• The KIPS Transactions:PartD
• /
• v.16D no.4
• /
• pp.507-516
• /
• 2009

In stream data processing system, generally the interval queries are in advance registered in the system. When a data is input to the system continuously, for realtime processing, a query indexing method is used to quickly search queries. Thus, a main memory-based query index with a small storage cost and a fast search time is needed for searching queries. In this paper, we propose a LVC-based(Limited Virtual Construct-based) query index method using a hashing to meet the both needs. In LVC-based query index, we divide the range of a stream into limited virtual construct, or LVC. We map each interval query to its corresponding LVC and the query ID is stored on each LVC. We have compared with the CEI-based query indexing method through the simulation experiment. When the range of values of input stream is broad and there are many short interval queries, the LVC-based indexing method have shown the performance enhancement for the storage cost and search time.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2008.11a
• /
• pp.686-689
• /
• 2008

In the past decade, various data mining techniques have been used in construction engineering as a means to make informed decisions through the aid of useful knowledge discovered from historical data. Researchers in the construction domain are often confronted with a challenge to derive a meaningful conclusion with a limited sample of data. However, when the data size is small, the proposed results are often illogical. Even if the derived results are technically flawless, sometimes it is difficult to reproduce these results by using the same analysis method when a different set of data is used. This paper reviews some problems that stem from limited data size, and discusses several recommendations for dealing with these problems.

• Database Research
• /
• v.34 no.3
• /
• pp.137-147
• /
• 2018

A key to a well-performing human activity recognition (HAR) system through machine learning technique is the availability of a substantial amount of labeled data. Collecting sufficient labeled data is an expensive and time-consuming task. To build a HAR system in a new environment (i.e., the target domain) with very limited labeled data, it is unfavorable to naively exploit the data or trained classifier model from the existing environment (i.e., the source domain) as it is due to the domain difference. While traditional machine learning approaches are unable to address such distribution mismatch, transfer learning approach leverages the utilization of knowledge from existing well-established source domains that help to build an accurate classifier in the target domain. In this work, we propose a transfer learning approach to create an accurate HAR classifier with very limited data through the multitask neural network. The classifier loss function minimization for source and target domain are treated as two different tasks. The knowledge transfer is performed by simultaneously minimizing the loss function of both tasks using a single neural network model. Furthermore, we utilize the unlabeled data in an unsupervised manner to help the model training. The experiment result shows that the proposed work consistently outperforms existing approaches.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.77-80
• /
• 2001

In general, the informations of the inner image that user interested in are limited to a special domain. In this paper, as using Wavelet Transform for dividing image into high frequency and low frequency, We can separate foreground including many data. After calculating object boundary of separated part, We extract special features using Color Coherence Vector. According to results of this experiment, the method of comparing data extracting foreground features is more effective than comparing data extracting features of entire image when we extract the image user interested in.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1295-1297
• /
• 2003

Owing to the higher temporal resolution, meteorological satellite data is widely used to monitor the disasters happened on the earth's surface. However, the precision of identifying disaster information is limited by the poor spatial resolution. As known, GIS technology is good at processing and analyzing the geographic information. The result shows, integrating with GIS technology, the ability of monitoring forest fire using meteorological satellite data has been greatly improved.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.1
• /
• pp.78.2-78.2
• /
• 2014

Many techniques for data analysis also based on gaussian noise assumption which is often valid in various situations. However, the sensitivity of gravitational wave searches are limited by their non-gaussian and non-stationary noise. We introduce various on-going efforts to overcome this limitation in Korean Gravitational Wave Group. First, artificial neural networks are applied to discriminate non-gaussian noise artefacts and gravitational-wave signals using auxiliary channels of a gravitational wave detector. Second, viability of applying Hilbert-Huang transform is investigated to deal with non-stationary data of gravitational wave detectors. We also report progress in acceleration of low-latency gravitational search using GPGPU.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.9
• /
• pp.33-40
• /
• 2017

