• Journal of Internet Computing and Services
• /
• v.25 no.4
• /
• pp.7-21
• /
• 2024

There are various machine learning techniques such as Reinforcement Learning, Deep Learning, Neural Network Learning, and so on. In recent, Large Language Models (LLMs) are popularly used for Generative AI based on Reinforcement Learning. It makes decisions with the most optimal rewards through the fine tuning process in a particular situation. Unfortunately, LLMs can not provide any explanation for how they reach the goal because the training is based on learning of black-box AI. Reinforcement Learning as black-box AI is based on graph-evolving structure for deriving enhanced solution through adjustment by human feedback or reinforced data. In this research, for mutually exclusive decision-making, Mutually Exclusive Learning (MEL) is proposed to provide explanations of the chosen goals that are achieved by a decision on both ends with specified conditions. In MEL, decision-making process is based on the tree-based structure that can provide processes of pruning branches that are used as explanations of how to achieve the goals. The goal can be reached by trade-off among mutually exclusive alternatives according to the specific contextual conditions. Therefore, the tree-based structure is adopted to provide feasible solutions with the explanations based on the pruning branches. The sequence of pruning processes can be used to provide the explanations of the inferences and ways to reach the goals, as Explainable AI (XAI). The learning process is based on the pruning branches according to the multi-dimensional contextual conditions. To deep-dive the search, they are composed of time window to determine the temporal perspective, depth of phases for lookahead and decision criteria to prune branches. The goal depends on the policy of the pruning branches, which can be dynamically changed by configured situation with the specific multi-dimensional contextual conditions at a particular moment. The explanation is represented by the chosen episode among the decision alternatives according to configured situations. In this research, MEL adopts the tree-based learning model to provide explanation for the goal derived with specific conditions. Therefore, as an example of mutually exclusive problems, employment process is proposed to demonstrate the decision-making process of how to reach the goal and explanation by the pruning branches. Finally, further study is discussed to verify the effectiveness of MEL with experiments.

• Journal of Information Technology Services
• /
• v.12 no.3
• /
• pp.273-287
• /
• 2013

The objective of this study is to empirically develop performance evaluation measures for LIS (logistics information systems) based upon BSC (balanced scorecard) model and determine the relative importance among four perspectives of BSC using AHP (analytic hierarchy process) methodology. Above all, forty nine probable measures for evaluating LIS performance were identified through reviewing the previous research related with SCM (supply chain management), logistics management, and information systems evaluation. And then, these probable measures were examined by means of coincidence analyses using three mutually exclusive criteria(validity of content, ease of measurement and reliability of measurement). Data for conducting the coincidence analyses were collected from LIS users and LIS development personnel. As the results of the coincidence analyses, it was found that 31 ones among 49 probable performance evaluation measures theoretically derived could be qualified for LIS performance evaluation. And AHP analysis showed that the weight of each perspective was respectively as follows : 46.8% for financial perspective, 31.3% for customer perspective, 14.8% for internal business process perspective, and 7.1% for learning and growth perspective. The academic contribution of this study is that performance evaluation measures for LIS could be systematically and empirically developed on the basis of BSC. Also, the results of this study are expected to be used as a practice guideline of evaluating and improving LIS.

• Proceedings of the National Institute of Ecology of the Republic of Korea
• /
• v.1 no.1
• /
• pp.1-8
• /
• 2020

Recent studies revealed that many animals identify individual humans. In this account, we review previous literatures on individual human recognition by wild or domestic animals and discuss the three hypotheses: "high cognitive abilities" hypothesis, "close human contact" and "pre-exposure to stimuli" hypothesis. The three hypotheses are not mutually exclusive. Close human contact hypothesis is an ultimate explanation for adaptive benefits whereas high cognitive abilities and pre-exposure to stimuli hypothesis are proximate explanations for mechanisms to perform such discriminatory behaviour. We report a case study of two bird species in a human-free habitat. Long-tailed skuas, which are known for having high cognitive abilities, exhibited the human discriminatory abilities whereas ruddy turnstones did not display such abilities toward approaching humans. This suggests that highly intelligent species may have this type of discriminatory ability so that they could learn to identify individual humans quickly by pre-exposure to stimuli, even in a human-free habitat. Here, we discuss that human recognition is more common in species with rapid learning ability and it could develop for a short period of time between an intelligent species and human.

• Structural Engineering and Mechanics
• /
• v.85 no.4
• /
• pp.469-484
• /
• 2023

A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

• Journal of Korean Home Economics Education Association
• /
• v.30 no.1
• /
• pp.65-77
• /
• 2018

The purpose of this study is to investigate how to strengthen the formation of relationship competence through the lesson in the housing of the 2015 revised Technology Home Economics curriculum. In order to achieve this goal, we analyzed curriculum documents, achievement standards of curriculum, achievement standards of assessment, and assessment standards of the 2015 revised Technology Home Economics curriculum. Then, we analyzed whether or not the curriculum could be truly reflected in the school site by comparing and analyzing verbs of the assessment standards and curriculum content system. In addition, we suggested the plan to ensure that 2015 revised curriculum can be implemented well by proposing a method of teaching-learning and assessment standards to cultivate the formation of relationship competence among the housing lessons. The results of the study are as follows. First, the 2015 revised Technology Home Economics curriculum are categorized into three subject competences according to the content, but eventually show the content is not mutually exclusive with the three subject competences. It can be said to demonstrate the characteristics of human ability development, that is, one learning can't be related only to one competency. Second, it is difficult to reflect the assessing subject competencies system in school education, where evaluation is carried out based on the assessment standards. This is because the skills of the 2015 revised curriculum documents are partially reflected in the assessment standards. Therefore, this study shows that teachers need to cultivate various subject competencies. In order to overcome the fundamental limitations of the competency-centered curriculum, it is necessary to retrain the teachers as well as to improve teaching and learning methods for operating the curriculum.

• Journal of The Korean Association For Science Education
• /
• v.40 no.2
• /
• pp.97-111
• /
• 2020

In the 2015 revised curriculum, 'Integrated Science' was established to increase convergent thinking and designated as a common subject for all students to learn, regardless of career. In addition, the 2015 revised curriculum introduced 'competence' as a distinctive feature from the previous curriculum. In the 2015 revised curriculum, competencies are divided into core competencies of cross-curricular character and subject competencies based on academic knowledge and skills of the subject. The science curriculum contains five subject competencies: scientific thinking, scientific inquiry, scientific problem solving, scientific communication, scientific participation and life-long learning. However, the description of competencies in curriculum documents is insufficient, and experts' perceptions of competencies are not uniform. Therefore, this study examines the perceptions of science subjects in science high school teachers by deciding that comprehension of competencies should be preceded in order for competency-based education to be properly applied to school sites. First, we analyzed the relationship between achievement standards and subject competencies of integrated science through the operation of an expert working group with a high understanding of the integrated science achievement standards. Next, 31 high school science teachers examined the perception of the five subject competencies through a descriptive questionnaire. The semantic network analysis has been utilized to analyze the teachers' responses. The results of the analysis showed that the three curriculum competencies of scientific inquiry, scientific communication, scientific participation and life-long learning ability are similar to the definitions of teachers and curriculum documents, but in the case of scientific thinking and scientific problem solving, there are some gaps in perception and definition in curriculum documents. In addition, the results of the comprehensive analysis of teachers' perceptions on the five competencies show that the five curriculum competencies are more relevant than mutually exclusive or independent.