• Analyses & Alternatives
• /
• v.8 no.3
• /
• pp.125-150
• /
• 2024

This paper explores the integration of artificial intelligence and causal inference in social science research, focusing on causal deep learning. We examine key theories including Pearl's Structural Causal Model, Rubin's Potential Outcomes Framework, and Schölkopf's Causal Representation Learning. Methodologies such as structural causal models with deep learning, counterfactual reasoning, and causal discovery algorithms are discussed. The paper presents applications in social media analysis, economic policy, public health, and education, demonstrating how causal deep learning enables nuanced understanding of complex social phenomena. Key challenges addressed include model complexity, causal identification, interpretability, and ethical considerations like fairness and privacy. Future research directions include developing new AI architectures, real-time causal inference, and multi-domain generalization. While limitations exist, causal deep learning shows significant potential for enhancing social science research and informing evidence-based policy-making, contributing to addressing complex social challenges globally.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.19 no.4
• /
• pp.45-52
• /
• 2023

With the advent of deep learning, Artificial Intelligence (AI) technology has experienced rapid advancements, extending its application across various industrial sectors. However, the focus has shifted from the independent use of AI technology to its dispersion and proliferation through the open AI ecosystem. This shift signifies the transition from a phase of research and development to an era where AI technology is becoming widely accessible to the general public. However, as this dispersion continues, there is an increasing demand for the verification of outcomes derived from AI technologies. Causal AI applies the traditional concept of causal inference to AI, allowing not only the analysis of data correlations but also the derivation of the causes of the results, thereby obtaining the optimal output values. Causal AI technology addresses these limitations by applying the theory of causal inference to machine learning and deep learning to derive the basis of the analysis results. This paper analyzes recent cases of causal AI technology and presents the major tasks and directions of causal AI, extracting patterns between data using the correlation between them and presenting the results of the analysis.

• Journal of The Korean Association For Science Education
• /
• v.26 no.1
• /
• pp.1-8
• /
• 2006

This study investigated the causal relationships between high school student multiple-choice algorithmic chemistry problem solving and 1) the motivational variables of achievement goal (task goal/performance goal/performance-avoidance) and perceived ability, and 2) the cognitive variables of learning strategy (deep learning/surface learning) and self-regulation. Path analysis supported a causal model in which perceived ability and task goal were found to positively influence algorithmic chemistry problem-solving ability via self-regulation. In particular it was found that perceived ability directly influenced algorithmic chemistry problem-solving ability. Moreover, deep learning was found to have been influenced by perceived ability and task goal, while surface learning was influenced by performance-avoidance goal. Lastly, there did not appear to be any causal relationship between learning strategy and algorithmic chemistry problem-solving ability.

• Korean System Dynamics Review
• /
• v.13 no.4
• /
• pp.33-55
• /
• 2012

The research is to identify important diffusion factors and their effects on green car diffusion process using system dynamics perspectives and a causal-loop analysis. Through a deep review on previous research, we have found the important factors of green car diffusion process. Price, driving range, network effect, recharge system, fuel cost had important facilitation on consumer attraction and green car diffusion. Based on the review, we have constructed a causal loop diagram explaining hybrid car diffusion process. We have found 3 important reinforcing loops in the causal loop diagram. Loop for learning & economies of scale(supply side), loop for network effect(consumer side), and loop for battery development(technology side) had most significant roles in the whole diffusion process. Through a deliberate analysis on the 3 causal loops, we have found meaningful results. First, there seems to exist a critical mass in the diffusion. Second, of the 3 loops, the battery technology had most significant role. Third, not consumer installed base but sales must be a standard to decide whether the critical mass is achieved or not. Based on these findings, several meaningful implications are suggested for the government and corporations related to the green car industries.

• Journal of the Korea Society for Simulation
• /
• v.30 no.4
• /
• pp.1-8
• /
• 2021

Deep learning is one of the most widely accepted methods for the forecasting of time series data which have the complexity and non-linear behavior. In this paper, we investigate the modification of a state-of-art WaveNet deep learning architecture and walk forward validation (WFV) in order to forecast electric power consumption data 24-hour-ahead. WaveNet originally designed for raw audio uses 1D dilated causal convolution for long-term information. First of all, we propose a modified version of WaveNet which activates real numbers instead of coded integers. Second, this paper provides with the training process with tuning of major hyper-parameters (i.e., input length, batch size, number of WaveNet blocks, dilation rates, and learning rate scheduler). Finally, performance evaluation results show that the prediction methodology based on WFV performs better than on the traditional holdout validation.

