• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.508-524
• /
• 2006

We present our experience of combining, in a realistic setting, a static analyzer with a statistical analysis. This combination is in order to reduce the inevitable false alarms from a domain-unaware static analyzer. Our analyzer named Airac(Array Index Range Analyzer for C) collects all the true buffer-overrun points in ANSI C programs. The soundness is maintained, and the analysis' cost-accuracy improvement is achieved by techniques that static analysis community has long accumulated. For still inevitable false alarms (e.g. Airac raised 970 buffer-overrun alarms in commercial C programs of 5.3 million lines and 737 among the 970 alarms were false), which are always apt for particular C programs, we use a statistical post analysis. The statistical analysis, given the analysis results (alarms), sifts out probable false alarms and prioritizes true alarms. It estimates the probability of each alarm being true. The probabilities are used in two ways: 1) only the alarms that have true-alarm probabilities higher than a threshold are reported to the user; 2) the alarms are sorted by the probability before reporting, so that the user can check highly probable errors first. In our experiments with Linux kernel sources, if we set the risk of missing true error is about 3 times greater than false alarming, 74.83% of false alarms could be filtered; only 15.17% of false alarms were mixed up until the user observes 50% of the true alarms.

• Journal of the Korean School Mathematics Society
• /
• v.22 no.1
• /
• pp.1-21
• /
• 2019

Mathematics is one of the subjects in which learners seriously experience under-achievements in school education. Middle school level mathematics especially plays such a role as a bridge between elementary level informal mathematics and high school level formal mathematics that learners' under-achievements in the middle school level mathematics may yield more serious under-achievements later. Therefore it is crucial to prevent learners' later under-achievements that we analyze the status of under achievements including analysing various under-achievers' errors in the middle school level mathematics. From this point of view, we analysed errors of mathematics under-achievers in understanding the concept of the square root of positive numbers and related calculations in this paper. As the results of our research, we found some unexpected errors of 'some mathematics under-achievers regarding the mathematical symbol ${\surd}$ of square root as a parenthesis ( ), and others interpret $x=-2{\pm}{\sqrt{10}}$ as x=-2 or ${\pm}{\sqrt{10}}$.' that suggest the necessity of more various and in-depth discussions and researches of analysis on learners' errors and misconceptions in all areas of school mathematics.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.1
• /
• pp.137-148
• /
• 1997

Diagnostic problem solving is a major application area of knowledge-based systems research. However, most of the current approaches, both heuristic and model-based, are designed to identify single faults, and do not generalize easily to multiple fault diagnosis without exhibiting exponential behavior in the amount of computation required. In this paper, we employ a decomposition approach based on system configuration to generate an efficient algorithm for multiple fault diagnosis. The basic idea of the algorithm is to reduce the inherent combinatorial explosion that occurs in generating multiple faults by partitioning the circuit into groups that correspond to output measurement points. Rules are multiple faults by partitioning the circuit into groups that correspond to output measurement points. rules are developed for combining candidates from individual groups, and forming consistent sets of minimal candidates.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2016.05a
• /
• pp.191-193
• /
• 2016

인적오류 예방은 해양사고 예방에 가장 중요한 이슈로 현재 인식되고 있다. 현재 이러한 인적오류를 예방하기 위한 다양한 과학적인 기법들이 등장하고 있으나, 실제 인적오류를 예방할 수 있는 기법은 아직 개발되어 있지 못한 실정이다. 그 이유는 인적오류의 발생 원인과 특징이 사람을 대상으로 하기 때문에 실로 방대하고 원인식별이 어려우며, 원인과 결과 사이의 인과관계 구축에는 한계가 있기 때문이다. 기존 개발된 다양한 기법들은 이론적으로는 완벽할 수 있으나, 실제 방대한 원인과 결과 사이에 형성된 연계체인을 모두 흡수하기가 곤란하기 때문이다. 현재 IMO의 공식안전성평가(FSA) 기법이 해상분야에 널리 적용되고 있으나 구체적으로 어떠한 기법을 적용하여 인적오류를 적용할 수 있는지에 대해서는 아직도 애매모호한 실정이다. FTA, ETA, FEMA, SWIFT 등 다양한 분석기법의 등장과 AI, Fuzzy, MMC, Kalman 등 기초과학분야의 기본적인 이론과 기술을 적용할 수 있으나 인간의 인적오류 식별과 분석 및 평가와 예측에는 한계가 있는 것이 현재의 실정이다. 한편 최근에는 기존에 많은 문제점을 내포하고 있는 것으로 고려되었던 베이지안 네트워크(Bayesian Network, BN)가 다시 분석과 예측 분야에 등장하고 있는데, BN의 장점을 수용하고 단점을 해결할 수 있는 방법들이 연구되고 있기 때문이다. BN의 장점은 전방추론과 후방추론을 적용하여 사고의 원인과 결과를 분석한 후, 이에 대한 해결 방안을 식별할 수 있기 때문이다. BN의 단점은 이진(binary) 구조의 데이터만을 수용할 수 있기 때문에 상관 변수들이 방대한 경우 계산시간이 방대해지고 이를 모두 수용할 수 있는 방법이 없기 때문이다. 따라서 BN 구조를 어떻게 설계하는냐가 최근의 이수로 등장하고 있다. 본 연구에서는 이러한 제 문제점을 고찰하고 인적오류 모델 개발에 최적인 방법 또는 기술을 모색하는데 있다.

