• Communications for Statistical Applications and Methods
• /
• v.24 no.6
• /
• pp.543-559
• /
• 2017

Tree-based regression and classification ensembles form a standard part of the data-science toolkit. Many commonly used methods take an algorithmic view, proposing greedy methods for constructing decision trees; examples include the classification and regression trees algorithm, boosted decision trees, and random forests. Recent history has seen a surge of interest in Bayesian techniques for constructing decision tree ensembles, with these methods frequently outperforming their algorithmic counterparts. The goal of this article is to survey the landscape surrounding Bayesian decision tree methods, and to discuss recent modeling and computational developments. We provide connections between Bayesian tree-based methods and existing machine learning techniques, and outline several recent theoretical developments establishing frequentist consistency and rates of convergence for the posterior distribution. The methodology we present is applicable for a wide variety of statistical tasks including regression, classification, modeling of count data, and many others. We illustrate the methodology on both simulated and real datasets.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.28 no.11
• /
• pp.51-57
• /
• 2014

This paper presents an algorithmic type computing technique of process coefficient in predicting model of temperature for reheating furnace and also suggests a design method of neural network model to find an adequate value of process coefficient for arbitrary operating conditions including test conditons. The proposed neural network use furnace temperature, line speed and slab information as input variables, and process coefficient is output variable. Reasonable process coefficients can be obtained by an algorithmic procedure proposed in this paper using process data gathered at test conditons. Also, neural network model output equal process coefficient under same input conditions. This means that adquate process coefficients can be found by only computing neural network model without additive test even if operating conditions vary.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2011.06a
• /
• pp.225-228
• /
• 2011

창의성이 강조되는 시대에 영재 교육의 중요성은 점차 높아지고 있다. 그러나 정보 영재를 위한 연구는 수학이나 과학 영재에 비해 미미한 수준이며, 특히 초등 정보영재를 위한 프로그래밍 교육은 창의적 알고리즘을 개발하는 능력을 기르는 것보다 학습자의 수준에 맞지 않는 특정 프로그래밍 언어의 사용법이나 문법 위주의 교육에 치중하고 있다는 우려의 목소리가 높았다. 이에 본 논문에서는 초등 정보영재의 알고리즘적 사고력을 향상시키기 위한 실생활 중심의 컨텐츠를 제안하고자 한다. 초등학생의 생활과 밀접하게 연관된 주제를 선정하여 학습 동기를 유발하고, Polya의 문제해결모형을 토대로 스스로 이야기를 만들고 그 안에서 알고리즘을 찾아가는 과정을 통해 알고리즘적 사고력을 향상시킬 수 있도록 컨텐츠를 설계하였다.

• International Journal of Computer Science & Network Security
• /
• v.24 no.6
• /
• pp.17-22
• /
• 2024

Software cost and schedule estimation is usually based on the estimated size of the software. Advanced estimation techniques also make use of the diverse factors viz, nature of the project, staff skills available, time constraints, performance constraints, technology required and so on. Usually, estimation is based on an estimation model prepared with the help of experienced project managers. Estimation of software cost is predominantly a crucial activity as it incurs huge economic and strategic investment. However accurate estimation still remains a challenge as the algorithmic models used for Software Project planning and Estimation doesn't address the true dynamic nature of Software Development. This paper presents an efficient approach using the contemporary Constructive Cost Model (COCOMO) augmented with the desirable feature of fuzzy logic to address the uncertainty and flexibility associated with the cost drivers (Effort Multiplier Factor). The approach has been validated and interpreted by project experts and shows convincing results as compared to simple algorithmic models.

