이미지 감성분류를 위한 CNN과 K-means RGB Cluster 이-단계 학습 방안 (A Two-Stage Learning Method of CNN and K-means RGB Cluster for Sentiment Classification of Images)
-
- 지능정보연구
- /
- 제27권3호
- /
- pp.139-156
- /
- 2021
이미지 분류에서 딥러닝 모형을 사용하는 가장 큰 이유는 이미지의 전체적인 정보에서 각 지역 특징을 추출하여 서로의 관계를 고려할 수 있기 때문이다. 하지만 이미지의 지역 특징이 없는 감정 이미지 데이터는 CNN 모델이 적합하지 않을 수 있다. 이러한 감정 이미지 분류의 어려움을 해결하기 위하여 매년 많은 연구자들이 감정 이미지에 적합한 CNN기반 아키텍처를 제시하고 있다. 색깔과 사람 감정간의 관계에 대한 연구들도 수행되었으며, 색깔에 따라 다른 감정이 유도된다는 결과들이 도출되었다. 딥러닝을 활용한 연구에서도 색깔정보를 활용하여 이미지 감성분류에 적용하는 연구들이 있어왔으며, 이미지만을 가지고 분류 모형을 학습한 경우보다 이미지의 색깔 정보를 추가로 활용한 경우가 이미지 감성 분류 정확도를 더 높일 수 있었다. 본 연구는 사람이 이미지의 감정을 분류하는 기준 중 많은 부분을 차지하는 색감을 이용하여 이미지 감성 분류 정확도를 향상시키는 방안을 제안한다. 이미지의 RGB 값에 K 평균 군집화 방안을 적용하여 이미지를 대표하는 색을 추출하여, 각 감성 클래스 별 해당 색깔이 나올 확률을 가중치 식으로 변형 후 CNN 모델의 최종 Layer에 적용하는 이-단계 학습방안을 구현하였다. 이미지 데이터는 6가지 감정으로 분류되는 Emotion6와 8가지 감정으로 분류되는 Artphoto를 사용하였다. 학습에 사용한 CNN 모델은 Densenet169, Mnasnet, Resnet101, Resnet152, Vgg19를 사용하였으며, 성능 평가는 5겹 교차검증으로 CNN 모델에 이-단계 학습 방안을 적용하여 전후 성과를 비교하였다. CNN 아키텍처만을 활용한 경우보다 색 속성에서 추출한 정보를 함께 사용하였을 때 더 좋은 분류 정확도를 보였다.
This project was a service-cum-research effort with a quasi-experimental study design to examine the health benefits of an integrated Family Planning (FP)/Maternal & Child health (MCH) Service approach that provides crucial factors missing in the present on-going programs. The specific objectives were: 1) To test the effectiveness of trained nurse/midwives (MW) assigned as change agents in the Health Sub-Center (HSC) to bring about the changes in the eight FP/MCH indicators, namely; (i)FP/MCH contacts between field workers and their clients (ii) the use of effective FP methods, (iii) the inter-birth interval and/or open interval, (iv) prenatal care by medically qualified personnel, (v) medically supervised deliveries, (vi) the rate of induced abortion, (vii) maternal and infant morbidity, and (viii) preinatal & infant mortality. 2) To measure the integrative linkage (contacts) between MW & HSC workers and between HSC and clients. 3) To examine the organizational or administrative factors influencing integrative linkage between health workers. Study design; The above objectives called for quasi-experimental design setting up a study and control area with and without a midwife. An active intervention program (FP/MCH minimum 'package' program) was conducted for a 2 year period from June 1982-July 1984 in Seosan County and 'before and after' surveys were conducted to measure the change. Service input; This study was undertaken by the Soonchunhyang University in collaboration with WHO. After a baseline survery in 1981, trained nurses/midwives were introduced into two health sub-centers in a rural setting (Seosan county) for a 2 year period from 1982 to 1984. A major service input was the establishment of midwifery services in the existing health delivery system with emphasis on nurse/midwife's role as the link between health workers (nurse aids) and village health workers, and the referral of risk patients to the private physician (OBGY specialist). An evaluation survey was made in August 1984 to assess the effectiveness of this alternative integrated approach in the study areas in comparison with the control area which had normal government services. Method of evaluation; a. In this study, the primary objective was first to examine to what extent the FP/MCH package program brought about changes in the pre-determined eight indicators (outcome and impact measures) and the following relationship was first analyzed; b. Nevertheless, this project did not automatically accept the assumption that if two or more activities were integrated, the results would automatically be better than a non-integrated or categorical program. There is a need to assess the 'integration process' itself within the package program. The process of integration was measured in terms of interactive linkages, or the quantity & quality of contacts between workers & clients and among workers. Intergrative linkages were hypothesized to be influenced by organizational factors at the HSC clinic level including HSC goals, sltrurture, authority, leadership style, resources, and personal characteristics of HSC staff. The extent or degree of integration, as measured by the intensity of integrative linkages, was in turn presumed to influence programme performance. Thus as indicated diagrammatically below, organizational factors constituted the independent variables, integration as the intervening variable and programme performance with respect to family planning and health services as the dependent