• BMB Reports
• /
• 제46권1호
• /
• pp.41-46
• /
• 2013

Identifying genes indispensable for an organism's life and their characteristics is one of the central questions in current biological research, and hence it would be helpful to develop computational approaches towards the prediction of essential genes. The performance of a predictor is usually measured by the area under the receiver operating characteristic curve (AUC). We propose a novel method by implementing genetic algorithms to maximize the partial AUC that is restricted to a specific interval of lower false positive rate (FPR), the region relevant to follow-up experimental validation. Our predictor uses various features based on sequence information, protein-protein interaction network topology, and gene expression profiles. A feature selection wrapper was developed to alleviate the over-fitting problem and to weigh each feature's relevance to prediction. We evaluated our method using the proteome of budding yeast. Our implementation of genetic algorithms maximizing the partial AUC below 0.05 or 0.10 of FPR outperformed other popular classification methods.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제3권2호
• /
• pp.133-137
• /
• 2003

Various bioinformatics tools for biological data processing have been developed and most of them are available in public. Most bioinformatics works are carried out by a composite application of those tools. Several integration approaches have been proposed for easy use of the tools. This paper proposes a new multi agent system to integrate bioinformatics tools in the perspective of workflow since the composite applications of tools can be regarded as workflows. For the easy integration, the proposed system employs wrapper agents for existing tools, uses XML-based messages in the inter-agent communication, and agents are supposed to extract necessary information from the received messages. This allows new tools to be easily added on the integration framework. The proposed method allows various control structures in workflow definition and provides the progress monitoring capability of the on-going workflows. In particular, agents in this system have the rule-based architecture which allows the defined rule set to be a special role agent. This feature provides fast and flexible agent development to aid in managing the complexity of bioinformatics application. This system has been partially implemented and has been proven to be a viable implementation for workflow-based bioinformatics tool integration.

• 산업경영시스템학회지
• /
• 제29권1호
• /
• pp.76-86
• /
• 2006

다양한 데이터 마이닝 기법들의 발전과 더불어, 속성(Feature 또는 Attribute)의 범위(Dimension)를 줄이기 위해 많은 요소 선택 방법이 개발되었다. 이는 확장성(Scalability)을 향상시킬 수 있고 학습 모델(Learning Model)을 더욱 쉽게 해석할 수 있도록 한다. 이 논문에서는 네스티드 분할(Nested Partition, 이하 NP)을 이용한 새로운 최적화 기반 속성 선택 방법을 NP 기본 구조와 다양한 실험 문제의 수치적 결과들과 함께 제시하여 어떻게 NP의 최적화 구조가 속성 선택 과정에 기여를 하고 있는지 보여준다. 그리고 이 새로운 지능적인 분할 방법이 어떻게 매우 효율적인 분할을 수행하는지를 제시한다. 이 새로운 속성 선택 방법은 필터(Filter)방법과 래퍼(Wrapper)방법 두 가지로 구현될 수 있다. 사례 연구로서, B2B e-비즈니스 시스템에서 효과적으로 사용될 수 있는 추천 시스템(Recommender System)을 제안하였다. 이 추천 시스템은 분류 기법(Classification Rule)과 제시된 NP 기반 요소 선택 방법을 사용하고 있다. 이 추천 시스템은 사용자의 인터넷 경매 참여를 추천하는데 사용되며, 이 때 제안된 요소 선택 앨고리듬은 추천 규칙들이 쉽게 이해될 수 있도록 모델을 간략화 하는데 사용된다.

• 한국CDE학회논문집
• /
• 제12권4호
• /
• pp.308-320
• /
• 2007

A multi-agent and web based engineering framework concerning the automation of fatigue durability analysis for welded bogie frame of railway vehicles is presented. Mostly, this kind of design or analysis includes complex workflow, huge amounts of information processing, and problem solving. Macro programs of I-DEAS, APDL of ANSYS, and in-house fatigue code are utilized for parametric geometry representation, automatic mesh generation, static stress analysis, fatigue durability analysis, post-processing, and data sorting. The engineering framework is implemented on the JADE. Since every task requires a fairly complex process and specialized knowledge, the multi-agent based framework is very useful to keep the independency among several disciplines or tasks and to use distributed hardware and software resources. All engineering programs are integrated by XML wrapper. Related database of the engineering framework and web based user interfaces are also developed. A parametric study is carried out to take into account the effect of geometrical change of transom support bracket on its cumulative fatigue damage. The developed engineering framework reduced remarkably the time and costs required in designing and solving engineering problems.

• Nuclear Engineering and Technology
• /
• 제50권1호
• /
• pp.25-34
• /
• 2018

The CINDER code has about 60 years of development history, and is thus one of the world's best transmutation computing codes to date. Unfortunately, it is complex and cumbersome to use. Preparing auxiliary input files for activation computation from MCNPX output and executing them using Perl script (activation script) is the first difficulty, and separation of gamma source computing script (gamma script), which analyzes the spectra files produced by CINDER code and creates source definition format for MCNPX code, is the second difficulty. In addition, for highly nonlinear problems, multiple human interventions may increase the possibility of errors. Postprocessing such as making plots with large text outputs is also time consuming. One way to improve these limitations is to make a graphical user interface wrapper that includes all codes, such as MCNPX and CINDER, and all scripts with a visual C#.NET tool. The graphical user interface merges all the codes and provides easy postprocessing of graphics data and Microsoft office tools, such as Excel sheets, which make the CINDER code easy to use. This study describes the VCINDER code (with visual C#.NET) and gives a typical application example.