• 대한방사선방어학회:학술대회논문집
• /
• 2010.04a
• /
• pp.70-71
• /
• 2010

• Journal of the Korean Chemical Society
• /
• v.44 no.1
• /
• pp.68-73
• /
• 2000

In this study, tlhe relationships of high school students' abilities to solve chemistry problems with cognitive variables (logical thinking ability, mental capacity. and learning strategy) and affective variables(self-efficacy, self-concept of ability, learning goal, and attitude toward science) were investigated. The proportion of variance due to the variables for algorithmic and conceptual problem solving ability was studied by a multiple regression analysis. The results indicated that, among the cognitive variables, the logical thinking ability significantly predicted the algorithmic problem solving ability, and the learning strategy was the best predictor of conceptual problem solving ability although not significant. Among the affective variables studied, the self-concept of alility was the significant predictor of both algorithmic and conceptual problem solving abilities. The seif-efficacy was significantly correlated with conceptual problem solving ability, but it had no predictive power.

• Communications of Mathematical Education
• /
• v.21 no.2 s.30
• /
• pp.153-176
• /
• 2007

The purpose of this paper is to analyze teaching-learning program which can be applied to mathematically gifted students in primary school, Our program is based on constructivist views on teaching and learning of mathematics. Mainly, we study the algorithmic thinking of mathematically gifted students in primary school in connection with the network problems; Eulerian graph problem, the minimum connector problem, and the shortest path problem, The above 3-subjects are not familiar with primary school mathematics, so that we adapt teaching-learning model based on the social constructivism. To achieve the purpose of this study, seventeen students in primary school participated in the study, and video type(observation) and student's mathematical note were used for collecting data while the students studied. The results of our study were summarized as follows: First, network problems based on teaching-learning model of constructivist views help students learn the algorithmic thinking. Second, the teaching-learning model based on constructivist views gives an opportunity of various mathematical thinking experience. Finally, the teaching-learning model based on constructivist views needs more the ability of teacher's research and the time of teaching for students than an ordinary teaching-learning model.

• Journal of The Korean Association of Information Education
• /
• v.17 no.3
• /
• pp.233-242
• /
• 2013

"Problem-solving methods and procedures" sections in the 2009 revised informatics curriculum emphasized active use of algorithmic thinking to solve problems. And it is proposed to solve the various problems of real life using programming language for the implementation of the algorithm. Recently, various Educational Programming Language has been developed for elementary programming activity and many researches showed that students' cognitive burden was reduced in learning programming language with Educational Programming Languages. However implementation of the algorithm is difficult for novice programmer. For the reason, effective way is required for elementary students to connect design of the algorithm and implementation of the algorithm. Therefore, in this study propose the algorithm design worksheets that it is possible to create an algorithm to describe the content needed to implementation in programming education. And this study proved the effect of the algorithm design learning tools through experiment.

• The Journal of Korean Association of Computer Education
• /
• v.22 no.6
• /
• pp.11-18
• /
• 2019

Software education is required from elementary schools to prepare students for the fourth industrial revolution, which aims to improve algorithmic thinking. In general, teaching is divided into two stages: using a flowchart to design algorithms and implementing them through programming. However, converting a flowchart into code and checking the results in an educational programming tool is time consuming and requires additional programming activities. This study proposes a tangible coding tool that enables elementary students to convert algorithms designed at the unplugged activity into educational programming tool codes. This tool was developed in order for students to design algorithms at the level of assembling paper blocks and input them into a programming tool by taking a picture. Sixth graders were participated in this activity to evaluate its usability.

• Journal of Convergence for Information Technology
• /
• v.10 no.8
• /
• pp.173-178
• /
• 2020

In the near future, as artificial intelligence and computing network technology develop, collaboration with artificial intelligence (AI) will become important. In an AI society, the ability to communicate and collaborate among people is an important element of talent. To do this, it is necessary to understand how artificial intelligence based on computer science works. An algorithmic education focused on problem solving and learning is efficient for computer science education. In this study, the results of an assessment of computational thinking at the beginning of the semester, a satisfaction survey at the end of the semester, and academic performance were compared and analyzed for 28 students who received algorithmic education focused on problem-solving learning. As a result of diagnosing students' computational thinking and problem-solving learning, teaching methods, lecture satisfaction, and other environmental factors, a correlation was found, and regression analysis confirmed that problem-solving learning had an effect on improving lecture satisfaction and computational thinking ability. For algorithmic education, if you pursue a problem-solving learning technique and a way to improve students' satisfaction, it will help students improve their problem-solving skills.