Parameter estimation and assessment of bias in genetic evaluation of carcass traits in Hanwoo cattle using real and simulated data

Mohammed Bedhane,Julius van der Werf,Sara de las Heras-Saldana,Leland Ackerson IV,Dajeong Lim,박병호,박미나,Seung-Hee Roh,Samuel Clark 한국축산학회 2023 한국축산학회지 Vol.65 No.6

Most carcass and meat quality traits are moderate to highly heritable, indicating that they can be improved through selection. Genetic evaluation for these types of traits is performed using performance data obtained from commercial and progeny testing evaluation. The performance data from commercial farms are available in large volume, however, some drawbacks have been observed. The drawback of the commercial data is mainly due to sorting of animals based on live weight prior to slaughter, and this could lead to bias in the genetic evaluation of later measured traits such as carcass traits. The current study has two components to address the drawback of the commercial data. The first component of the study aimed to estimate genetic parameters for carcass and meat quality traits in Korean Hanwoo cattle using a large sample size of industry-based carcass performance records (n = 469,002). The second component of the study aimed to describe the impact of sorting animals into different contemporary groups based on an early measured trait and then examine the effect on the genetic evaluation of subsequently measured traits. To demonstrate our objectives, we used real performance data to estimate genetic parameters and simulated data was used to assess the bias in genetic evaluation. The results of our first study showed that commercial data obtained from slaughterhouses is a potential source of carcass performance data and useful for genetic evaluation of carcass traits to improve beef cattle performance. However, we observed some harvesting effect which leads to bias in genetic evaluation of carcass traits. This is mainly due to the selection of animal based on their body weight before arrival to slaughterhouse. Overall, the non-random allocation of animals into a contemporary group leads to a biased estimated breeding value in genetic evaluation, the severity of which increases when the evaluation traits are highly correlated.

데이터 불균형 해소를 위한 유전알고리즘 기반 최적의 오버샘플링 비율

신승수,조휘연,김용혁 한국융합학회 2021 한국융합학회논문지 Vol.12 No.1

최근에는 데이터베이스의 발달로 금융, 보안, 네트워크 등에서 생성된 많은 데이터가 저장 가능하며, 기계학습 기반 분류기를 통해 분석이 이루어지고 있다. 이 때 주로 야기되는 문제는 데이터 불균형으로, 학습 시 다수 범주의 데이터들로 과적합이 되어 분류 정확도가 떨어지는 경우가 발생한다. 이를 해결하기 위해 소수 범주의 데이터 수를 증가 시키는 오버샘플링 전략이 주로 사용되며, 데이터 분포에 적합한 기법과 인자들을 다양하게 조절하는 과정이 필요하다. 이러한 과정의 개선을 위해 본 연구에서는 스모트와 생성적 적대 신경망 등 다양한 기법 기반의 오버샘플링 조합과 비율을 유전알고리즘을 통해 탐색하고 최적화 하는 전략을 제안한다. 제안된 전략과 단일 오버샘플링 기법으로 신용카 드 사기 탐지 데이터를 샘플링 한 뒤, 각각의 데이터들로 학습한 분류기의 성능을 비교한다. 그 결과 유전알고리즘으로 기법별 비율을 탐색하여 최적화 한 전략의 성능이 기존 전략들 보다 우수했다. Recently, with the development of database, it is possible to store a lot of data generated in finance, security, and networks. These data are being analyzed through classifiers based on machine learning. The main problem at this time is data imbalance. When we train imbalanced data, it may happen that classification accuracy is degraded due to over-fitting with majority class data. To overcome the problem of data imbalance, oversampling strategy that increases the quantity of data of minority class data is widely used. It requires to tuning process about suitable method and parameters for data distribution. To improve the process, In this study, we propose a strategy to explore and optimize oversampling combinations and ratio based on various methods such as synthetic minority oversampling technique and generative adversarial networks through genetic algorithms. After sampling credit card fraud detection which is a representative case of data imbalance, with the proposed strategy and single oversampling strategies, we compare the performance of trained classifiers with each data. As a result, a strategy that is optimized by exploring for ratio of each method with genetic algorithms was superior to previous strategies.

