• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.402-405
• /
• 2023

본 연구의 목적은 발생한 거래에 대해 적합한 회계 계정을 선택하는 예측 모델을 생성하는 것이다. 이를 통해 단기적으로 보조 수단으로 활용하여 회계 전표 승인에 대한 소요 시간을 단축하거나, 장기적으로 손익계산서가 일정 오차율 이내에서 자동으로 빠르게 작성됨으로써 재무 정보의 적시성을 올려주고, 기업의 실적을 나타냄에 있어 개별 담당자의 주관이 개입될 여지가 줄어든다는 면에서 재무 정보의 객관성을 올려줄 것으로 기대한다. 제안하는 모델은 비용 전표 이력의 적요를 자연어 처리하고 학습한 모델을 통해 1 차(공시용) 비용 계정을 분류한다. 분류 결과를 범위로 활용하는 기계 학습을 활용하여 좀 더 세밀한 범위의 2 차(관리용) 비용 계정을 분류하였다.

• The Journal of Society for e-Business Studies
• /
• v.25 no.3
• /
• pp.131-145
• /
• 2020

The numbers of SNS (Social Network Service) users and usage amounts are increasing every year. The influence of SNS is increasing also. SNS has a wide range of influences from daily decision-making to corporate management activities. Therefore, proper analysis of SNS can be a very meaningful work, and many studies are making a lot of effort to look into various activities and relationships in SNS. In this study, we analyze the SNS following relationships using Twitter, one of the representative SNS services. In other words, unlike the existing SNS analysis, our intention is to analyze the interests of the accounts by extracting and visualizing the accounts that two accounts follow in common. For this, a common following account was extracted using Microsoft Excel macros, and the relationship between the extracted accounts was defined using an adjacency matrix. In addition, to facilitate the analysis of the following relationships, a direction graph was used for visualization, and R programming was used for such visualization.

• Journal of Korea Game Society
• /
• v.11 no.6
• /
• pp.81-93
• /
• 2011

Identity theft happens frequently in popular MMORPG(Massively Multi-player Online Role Playing Games) where profits can be gained easily. In spite of the importance of security about identity theft in MMORPG, few methods to prevent and detect identity theft in online games have been proposed. In this study, we investigate real identity theft cases of an online game and define the representative patterns of identity theft as the speedy type, cautious type, and bold type. We then propose the automatic identity theft detection model based on the multi-class classification. We verify the system with one of the leading online games in Korea. The multi-class detection model outperforms the existing binary-class one(hacked or not).

• Journal of Internet of Things and Convergence
• /
• v.7 no.1
• /
• pp.71-77
• /
• 2021

As the IOT technology has developed, it becomes more big issues about IOT device security. An account management is a nerve-in-the-box routine job for the system administrator and users who manage the several hundreds IOT devices(IP camera, Home IOT, the various measuring equipment). This study is to propose the account management system by the hierarchical design and it is efficient for the user to manipulate the account management. The designed system supports 6 functions for the account management and apply the advanced technologies for the existed system. After the performance of the designed system is validated successfully by the authoritative test, the designed system is applied for the relative fields. And it is on going for the development of the smart account management system by applying the AI technique.

• Journal of the Korean Society for information Management
• /
• v.39 no.3
• /
• pp.133-163
• /
• 2022

This study aims to analyze the effect of marketing using Chinese public library WeChat official accounts on the perception of the library and loyalty. For this purpose, research possibilities were explored, and a questionnaire was developed through in-depth interviews with users who subscribed to the Chinese public library WeChat official accounts and those who did not. A survey was conducted with the questionnaire for public library users in six regions of China. As a result, it was confirmed that the characteristics of the official account operation had a positive effect on the perception of the libraries and users' loyalty. This study is meaningful in that it verified how Chinese public library marketing using SNS affected users' purpose on the use of libraries, perception of libraries, and users' loyalty, and suggested how to deal with WeChat official accounts among SNS types.