• 대한전자공학회논문지SD
• /
• 제40권5호
• /
• pp.321-334
• /
• 2003

GALS(Globally-Asynchronous Locally-Synchronous) 시스템은 대규모의 칩 설계 시에 설계의 용이성과 신뢰성을 확보할 수 있는 구조로 주목 받고 있다. 본 논문에서는 GALS 시스템에 필수적인 비동기 접속장치를 제안한다. 접속 장치는 크게 센더 모듈과 리시버 모듈로 구성되어 있으며, 센더 모듈에서는 부분적으로 내부 클록과는 무관하게 데이터 전송이 가능하다. 0.25um 공정의 게이트 레벨 표준 셀 라이브러리를 사용하여 설계하였고, 성능 향상 정도를 시뮬레이션을 통하여 예측할 수 있었다. 마지막으로, 접속장치를 장착한 GALS 구조의 예제 회로를 설계하여 올바르게 동작함을 확인하였다.

• Journal of information and communication convergence engineering
• /
• 제5권3호
• /
• pp.233-238
• /
• 2007

An approach to identify Chinese event types is proposed in this paper which combines a good feature selection policy and a Maximum Entropy (ME) model. The approach not only effectively alleviates the problem that classifier performs poorly on the small and difficult types, but improve overall performance. Experiments on the ACE2005 corpus show that performance is satisfying with the 83.5% macro - average F measure. The main characters and ideas of the approach are: (1) Optimal feature set is built for each type according to local feature selection, which fully ensures the performance of each type. (2) Positive and negative features are explicitly discriminated and combined by using one - sided metrics, which makes use of both features' advantages. (3) Wrapper methods are used to search new features and evaluate the various feature subsets to obtain the optimal feature subset.

• 산업공학
• /
• 제22권2호
• /
• pp.126-134
• /
• 2009

As product quality and yield are essential factors in semiconductor manufacturing, monitoring the main manufacturing steps is a critical task. For the purpose, FDC(Fault detection and classification) is used for diagnosing fault states in the processes by monitoring data stream collected by equipment sensors. This paper proposes an FDC model based on decision tree which provides if-then classification rules for causal analysis of the processing results. Unlike previous decision tree approaches, we reflect the structural aspect of the data stream to FDC. For this, we segment the data stream into multiple subregions, define structural features for each subregion, and select the features which have high relevance to results of the process and low redundancy to other features. As the result, we can construct simple, but highly accurate FDC model. Experiments using the data stream collected from etching process show that the proposed method is able to classify normal/abnormal states with high accuracy.

• 한국정보과학회논문지:컴퓨팅의 실제 및 레터
• /
• 제6권6호
• /
• pp.569-579
• /
• 2000

웹에서 정보 전달의 수단으로 등장한 XML은 많은 응용 분야에서 사용될 것이다. 많은 양의 XML 문서를 효율적으로 다루기 위하여 데이타베이스의 지원은 필수적이다. 데이터베이스를 이용하여 XML 데이타를 처리할 때 데이타베이스이 종류뿐만 아니라 그 인터페이스도 중요한 문제이다. 본 논문에서는 관계형 데이터베이스, 객체지향형 데이터베이스 및 랩포를 이용하여 XML 데이타를 저장, 질의하며 그 인터페이스로 XML-뷰, ODMG C++바인딩, OQL, DOM을 사용하였다. 또한 각각의 시스템의 구현을 통하여 각 방법들의 장단점을 논하고, 효율적인 XML 문서 처리에 대한 방법을 제시한다.

• Journal of Multimedia Information System
• /
• 제8권4호
• /
• pp.233-242
• /
• 2021

Various anatomical MRI imaging biomarkers for Alzheimer's Disease (AD) identification have been recognized so far. Cortical and subcortical volume, hippocampal, amygdala volume, and genetics patterns have been utilized successfully to diagnose AD patients from healthy. These fundamental sMRI bio-measures have been utilized frequently and independently. The entire possibility of anatomical MRI imaging measures for AD diagnosis might thus still to analyze fully. Thus, in this paper, we merge different structural MRI imaging biomarkers to intensify diagnostic classification and analysis of Alzheimer's. For 54 clinically pronounce Alzheimer's patients, 58 cognitively healthy controls, and 99 Mild Cognitive Impairment (MCI); we calculated 1. Cortical and subcortical features, 2. The hippocampal subfield, amygdala nuclei volume using Freesurfer (6.0.0) and 3. Genetics (APoE ε4) biomarkers were obtained from the ADNI database. These three measures were first applied separately and then combined to predict the AD. After feature combination, we utilize the sequential feature selection [SFS (wrapper)] method to select the top-ranked features vectors and feed them into the Multi-Kernel SVM for classification. This diagnostic classification algorithm yields 94.33% of accuracy, 95.40% of sensitivity, 96.50% of specificity with 94.30% of AUC for AD/HC; for AD/MCI propose method obtained 85.58% of accuracy, 95.73% of sensitivity, and 87.30% of specificity along with 91.48% of AUC. Similarly, for HC/MCI, we obtained 89.77% of accuracy, 96.15% of sensitivity, and 87.35% of specificity with 92.55% of AUC. We also presented the performance comparison of the proposed method with KNN classifiers.