• Journal of Internet Computing and Services
• /
• v.24 no.4
• /
• pp.127-135
• /
• 2023

This research paper explores the application of FinBERT, a variational BERT-based model pre-trained on financial domain, for sentiment analysis in the financial domain while focusing on the process of identifying suitable training data and hyperparameters. Our goal is to offer a comprehensive guide on effectively utilizing the FinBERT model for accurate sentiment analysis by employing various datasets and fine-tuning hyperparameters. We outline the architecture and workflow of the proposed approach for fine-tuning the FinBERT model in this study, emphasizing the performance of various datasets and hyperparameters for sentiment analysis tasks. Additionally, we verify the reliability of GPT-3 as a suitable annotator by using it for sentiment labeling tasks. Our results show that the fine-tuned FinBERT model excels across a range of datasets and that the optimal combination is a learning rate of 5e-5 and a batch size of 64, which perform consistently well across all datasets. Furthermore, based on the significant performance improvement of the FinBERT model with our Twitter data in general domain compared to our news data in general domain, we also express uncertainty about the model being further pre-trained only on financial news data. We simplify the complex process of determining the optimal approach to the FinBERT model and provide guidelines for selecting additional training datasets and hyperparameters within the fine-tuning process of financial sentiment analysis models.

• Journal of the Korean Institute of Traditional Landscape Architecture
• /
• v.41 no.2
• /
• pp.26-34
• /
• 2023

This study conducted interviews with former and current managers of Cheongwadae landscape architecture to provide basic information necessary to preserve, manage, and utilize Gyeongbokgung Palace's back garden, and the main conclusions summarized are as follows. First, the topography changed a lot with the construction of the main building and the official residence under President Roh Tae-woo. The water system was connected to Gyeongbokgung Palace in the past, but is now disconnected. Second, in the case of planting, the most important principles were the president's security and protocol, and accordingly, trees were placed or managed. Trees were planted by introducing excellent trees in various regions, and wildflowers and ground cover plants were frequently replaced according to the season. Third, facilities and roads were arranged for the president's protocol, hobbies, and workers' rest, and fire-fighting facilities were installed to prevent disaster in the forest area of Baegaksan Mountain. Fourth, the biggest inflection point of Gyeongbokgung Palace's back garden during Cheongwadae period was the change in topography due to the new construction of the main building and official residence during President Roh Tae-woo, the removal of A and B barbed wire fences that separated space during President Roh Moo-hyun, and the extensive landscaping carried out for the G20 Summit under President Lee Myung-bak. The area of Gyeongbokgung Palace's back garden is expected to face another inflection point due to the opening of Cheongwadae on May 10, 2022, and the work of evaluating the historical, academic, and landscape values of Gyeongbokgung Palace's back garden should be carried out while preserving the status.

• Journal of Korea Entertainment Industry Association
• /
• v.13 no.2
• /
• pp.79-91
• /
• 2019

In the entertainment industry, risk management is crucial for securing competitiveness due to the risk of investment. The competitiveness of contents is reinforced when external factors such as industrial environment and internal factors centering on involving mechanism are simultaneously provided. The involving mechanism is a form of cognitive response behavior of the audience and occurs through signal processing of the brain when watching the image contents. The signal processing of the brain related to the contents watching is mainly performed in the working memory area, and in the case of the captivating movie, the information other than the contents transmitted to the audience is blocked to generate a temporary dissociation state. A dissociation state similar to a symptom such as hypnosis or amnesia occurs when the audience's level of involving is high. On the other hand, contents information in which the audience is concentrating his attention is used intensively for constructing future thinking through an episodic buffer while the inflow of external information is relatively blocked or delayed. The spectator's future thinking configuration takes the form of a hypothesis-forming activity and is based on the predictability of the brain. When these hypothesized behaviors correspond to the problem solving simulation of story and predictability which is an evolutionary function of the brain, the audience' s brain is involved in the contents at a high level. In order for the act to be effective, the factors such as the background of the hypothesis, the subject of the hypothesis, the internal information of the person, the type and position and quantity of the hypothesis information, and the hypothesis relevance and type of information are important. Based on these factors, analysis of the Kim Eun Sook Drama 'Goblin' shows that the above elements are operated in a very organic and meaningful way.