• Journal of The Korean Association For Science Education
• /
• v.17 no.4
• /
• pp.415-423
• /
• 1997

The purposes of this study were to investigate the intercorrelations among various motivational patterns and learning strategies and to examine the differences in motivation and strategy usage in terms of students' science achievement level, gender, and grade. A questionnaire on achievement goal, self-efficacy, self-concept of ability, expectancy, value, causal attributions, and learning strategies was administered to 360 junior high/high school students (178 males, 182 females). Students who adopted performance-oriented goal tended not to be task oriented. Task-oriented students had high levels of self-efficacy, high self-concept of ability, and expectancies for future performance in science. They also valued science and attributed thier failures to the lack of effort. However, performance-oriented students evaluated their ability negatively, did not value science, and attributed thier failures to uncontrollable causes. With respect to learning strategy, task-oriented students tended to use deep-level strategy, whereas performance-oriented students tended to use surface-level strategy and not to use deep-level strategy. High-achieving students, boys, and junior high school students were more task-oriented, evaluated their ability more positively, and valued science more than low-achieving students, girls, and high school students, respectively. High-achieving students and boys also used deep-level strategy more than each of their counterparts. However, no significant difference in learning strategy was found between junior high school students and high school students. Educational implications of these findings are discussed.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.38 no.4
• /
• pp.621-625
• /
• 2018

This paper proposes a method for bridge damage factor recognition from inspection reports using deep learning. Bridge inspection reports contains inspection results including identified damages and causal analysis results. However, collecting such information from inspection reports manually is limited due to their considerable amount. Therefore, this paper proposes a model for recognizing bridge damage factor from inspection reports applying Named Entity Recognition (NER) using deep learning. Named Entity Recognition, Word Embedding, Recurrent Neural Network, one of deep learning methods, were applied to construct the proposed model. Experimental results showed that the proposed model has abilities to 1) recognize damage and damage factor included in a training data, 2) distinguish a specific word as a damage or a damage factor, depending on its context, and 3) recognize new damage words not included in a training data.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.7
• /
• pp.307-314
• /
• 2022

Recently, researches using deep learning-based models are being actively conducted to replace statistical-based time series forecast techniques to predict electric power demand. The result of analyzing the researches shows that the performance of the LSTM-based prediction model is acceptable, but it is not sufficient for long-term regional-wide power demand prediction. In this paper, we propose a WaveNet deep learning model to predict electric power demand 24-hour-ahead with temperature data in order to achieve the prediction accuracy better than MAPE value of 2% which statistical-based time series forecast techniques can present. First of all, we illustrate a delated causal one-dimensional convolutional neural network architecture of WaveNet and the preprocessing mechanism of the input data of electric power demand and temperature. Second, we present the training process and walk forward validation with the modified WaveNet. The performance comparison results show that the prediction model with temperature data achieves MAPE value of 1.33%, which is better than MAPE Value (2.33%) of the same model without temperature data.

• Journal of Information Technology Applications and Management
• /
• v.17 no.3
• /
• pp.57-69
• /
• 2010

Many organizations have had deep interests in studies concerning leadership and in academic areas, in not only management but also psychology. Until now, leadership has been accentuated by managers or team leaders especially. Recently, however, the concept of self-leadership directing one's own activities through self-control or self-management is being focused on practices and in academia. This study is to investigate the influence between self-leadership strategies and learning performance in IT classes mediated by attitude of attendance focused on the social science students in a university. Research results can give us direction of task-taking attitudes in firms or learning attitudes in teaching organizations and implications to human resource managers who are in charge of improving learning performance or productivity.

• Journal of dental hygiene science
• /
• v.24 no.1
• /
• pp.22-28
• /
• 2024

Background: Particulate matter (PM) has been extensively observed due to its negative association with human health. Previous research revealed the possible negative effect of air pollutant exposure on oral health. However, the predictive model between air pollutant exposure and the prevalence of periodontitis has not been observed yet. Therefore, this study aims to propose a predictive model for the number of patients with periodontitis exposed to PM and atmospheric factors in South Korea using deep learning. Methods: This study is a retrospective cohort study utilizing secondary data from the Korean Statistical Information Service and the Health Insurance Review and Assessment database for air pollution and the number of patients with periodontitis, respectively. Data from 2015 to 2022 were collected and consolidated every month, organized by region. Following data matching and management, the deep neural networks (DNN) model was applied, and the mean absolute percentage error (MAPE) value was calculated to ensure the accuracy of the model. Results: As we evaluated the DNN model with MAPE, the multivariate model of air pollution including exposure to PM_2.5, PM₁₀, and other atmospheric factors predict approximately 85% of the number of patients with periodontitis. The MAPE value ranged from 12.85 to 17.10 (mean±standard deviation=14.12±1.30), indicating a commendable level of accuracy. Conclusion: In this study, the predictive model for the number of patients with periodontitis is developed based on air pollution, including exposure to PM_2.5, PM₁₀, and other atmospheric factors. Additionally, various relevant factors are incorporated into the developed predictive model to elucidate specific causal relationships. It is anticipated that future research will lead to the development of a more accurate model for predicting the number of patients with periodontitis.