• Proceedings of the Safety Management and Science Conference
• /
• 2009.04a
• /
• pp.371-380
• /
• 2009

본 연구는 국내에 있는 한 금융사에서 사용하고 있는 기존 인사평가시스템을 사례로 하여 시스템 운영자와 사용자(모든 임직원)가 시스템을 효율적으로 운영하고 사용할 수 있도록 인사평가시스템을 재구축하였다. 새로 구축될 인사평가시스템의 효율을 보다 높이기 위해 인사평가 업무와 평가 절차를 분석하고 그러한 평가체계를 기존 legacy 시스템(인사평가시스템)에 어떻게 반영이 되었는지도 분석하였다. 해당 금융사는 기존 평가시스템을 통하여 임원과 직원뿐 아니라, 비정규직원까지 년마다 상반기와 하반기로 나누어 업적과 역량평가를 각각 두번씩 실시하고 있었다. 평가결과는 나중에 보상이나 승진 시에 중요한 고과자료로 활용하고 있었다. 업적과 역량평가 뿐만 아니라, 신업사원을 대상으로 하는 정규직 수습평가와 계약직을 대상으로 한 역량평가 그리고 전직원이 대상인 다면평가도 실시하고 있었다. 그런데 여러 해에 걸쳐 인사평가를 실시하는 과정에서 평가프로세스가 여러 차례 변하였고, 많은 평가대상자를 평가하다 보니 예외적으로 처리해야 하는 경우도 발생하였다. 이런 변화와 예외적인 사항을 기존 Lagacy시스템에 반영하다 보니 평가데이터를 관리하는 테이블이 처음보다는 많이 늘어나게 되어 현재 약 170개에 달하고 있었다. 이렇게 많은 테이블 중에는 당시에는 사용했으나, 현재는 사용하지 않게 된 것도 포함되어 있었다. 프로그램 소스도 마찬가지로 새로운 요구사항과 많은 예외사항 처리로 인해 복잡해지고 프로그램 소스의 수도 늘어나게 되었다. 이로 인해 평가시스템 담당자는 시스템을 관리하기가 복잡해지고, 새로운 변화나 요구사항에 대응하기가 어려운 사항이었다. 그리고 담당자가 시스템 관리에 더 더욱 어려움을 겪게된 것은 시스템 담당자가 여러 번 바뀌는 과정에서 인수인계에 제대로 이루어 지지 않게 되었고 인사평가시스템 관리문서도 시스템 변화에 따른 히스토리 내용이 제대로 관리되지 못하고 있었기 때문이다. 그래서 이번 연구에서는 기존 legacy 시스템의 테이블과 프로그램 소스를 분석하여 방만하게 늘어날 데이터 관리 테이블을 효율적으로 줄이고, 예외적인 사항을 새로운 인사평가시스템에서 포괄적으로 수용할 수 있도록 재설계하였다. 또한 인사평가시스템 운영중에도 사용자(임직원)들로부터 있을 수 있는 예외적 요구사항을 미리 예측하여 운영담당자가 신속하고 정확하게 대응 할 수 있도록 하였다.

• Journal of Broadcast Engineering
• /
• v.16 no.2
• /
• pp.237-246
• /
• 2011

In order to produce the variety of UCC, the recommendation service is required which considers the copyright of UCC producer discriminated from one for UCC consumers and the purpose of its production. The recommender system designed in this thesis enables UCC which is much similar to one UCC producer utilizes to be used with custom-made when recommending and producing based on UCC view history and production list, etc. of its producer. The recommender system is largely divided into filtering based on the preferred tag, UCC filtering used when producing the preferred UCC and creating process of recommended UCC using the Pearson formula. The recommender system in this thesis requires the data which were used when producing UCC. For that, we added the reference factor so that the data of UCC which were utilized when producing UCC into the existing metadata can be recorded. If the recommender system suggested in this thesis is used, the more effective and convenient UCC recommendation services with custom-made for producers can be provided.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.9 no.10
• /
• pp.797-808
• /
• 2019

In a closed workplace, the management of the workplace is important because the environmental data at the workplace has a great influence on the safety of workers. Today's industrial sites are transformed into data-based factories that collect and analyze data through sensors in those sites, requiring a management system to ensure safety. In general, a safety management system stores and manages data on a central server associated with a database. Since such management system introduces high possibility of forgery and loss of data, workers often suspect the reliability of the information on the management system. In this paper, we present a worker safety management system based on the EOS blockchain which is considered as third-generation blockchain technology. The developed system consists of a set of smart contracts on the EOS blockchain and 3 decentralized applications associated with the blockchain. According to the roles of users, the worker and manager applications respectively perform the process of initiating or terminating tasks as blockchain transactions. The entire transaction history is distributed and stored in all nodes participating in the blockchain network, so forgery and loss of data is practically impossible. The system administrator application assigns the account rights of workers and managers appropriate for performing the functions, and specifies the safety standards of IoT data for ensuring workplace safety. The IoT data received from sensor platforms in workplaces and the information on initiation, termination or approval of tasks assigned to workers, are explicitly stored and managed in the EOS smart contracts.

• The Journal of Korean Association of Computer Education
• /
• v.8 no.1
• /
• pp.117-127
• /
• 2005

Numerous internet pop advertisement are being provided according to the rapid development of e-commercial and a rise in users. However, it has not been based on analysis of users' inclination but just one-sided providing. With that reason, many web-site provider want to advertis e more efficient and distinguished Internet-advertisement as analyzing Server's Log accessed. In this thesis, we have studied and tested relatively simply adoption system to provide personalized advertisement service. In order to influence personal disposition to system as the most effective way, it first of all uses History files as source data and after refining it, it can search not only visitors' inclination but also the others' visit-list on the other server. As a result of it, it can make advertisement more reality and activity.