variable: Concerning organizational factors, however, due to the limited number of HSCs (2 in the study area and 3 in the control area), they were studied by participatory observation of an anthropologist who was independent of the project. In this observation, we examined whether the assumed integration process actually occurred or not. If not, what were the constraints in producing an effective integration process. Summary of Findings; A) Program effects and impact 1. Effects on FP use: During this 2 year action period, FP acceptance increased from 58% in 1981 to 78% in 1984 in both the study and control areas. This increase in both areas was mainly due to the new family planning campaign driven by the Government for the same study period. Therefore, there was no increment of FP acceptance rate due to additional input of MW to the on-going FP program. But in the study area, quality aspects of FP were somewhat improved, having a better continuation rate of IUDs & pills and more use of effective Contraceptive methods in comparison with the control area. 2. Effects of use of MCH services: Between the study and control areas, however, there was a significant difference in maternal and child health care. For example, the coverage of prenatal care was increased from 53% for 1981 birth cohort to 75% for 1984 birth cohort in the study area. In the control area, the same increased from 41% (1981) to 65% (1984). It is noteworthy that almost two thirds of the recent birth cohort received prenatal care even in the control area, indicating that there is a growing demand of MCH care as the size of family norm becomes smaller 3. There has been a substantive increase in delivery care by medical professions in the study area, with an annual increase rate of 10% due to midwives input in the study areas. The project had about two times greater effect on postnatal care (68% vs. 33%) at delivery care(45.2% vs. 26.1%). 4. The study area had better reproductive efficiency (wanted pregancies with FP practice & healthy live births survived by one year old) than the control area, especially among women under 30 (14.1% vs. 9.6%). The proportion of women who preferred the 1st trimester for their first prenatal care rose significantly in the study area as compared to the control area (24% vs 13%). B) Effects on Interactive Linkage 1. This project made a contribution in making several useful steps in the direction of service integration, namely; i) The health workers have become familiar with procedures on how to work together with each other (especially with a midwife) in carrying out their work in FP/MCH and, ii) The health workers have gotten a feeling of the usefulness of family health records (statistical integration) in identifying targets in their own work and their usefulness in caring for family health. 2. On the other hand, because of a lack of required organizational factors, complete linkage was not obtained as the project intended. i) In regards to the government health worker's activities in terms of home visiting there was not much difference between the study & control areas though the MW did more home visiting than Government health workers. ii) In assessing the service performance of MW & health workers, the midwives balanced their workload between 40% FP, 40% MCH & 20% other activities (mainly immunization). However,
세계는 신형 코로나바이러스 감염증(COVID-19)으로 수 많은 인명 피해와 경제적 손실을 기록하고 있는 상황이다. 우리나라 정부는 연구개발(Research & Development)을 통해 국가 감염병 위기를 극복하려는 전략을 수립하고 실행하기 위한 투자방향을 수립하였다. 기존 기술분류나 과학기술 표준분류에 따른 통계를 활용하면 특정 R&D 분야의 특이점 및 변화를 발견하기 어렵다는 한계가 존재해왔다. 최근 우리나라 감염병 연구개발 과제를 대상으로 수요자의 목적에 맞게 분류체계를 수립하고 연구비 비교 분석을 통해 투자가 요구되는 연구 분야를 제시하는 연구들이 진행되었다. 하지만 현재 국가 보건 안보와 신성장 산업육성이라는 목표를 달성하기 위한 실행방안으로 요구되고 있는 전염병 연구분야의 국가간 협력전략 수립에 필요한 정보를 체계적으로 제공하고 있지 못한 상황이다. 따라서 국가 공동 연구개발 전략 수립을 위한 분류체계와 분류모델기반의 정보서비스에 대한 연구가 요구되고 있다. 우선 감염병관련 NTIS 과제데이터를 기반으로 정성분석을 통해 7개의 분류체계를 도출하였다. 스코퍼스(Scopus) 데이터와 양방향 RNN모델을 사용하여, 분류체계 모델을 학습시켰다. 최종적인 모델의 분류 성능은 90%이상의 높은 정확도와 강건성을 확보하였다. 실증연구를 위해 주요 국가의 코로나바이러스 연구개발 과제를 대상으로 전염병 분류체계를 적용하였다. 주요 국가의 감염병(코로나바이러스) 연구개발 과제를 분류체계별로 분석한 결과, 세계적으로 유행하는 바이러스의 예상치 못한 창궐이 확산되는 속도에 비해 백신과 치료제 개발이 제대로 이뤄지지 않는 원인의 배경을 간접적으로 확인할 수 있었다. 국가별 비교분석을 통해 미국과 일본은 상대적으로 모든 영역에 골고루 연구개발 투자를 하고 있는 것으로 나타난 반면, 유럽은 상대적으로 특정 연구분야에 많은 투자를 하는 집중화 전략을 취하는 것으로 나타났다. 동시에 주요 국가의 코로나 바이러스 주요 연구조직에 대한 정보를 분류체계별로 제공하여 국제 공동R&D 전략의 기초정보를 제공하였다. 본 연구 결과를 통해 세 가지 정책적 의미를 도출할 수 있다. 첫째, 데이터기반 과학기술정책 관점에서 수요자 관심분야에 대한 국가 R&D사업의 정보를 글로벌 기준으로 문서를 분류하는 방안을 제시하였다. 둘째, 감염병관련 국가 R&D사업 영역에 대한 정보분석 서비스 기획의 기반을 마련하였다. 마지막으로 국가 감염병 R&D 분류체계 수립을 통해 분류 체계의 궁극적 목표인 산업, 기업, 정책 정보를 제공할 수 있는 기반을 마련한 것이다.