데이터 기반 설계를 활용한 최적 파빌리온 배치 및 적용 설계안

손상현 ( Son Sanghyun ) 한국공간디자인학회 2021 한국공간디자인학회논문집 Vol.16 No.1

(연구배경 및 목적) 최근 디지털 기술의 눈부신 발전은 우리 사회 전반에 거대한 변화를 초래하고 있다. 건축 분야에서도 4차산업 기반으로 건축 및 도시 데이터를 수집, 처리, 평가하고 이를 시각화하여 객관적인 의사결정을 돕는 데이터 기반 설계에 대한 관심이 증가하고 있다. 본 디자인에서는 구체적인 조건의 대지를 설정하고, 건물의 성능을 평가하여, 그 결과 데이터를 기초로 유전알고리즘이 적용된 최적의 설계안을 탐색한다. 이는 과거 건축가의 경험, 직관, 개인의 재능에 치우쳤던 의사결정 방식에 대한 대안이 될 수 있으며 실질적 데이터 분석을 토대로 이를 설계 과정과 접목하는 새로운 설계 영역으로 발전될 수 있다. (연구 방법) 도심 속 오픈스페이스를 대상지로 선정하고, 그 시설의 접근성과 활용도를 높이기 위해 주요 가로변으로부터 최대로 건물을 가시할 수 있으며 최소의 공사비가 예상되는 배치안을 도출한다. 이 도출된 배치안을 기준으로 해당 대지에 다양한 프로그램이 삽입 가능한 파빌리온을 계획한다. 이를 위해 3D 모델링 툴인 라이노(Rhino3d), 패러매트릭 디자인 툴인 그래스호퍼(Grasshopper), 유전자 알고리즘 기반의 다중목적함수 최적화(Multi-Objective Optimization, MOO) 애드온(Add-On)인 옥토퍼스(Octopus)를 활용한다. 먼저 패러매트릭 개념의 매스 대안을 생성하고 대안별 가시성과 총 외피 면적을 분석한다. 옥토퍼스로 도출된 최적 범위의 대안을 각각 실시간으로 시각화하여 최적 배치 및 형태 대안을 최종 선정할 수 있도록 한다. 선정 결과가 반영된 설계안으로 최종 제안한다. (결과) 도출된 디자인은 최적 배치안이 반영된 결과이며 도심 속 오픈 스페이스 활용법에 대한 새로운 제안을 포함하는 파빌리온 설계안이다. 먼저 도심의 주요 가로변에 면한 오픈 스페이스에 다중목적함수 최적화(MOO) 알고리즘을 적용한 최적 파빌리온을 배치하고 대상지에서 요구된 다양한 프로그램과 이벤트를 위해 실내외 전시공간, 소규모 전시공간, 상점, 카페, 소규모 정원, 잔디밭, 바닥 분수 등의 프로그램을 파빌리온 기본모듈에 담는다, 또한 이를 모듈화된 하나의 거리로 연결하여 미래 프로그램적 변화에 대응하면서 연속적으로 경험할 수 있는 새로운 도심 속 파빌리온으로 제안한다. (결론) 본 디자인은 건물 성능의 분석, 최적화 결과 데이터의 도출 및 시각화 그리고 이를 반영한 최종 설계안의 제안을 포함한다. 이를 통해 건축가의 자의적 방식에 의존했던 과거 건축 분야 의사결정 방식이 가지는 한계를 극복하고, 정량적 데이터를 기반으로 객관성을 추구하며, 설계 초기 단계에서 계획안의 수정 및 재생산을 반복했던 시간 소모적 과정의 최소화를 이루는 새로운 데이터 기반 설계의 가능성을 탐색한다. (Background and Purpose) The recent remarkable development of digital technology is causing rapid change in our society. In architecture, there is increased interest in data-based design that supports decision-making through collecting, filtering, and evaluating building and urban-related data, as well as visualizing this in real-time. This design proposal set up an example with specific site conditions and evaluates the performance of the building, explores an optimal design that reflects the computation result derived from the optimization genetic algorithm based on the evaluation data. This can be a new way of making decisions as opposed to relying on the experience, intuition, and personal talent of the architect as it was in the past, and develop into a new design approach that combines design process with actual data. (Method) An open space in an urban environment can be selected as an example, and the design layout can be derived that maximizes visibility of the building, as well as minimizes total surface area. Based on this layout, a pavilion with various programs can be proposed on this open space. For this, the 3D modeling tool Rhino 3D, the parametric design tool Grasshopper, and the genetic algorithm-based Multi-Objective Optimization add-on Octopus were used. First, various parametric alternatives for the pavilion layout were considered and the visibility and total surface area of each alternative were analyzed. The optimized result data linked to each alternative was visualized in real-time to select the optimal layout. Then the final proposal was drawn from the selected layout and form. (Results) The final design is the result derived from the optimal layout and is an architectural design that considers a new way of utilizing open space in the developed city. First, the pavilion layout derived from the optimization genetic algorithm was propose on the open space that is adjacent to the main street in the city. Diverse programs and event spaces such as indoor and outdoor exhibition spaces, small exhibition spaces, shops, cafes, small gardens, lawns, and fountains were included in the pavilion with basic modules. In addition, this was connected to a modular street, proposing a new open space in the city that can respond to future program changes and be experienced continuously. (Conclusions) This design proposal evaluated building performance using computation, visualized and optimized the results, and proposed a final pavilion design reflecting this design process. Through this process, the limitations of the traditional architectural decision-making method which relied on the experience, intuition of the architect can be overcome. The proposal pursues objectivity based on quantitative data and presented the possibility of a new data-based architectural design which minimizes the time-consuming process of revising and reproducing alternatives in the early stage of design.