• Journal of Software Engineering Society
• /
• v.28 no.2
• /
• pp.1-6
• /
• 2019

There are various stakeholders in the blockchain ecosystem. Since the emergence of Ethereum, many transactions have been made using smart contracts, and a wider range of stakeholders are participating, including not only developers, but also investors, banks, companies, and general users. However, various stakeholders have a problem in that it is difficult and complicated to check the state of smart contracts. If it becomes difficult to check the state, the reliability of the smart contract will be lowered and the utilization will be lowered. Also, if the state check is difficult and complicated for the developer, it will be difficult to provide high quality due to the difficulty of testing and debugging the smart contract developed by the developer. In this research, we propose a design and implementation method of the Ethereum Smart Contract State Monitoring System that enables various stakeholders and developers to easily and continuously check the state of smart contracts and analyze them using historical data.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.10
• /
• pp.415-420
• /
• 2017

Virtualization technique of OS-level is a new paradigm for deploying applications, and is attracting attention as a technology to replace traditional virtualization technique, VM (Virtual Machine). Especially, docker containers are capable of distributing application images faster and more efficient than before by applying layered image structures and union mount point to existing linux container. These characteristics of containers can only be used in layered file systems that support snapshot functionality, so it is required to select appropriate layered file systems according to the characteristics of the containerized application. We examine the characteristics of representative layered file systems and conduct write performance evaluations of each layered file systems according to the operating principles of the layered file system, Allocate-on-Demand and Copy-up. We also suggest the method of determining a appropriate layered file system principle for unknown containerized application by learning block I/O usage history of each layered file system principles in artificial neural network. Finally we validate effectiveness of artificial neural network created from block I/O history of each layered file system principles.

• The KIPS Transactions:PartC
• /
• v.11C no.5
• /
• pp.659-670
• /
• 2004

Handoff is one of the most important features for the user's mobility in a wireless cellular communication system. It is related to resource reservation at nearby cells. Resource reservation to the new connection point should occur prior to handoff to enable the user to receive the data or services at the new location, at the same level of service as at the previous location. For the efficient resource reservation, mobility prediction has been reported as an effective means to decrease the call dropping probability and to shorten the handoff latency in a wireless cellular environment. A recently proposed algorithm, ZMHB, makes use of the history of the user's positions within the current cell to predict the next cell. But, the prediction of the ZMHB algorithm is found to be 80∼85% accurate for regular and random movements. In this paper, we propose a new improved ZMHB mobility prediction algorithm, which is called Detailed-ZMHB that uses detailed-zone-based tracking of mo-bile users to predict user movements. The effectiveness of the proposed algorithm is then demonstrated through a simulation.

• Journal of Intelligence and Information Systems
• /
• v.8 no.2
• /
• pp.91-103
• /
• 2002

A collaborative filtering which supports personalized services of users has been common use in existing web sites for increasing the satisfaction of users. A collaborative filtering is demanded that items are estimated more than specified number. Besides, it tends to ignore information of other users as recommending them on the basis of information of partial users who have similar inclination. However, there are valuable hidden information into other users' one. In this paper, we use Association Rule, which is common wide use in Data Mining, with collaborative filtering for the purpose of discovering those information. In addition, this paper proved that Association Rule applied to Recommender System has a effects to recommend users by the relation between groups. In other words, Association Rule based on the history of all users is derived from. and the efficiency of Recommender System is improved by using Association Rule with collaborative filtering.

• KNOM Review
• /
• v.23 no.2
• /
• pp.11-17
• /
• 2020

Currently, the web page that provides residual substance information provides information on residual acceptance criteria in food for pesticides and veterinary drugs. Residual substances refer to pesticides or veterinary drugs that are left in agricultural, livestock, or marine products after being used by diluting them thousands of times. However, users are experiencing inconvenience due to the lack of information on pesticides and veterinary drugs, delays in search time, and Web page errors. In addition, the manager has the inconvenience of manually entering information such as residual acceptance criteria and analysis methods. Thus, this paper proposes a system that can efficiently manage and update the history of changes in information, such as residual material standards for pesticides, animal medicines and the characteristics of drugs.

• Journal of the Korea Convergence Society
• /
• v.8 no.8
• /
• pp.255-265
• /
• 2017

The study analyzed previous studies on traditional markets and market modernization projects to analyze research trends and content related to traditional markets. Next, the study selected 4 traditional markets throughout the nation where there was promotion of market modernization projects to conduct field research and interviews about store modules, line of flow, facility configuration, and types of businesses. Empirical comparative analysis was conducted on construction hardware status through measurement and observation and data was collected on business environment and requirement characteristics by business type through interviews with merchant associations of the corresponding markets. Consistent standard was applied as much it was possible to comparatively analyze the 4 market modernization cases and on the unique characteristics of individual markets, the cause was determined in conjunction to the history of the business promotion process and regional characteristics. After the study, basic data to suggest guidelines in store modules by business type in traditional markets could be acquired and the study identified the facilities equipment standards that must be considered in future market modernization projects. Through this it will be possible to derive policy implications to minimize trial and error and guarantee business efficiency in future market modernization projects.