• Journal of Service Research and Studies
• /
• v.12 no.3
• /
• pp.158-175
• /
• 2022

This study discusses mass unemployment and job insecurity due to the 4th industrial revolution and technological progress. In particular, the construction automation service method can contribute to increasing work productivity, preventing on-site safety accidents, and enhancing the competitiveness of the construction industry according to rapid development and convergence between technologies. However, there is great concern that the position of workers will decrease and the income distribution will deteriorate. Therefore, this study is necessary to alleviate the anxiety of the labor market and to find a direction for the government and all walks of life to ponder. To carry out this study, in-depth interviews were conducted with two experts currently engaged in the construction field, and through analysis, we intend to derive meaning and identify current trends, identify necessary improvement measures and institutional areas and suggest research directions. As a result of the analysis, it is possible to suggest a response strategy in a total of three themes: purpose, implication, and strategy. Based on this, there are response strategies in four areas: (1) industrial site response, (2) worker response, (3) education, and (4) training response, and government and corporate response. Through this study, it is necessary to revitalize economic and sociological discussions in the future so that the improvement in productivity and efficiency of society as a whole due to technological innovation of construction automation services does not lead to social problems such as an increase in the unemployment rate and a decrease in jobs in the labor market.

• Journal of Popular Narrative
• /
• v.25 no.4
• /
• pp.111-134
• /
• 2019

Since This study is about game novel. I'm going to discuss the flow and features of game novels. Game novels developed in Korea, the game industry developed and made it possible. Fantasy novels, Muhyeop novels, and the generation of games met, so game novels could develop. By looking at the trends of game novels, I want to lay the foundation for the study of popular literature genre. The reason behind the emergence of game novels is the meeting of the game generation and the novel. The types of game novels can be distinguished according to the 'location of the main characters' and 'the narrative style'. It is not the best, but it is the minimum work to identify the genre. The biggest characteristic of game novels is their connection to games. It also uses the system of the real game. Another characteristic is that a variety of variations are possible with other genres. Game novels were influenced by fantasy novels and Muhyeop novels. Game novels now hold a solid place in the realm of popular literature. But the limits are also clear. It is the repeated use of game novel progression. Here are the limitations that game have to overcome. In addition, active research should be carried out. This study is intended to promote research on these game novel genres.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.28 no.5C
• /
• pp.293-301
• /
• 2008

Soil profiling is one of the most important work among geotehnical engineering practice. Generally, soil profile is estimated from the observation of soil samples during subsurface exploration but such estimation also includes some experiencing aspects such as flushed water from the borehole, slime colour, boring speed and so on. In addition, since the capacity of hydraulic drill rig is significantly increased, thin layers might be easily missed. So, continuous soil profile is almost impossible over all depth to be bored from conventional subsurface exploration. While CPT or CPTu can serve continuous soil profile information over all depth generally in 5cm interval. Many charts or methods for soil profile from CPT result have been proposed during last several decades over the world. However they have not been verified in local ground condition in Korea. In this research, CPT results and soil classification results based on USCS were compiled from 17 sites over the Korea. Soil classification results by using 7 CPT soil classification charts were compared with those of USCS for the compiled database. Most proper CPT soil classification chart for Korean soil characteristics was evaluated and effective parameters for the soil classification from CPT were discussed. Finally interrelationship between CPT soil classification chart and USCS soil classification was evaluated.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.105-112
• /
• 2023