‘보건의료 데이터 활용 가이드라인’의 현행법상 문제점

이석배 대한의료법학회 2021 의료법학 Vol.22 No.4

민간과 공공이 생산해내는 정보의 홍수속에서, 이 방대한 분량의 정보는 빅데이터로 대표되는 제4차 산업혁명시대의 핵심자원으로 간주되고 있다. 전 세계적으로 이 빅데이터에 대한 관심이 높아지고 데이터의 확보와 축적, 축적된 데이터의 안전하면서도 유용하게 활용하는 방안에 대한 논의가 활발하다. 특히 보건의료 데이터는 빅데이터 기술이 활용될 가장 가치있는 자원으로 평가되고 있다. 이러한 보건의료 데이터를 유용하게 활용하기 위해서는 분산된 보건의료 데이터를 통합하여 조사나 연구에 활용가능한 형태로 이용자에게 제공되어야 한다. 주요 국가들이 데이터 경제의 주도권을 확보하기 위해 경쟁하는 상황에서 우리나라도 2020년 8월 「개인정보보호법」등 소위 ‘데이터 3법’이 개인정보의 활용방향으로 개정되었다. ‘데이터 3법’의 개정은 개인정보 정의의 판단기준을 명확하게 하고, 가명정보의 개념을 도입하여 개인정보의 안전한 활용을 뒷받침하기 위한 제도적 기반이라 할 수 있다. 최근에는 그 후속 조치로 개인정보보호위원회가 ‘가명정보 처리 가이드라인’을 발표하였고, 보건복지부는 이와 별도로 ‘보건의료 데이터 활용 가이드라인’을 발표하였다. 하지만 여전히 풀어야 할 숙제는 남아있다. 우리나라는 「국민건강보호법」에 따라 전국민의 건강보험 가입이 의무화되어 있고, 모든 국민의 보건의료정보는 국민건강보험공단, 국민건강보험심사평가원 등 공공기관이 보유, 관리하고 있다. 이러한 데이터는 보건의료와 관한 빅데이터를 구성하게 되는데, 특히 모든 국민이 단일 건강보험에 모두 가입되어 있다는 점에서 보건의료 영역에서 빅데이터로서 그 가치와 잠재력은 어느 나라에서도 찾기 어려운 것도 사실이다. 반면 안정성의 측면에서는 그만큼 위험을 가지고 있다고 볼 수 있다. 보건의료데이터는 사람의 생명이나 신체와 직결되고 그와 관련된 수많은 민감정보를 포함하고 있어, 다른 분야보다 세심하고 보수적인 관점에서 개인정보를 보다 안전하게 보호하는 것을 전제로 그 안에서 활용이 될 수 있도록 제도가 마련되어야 할 것이다. 이 글에서는 개인정보보호위원회와 보건복지부가 제시한 ‘보건의료데이터 활용 가이드라인’의 주요내용을 분석하기 위하여 우선 개정된 「개인정보보호법」의 주요내용을 검토하고, 그에 따라 ‘보건의료 데이터 활용 가이드라인’의 주요내용을 분석하여 타법률과 충돌문제 등 그 문제점과 개선방안을 검토하였다. ‘보건의료 데이터 활용 가이드라인’은 그 성격상 현행 「개인정보보호법」의 해석을 보충하고, 보건의료 분야에 특화된 데이터 활용의 관점에서 「개인정보보호법」이 내다보지 못했던 상황에 관해 법의 해석・적용과 실무상의 지침을 제시하려 하였으나, 가이드라인의 제목에서 나타나듯이 ‘활용’에 초점을 두어 개인정보보호와 균형을 이루는 데에는 실패한 것으로 보인다. ‘보건의료 데이터 활용 가이드라인’은 「개인정보보호법」의 내재적인 문제점과 「의료법」, 「생명윤리법」과 충돌문제나 실효성 문제, 법률에 규정할 네용을 법률에 근거없이 가이드라인에 담고 있는 등 아직까지 미흡한 부분이 많고, 여러 가지 문제점을 가지고 있다는 점을 확인하였다. In the midst of the flood of private and public information, the huge amount of information is a key resource in the age of the 4th industrial revolution, represented by big data. Interest in these is growing worldwide. There is an active discussion about how to backup and accumulate data and how to use the collected data safely and effectively. Above all, health data is valued as the most valuable resource for which big data technology is used. To make good use of health data, distributed health data must be integrated and made available to users in a form that can be used for research or inspection. In a situation in which large countries are competing for the establishment or management of the data economy, the so-called 3 data laws, which contain the PERSONAL INFORMATION PROTECTION ACT(PIPA)), were also changed in South Korea in August 2020. The PIPA introduced the concept of pseudonymous information and established a legal basis for its use. As a follow-up action, the 'Personal Information Protection Commission (PIPC)' announced the 'Guidelines for Handling Pseudonymous Information' and 'Ministry of Health and Welfare' announced the 'Guidelines for the Use of Health Data'. Health data are directly related to human life and body and therefore contain a lot of sensitive data. So it is a system that can be used from a more cautious and conservative point of view, provided that personal data is more securely protected. In order to analyze the main content of the “Guidelines for the Use of Health Data”, we first checked the main content of the revised DSG. Afterwards, by analyzing the essential contents of the “Guidelines for Use of Health Data”, problems such as conflicts with other laws and improvement measures were checked.