Personalized services directly and indirectly acquire personal data, in part, to provide customers with higher-value services that are specifically context-relevant (such as place and time). Information technologies continue to mature and develop, providing greatly improved performance. Sensory networks and intelligent software can now obtain context data, and that is the cornerstone for providing personalized, context-specific services. Yet, the danger of overflowing personal information is increasing because the data retrieved by the sensors usually contains privacy information. Various technical characteristics of context-aware applications have more troubling implications for information privacy. In parallel with increasing use of context for service personalization, information privacy concerns have also increased such as an unrestricted availability of context information. Those privacy concerns are consistently regarded as a critical issue facing context-aware personalized service success. The entire field of information privacy is growing as an important area of research, with many new definitions and terminologies, because of a need for a better understanding of information privacy concepts. Especially, it requires that the factors of information privacy should be revised according to the characteristics of new technologies. However, previous information privacy factors of context-aware applications have at least two shortcomings. First, there has been little overview of the technology characteristics of context-aware computing. Existing studies have only focused on a small subset of the technical characteristics of context-aware computing. Therefore, there has not been a mutually exclusive set of factors that uniquely and completely describe information privacy on context-aware applications. Second, user survey has been widely used to identify factors of information privacy in most studies despite the limitation of users' knowledge and experiences about context-aware computing technology. To date, since context-aware services have not been widely deployed on a commercial scale yet, only very few people have prior experiences with context-aware personalized services. It is difficult to build users' knowledge about context-aware technology even by increasing their understanding in various ways: scenarios, pictures, flash animation, etc. Nevertheless, conducting a survey, assuming that the participants have sufficient experience or understanding about the technologies shown in the survey, may not be absolutely valid. Moreover, some surveys are based solely on simplifying and hence unrealistic assumptions (e.g., they only consider location information as a context data). A better understanding of information privacy concern in context-aware personalized services is highly needed. Hence, the purpose of this paper is to identify a generic set of factors for elemental information privacy concern in context-aware personalized services and to develop a rank-order list of information privacy concern factors. We consider overall technology characteristics to establish a mutually exclusive set of factors. A Delphi survey, a rigorous data collection method, was deployed to obtain a reliable opinion from the experts and to produce a rank-order list. It, therefore, lends itself well to obtaining a set of universal factors of information privacy concern and its priority. An international panel of researchers and practitioners who have the expertise in privacy and context-aware system fields were involved in our research. Delphi rounds formatting will faithfully follow the procedure for the Delphi study proposed by Okoli and Pawlowski. This will involve three general rounds: (1) brainstorming for important factors; (2) narrowing down the original list to the most important ones; and (3) ranking the list of important factors. For this round only, experts were treated as individuals, not panels. Adapted from Okoli and Pawlowski, we outlined the process of administrating the study. We performed three rounds. In the first and second rounds of the Delphi questionnaire, we gathered a set of exclusive factors for information privacy concern in context-aware personalized services. The respondents were asked to provide at least five main factors for the most appropriate understanding of the information privacy concern in the first round. To do so, some of the main factors found in the literature were presented to the participants. The second round of the questionnaire