• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.185-207
• /
• 2022

Recently, with the development of deep learning technology, researches to apply a deep learning algorithm to analyze unstructured data such as text and images are being actively conducted. Text classification has been studied for a long time in academia and industry, and various attempts are being performed to utilize data characteristics to improve classification performance. In particular, a hierarchical relationship of labels has been utilized for hierarchical classification. However, the top-down approach mainly used for hierarchical classification has a limitation that misclassification at a higher level blocks the opportunity for correct classification at a lower level. Therefore, in this study, we propose a methodology for classifying hierarchical data using the autoencoder-based deeply supervised network that high-level classification does not block the low-level classification while considering the hierarchical relationship of labels. The proposed methodology adds a main classifier that predicts a low-level label to the autoencoder's latent variable and an auxiliary classifier that predicts a high-level label to the hidden layer of the autoencoder. As a result of experiments on 22,512 academic papers to evaluate the performance of the proposed methodology, it was confirmed that the proposed model showed superior classification accuracy and F1-score compared to the traditional supervised autoencoder and DNN model.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.81-103
• /
• 2022

It recognizes the importance of mental health and well-being worldwide and consistently records public happiness figures through the World Happiness Report. COVID-19, which occurred in China in 2019, has changed people's daily lives a lot. The accumulation of stress caused by the prolonged epidemic is affecting people's happiness. The present research has revealed negative mental health effects such as "depression" and "anxiety" after the pandemic. In this regard, it was revealed that the happiness index was also lowered numerically. It is insufficient to analyze specific issues about changes in the issue of happiness felt by the public in Korean society after the epidemic. Therefore, this study aims to identify changes in the happiness issue of Koreans after COVID-19 and find ways to improve happiness. Data were collected from various aspects by searching 32 sub keywords based on ERG theory by dividing the period before and after COVID-19. The results of topic modeling before and after COVID-19 were classified into seven areas of happiness index 2.0 published by the National Assembly Future Research Institute and compared and analyzed. Based on the results of comparing the results of the before and after topic from the perspective of each area, a plan to improve happiness was presented. The academic implications of this paper are that the research on psychological changes caused by COVID-19 was expanded by mining the opinions of the actual public on 'happiness'. In addition, it has practical implications in that it specifically presented measures to promote happiness by utilizing the area of objective happiness indicators based on the existing research on ways to reduce happiness promotion unhappiness.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.265-288
• /
• 2023

Process mining has generally analyzed the default logs of Information Systems such as SAP ERP, but as the use of automation software called RPA expands, the logs by RPA bots can be utilized. In this study, the actual status of RPA automation in the field was identified by applying RPA bots to the work of three domestic manufacturing companies (cosmetic field) and analyzing them after leaving logs. Using Uipath and Python, we implemented RPA bots and wrote logs. We used Disco, a software dedicated to process mining to analyze the bot logs. As a result of log analysis in two aspects of bot utilization and performance through process mining, improvement requirements were found. In particular, we found that there was a point of improvement in all cases in that the utilization of the bot and errors or exceptions were found in many cases of process. Our approach is very scientific and empirical in that it analyzes the automation status and performance of bots using data rather than existing qualitative methods such as surveys or interviews. Furthermore, our study will be a meaningful basic step for bot behavior optimization, and can be seen as the foundation for ultimately performing process management.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.397-412
• /
• 2023

In order for social robots to take root in human life, it is important to consider the technical implementation of social robots and human psychology toward social robots. This study aimed to derive potential social robot clusters based on the emotions consumers feel about social robot appearance design, and to identify and compare important design characteristics and emotional differences of each cluster. In our study, we established a social robot emotion framework to measure and evaluate the emotions consumers feel about social robots, and evaluated the emotions of social robot designs based on the semantic differential method, an kansei engineering approach. We classified 30 social robots into 4 clusters by conducting a multidimensional scaling method and K-means cluster analysis based on the emotion evaluation results, confirmed the characteristics of design elements for each cluster, and conducted a comparative analysis on consumer emotions. We proposed a strategic direction for successful social robot design and development from a human-centered perspective based on the design characteristics and emotional differences derived for each cluster.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.19-36
• /
• 2023