Genomic Common Data Model for Seamless Interoperation of Biomedical Data in Clinical Practice: Retrospective Study

Shin, Seo Jeong,You, Seng Chan,Park, Yu Rang,Roh, Jin,Kim, Jang-Hee,Haam, Seokjin,Reich, Christian G,Blacketer, Clair,Son, Dae-Soon,Oh, Seungbin,Park, Rae Woong JMIR Publications 2019 Journal of medical Internet research Vol.21 No.3

<P><B>Background</B></P><P>Clinical sequencing data should be shared in order to achieve the sufficient scale and diversity required to provide strong evidence for improving patient care. A distributed research network allows researchers to share this evidence rather than the patient-level data across centers, thereby avoiding privacy issues. The Observational Medical Outcomes Partnership (OMOP) common data model (CDM) used in distributed research networks has low coverage of sequencing data and does not reflect the latest trends of precision medicine.</P><P><B>Objective</B></P><P>The aim of this study was to develop and evaluate the feasibility of a genomic CDM (G-CDM), as an extension of the OMOP-CDM, for application of genomic data in clinical practice.</P><P><B>Methods</B></P><P>Existing genomic data models and sequencing reports were reviewed to extend the OMOP-CDM to cover genomic data. The Human Genome Organisation Gene Nomenclature Committee and Human Genome Variation Society nomenclature were adopted to standardize the terminology in the model. Sequencing data of 114 and 1060 patients with lung cancer were obtained from the Ajou University School of Medicine database of Ajou University Hospital and The Cancer Genome Atlas, respectively, which were transformed to a format appropriate for the G-CDM. The data were compared with respect to gene name, variant type, and actionable mutations.</P><P><B>Results</B></P><P>The G-CDM was extended into four tables linked to tables of the OMOP-CDM. Upon comparison with The Cancer Genome Atlas data, a clinically actionable mutation, p.Leu858Arg, in the <I>EGFR</I> gene was 6.64 times more frequent in the Ajou University School of Medicine database, while the p.Gly12Xaa mutation in the <I>KRAS</I> gene was 2.02 times more frequent in The Cancer Genome Atlas dataset. The data-exploring tool GeneProfiler was further developed to conduct descriptive analyses automatically using the G-CDM, which provides the proportions of genes, variant types, and actionable mutations. GeneProfiler also allows for querying the specific gene name and Human Genome Variation Society nomenclature to calculate the proportion of patients with a given mutation.</P><P><B>Conclusions</B></P><P>We developed the G-CDM for effective integration of genomic data with standardized clinical data, allowing for data sharing across institutes. The feasibility of the G-CDM was validated by assessing the differences in data characteristics between two different genomic databases through the proposed data-exploring tool GeneProfiler. The G-CDM may facilitate analyses of interoperating clinical and genomic datasets across multiple institutions, minimizing privacy issues and enabling researchers to better understand the characteristics of patients and promote personalized medicine in clinical practice.</P>

