• Title/Summary/Keyword: 검증 소프트웨어

Search Result 2,336, Processing Time 0.031 seconds

A Study of Metrics for Test Organization Improvement by TMM in Web Application Product (웹 서비스 제품에서의 테스트 조직 향상을 위한 TMM 적용과 메트릭 연구)

  • Kwon, Hyo-Jin;Lee, Jin-Ho;Choi, Jin-Young
    • Annual Conference of KIPS
    • /
    • 2008.05a
    • /
    • pp.367-370
    • /
    • 2008
  • 소프트웨어 제품의 신뢰성과 안정성 확보를 위한 품질 보증이 중요해짐에 따라 독립된 테스트 조직이 증가되고 이러한 테스트 조직은 효과적 효율적으로 테스트 활동을 수행하기 위해 테스트 프로세스 개선을 필요로 한다. 그러나, 프로세스는 제품의 도메인 특성으로부터 많은 영향을 받아 특화되므로 실시간(Time to Market) 서비스가 강조되는 웹 서비스 시스템에서의 테스트 프로세스에 대한 연구가 부족하므로 테스트 조직 향상을 위한 방향 모델과 이를 이행할 수 있는 접근 방법을 필요로 한다. 본 연구에서는 웹 서비스 제품의 품질을 보증하기 위해 검증 작업을 수행하고 있는 테스트 조직이 향상되기 위한 방안으로 TMM을 적용하였고, 성숙된 프로세스 구축을 위한 방법으로 GQM을 이용하여 메트릭을 도출한 사례를 제시하였다. 이런 연구결과는 도메인 특성에 맞게 도출된 매트릭을 통해 제품과 테스터의 능력을 평가하고 결과 값을 프로세스 개선에 반영하여 정량적인 데이터 기반으로 관리되는 테스트 조직으로 발전할 수 있을 것이다.

A Study on Criteria Establishment for Efficient Code Inspection (효율적인 코드 인스펙션을 위한 규칙 수립에 대한 연구: A 사(社) 정보시스템 사례)

  • Kyung, Tae-Won;Kim, Sang-Kuk
    • Annual Conference of KIPS
    • /
    • 2007.11a
    • /
    • pp.407-410
    • /
    • 2007
  • 본 연구에서는 소프트웨어의 완성도와 품질을 높이기 위한 방법들 중 하나인 인스펙션과 기존 검토 기법들을 정리 하였다. 그리고 코드 인스펙션(Code Inspection)을 위한 규칙들을 수립하고 실제 프로젝트에 적용함으로써 그 효율성과 효과성을 검증하였다. 본 연구를 통해 다음과 같은 결과를 얻었다. 첫째, 소프트웨어 검토 방법들에 대한 이론적 내용을 정리하였다. 둘째, 코드 인스펙션을 위한 규칙을 수립하고 사례를 통해 성과를 증명하였다. 셋째, 코드 인스펙션을 통해 정량적 데이터 관리를 수행하였다.

Research on Performance Improvement Using LoRA Techniques in RAG End2End Models (RAG End2End 모델에서 LoRA기법을 이용한 성능 향상에 관한 연구)

  • Min-Chang Kim;Sae-Hun Yeom
    • Annual Conference of KIPS
    • /
    • 2024.05a
    • /
    • pp.600-601
    • /
    • 2024
  • 본 논문은 RAG(Retrieval-Augmented Generation) End2End의 리소스(Resource) 과부하 문제를 해결하는 동시에 모델 성능을 향상 시키기 위해 PEFT(Parameters-Efficient Fine-Tuning)기술인 LoRA(Low Rank Adaptation)적용에 관한 연구이다. 본 논문에서는 RAG End2End 모델의 파라미터 값과 개수를 유지하면서, LRM(Low Rank Matrices)을 이용하여 추가적인 파라미터만을 미세 조정하는 방식으로, 전반적인 모델의 효율성을 극대화하는 방안을 제시하였다. 본 논문에서 다양한 도메인에 데이터 셋에 대한 제안 방식의 성능을 검증하고자 Conversation, Covid-19, News 데이터 셋을 사용하였다. 실험결과, 훈련에 필요한 파라미터의 크기가 약 6.4억개에서 180만개로 감소하였고, EM(Exact Match)점수가 유사하거나 향상되었다. 이는 LoRA를 통한 접근 법이 RAG End2End 모델의 효율성을 개선할 수 있는 효과적인 전략임을 증명하였다.

Food Image Classification using Deep Learning (딥러닝을 이용한 음식 이미지 분류 기술 개발)

  • Gagyeong Lee;Seyeon Im;Jini Yang;Minjung Yoo;Sunok Kim
    • The Journal of Bigdata
    • /
    • v.8 no.2
    • /
    • pp.133-140
    • /
    • 2023
  • This study was conducted with the aim of improving the food image classification model of a health care application targeting Koreans in their twenties. 546,194 images were collected from the Public Data Portal and AI Hub, and 175 food classes were constructed. The ResNet artificial intelligence model was trained and validated. Additionally, we deeply investigated the reasons for the relatively lower recognition accuracy of the actual food images, and we attempted various methods to optimize the model's performance as a solution.