As advances in information and communication technology have made it easier for anyone to produce and disseminate information, a new problem has emerged: fake news, which is false information intentionally shared to mislead people. Initially spread mainly through text, fake news has gradually evolved and is now distributed in multimedia formats. Since its founding in 2005, YouTube has become the world's leading video platform and is used by most people worldwide. However, it has also become a primary source of fake news, causing social problems. Various researchers have been working on detecting fake news on YouTube. There are content-based and background information-based approaches to fake news detection. Still, content-based approaches are dominant when looking at conventional fake news research and YouTube fake news detection research. This study proposes a fake news detection method based on background information rather than content-based fake news detection. In detail, we suggest detecting fake news by utilizing related video information from YouTube. Specifically, the method detects fake news through CNN, a deep learning network, from the vectorized information obtained from related videos and the original video using Doc2vec, an embedding technique. The empirical analysis shows that the proposed method has better prediction performance than the existing content-based approach to detecting fake news on YouTube. The proposed method in this study contributes to making our society safer and more reliable by preventing the spread of fake news on YouTube, which is highly contagious.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.211-227
• /
• 2023

Personal service robots, a type of social robot that has emerged with the aging population and technological advancements, are undergoing a transformation centered around technologies that can extend independent living for older adults in their homes. For older adults to accept and use social robot innovations in their daily lives on a long-term basis, it is crucial to have a deeper understanding of user perspectives, contexts, and emotions. This research aims to comprehensively understand older adults by utilizing a mixed-method approach that integrates quantitative and qualitative data. Specifically, we employ the Van Kaam phenomenological methodology to group conversations into nine categories based on emotional cues and conversation participants as key variables, using voice conversation records between older adults and social robots. We then personalize the conversations based on frequency and weight, allowing for user segmentation. Additionally, we conduct profiling analysis using demographic data and health indicators obtained from pre-survey questionnaires. Furthermore, based on the analysis of conversations, we perform K-means cluster analysis to classify older adults into three groups and examine their respective characteristics. The proposed model in this study is expected to contribute to the growth of businesses related to understanding users and deriving insights by providing a methodology for segmenting older adult s, which is essential for the future provision of social robots with caregiving functions in everyday life.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.267-286
• /
• 2023

The conversational agents such as AI speakers utilize voice conversation for human-computer interaction. Voice recognition errors often occur in conversational situations. Recognition errors in user utterance records can be categorized into two types. The first type is misrecognition errors, where the agent fails to recognize the user's speech entirely. The second type is misinterpretation errors, where the user's speech is recognized and services are provided, but the interpretation differs from the user's intention. Among these, misinterpretation errors require separate error detection as they are recorded as successful service interactions. In this study, various text separation methods were applied to detect misinterpretation. For each of these text separation methods, the similarity of consecutive speech pairs using word embedding and document embedding techniques, which convert words and documents into vectors. This approach goes beyond simple word-based similarity calculation to explore a new method for detecting misinterpretation errors. The research method involved utilizing real user utterance records to train and develop a detection model by applying patterns of misinterpretation error causes. The results revealed that the most significant analysis result was obtained through initial consonant extraction for detecting misinterpretation errors caused by the use of unregistered neologisms. Through comparison with other separation methods, different error types could be observed. This study has two main implications. First, for misinterpretation errors that are difficult to detect due to lack of recognition, the study proposed diverse text separation methods and found a novel method that improved performance remarkably. Second, if this is applied to conversational agents or voice recognition services requiring neologism detection, patterns of errors occurring from the voice recognition stage can be specified. The study proposed and verified that even if not categorized as errors, services can be provided according to user-desired results.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.2
• /
• pp.57-65
• /
• 2019