Genetic Algorithm Application to Machine Learning

Myung-mook Han,Yill-byung Lee 한국지능시스템학회 2001 한국지능시스템학회논문지 Vol.11 No.7

In this paper we examine the machine learning issues raised by the domain of the Intrusion Detection Systems(IDS), which have difficulty successfully classifying intruders. These systems also require a significant amount of computational overhead making it difficult to create robust real-time IDS. Machine learning techniques can reduce the human effort required to build these systems and can improve their performance, Genetic algorithms are used to improve the performance of search problems, while data mining has been used for data analysis. Data Mining is the exploration and analysis of large quantities of data to discover meaningful patterns and rules. Among the tasks for data mining, we concentrate the classification task. Since classification is the basic element of humans way of thinking, it is a well-studied problem in a wide variety of application. In this paper, we propose a classifier system based on genetic algorithm, and the proposed system is evaluated by applying it to IDS problem related to classification task in data mining. We report our experiments in using these methods on KDD audit data.

Enhanced Genetic Programming Approach for a Ship Design

Lee, Kyung-Ho,Han, Young-Soo,Lee, Jae-Joon The Society of Naval Architects of Korea 2007 Journal of ship and ocean technology Vol.11 No.4

Recently the importance of the utilization of engineering data is gradually increasing. Engineering data contains the experiences and know-how of experts. Data mining technique is useful to extract knowledge or information from the accumulated existing data. This paper deals with generating optimal polynomials using genetic programming (GP) as the module of Data Mining system. Low order Taylor series are used to approximate the polynomial easily as a nonlinear function to fit the accumulated data. The overfitting problem is unavoidable because in real applications, the size of learning samples is minimal. This problem can be handled with the extended data set and function node stabilization method. The Data Mining system for the ship design based on polynomial genetic programming is presented.

Genetic Algorithm Application to Machine Learning

Han, Myung-mook,Lee, Yill-byung Korean Institute of Intelligent Systems 2001 한국지능시스템학회논문지 Vol.11 No.7

In this paper we examine the machine learning issues raised by the domain of the Intrusion Detection Systems(IDS), which have difficulty successfully classifying intruders. There systems also require a significant amount of computational overhead making it difficult to create robust real-time IDS. Machine learning techniques can reduce the human effort required to build these systems and can improve their performance. Genetic algorithms are used to improve the performance of search problems, while data mining has been used for data analysis. Data Mining is the exploration and analysis of large quantities of data to discover meaningful patterns and rules. Among the tasks for data mining, we concentrate the classification task. Since classification is the basic element of human way of thinking, it is a well-studied problem in a wide variety of application. In this paper, we propose a classifier system based on genetic algorithm, and the proposed system is evaluated by applying it to IDS problem related to classification task in data mining. We report our experiments in using these method on KDD audit data.

유전적 프로그래밍과 SOM을 결합한 개선된 선박 설계용 데이터 마이닝 시스템 개발

이경호(Kyungho Lee),박종훈(Jonghoon Park),한영수(Youngsoo Han),최시영(Siyoung Choi) (사)한국CDE학회 2009 한국CDE학회 논문집 Vol.14 No.6

Recently, knowledge management has been required in companies as a tool of competitiveness. Companies have constructed Enterprise Resource Planning(ERP) system in order to manage huge knowledge. But, it is not easy to formalize knowledge in organization. We focused on data mining system by genetic programming(GP). Data mining system by genetic programming can be useful tools to derive and extract the necessary information and knowledge from the huge accumulated data. However when we don"t have enough amounts of data to perform the learning process of genetic programming, we have to reduce input parameter(s) or increase number of learning or training data. In this study, an enhanced data mining method combining Genetic Programming with Self organizing map, that reduces the number of input parameters, is suggested. Experiment results through a prototype implementation are also discussed.

유전프로그래밍과 BEMS 데이터를 이용한 실내온도 기계학습 모델

서원준(Suh, Won Jun),박철수(Park, Cheol Soo) 대한건축학회 2016 대한건축학회논문집 Vol.32 No.6

Recently, BEMS(Building Energy Management Systems) are widely adopted in large existing buildings and there is a growing interest in applying model-assisted optimal control based on the BEMS data. Unfortunately, current BEMS are used only for measurement, data collection and rule-based operation. It would be ideal if a building’s data-driven energy model can be automatically generated out of BEMS data and is used for real-time optimal control. This paper presents such approach that a data-driven genetic programming can be beneficially utilized for automatic development of a room air temperature prediction model. In this study, the room air temperature prediction model was developed and successfully validated using the genetic programming and actual BEMS data. In the paper, pros and cons of the genetic programming approach is discussed.