• Journal of Intelligence and Information Systems
• /
• 제29권3호
• /
• pp.125-144
• /
• 2023

Recently, the use of Large Language Models (LLMs) such as ChatGPT has been increasing in various fields such as interactive commerce and mobile financial services. However, LMMs, which are mainly created by learning existing documents, can also learn various human biases inherent in documents. Nevertheless, there have been few comparative studies on the aspects of bias and discrimination in LLMs. The purpose of this study is to examine the existence and extent of nine types of discrimination (Age, Disability status, Gender identity, Nationality, Physical appearance, Race ethnicity, Religion, Socio-economic status, Sexual orientation) in LLMs and suggest ways to improve them. For this purpose, we utilized BBQ (Bias Benchmark for QA), a tool for identifying discrimination, to compare three large-scale language models including ChatGPT, GPT-3, and Bing Chat. As a result of the evaluation, a large number of discriminatory responses were observed in the mega-language models, and the patterns differed depending on the mega-language model. In particular, problems were exposed in elder discrimination and disability discrimination, which are not traditional AI ethics issues such as sexism, racism, and economic inequality, and a new perspective on AI ethics was found. Based on the results of the comparison, this paper describes how to improve and develop large-scale language models in the future.

• Journal of Intelligence and Information Systems
• /
• 제29권3호
• /
• 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
• /
• 제29권3호
• /
• 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 Wetlands Research
• /
• 제25권3호
• /
• pp.167-175
• /
• 2023

Precipitation data is one of the essential input datasets used in various fields such as wetland management, hydrological simulation, and water resource management. In order to efficiently manage water resources using precipitation data, it is essential to secure as much data as possible by minimizing the missing rate of data. In addition, more efficient hydrological simulation is possible if precipitation data for ungauged areas are secured. However, missing precipitation data have been estimated mainly by statistical equations. The purpose of this study is to propose a new method to restore missing precipitation data using machine learning algorithms that can predict new data based on correlations between data. Moreover, compared to existing statistical methods, the applicability of machine learning techniques for restoring missing precipitation data is evaluated. Representative machine learning algorithms, Artificial Neural Network (ANN) and Random Forest (RF), were applied. For the performance of classifying the occurrence of precipitation, the RF algorithm has higher accuracy in classifying the occurrence of precipitation than the ANN algorithm. The F1-score and Accuracy values, which are evaluation indicators of the classification model, were calculated as 0.80 and 0.77, while the ANN was calculated as 0.76 and 0.71. In addition, the performance of estimating precipitation also showed higher accuracy in RF than in ANN algorithm. The RMSE of the RF and ANN algorithms was 2.8 mm/day and 2.9 mm/day, and the values were calculated as 0.68 and 0.73.

• Economic and Environmental Geology
• /
• 제56권4호
• /
• pp.385-396
• /
• 2023

The Mesozoic Bansong Group, distributed along the NE-SW thrust fault zone of the Okcheon Fold Belt in the Danyang-Yeongwol-Jeongseon areas, contains important information on the two Mosozoic orogenic cycles in the Koran Peninsula, the Permian-Triassic Songrim Orogeny and the Jurassic Daebo Orogeny. This study aims to review previous studies on the stratigraphy, depositional period, and basin evolution of the Bansong Group and to suggest future research directions. The perspective on the implication of the Bansong Group in the context of the tectonic evolution of the Korean Peninsula is largely divided into two points of view. The traditional view assumes that it was deposited as a product of the post-collisional Songrim Orogeny and then subsequently deformed by the Daebo Orogeny. This interpretation is based on the stratigraphic, paleontologic, and structural geologic research carried out in the Danyang Coalfield area. On the other hand, recent research regards the Bansong Group as a product of syn-orogenic sedimentation during the Daebo Orogeny. This alternative view is based on the zircon U-Pb ages of pyroclastic rocks distributed in the Yeongwol area and their structural position. However, both models cannot comprehensively explain the paleontological and geochronological data derived from Bansong Group sediments. This suggests the need for a new basin evolution model integrated from multidisciplinary data obtained through sedimentology, structural geology, geochronology, petrology, and geochemistry studies.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• 제43권5호
• /
• pp.587-599
• /
• 2023

As climate change increasesthe frequency and risk of flooding in major cities around theworld, the importance ofsimulation technology that can quickly and accurately analyze high-resolution 2D flooding information in large-scale areasis emerging. The physically-based approaches based on the Shallow Water Equations (SWE) often requires huge computer resources hindering high-resolution flood prediction. This study investigated the theoretical background of Weighted Cellular Automata 2D (WCA2D), which simulates spatio-temporal changes offlooding using transition rules and weight-based system, and assessed feasibility to simulate pluvial flooding in the urbancatchment, theOncheon-cheon catchmentinBusan, SouthKorea.Inaddition,the computation performancewas compared by applying versions using OpenComputing Language (OpenCL) andOpenMulti-Processing (OpenMP) parallel computing techniques. Simulationresultsshowed that the maximuminundation depthmap by theWCA2Dmodel cansimilarly reproduce historical inundation maps. Also, it can precisely simulate spatio-temporal changes of flooding extent in the urban catchment with complex topographic characteristics. For computation efficiency, parallel computing schemes, theOpenCLandOpenMP, improved the computation by about 8~14 and 5~6 folds respectively, compared to the sequential computation.