In this paper, we propose a new method for displaying colored defects by measuring the relative density with the wide-area and local densities of X-ray. The relative density of one pixel represents a relative difference from the surrounding pixels, and we also suggest a colorization of X-ray images representing these pixels as normal and defective. The traditional method mainly inspects materials such as plastics and metals, which have large differences in transmittance to the object. Our proposed method can be used to detect defects such as sprouts or holes in images obtained by an inspection machine that detects X-rays. In the experiment, the products that could not be seen with the naked eye were colored with pests or sprouts in a specific color so that they could be used in the agricultural product selection system. Products that are uniformly filled with a single ingredient inside, such as potatoes, carrots, and apples, can be detected effectively. However, it does not work well with bumpy products, such as peppers and paprika. The advantage of this method is that, unlike machine learning, it doesn't require large amounts of data. The proposed method could be applied to a screening system using X-rays and used not only in agricultural product screening systems but also in manufacturing processes such as processed food and parts manufacturing, so that it can be actively used to select defective products.

• Journal of Internet Computing and Services
• /
• v.24 no.6
• /
• pp.107-117
• /
• 2023

It must be possible to assess how combat actions taking place in cyberspace affect the military's major mission systems and weapon systems. In order to analyze the mission impact caused by a cyber attack through cyber M&S, the target mission system and cyber warfare elements must be built as a model and a scenario for simulation must be authored. Many studies related to mission impact analysis due to cyber warfare have been conducted focusing on the United States, and existing studies have authored separate scenarios for physical battlefields and cyber battlefields. It is necessary to build a simulation environment that combines a physical battlefield model and a cyber battlefield model, and be able to integrate and author mission scenarios and cyber attack/defense scenarios. In addition, the physical battlefield and cyber battlefield are different work areas, so authoring two types of scenarios for simulation is very complicated and time-consuming. In this paper, we propose a method of using mission system information to prepare the data needed for scenario authoring in advance and using the pre-worked data to author an integrated scenario. The proposed method is being developed by reflecting it in the design of the scenario authoring tool, and an integrated scenario authoring in the field of counter-fire warfare is being performed to prove the proposed method. In the future, by using a scenario authoring tool that reflects the proposed method, it will be possible to easily author an integrated scenario for mission impact analysis in a short period of time.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.73-83
• /
• 2023

This study is a descriptive survey to develop a gamification e-learning class based on self-determination theory and to check its effectiveness. The data collection period was from March 1 to June 15, 2023, and 59 students at G University in G Metropolitan City were surveyed on class participation, learning immersion, and teaching presence before and after the course. IBM SPSS/Win 26.0 was used to analyze the collected data, and descriptive statistics, analysis of variance (ANOVA), and analysis of covariance (ANCOVA) were conducted. The results showed that the self-determination-based gamification class significantly improved students' class participation, learning engagement, and teaching presence (p<.05). An analysis of covariance (ANCOVA) was conducted to determine whether the general characteristics of the participants affected the results of the post-test, and gender affected the post-test results of learning engagement, with an effect of 7.9%. Based on the results of this study, it can be seen that self-determination-based gamification e-learning class is effective in improving learners' class participation, learning engagement, and teaching presence. As the demand for e-learning in universities is expanding, self-determination-based gamification e-learning classes should be developed in various fields of liberal arts and majors.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.785-796
• /
• 2023

In this study, a methodology for real-time classification and prediction of defects that may appear in PVDF(Polyvinylidene fluoride) sensors, which are widely used for structural integrity monitoring, is proposed. The types of sensor defects appearing according to the sensor attachment environment were classified, and an impact test using an impact hammer was performed to obtain an output signal according to the defect type. In order to cleary identify the difference between the output signal according to the defect types, the time domain statistical features were extracted and a data set was constructed. Among the machine learning based classification algorithms, the learning of the acquired data set and the result were analyzed to select the most suitable algorithm for detecting sensor defect types, and among them, it was confirmed that the highest optimization was performed to show SVM(Support Vector Machine). As a result, sensor defect types were classified with an accuracy of 92.5%, which was up to 13.95% higher than other classification algorithms. It is believed that the sensor defect prediction technique proposed in this study can be used as a base technology to secure the reliability of not only PVDF sensors but also various sensors for real time structural health monitoring.