• Smart Media Journal
• /
• v.11 no.1
• /
• pp.58-66
• /
• 2022

As Internet technology develops, SNS users are increasing. As SNS becomes popular, SNS-type crimes using the influence and anonymity of social networks are increasing day by day. In this paper, we propose a fake account classification method that applies machine learning and deep learning to statistical and image data for fake accounts classification. SNS account data used for training was collected by itself, and the collected data is based on statistical data and image data. In the case of statistical data, machine learning and multi-layer perceptron were employed to train. Furthermore in the case of image data, a convolutional neural network (CNN) was utilized. Accordingly, it was confirmed that the overall performance of account classification was significantly meaningful.

• Management & Information Systems Review
• /
• v.35 no.2
• /
• pp.83-97
• /
• 2016

The purpose of this study is to evaluate our system and consider how the tax rules on corporate reorganization and asset adjustment account can be improved. The scope of this study includes the Korean tax rules on corporate reorganizations, as well as key tax benefits provided by the Special Tax Treatment Control Act. In case of Korea, The relevant regulations and system of taxation respecting the merger has been made the rapid progress in several respects in this situation, there are capable of improvement. This paper suggests improve some problems on tax avoidance abusing incomplete tax law. First, the asset adjustment account is the difficulty that it needs to follow-up during this period of time, as well as the complexity of the accounting and tax adjustments on the practice. If it is permitted to succeed asset-liability as market value, the complexity of asset adjustment account in corporate tax accounting also disappear. Second, in case that controlling shareholders possess more than 20% of merged entity, they could not get the tax deferral until after the time of two years has elapsed. It needs to further alleviate the merged entity ratio than the present level. Finally, after the merger it will be to strengthen the surveillance provisions of five years from the current two years. In addition, continuity of shareholder's requirements and business requirements, it is also necessary to strengthen the requirements of the follow-up provided by a separate regulation.

• Journal of Environmental Impact Assessment
• /
• v.32 no.6
• /
• pp.488-502
• /
• 2023

Ecosystem service accounting must measure ecosystem supply functions, demand, and the actual service flows that occur between them. In order to measure flows, supply and demand relationships must be defined, and a methodology that can objectify complex connections is needed. Although various studies on ecosystem services have been conducted in Korea, but researches on accounting for ecosystem services are not enough. The purpose of this study is to evaluate flood control ecosystem services by applying the EU methodology studied in the Experimental Ecosystem Account (EEA) of System of Environmental Economy Account (SEEA) and explore ways to introduce ecosystem account. To conduct the study, the ecosystem's runoff retention potential, social and economic demand for flood control, and actual service benefit flows formed from the relationships between them were modeled and quantified on a spatial basis. As a result of calculating the actual flow of flood control ecosystem services, the total domestic service amount was calculated to be 165,595 (ha), and it was confirmed that much of it was concentrated in agricultural land. In order to account for domestic flood control services in the future, key spatial data such as land cover maps must be continuously established and managed, and researches on input data and methodologies applicable to various spatial scopes such as national, regional, and unit watersheds are expected to be necessary.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.17 no.6
• /
• pp.595-603
• /
• 1992

It has been establisued that the reliability of communication networks is NP hard problem owing to computationally and complexity as the number of componeuts is Increased in large networks. This paper proposed an algorithm for determining upper and lower bounds In the reliability of source-to-terminal in communication networks to solve this problem. The evaluation method follows the next procedures. First, minimal pathset and minimal cut set are serched. Second, it is sorted that the number of components is the same events and the reliability bounds Is evaluated by the section function to extract common variable. The performance of proposed algorithm is also estimate(1 as compared to the reliability of Esary-Proschan, Shogan and Copal.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.14 no.6
• /
• pp.7-12
• /
• 2014

It has good features for social networking with friends SNS is maintained. However, various enterprises, individuals invading the inconvenience spammers have exposure to a number of users to tweet spam. The study was conducted in the existing research on these spam tweets. However, the results showed a more accurate classification and detection is difficult because of the lack of precision and different causes. In this paper, we describe how to classify the characteristics of spammers, classification criteria. Also has a link rate and difference between followers and following, these features were present classification criteria for spammers account. This experiment was performed according to the criteria. Randomized trial of spam and non-spam accounts were selected and account type was conducted according to the criteria 68% of the link ratio of spam accounts. Followers / Following ratio was 27581.5. Non-spam accounts was 6.12%. Followers / Following ratio was 1.26.