In this paper, we propose a sink-location management and data-routing scheme to effectively support the mobile sink in solar-powered WSN. Battery-based wireless sensor networks (WSNs) have a limited lifetime due to their limited energy, but solar energy-based WSNs can be supplied with energy periodically and can operate forever. On the other hand, introduction of mobile sink in WSNs can solve some energy unbalance problem between sink-neighboring nodes and outer nodes which is one of the major challenges in WSNs. However, there is a problem that additional energy should be consumed to notify each sensor node of the location of the randomly moving mobile sink. In the proposed scheme, one of the nodes that harvests enough energy in each cluster are selected as the cluster head, and the location information of the mobile sink is shared only among the cluster heads, thereby reducing the location management overhead. In addition, the overhead for setting the routing path can be removed by transferring data in the opposite direction to the path where the sink-position information is transferred among the heads. Lastly, the access node is introduced to transmit data to the sink more reliably when the sink moves frequently.

• Particle and aerosol research
• /
• v.7 no.3
• /
• pp.99-108
• /
• 2011

To establish area specific control strategies for the reduction of the ozone concentration, the Ozone Isopleth Plotting Package for Research(OZIPR) model has been widely used. However, the model results tend to changed by various input parameters such as the background concentration, emission amount of NOx and volatile organic compounds (VOCs), and meteorological condition. Thus, sensitivity analysis should be required to ensure the reliability of the result. The OZIPR modeling results for five local government districts in the Seoul Metropolitan Area (SMA) in June 2000 were used for the sensitivity analysis. The sensitivity analysis result showed that the modeling result of the SMA being VOC-limited region be still valid for a wide range of input parameters' variation. The estimated ozone concentrations were positively related with the initial VOCs concentrations while were negatively related with the initial NOx concentrations. But, the degree of the variations at each local district was different suggesting area specific characteristics being also important. Among the five local governments, Suwon was chosen to identify other variance through the period from April to September in 2000. The monthly modeling results show different ozone values, but still showing the characteristics of VOCs-limited region. Limitations due to not considering long range transport and transfer from neighbor area, limitation of input data, error between observed data and estimated data are all discussed.

• Journal of information and communication convergence engineering
• /
• v.12 no.3
• /
• pp.181-185
• /
• 2014

The lifetime of a solid-state drive (SSD) is limited because of the number of program and erase cycles allowed on its NAND flash blocks. Data cannot be overwritten in an SSD, leading to an out-of-place update every time the data are modified. This result in two copies of the data: the original copy and a modified copy. This phenomenon is known as write amplification and adversely affects the endurance of the memory. In this study, we address the issue of reducing wear leveling through efficient handling of write requests. This results in even wearing of all the blocks, thereby increasing the endurance period. The focus of our work is to logically divert the write requests, which are concentrated to limited blocks, to the less-worn blocks and then measure the maximum number of write requests that the memory can handle. A memory without the proposed algorithm wears out prematurely as compared to that with the algorithm. The main feature of the proposed algorithm is to delay out-of-place updates till the threshold is reached, which results in a low overhead. Further, the algorithm increases endurance by a factor of the threshold level multiplied by the number of blocks in the memory.

• Journal of Environmental Science International
• /
• v.22 no.9
• /
• pp.1171-1180
• /
• 2013

Recent climate change has led to fluctuations in agricultural production, and as a result national food supply has become an important strategic factor in economic policy. As such, in this study, panel data was collected to analyze the effects of seven meteorological elements and using the Lagrange multipliers method, the fixed-effects model for the production of five types of food crop and the seven meteorological elements were analyzed. Results showed that the key factors effecting increases in production of rice grains were average temperature, average relative humidity and average ground surface temperature, while wheat and barley were found to have positive correlations with average temperature and average humidity. The implications of this study are as follow. First, it was confirmed that the meteorological elements have profound effects on the production of food crops. Second, when compared to existing studies, the study was not limited to one food crop but encompassed all five types, and went beyond other studies that were limited to temperature and rainfall to include various meterological elements.