• Journal of Information Technology Applications and Management
• /
• 제14권4호
• /
• pp.175-198
• /
• 2007

Technology Acceptance Model (TAM) has been a basis model for testing technology use. Post researches of TAM have been conducted with the updating the TAM by adding new independent variables in order to increase the explanation power of the model. However, one problem is that different independent variables have to be introduced to keep the explanation power whenever applying to particular technology. This reduces the generality of the research model. Thus in order to increase the generality of the model, this study reviewed the previous researches and collected the independent variables used, and regrouped them into three basic independent constructs. New research model was designed with three basic independent constructs with four constructs selected for the mandatory IT environment and voluntary IT environment, and the structured equations analysis(AMOS) was applied to find the significant causal effect relationships between constructs in addition to the explanation power of the model. Finally, this study concluded that new TAM could be used to explain the users' adopting new technology without any adding new particular independent variables.

• 대한지역사회영양학회지
• /
• 제20권1호
• /
• pp.1-10
• /
• 2015

Objectives: This study was conducted to develop and validate Eating Behaviors Test form (EBT) for infants and young children, including eating behaviors of their parents and parental feeding practices. Methods: Draft version of EBT form was developed after a pretest on 83 mothers. It was consisted of 42 questions including 3 components; eating behavior of children, eating behavior of parents, and parental feeding practices. Using these questionnaires, the first survey was conducted on 320 infants and children, 1 to 6 year old, for exploratory factor analysis, and the second survey was collected on 731 infants and children for confirmatory factor analysis. Results: Exploratory factor analysis on 42 questions of EBT form resulted in 3 factor model for children's eating behavior, 3 factor model for parents' eating behavior, and 1 factor model for parental feeding practices. Three factors for children's eating behavior could be explained as follows; factor 1, pickiness (reliability ${\alpha}=0.89$; explanation of variance=27.79), factor 2, over activity (${\alpha}=0.80$, explanation of variance=16.51), and factor 3, irregularity (${\alpha}=0.59$, explanation of variance=10.01). Three factors for mother's eating behavior could be explained as follows; factor 1,irregularities (${\alpha}=0.73$, explanation of variance=21.73), factor 2, pickiness (${\alpha}=0.65$, explanation of variance= 20.16), and factor 3, permissiveness (${\alpha}=0.60$, explanation of variance=19.13). Confirmatory factor analysis confirmed an acceptance fit for these models. Internal consistencies for these factors were above 0.6. Conclusions: Our results indicated that EBT form is a valid tool to measure comprehensive eating and feeding behaviors for infants and young children.

• Journal of Information Science Theory and Practice
• /
• 제1권3호
• /
• pp.6-11
• /
• 2013

A new explanation, using exponential functions, is given for the S-shaped functional relation between the mean citation score and the proportion of top 10% (and other percentages) publications for the 500 Leiden Ranking universities. With this new model we again obtain an explanation for the concave or convex relation between the proportion of top $100{\theta}%$ publications, for different fractions of ${\theta}$.

• BMB Reports
• /
• 제37권4호
• /
• pp.394-401
• /
• 2004

The purpose of this paper is to suggest that the prominence of Haldane's explanation for enzyme catalysis significantly hinders investigations in understanding enzyme structure and function. This occurs despite the existence of much evidence that the Haldane model cannot embrace. Some of the evidence, in fact, disproves the model. A brief history of the explanation of enzyme catalysis is presented. The currently accepted view of enzyme catalysis -- the Haldane model -- is examined in terms of its strengths and weaknesses. An alternate model for general enzyme catalysis (the Shifting Specificity model) is reintroduced and an assessment of why it may be superior to the Haldane model is presented. Finally, it is proposed that a re-examination of many current aspects in enzyme structure and function (specifically, protein folding, x-ray and NMR structure analyses, enzyme stability curves, enzyme mimics, catalytic antibodies, and the loose packing of enzyme folded forms) in terms of the new model may offer crucial insights.

• 한국경영과학회지
• /
• 제37권2호
• /
• pp.113-125
• /
• 2012

The logistic model and the Bass model have diverse names and formulae in diffusion theory. This diversity makes users or readers confused while it also contributes to the flexibility of modeling. The method of handling the integration constant, which is generated in process of deriving the closed form solution of the differential equation for a diffusion model, results in two different 'actual' models. We rename the actual four models and propose the usage of the models with respect to the purpose of model applications. The application purpose would be the explanation of historical diffusion pattern or the forecasting of future demand. Empirical validation with 86 historical diffusion data shows that misuse of the models can draw improper conclusions for the explanation of historical diffusion pattern.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제42권3호
• /
• pp.303-325
• /
• 2003

This study sought to provide an explanation of university students' concept understanding on the infinity and infinite process and utilized a psychological constructivist perspective to examine the differences in transitions that students make from static concept of limit to actualized infinity stage in context of problems. Open-ended questions were used to gather data that were used to develop an explanation concerning student understanding. 47 university students answered individually and were asked to solve 16 tasks developed by Petty(1996). Microgenetic method with two cases from the expert-novice perspective were used to develop and substantiate an explanation regarding students' transitions from static concept of limit to actualized infinity stage. The protocols were analyzed to document student conceptions. Cifarelli(1988)'s levels of reflective abstraction and Robert(1982) and Sierpinska(1985)'s three-stage concept development model of infinity and infinite process provided a framework for this explanation. Students who completed a transition to actualized infinity operated higher levels of reflective abstraction than students who was unable to complete such a transition. Developing this ability was found to be critical in achieving about understanding the concept of infinity and infinite process.

