• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.6
• /
• pp.237-244
• /
• 2022

In this paper, we present VAE-based recommendation using online behavior log and purchase history to overcome data sparsity and cold start. To generate a variable for customers' purchase history, embedding and dimensionality reduction are applied to the customers' purchase history. Also, Variational Autoencoders are applied to online behavior and purchase history. A total number of 12 variables are used, and nDCG is chosen for performance evaluation. Our experimental results showed that the proposed VAE-based recommendation outperforms SVD-based recommendation. Also, the generated purchase history variable improves the recommendation performance.

• Journal of Broadcast Engineering
• /
• v.7 no.2
• /
• pp.88-95
• /
• 2002

According to an evolution of digital broadcasting, it is possible that terrestrial and satellite broadcasting media provide multi-channel services. CATV and satellite media have been also extended to hundreds of channels. As the result of channel expanding, viewers came to select lots of channels. But it is difficult that they select the favorite channel among hundreds of channels. In this paper, we propose an efficient automatic method to recommend channels and programs on a viewer's preference in a multi-channel broadcasting receiver like a Set ToP Box(STB). The proposed algorithm selects channels based on the following method. It makes and saves user history data by using MPEG-7 MDS based on the program information a viewer had watched. It recommends programs similar to a viewer's preference based on user history data. It selects the channel in the recommended genre based on the viewer's channel preference. The experimental result shows that the proposed scheme is efficient to select the user preference channel.

• Journal of the Korean housing association
• /
• v.6 no.2
• /
• pp.213-223
• /
• 1995

The purpose of this study is to investigate the housing behavior through the code's housing history. The data were collected by open descriptive answering of 28 students in Y.University. The major findings are as follows: The characteristics of code's housing history were changed by stages of family life. The greastes changes are in the stage of elementary school(3rd stage). Before the 3rd stage, living in out of town, detached house, rental house is to superior than town(Dae Gu). Apt.. own house. And housing adaptation is to superior than housing adjustment. After 3rd stage go by contraies.

• 시스템엔지니어링워크숍
• /
• s.1
• /
• pp.60-64
• /
• 2003

In this study, we have analyzed the cost of korea high-speed railway system. The predicted cost in planning phase and adjustment data to 5th year are collected. Then, predicted cost is compared with adjustment in year/item/system base. We make a project history table for criteria to review project history and research & development activity. We have developed CBS(cost breakdown structure) and allocated adjustment data to them. It is shown that cost prediction related to research & development activity in planning phase is relatively correct.

• The Korean Journal of Applied Statistics
• /
• v.25 no.4
• /
• pp.543-552
• /
• 2012

Parrondo paradox is a counter-intuitive phenomenon where two losing games can be combined to win or two winning games can be combined to lose. When we trade stocks with a history-dependent Parrondo game rule (where we buy and sell stocks based on recent investment outcomes) we found Parrondo paradox in stock trading. Using stock data of the KRX from 2008 to 2010, we analyzed the Parrondo paradoxical cases in the Korean stock market.

• Proceedings of the KSR Conference
• /
• 2002.10a
• /
• pp.376-381
• /
• 2002

In this study, we have analyzed the cost of korea high-speed rail project. The predicted cost in planning phase and adjustment data to 5th year are collected. Then, predicted cost is compared with adjustment in year/item/system base. We make a project history table for criteria to review project history and research ＆ development activity. We have developed CBS(cost breakdown structure) and allocated adjustment data to them. It is shown that cost prediction related to research St development activity in planning phase is relatively correct.

• International Journal of Air-Conditioning and Refrigeration
• /
• v.11 no.1
• /
• pp.32-39
• /
• 2003

Though conventional calorimetry methods such as differential scanning calorimetry and differential thermal analysis are used generally in measuring heat of fusion, T-history method has advantages of a simple experimental apparatus and no requirements of sampling process, which is particularly useful for measuring thermophyical properties of in-homogeneous phase change materials in sealed tubes. However, the degree of supercooling used in selecting a range of latent heat release and neglecting sensible heat during the phase change process can cause significant errors in determining the heat of fusion. In the present study, it was shown that a 40% discrepancy exists between the original T-history and the present methods when analyzing the same experimental data. As a result, a reasonable modification to the original T-history method is proposed.

• Journal of Korea Multimedia Society
• /
• v.14 no.7
• /
• pp.933-939
• /
• 2011

Recent heavy rains have caused natural disasters such as flooding and landslides nationwide. Because of flooding occurrence in most of the roads, traffic congestion and isolation caused many loss especially at rush hour. Constant monitoring and analysis of past disaster history data are needed to prevent disasters on areas prone to floods and disaster risk areas. If we managed to obtain traffic volume, speed, phase around intersection using disaster history data when disasters occurred, we can analyse traffic congestion, change of disaster scale and rainfall. In this study, We select a target district to develop by using a route from Dae-nam intersection in Busan Namgu Daeyoeon-dong, over Gwangan large bridge up until Haeundae Olympic intersection, We developed a system which searches disaster history information, traffic volume using disaster history data based on user selection of the road.

• KSCI Review
• /
• v.14 no.2
• /
• pp.245-254
• /
• 2006

Web Forensics algorithm used to an extraction of technical Web Forensics data to be adopted to proof data regarding a crime cyber a computer at data of a great number of log History is an essential element. Propose Web Forensics algorithm, and design at these papers, and try to implement in a Web server system of an actual company. And make the Web dispatch Loging system configuration experiment that applied integrity regarding Web log History information or authentication regarding an information source. Design Web Forensics algorithm and the Flow which used for Web log History analyses at server of e-mail, webmail, HTTP (Web BBS, Blog etc.), FTP, Telnet and messengers (MSN, NateOn, Yahoo, DaumTouch. BuddyBuddy, MsLee, AOL, SoftMe) of a company, and implement through coding. Therefore have a purpose of these paper to will contribute in scientific technical development regarding a crime cyber a computer through Web Forensics.

• Journal of Biosystems Engineering
• /
• v.42 no.4
• /
• pp.251-257
• /
• 2017

Purpose: A chisel mounted on working implement, such as agricultural machinery used in irregular farming conditions, is subjected to highly variable fatigue loading during work. To ensure the safety of the chisel on a working implement for the duration of its service life, fatigue testing must be performed with the proper fatigue test load conditions. In this study, working loads for a chisel were developed by reconstructing loads from strain gage data collected during field tests and used to conduct fatigue tests on the chisel component. Methods: FE analysis with nCode software was utilized to select the proper quantity and locations of strain gages for load measurements. A fatigue test was performed to experimentally verify the fatigue strength of the chisel and to evaluate the validity of the load history developed with the load reconstruction technique. Results: A strain history for the chisel was obtained from data collected during field tests. The data was filtered for the 14-16 km/h speed range, connected, and merged. The chisel load history was developed using the load reconstruction technique. The resulting load history was expressed as a load spectrum using the rain-flow counting method. Conclusions: A fatigue test was conducted on a chisel under a constant load condition with an equivalent load amplitude and number of cycles, as calculated by Miner's Rule for linear damage accumulation. During the fatigue test, there were no cracks at any position. It is concluded that the fatigue test method proposed in this study can be utilized successfully as a durability evaluation method for the chisel.