• Title/Summary/Keyword: C4.5 algorithm

Search Result 296, Processing Time 0.031 seconds

Usability Analysis of Algorithm Visualization Tool for Learning Basic Algorithms (기초 알고리즘 학습을 위한 알고리즘 시각화 시스템의 효용성 분석)

  • Oh, Kyeong-Sug;Lee, Sang-Jin;Kim, Eung-Kon;Park, Kyoung-Wook;Ryu, Nam-Hoon;Lee, Hye-Mi
    • The Journal of the Korea institute of electronic communication sciences
    • /
    • v.6 no.2
    • /
    • pp.212-218
    • /
    • 2011
  • The curriculum of programming education including algorithm has been recognized as a very important subject to many students majoring in natural sciences and engineering including electronic engineering and computer related departments. This study analyzed usability of the learning system of programming languages using basic algorithms so as for students to easily learn basic algorithm among the entire programming curriculum. The results show that the grade of learning achievement of experimental group using the learning system is 15 points higher than that of non-experimental group and the grade of interest, concentration, effectiveness, understanding, convenience, suitability, and attending a lecture in the next semester are 4 points or more of 5 points criterion.

High-Performance and Low-Complexity Decoding of High-Weight LDPC Codes (높은 무게 LDPC 부호의 저복잡도 고성능 복호 알고리즘)

  • Cho, Jun-Ho;Sung, Won-Yong
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.34 no.5C
    • /
    • pp.498-504
    • /
    • 2009
  • A high-performance low-complexity decoding algorithm for LDPC codes is proposed in this paper, which has the advantages of both bit-flipping (BF) algorithm and sum-product algorithm (SPA). The proposed soft bit-flipping algorithm requires only simple comparison and addition operations for computing the messages between bit and check nodes, and the amount of those operations is also small. By increasing the utilization ratio of the computed messages and by adopting nonuniform quantization, the signal-to-noise ratio (SNR) gap to the SPA is reduced to 0.4dB at the frame error rate of 10-4 with only 5-bit assignment for quantization. LDPC codes with high column or row weights, which are not suitable for the SPA decoding due to the complexity, can be practically implemented without much worsening the error performance.

Optimi Design for R.C. Beam with Discrete Variables (이산형 설계변수를 갖는 철그콘크리트보의 최적설계)

  • 구봉근;한상훈;김홍룡
    • Magazine of the Korea Concrete Institute
    • /
    • v.5 no.4
    • /
    • pp.167-178
    • /
    • 1993
  • The objective of this paper is to look into the possibility of the detailed and practical optimum design of rt:inforced concrete beam using methods oi discrete mathematical programming. In this discrete optimum formulation, the design variables are the overall depth, width and effective depth of members, and area of longitudinal reinforcement. In addition, the details such as the amount of web reinforcement and cutoff points of longitudinal reinforcement are also considered as variables. Total cost has been used as the objective function. The constraints include the code requirments such as flexural strength, shear strength, ductility, serviceability, concrete cover. spacing, web reinforcement, and development length and cutoff points of longitudinal renforcement. An optimization algorithm is presented for effective optimum design of R.C. beam with discrete de sign variables. First, the continuous variable optimization can be achieved by Feasible Direction Method. Using the results obtained from the continuous variable optimization, a branch and bound method is used to obtained the discrete design values. The proposed algorithm is applied to test problem for reliability, and the results are compared with those of graphical method and rounded-up method. And a simply supported R.C. beam and a two-span continuous R.C. beam are presented as numerical examples for effectiveness and applicability. It is considered that the presented algorithm can be effectively applied to the discrete optimum design of R.C. beams.

Estimation of Aboveground Biomass Carbon Stock in Danyang Area using kNN Algorithm and Landsat TM Seasonal Satellite Images (kNN 알고리즘과 계절별 Landsat TM 위성영상을 이용한 단양군 지역의 지상부 바이오매스 탄소저장량 추정)

  • Jung, Jae-Hoon;Heo, Joon;Yoo, Su-Hong;Kim, Kyung-Min;Lee, Jung-Bin
    • Journal of Korean Society for Geospatial Information Science
    • /
    • v.18 no.4
    • /
    • pp.119-129
    • /
    • 2010
  • The joint use of remotely sensed data and field measurements has been widely used to estimate aboveground carbon stock in many countries. Recently, Korea Forest Research Institute has developed new carbon emission factors for kind of tree, thus more accurate estimate is possible. In this study, the aboveground carbon stock of Danyang area in South Korea was estimated using k-Nearest Neighbor(kNN) algorithm with the 5th National Forest Inventory(NFI) data. Considering the spectral response of forested area under the climate condition in Korea peninsular which has 4 distinct seasons, Landsat TM seasonal satellite images were collected. As a result, the estimated total carbon stock of Danyang area was ranged from 3542768.49tonC to 3329037.51tonC but seasonal trends were not found.

The empirical comparison of efficiency in classification algorithms (분류 알고리즘의 효율성에 대한 경험적 비교연구)

  • 전홍석;이주영
    • Journal of the Korea Safety Management & Science
    • /
    • v.2 no.3
    • /
    • pp.171-184
    • /
    • 2000
  • We may be given a set of observations with the classes or clusters. The aim of this article is to provide an up-to-date review of different approaches to classification, compare their performance on a wide range of challenging data-sets. In this paper, machine learning algorithm classifiers based on CART, C4.5, CAL5, FACT, QUEST and statistical discriminant analysis are compared on various datasets in classification error rate and algorithms.

