about clients, role conflict on emotional burnout, and role overload on depersonalization about clients. Among the sub-factors of personal tendency, cognitive empathy and internal attribution had a significant negative effect on low personal achievement, and external attribution had a significant positive effect on emotional burnout and depersonalization about clients. This study is meaningful in that it illuminated social workers' burnout not only from the aspect of work environment but also from that of personal tendency.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.7
• /
• pp.109-118
• /
• 2021

Different from previous studies, this study focuses on accommodation providers' personalization services and platform convenience variables, identifying how these prior factors affect perceived value and trust in accommodation services on a shared homestay platform, and how consumers' innovation plays a role in the process. Through this, we would like to identify the mechanism of interaction between accommodation service providers and consumers mediated by the shared homestay platform and present implications for a more customer-centered platform operation strategy. This study has an extended meaning for prior research, empirically confirming that the increase in personalized offline service quality of personalized hosts in shared economic models has a positive impact on perceived value and platform trust of consumers. At the same time, we confirm that under the shared economy model, consumers' innovation propensity plays an important positive role in regulating their perceived value aspects as well as their confidence in the platform.

• 한국경영정보학회:학술대회논문집
• /
• 2007.11a
• /
• pp.597-602
• /
• 2007

본 연구는 추천시스템에서 협업필터링 알고리즘을 이용하여 특정 상품에 대한 고객의 선호도를 예측함에 있어 고객이 상품에 대해 평가한 선호도 평가치를 고객별로 표준화시켜 예측하여 기존의 예측 정확도를 향상시키는 방법에 대하여 연구하였다. 일반적으로 상품에 대한 고객의 선호도를 평가하기 위하여 절대적 기준의 수치적 척도가 제공되지만 개인에 따라서는 상품에 대한 선호 정도가 절대적 척도에 다르게 반영되어 개인별 선호도에 차이가 발생할 수 있다. 이러한 개인적 특성이 동일한 척도의 평가치로 예측되면 예측 결과의 오차를 크게 할 가능성이 있다. 또한 개인이 평가한 선호도 평가치의 편차가 협업필터링 알고리즘을 통한 선호도 예측 정확도와 밀접한 관계를 가지고 있음을 알 수 있었으며 이러한 문제를 해결하기 위하여 개별 고객이 평가한 선호도 평가치를 표준화시켜 표준화된 선호도 평가치를 이용한 선호도 예측을 실시하였다. 분석결과 표준화된 선호도 평가치를 이용한 예측 결과가 비표준화 선호도 평가치를 이용한 예측 결과보다 예측력이 우수함을 알 수 있었으며 결과에 대한 통계적 분석을 통하여 표준화된 선호도 평가치를 이용한 선호도 예측 방법과 비 표준화 선호도 평가치를 이용한 선호도 예측 방법을 혼합할 경우 선호도 예측 정확도를 더 향상시킬 수 있음을 알 수 있었다.