Procedure Development and Qualification of the Phased Array Ultrasonic Testing for the Nuclear Power Plant Piping Weld (원자력발전소 배관 용접부 위상배열 초음파검사 절차서 개발 및 기량검증)

  • Yoon, Byung-Sik;Yang, Seung-Han;Kim, Yong-Sik;Lee, Hee-Jong
    • Journal of the Korean Society for Nondestructive Testing
    • /
    • v.30 no.4
    • /
    • pp.317-323
    • /
    • 2010
  • The manual ultrasonic examination for the nuclear power plant piping welds has been demonstrated by using KPD(Korean Performance Demonstration) generic procedure. For automated ultrasonic examination, there is no generic procedure and it should be qualified by using applicable automated equipment. Until now, most of qualified procedures used pulse-echo technique and there is no qualified procedure using phased array technique. In this study, data acquisition and analysis software were developed and phased-array transducer and wedge were designed to implement phased array technique for nuclear power plant in-service inspection. The developed procedure are qualified for performance demonstration for the flaw detection, length sizing and depth sizing. The qualified procedure will be applied for the field examination in the nuclear power plant piping weld inspection.

Virtual Flight Test for Conceptual Lunar Lander Demonstrator (달 착륙선 개념설계형상 검증모델 가상비행시험)

  • Lee, Won-Beom;Rew, Dong-Young
    • Aerospace Engineering and Technology
    • /
    • v.12 no.1
    • /
    • pp.87-93
    • /
    • 2013
  • The conceptual design lunar lander demonstrator has been developed to use as a test bed for advanced spacecraft technologies and to test a prototype planetary lander capable of vertical takeoff and landing. Size of the lunar lander demonstrator is the same as that of lunar lander conceptually designed, however, the weight of lunar lander demonstrator is designed in 1/6 scale in consideration of gravity difference between moon and earth. The thruster clustering and virtual flight test were performed in the demonstrator fixed on the ground. The demonstrator ground test has been conducted for two months in the test site for the solid motor combustion of the Goheung Flight Center. The purposes of ground test of demonstrator are to demonstrate and verify essential electronics, propulsion system, control algorithm, embedded software, structure and system operation technologies before developing the flight model lander. This paper is described about the virtual flight test including test configuration, test aims and test facilities

Analysis of Saccharomyces Cell Cycle Expression Data using Bayesian Validation of Fuzzy Clustering (퍼지 클러스터링의 베이지안 검증 방법을 이용한 발아효모 세포주기 발현 데이타의 분석)

  • Yoo Si-Ho;Won Hong-Hee;Cho Sung-Bae
    • Journal of KIISE:Software and Applications
    • /
    • v.31 no.12
    • /
    • pp.1591-1601
    • /
    • 2004
  • Clustering, a technique for the analysis of the genes, organizes the patterns into groups by the similarity of the dataset and has been used for identifying the functions of the genes in the cluster or analyzing the functions of unknown gones. Since the genes usually belong to multiple functional families, fuzzy clustering methods are more appropriate than the conventional hard clustering methods which assign a sample to a group. In this paper, a Bayesian validation method is proposed to evaluate the fuzzy partitions effectively. Bayesian validation method is a probability-based approach, selecting a fuzzy partition with the largest posterior probability given the dataset. At first, the proposed Bayesian validation method is compared to the 4 representative conventional fuzzy cluster validity measures in 4 well-known datasets where foray c-means algorithm is used. Then, we have analyzed the results of Saccharomyces cell cycle expression data evaluated by the proposed method.

The Verification Test of Launch Control System Algorithms Using Automated Verification System (자동화 검증시스템을 이용한 발사관제시스템 알고리즘 검증시험)

  • An, Jae-Chel;Moon, Kyung-Rok;Oh, Il-Seok
    • Journal of the Korean Society of Propulsion Engineers
    • /
    • v.25 no.3
    • /
    • pp.127-137
    • /
    • 2021
  • The launch complex(LC) is composed of various facilities. The launch control system that operates remotely those of LC spends much time and labor for developing and verifying its control algorithms. The verification of algorithms is performed by the software developer entering simulated state values based on the test procedure and checking the output result according to the algorithm flow. These verification processes should be performed repeatedly, thus the human errors are easily occurred. In this paper, an efficient automated verification method with a script test procedure is proposed to minimize human errors and shorten the verification duration. We also present the results of the algorithm verification tests for the cases of the compressed gases supply system and the electro pneumatic panel system of LC.

Testing Transactions based on Verification of Isolation Levels (고립화 수준을 검증하기 위한 트랜잭션의 시험)

  • Hong, Seok-Hee
    • The Journal of the Korea Contents Association
    • /
    • v.8 no.7
    • /
    • pp.75-84
    • /
    • 2008
  • Concurrency and synchronization problems are often caused by database applications concurrently accessing databases managed by DBMS. Most commercial DBMSs support isolation levels to resolve these problems. Verification of isolation levels are most important because consistency and integrity constraints of the database can be violated according to isolation levels of transactions that consists of database applications. We propose a test tool set to verify and reveal faulty settings of isolation levels and implement a prototype of the test tool set. The proposed tool set analyzes the SQL statements of ESQL/C programs, attaches the test codes to verify isolation levels, runs the test transactions and detects errors.

On-line Signature Verification using Segment Matching and LDA Method (구간분할 매칭방법과 선형판별분석기법을 융합한 온라인 서명 검증)

  • Lee, Dae-Jong;Go, Hyoun-Joo;Chun, Myung-Geun
    • Journal of KIISE:Software and Applications
    • /
    • v.34 no.12
    • /
    • pp.1065-1074
    • /
    • 2007
  • Among various methods to compare reference signatures with an input signature, the segment-to-segment matching method has more advantages than global and point-to-point methods. However, the segment-to-segment matching method has the problem of having lower recognition rate according to the variation of partitioning points. To resolve this drawback, this paper proposes a signature verification method by considering linear discriminant analysis as well as segment-to-segment matching method. For the final decision step, we adopt statistical based Bayesian classifier technique to effectively combine two individual systems. Under the various experiments, the proposed method shows better performance than segment-to-segment based matching method.