This study is a scientific approach to psychological factors such as emotional stability among various effects of forest resources. In order to carry out this study, the experiment was conducted on the subjects by setting the forest healing space as various spaces. The subjects who participated in this experiment were the students in their twenties and the average age was 22±1.25 years. The subjects were assessed for emotional words through subjective sequence evaluation in different designated forest healing spot. In addition, the emotional states that they actually perceived were measured by measuring the bio-signals to their perceived emotions. BMP, SDNN, VLF, LF, HF, Amplitude, and PPI were used for the bio-signal reaction experiment applied to this study. The results of this experiment were measured by Friedman test and Wilcoxon test for statistical analysis. n this study, 'good', 'clear', and 'uncomfortable' words were found statistically significant at the spot of forest healing space for subjective emotional vocabulary. In addition, SDNN, HF and Amplitude were statistically significant in the results of quantitative bio-signal measurement at each spot in the forest healing space. Based on the results of this study, we can suggest the application direction and strategic utilization plan of forest healing spot and forest resource utilization field. This is not only a guide for the users who use the facility through the spatial facilities and physical requirements for the emotion based forest-healing, but also can be used as a personalized emotional space design aspect.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.241-265
• /
• 2023

A corporate insolvency prediction model serves as a vital tool for objectively monitoring the financial condition of companies. It enables timely warnings, facilitates responsive actions, and supports the formulation of effective management strategies to mitigate bankruptcy risks and enhance performance. Investors and financial institutions utilize default prediction models to minimize financial losses. As the interest in utilizing artificial intelligence (AI) technology for corporate insolvency prediction grows, extensive research has been conducted in this domain. However, there is an increasing demand for explainable AI models in corporate insolvency prediction, emphasizing interpretability and reliability. The SHAP (SHapley Additive exPlanations) technique has gained significant popularity and has demonstrated strong performance in various applications. Nonetheless, it has limitations such as computational cost, processing time, and scalability concerns based on the number of variables. This study introduces a novel approach to variable selection that reduces the number of variables by averaging SHAP values from bootstrapped data subsets instead of using the entire dataset. This technique aims to improve computational efficiency while maintaining excellent predictive performance. To obtain classification results, we aim to train random forest, XGBoost, and C5.0 models using carefully selected variables with high interpretability. The classification accuracy of the ensemble model, generated through soft voting as the goal of high-performance model design, is compared with the individual models. The study leverages data from 1,698 Korean light industrial companies and employs bootstrapping to create distinct data groups. Logistic Regression is employed to calculate SHAP values for each data group, and their averages are computed to derive the final SHAP values. The proposed model enhances interpretability and aims to achieve superior predictive performance.

• Journal of Intelligence and Information Systems
• /
• v.29 no.4
• /
• pp.185-205
• /
• 2023

As the tourism industry recovers post the COVID-19 pandemic, an increasing number of tourists are utilizing various platforms to leave reviews. However, amidst the vast amount of data, finding useful information remains challenging, often leading to time and cost inefficiencies in selecting travel destinations. Despite ongoing research, there are limitations due to the absence of ratings or the presence of different rating formats across platforms. Moreover, inconsistencies between ratings and the content of reviews pose challenges in developing recommendation models. To address these issues, this study utilized 7,104 reviews of tourist spots in Jeju Island to develop a specialized satisfaction index for Jeju tourist attractions and employed this index to construct a 'Rating Prediction Model.' To validate the model's performance, we predicted the ratings of 700 experimental data points using both the developed model and an LSTM approach. The proposed model demonstrated superior performance with a weighted accuracy of 73.87%, which is approximately 4.67% higher than that of the LSTM. The results of this study are expected to resolve the discrepancies between ratings and review contents, standardize ratings in reviews without ratings or in various formats, and provide reliable rating indicators applicable across all areas of travel in different domains.