  • PDF

A study on data mining techniques for soil classification methods using cone penetration test results

  • Junghee Park;So-Hyun Cho;Jong-Sub Lee;Hyun-Ki Kim
    • Geomechanics and Engineering
    • /
    • v.35 no.1
    • /
    • pp.67-80
    • /
    • 2023
  • Due to the nature of the conjunctive Cone Penetration Test(CPT), which does not verify the actual sample directly, geotechnical engineers commonly classify the underground geomaterials using CPT results with the classification diagrams proposed by various researchers. However, such classification diagrams may fail to reflect local geotechnical characteristics, potentially resulting in misclassification that does not align with the actual stratification in regions with strong local features. To address this, this paper presents an objective method for more accurate local CPT soil classification criteria, which utilizes C4.5 decision tree models trained with the CPT results from the clay-dominant southern coast of Korea and the sand-dominant region in South Carolina, USA. The results and analyses demonstrate that the C4.5 algorithm, in conjunction with oversampling, outlier removal, and pruning methods, can enhance and optimize the decision tree-based CPT soil classification model.

A Study on the computer-aided synthesis of TANT network (TANT회로망의 계산기 이용 합성에 관한 연구)

  • 안광선;박규태
    • Journal of the Korean Institute of Telematics and Electronics
    • /
    • v.17 no.6
    • /
    • pp.51-57
    • /
    • 1980
  • Any switching function can be constructed with universal building block of MAND gate. Threelevel AND-NOT logic networks with only true inputs are called TANT networks. Systematic approach to TANT minimization starts from the UF type minterm with the smallest subscript and ends when UF type minterms are all covered. Optinal PEI is composed of CPPI or EPPi without C-C table. The algorithm in this work is usful in solving TANT optimization porblem of four or five variables by hand solution. When variable are six or more, it is required to be solved by computer, A CAD software package of this algorithm with FORTRAN IV language is made to solve such problems.

  • PDF

A MAC System Design for High-speed UWB SoC (고속 UWB SoC의 MAC 시스템 설계)

  • Kim, Do-Hoon;Wee, Jeong-Wook;Lee, Chung-Yong
    • Journal of the Institute of Electronics Engineers of Korea TC
    • /
    • v.48 no.4
    • /
    • pp.1-5
    • /
    • 2011
  • We present the implementation of MAC system for MBOA UWB SoC. The implemented MBOA MAC algorithm is not master control mechanism, but distributed network mechanism. Therefore, mesh network can be easily constructed because MAC consists of distributed network and administrates network. The ARM926EJ with cache is adopted for high performnace and AMBA bus is applied for system design and reuse. In addition, the system operating clock management algorithm is implemented for low power consumption. The dedicated DMA for MAC is designed between the system memory buffer and MAC hardware, and the dedicated DMA for USB 2.0 is also implemented between system memory buffer and host for high data transaction.

A Study on Geometric Modeling and Generation of 4-axis NC Data for Single Setup of Small Marine Propeller (선박용 소형 프로펠러의 곡면 모델링 및 단일 셋업에 의한 4축 NC가공 데이터 생성에 관한 연구)

  • 이재현;이철수
    • Korean Journal of Computational Design and Engineering
    • /
    • v.7 no.4
    • /
    • pp.254-261
    • /
    • 2002
  • Small marine propeller is generally machined by 5-axis machining. This paper suggests a method to create geometric model from point array data and 4-axis machining NC data for propeller. With conventional method, the setting posture should be changed, because propeller has front and back surface of wing. The change of setting posture has a bad influence on precision of propeller. So this paper pro-poses a method to machine propeller by single setup for 4-axis machining. The cutter moves to parallel direction of the XY plane. To determine the cutter orientation efficiently, the' tilting guiding line' is proposed. A proposed algorithm is written in C language and successfully applied to the 5-axis milling machine of industrial field.

Haplotype Analysis and Single Nucleotide Polymorphism Frequency of PEPT1 Gene (Exon 5 and 16) in Korean (한국인에 있어서 PEPT1 유전자(exon 5 및 16)의 단일염기변이 빈도 및 일배체형 분석)

  • Kim, Se-Mi;Lee, Sang-No;Kang, Hyun-Ah;Cho, Hea-Young;Lee, Il-Kwon;Lee, Yong-Bok
    • Journal of Pharmaceutical Investigation
    • /
    • v.39 no.6
    • /
    • pp.411-416
    • /
    • 2009
  • The aim of this study was to investigate the frequency of the SNPs on PEPT1 exon 5 and 16 and to analyze haplotype frequency on PEPT1 exon 5 and 16 in Korean population. A total of 519 healthy subjects was genotyped for PEPT1, using pyrosequencing analysis and polymerase chain reaction-based diagnostic tests. Haplotype was statistically inferred using an algorithm based on the expectation-maximization (EM). PEPT1 exon 5 G381A genotyping revealed that the frequency for homozygous wild-type (G/G), heterozygous (G/A) and homozygous mutant-type (A/A) was 30.4, 53.4 and 16.2%, respectively. PEPT1 exon 16 G1287C genotyping revealed that the frequency for homozygous G/G, heterozygous G/C and homozygous C/C type was 88.8, 10.0 and 1.2%, respectively. Based on these genotype data, haplotype analysis between PEPT1 exon 5 G381A and exon 16 G1287C using HapAnalyzer and PL-EM has proceeded. The result has revealed that linkage disequilibrium between alleles is not obvious (|D'|=0.3667).