• Journal of Satellite, Information and Communications
• /
• v.1 no.2
• /
• pp.51-55
• /
• 2006

Digital satellite communication systems are usually integrated with terrestrial systems to provide various services. In these cases, they should satisfy the performance objectives defined by the terrestrial systems. Recommendation ITU-R S.1062 specifies the error performance objectives of digital satellite communication systems operating below 15 GHz. The error performance are given in terms of bit error probability divided by the number of the average bit errors in the burst ($\alpha$). This paper presents a theoretical method to estimate $\alpha$ that is a very important parameter in the satellite communication systems to analyze the error performance objectives. We show performance estimation result of DVB-RCS turbo code using the presented method, and verify them by comparing to the simulation results.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1999.11a
• /
• pp.270-273
• /
• 1999

Y2k will be able to enormous disaster. The many make an effort to find a solution to problem of Y2k. Problem of Y2k must solution to as follow. First, problem of Y2k solution organization must constructed. Second, in step with each stage-the first, developing and complete stage, stage of Y2k solution must be constructed. Third, solution of Y2k must construct to hierarchy. hierarchy structure constructed form six stage to first stage, first stage is investigation resources, second stage is estimation influence, third stage is planing conversion, fourth stage is working conversion, fifte spot, sixth stage is diffusion on the spot.

• Education of Primary School Mathematics
• /
• v.22 no.4
• /
• pp.281-294
• /
• 2019

The purpose of this study is to analyze errors and treatment behavior during the course of mathematics learning of academic achievement by applying reflective activities in the second semester of the third year of elementary school. The study participants are students from two classes, 21 from the third-grade S elementary school in Seoul and 20 from the comparative class. In the case of the experiment group, the multiplication unit was reconstructed into a mathematics class that applied reflective activities. They were pre-post-test to examine the changes in students' mathematics performance, and mathematical communication was recorded and analyzed for the focus group to analyze the patterns of learners' error handling in the reflective activities. In addition, they recorded and analyzed students' activities and conversations for error type and error handling. As a result of the study, the student's mathematics achievement was increased using reflective activities. When learning double digit multiplication, the error types varied. It was also confirmed that the reflective activities helped learners reflect on the multiplication algorithm and analyze the error-ridden calculations to reflect on and deal with their errors.

• Proceedings of the Korea Society of Mathematical Education Conference
• /
• 1990.07a
• /
• pp.62-62
• /
• 1990

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.130-137
• /
• 2001

최대 엔트로피 모델(maximum entropy model)은 여러 가지 자연언어 문제를 학습하는데 성공적으로 적용되어 왔지만, 두 가지의 주요한 문제점을 가지고 있다. 그 첫번째 문제는 해당 언어에 대한 많은 사전 지식(prior knowledge)이 필요하다는 것이고, 두번째 문제는 계산량이 너무 많다는 것이다. 본 논문에서는 텍스트 단위화(text chunking)에 최대 엔트로피 모델을 적용하는 데 나타나는 이 문제점들을 해소하기 위해 새로운 방법을 제시한다. 사전 지식으로, 간단한 언어 모델로부터 쉽게 생성된 결정트리(decision tree)에서 자동적으로 만들어진 규칙을 사용한다. 따라서, 제시된 방법에서의 최대 엔트로피 모델은 결정트리를 보강하는 방법으로 간주될 수 있다. 계산론적 복잡도를 줄이기 위해서, 최대 엔트로피 모델을 학습할 때 일종의 능동 학습(active learning) 방법을 사용한다. 전체 학습 데이터가 아닌 일부분만을 사용함으로써 계산 비용은 크게 줄어 들 수 있다. 실험 결과, 제시된 방법으로 결정트리의 오류의 수가 반으로 줄었다. 대부분의 자연언어 데이터가 매우 불균형을 이루므로, 학습된 모델을 부스팅(boosting)으로 강화할 수 있다. 부스팅을 한 후 제시된 방법은 전문가에 의해 선택된 자질로 학습된 최대 엔트로피 모델보다 졸은 성능을 보이며 지금까지 보고된 기계 학습 알고리즘 중 가장 성능이 좋은 방법과 비슷한 성능을 보인다 텍스트 단위화가 일반적으로 전체 구문분석의 전 단계이고 이 단계에서의 오류가 다음 단계에서 복구될 수 없으므로 이 성능은 텍스트 단위화에서 매우 의미가 길다.

• Education of Primary School Mathematics
• /
• v.14 no.3
• /
• pp.317-328
• /
• 2011

We analyzed errors committed by Korean prospective elementary teachers in finding and interpreting quotient and remainder within measurement division of fractions. 65 prospective elementary teachers were participated in this study. They solved a word problem about measurement division of fractions. We analyzed solutions of all participants, and interviewed 5 participants of them. The results reveal many of these prospective teachers could not tell what fractional part of division result means. Thses results suggest that teacher preparation program should emphasize interpreting calculation results within given situations.