• Journal of Information Processing Systems
• /
• 제19권6호
• /
• pp.803-816
• /
• 2023

Explainable artificial intelligence is a method that explains how a complex model (e.g., a deep neural network) yields its output from a given input. Recently, graph-type data have been widely used in various fields, and diverse graph neural networks (GNNs) have been developed for graph-type data. However, methods to explain the behavior of GNNs have not been studied much, and only a limited understanding of GNNs is currently available. Therefore, in this paper, we propose an explanation method for node classification using graph convolutional networks (GCNs), which is a representative type of GNN. The proposed method finds out which features of each node have the greatest influence on the classification of that node using GCN. The proposed method identifies influential features by backtracking the layers of the GCN from the output layer to the input layer using the gradients. The experimental results on both synthetic and real datasets demonstrate that the proposed explanation method accurately identifies the features of each node that have the greatest influence on its classification.

• 경영정보학연구
• /
• 제24권1호
• /
• pp.105-123
• /
• 2022

신용리스크 관리에 해당하는 부도예측모형은 기업에 대한 신용평가라고도 볼 수 있으며 은행을 비롯한 금융기관의 신용평가모형의 기본 지식기반으로 새로운 인공지능 기술을 접목할 수 있는 유망한 분야로 손꼽히고 있다. 고도화된 모형의 실제 응용은 사용자의 수용도가 중요하나 부도예측모형의 경우, 금융전문가 혹은 고객에게 모형의 결과에 대한 설명이 요구되는 분야로 설명력이 없는 모형은 실제로 도입되고 사용자들에게 수용되기에는 어려움이 있다. 결국 모형의 결과에 대한 설명은 모형의 사용자에게 제공되는 것으로 사용자가 납득할 수 있는 설명을 제공하는 것이 모형에 대한 신뢰와 수용을 증진시킬 수 있다. 본 연구에서는 머신러닝 기반 모형에 설명력을 제고하는 방안으로 설명대상 인스턴스에 대하여 로컬영역에서의 설명을 제공하고자 한다. 이를 위해 설명대상의 로컬영역에 유전알고리즘(GA)을 이용하여 가상의 데이터포인트들을 생성한 후, 로컬 대리모델(surrogate model)로 연관규칙 알고리즘을 이용하여 설명대상에 대한 규칙기반 설명(rule-based explanation)을 생성한다. 해석 가능한 로컬 모델의 활용으로 설명을 제공하는 기존의 방법에서 더 나아가 본 연구는 부도예측모형에 이용된 재무변수의 특성을 반영하여 연관규칙으로 도출된 설명에 도메인 지식을 통합한다. 이를 통해 사용자에게 제공되는 규칙의 현실적 가능성(feasibility)을 확보하고 제공되는 설명의 이해와 수용을 제고하고자 한다. 본 연구에서는 대표적인 블랙박스 모형인 인공신경망 기반 부도예측모형을 기반으로 최신의 규칙기반 설명 방법인 Anchor와 비교하였다. 제안하는 방법은 인공신경망 뿐만 아니라 다른 머신러닝 모형에도 적용 가능한 방법(model-agonistic method)이다.

• 한국지능시스템학회논문지
• /
• 제19권2호
• /
• pp.218-223
• /
• 2009

이 논문은 개별화된 오답노트 모형 개발에 관한 문제를 고찰한다. 여기에서 사용된 방법은 집단지성을 이용하여 오답노트해설을 추가하고, 오답분석을 통하여 유사한 패턴의 오답자 해설노트를 참조한다. 이 논문의 주된 결과는 정답에 대한 해설이 아닌 오답에 이르는 틀린 과정을 찾고 오답을 정리하는 것이다. 제안된 방법으로 기존 오답노트 시스템의 개선된 해결책을 찾을 수 있다.

• 대한임베디드공학회논문지
• /
• 제16권4호
• /
• pp.145-151
• /
• 2021

In this paper, we propose a novel visualization technique to explain the predictions of deep neural networks. We use knowledge distillation (KD) to identify the interior of a black-box model for which we know only inputs and outputs. The information of the black box model will be transferred to a white box model that we aim to create through the KD. The white box model will learn the representation of the black-box model. Second, the white-box model generates attention maps for each of its layers using Grad-CAM. Then we combine the attention maps of different layers using the pixel-wise summation to generate a final saliency map that contains information from all layers of the model. The experiments show that the proposed technique found important layers and explained which part of the input is important. Saliency maps generated by the proposed technique performed better than those of Grad-CAM in deletion game.