• Journal of Korean Philosophical Society
• /
• 제93호
• /
• pp.231-260
• /
• 2011

Simondon's theory of individuation and methodology of transduction presents a possibility of contemporary natural philosophy and a new perspective about the relation between philosophy and sciences. According to Simondon's anti-substantial viewpoint, being, as a metastable system charged with potential energy, complicates itself with quantum leaps transversing successive equilibriums. Individuation is the becoming of phases of being which transits from preindividual state to individuated states. Physical individuation as a paradigmatic model of individuation in general demonstrates not only insufficiency of form-oriented hylomorphism, but also spontaneous formational capacity of matter and reality of energetic relational operation immanent in matter. Genesis of a individual (structure or form) occurs as a resolution of the disparation between orders of magnitude, that is, the difference of potentials immanent in nature through the internal resonance, communication by information, transductive relation between the opposites. I'm trying to show that Simondon revives 'physis' of ancient natural philosophy by his own transductive applications of contemporary physics' conceptions, and therefore suggest a new non-reductive materialism. Especially Simondon's 'transduction' which is neither induction, deduction, nor dialectic, but an original ontological process and a peculiar method of thinking, I think, is worthy of note in order to construct network of knowledge and inter-relation between various sciences.

• Journal of Korea Entertainment Industry Association
• /
• 제13권1호
• /
• pp.47-57
• /
• 2019

The end of the 20th century, along with the popularity of new media technology and the rise of new media art, dance as a visual art, and body language art, has the features of more and more rich and changeful. In today's Internet booming new media environment, many different fields, such as film and theater, computer technology, digital art, etc.) with its commonness and characteristics of all kinds of interaction between the creation, produced a new interdisciplinary research with theoretical model. When cross-border interactions between various areas become a hot topic at the same time, the traditional form of dance performances are also seeking new breakthrough. Canada's famous social psychologist McLuhan believes that modern is retrieving lost over a long period of time "overall" feel, return to a feeling of equilibrium. The audience how to have the characteristics of focus on details of visual art back to the "overall" feel worthy of study. At the same time, the new media in today's digital dance teaching in colleges and universities dancing education remains to be perfect and popular, if continue to use the precept of the traditional teaching way blindly, so it is difficult to get from the development of the current domestic dance overall demand. In this paper, the main body is divided into two parts, the first chapter is the study of image device dance performance art, the second chapter is the research of digital dance teaching application system, thus further perspective of media technology to explore dance art crossover practice under the new media environment and mode of transmission.

• Economic Analysis
• /
• 제27권3호
• /
• pp.1-32
• /
• 2021

This study empirically analyzes whether bank capital channel and risk-taking channel for monetary policy work for domestic banks in South Korea by analyzing the impact of the expansionary monetary policy on the rate spread between deposit and loan, capital ratio, and loan amount. For the empirical analysis, the Uhlig (2005)'s sign-restricted SVAR(Structural Vector Auto-Regression) model is used. The empirical results are as follows: the bank's interest rate margin increases, the capital ratio improves, risk-weighted asset ratio increases, and the amount of loans increases in response to expansionary monetary shock. This empirical results confirm that bank capital channel and risk-taking channel work in domestic banks, similar to the previous research results. The implications of this study are as follows. Although the expansionary monetary policy has the effect of improving the bank's financial soundness and profitability in the short term as bank capital channel works, it could negatively affect the soundness of banks by encouraging banks to pursue risk in the long run as risk-taking channel works. It is necessary to note that the capital ratio according to the BIS minimum capital requirement of individual banks may cause an illusion in supervising the soundness of the bank. So, the bank's aggressive lending expansion may lead to an inherent weakness in the event of a crisis. Since the financial authority may have an illusion about the bank's financial soundness if the low interest rate persists, the authority needs to be actively interested in stress tests and concentration risk management in the pillar 2 of the BIS capital accord. In addition, since system risk may increase, it is necessary to conduct regular stress tests or preemptive monitoring of assets concentration risk.

• Journal of Broadcast Engineering
• /
• 제28권3호
• /
• pp.285-292
• /
• 2023

Recently, the impact of fine dust on health has become a major topic. Fine dust is dangerous because it can penetrate the body and affect the respiratory system, without being filtered out by the mucous membrane in the nose. Since fine dust is directly related to the industry, it is practically impossible to completely remove it. Therefore, if the concentration of fine dust can be predicted in advance, pre-emptive measures can be taken to minimize its impact on the human body. Fine dust can travel over 600km in a day, so it not only affects neighboring areas, but also distant regions. In this paper, wind direction and speed data and a time series prediction model were used to predict the concentration of fine dust in Seoul, and the correlation between the concentration of fine dust in Seoul and the concentration in each region was confirmed. In addition, predictions were made using the concentration of fine dust in each region and in Seoul. The lowest MAE (mean absolute error) in the prediction results was 12.13, which was about 15.17% better than the MAE of 14.3 presented in previous studies.