discussed the main factor provided in the first round, fleshed out with relevant sub-factors. Respondents were then requested to evaluate each sub factor's suitability against the corresponding main factors to determine the final sub-factors from the candidate factors. The sub-factors were found from the literature survey. Final factors selected by over 50% of experts. In the third round, a list of factors with corresponding questions was provided, and the respondents were requested to assess the importance of each main factor and its corresponding sub factors. Finally, we calculated the mean rank of each item to make a final result. While analyzing the data, we focused on group consensus rather than individual insistence. To do so, a concordance analysis, which measures the consistency of the experts' responses over successive rounds of the Delphi, was adopted during the survey process. As a result, experts reported that context data collection and high identifiable level of identical data are the most important factor in the main factors and sub factors, respectively. Additional important sub-factors included diverse types of context data collected, tracking and recording functionalities, and embedded and disappeared sensor devices. The average score of each factor is very useful for future context-aware personalized service development in the view of the information privacy. The final factors have the following differences comparing to those proposed in other studies. First, the concern factors differ from existing studies, which are based on privacy issues that may occur during the lifecycle of acquired user information. However, our study helped to clarify these sometimes vague issues by determining which privacy concern issues are viable based on specific technical characteristics in context-aware personalized services. Since a context-aware service differs in its technical characteristics compared to other services, we selected specific characteristics that had a higher potential to increase user's privacy concerns. Secondly, this study considered privacy issues in terms of service delivery and display that were almost overlooked in existing studies by introducing IPOS as the factor division. Lastly, in each factor, it correlated the level of importance with professionals' opinions as to what extent users have privacy concerns. The reason that it did not select the traditional method questionnaire at that time is that context-aware personalized service considered the absolute lack in understanding and experience of users with new technology. For understanding users' privacy concerns, professionals in the Delphi questionnaire process selected context data collection, tracking and recording, and sensory network as the most important factors among technological characteristics of context-aware personalized services. In the creation of a context-aware personalized services, this study demonstrates the importance and relevance of determining an optimal methodology, and which technologies and in what sequence are needed, to acquire what types of users' context information. Most studies focus on which services and systems should be provided and developed by utilizing context information on the supposition, along with the development of context-aware technology. However, the results in this study show that, in terms of users' privacy, it is necessary to pay greater attention to the activities that acquire context information. To inspect the results in the evaluation of sub factor, additional studies would be necessary for approaches on reducing users' privacy concerns toward technological characteristics such as highly identifiable level of identical data, diverse types of context data collected, tracking and recording functionality, embedded and disappearing sensor devices. The factor ranked the next highest level of importance after input is a context-aware service delivery that is related to output. The results show that delivery and display showing services to users in a context-aware personalized services toward the anywhere-anytime-any device concept have been regarded as even more important than in previous computing environment. Considering the concern factors to develop context aware personalized services will help to increase service success rate and hopefully user acceptance for those services. Our future work will be to adopt these factors for qualifying context aware service development projects such as u-city development projects in terms of service quality and hence user acceptance.