• Korean Journal of Remote Sensing
• /
• v.32 no.4
• /
• pp.365-382
• /
• 2016

The results from the Global Positioning System (GPS) measurements of the Mobile Observation Vehicle (MOVE) on the eastern coast of Korea have been compared with REFerence (REF) values from the fixed GPS sites to assess the performance of Precipitable Water Vapor (PWV) retrievals in a kinematic environment. MOVE-PWV retrievals had comparatively similar trends and fairly good agreements with REF-PWV with a Root-Mean-Square Error (RMSE) of 7.4 mm and $R^2$ of 0.61, indicating statistical significance with a p-value of 0.01. PWV retrievals from the June cases showed better agreement than those of the other month cases, with a mean bias of 2.1 mm and RMSE of 3.8 mm. We further investigated the relationships of the determinant factors of GPS signals with the PWV retrievals for detailed error analysis. As a result, both MultiPath (MP) errors of L1 and L2 pseudo-range had the best indices for the June cases, 0.75-0.99 m. We also found that both Position Dilution Of Precision (PDOP) and Signal to Noise Ratio (SNR) values in the June cases were better than those in other cases. That is, the analytical results of the key factors such as MP errors, PDOP, and SNR that can affect GPS signals should be considered for obtaining more stable performance. The data of MOVE can be used to provide water vapor information with high spatial and temporal resolutions in the case of dramatic changes of severe weather such as those frequently occurring in the Korean Peninsula.

• The KIPS Transactions:PartC
• /
• v.13C no.6 s.109
• /
• pp.709-716
• /
• 2006

According to the increase of various AV media devices and contents in the digital home, the DLNA becomes to play an important role as the interoperability standard between then Since this guideline only focuses on the interoperability among home networked devices, media players, and media contents existing inside of the homenetwork, there is no retrieval and transmission method for sharing multimedia contents located over several homes via Internet. Additionally, this guideline lets device-detection and notification messages to be transmitted using W multicast methods, and current Internet environment cannot guarantee consistent IP multicast services, it has the limitation that it cannot retrieve and control DLNA devices in other digital homes remotely via the Internet. Therefore, in this paper, we propose the IHM(Inter-Home Media) proxy system and its operating mechanism to provide a way of sharing media contents distributed over multiple DLNA-based homes, through analyzing these limitations and building up a sharing method for A/V media contents distributed over the DLNA homes based on the dynamic or private IP networks. Our method removes the limitation on the user locations through sharing distributed media contents, and also makes cost-downs for storing media contents, from the view point of individual residents.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.47 no.1
• /
• pp.80-91
• /
• 2010

As the volume of image data increases dramatically, its good organization of image data is crucial for efficient image retrieval. Clustering is a typical way of organizing image data. However, traditional clustering methods have a difficulty of requiring a user to provide the number of clusters as a parameter before clustering. In this paper, we discuss an approach for clustering image data that does not require the parameter. Basically, the proposed approach is based on Cross-Association that finds a structure or patterns hidden in data using the relationship between individual objects. In order to apply Cross-Association to clustering of image data, we convert the image data into a graph first. Then, we perform Cross-Association on the graph thus obtained and interpret the results in the clustering perspective. We also propose the method of hierarchical clustering and the method of outlier detection based on Cross-Association. By performing a series of experiments, we verify the effectiveness of the proposed approach. Finally, we discuss the finding of a good value of k used in k-nearest neighbor search and also compare the clustering results with symmetric and asymmetric ways used in building a graph.

• 대한공업교육학회지
• /
• v.41 no.2
• /
• pp.1-23
• /
• 2016

This study was conducted to analyze the effect Meister high school students' career maturity with respect to the impact on school maladjustment. Also, this study clarify the relationship. This study purpose is to permanently provide Meister as the basis for the vocational education sector career education needed to faithfully serve as a special purpose high schools. Tools used for the survey is maladaptive measurement tools developed by Leegyumi (2004) and Career maturity measurement tools developed at Korea Research Institute for Vocational Education and Training (2012). Using these tools, a reliability test was conducted. Meister students' career maturity was conducted correlation analysis and multiple regression analysis to analyze the impact of school maladjustment. Independent variables are consisted of career maturity and independence, attitude toward the job, planning, self-understanding, rational decision-making, information retrieval, knowledge of the desired job, career exploration and ready for action. Meister high school student's career maturity according to the students' background variables are little girls was higher than boys, but it was not statistically significant. T-test was conducted to ascertain the career maturity and school maladjustment differences of adaptation groups and maladaptive group in meister school students in background variables. A career maturity and school maladjustment between adaptive and maladaptive population groups showed a statistically significant difference in background variables.

• The KIPS Transactions:PartD
• /
• v.16D no.4
• /
• pp.475-486
• /
• 2009

Data cube plays an important role in multi-dimensional, multi-level data analysis. Meeting on-line analysis requirements of data stream, several cube structures have been proposed for OLAP on data stream, such as stream cube, flowcube, S-cube. Since it is costly to construct data cube and execute ad-hoc OLAP queries, more research works should be done considering efficient data structure, query method and algorithms. Stream cube uses H-cubing to compute selected cuboids and store the computed cells in an H-tree, which form the cuboids along popular-path. However, the H-tree layoutis disorderly and H-cubing method relies too much on popular path.In this paper, first, we propose $H^*$-tree, an improved data structure, which makes the retrieval operation in tree structure more efficient. Second, we propose an improved cubing method, $H^*$-cubing, with respect to computing the cuboids that cannot be retrieved along popular-path when an ad-hoc OLAP query is executed. $H^*$-tree construction and $H^*$-cubing algorithms are given. Performance study turns out that during the construction step, $H^*$-tree outperforms H-tree with a more desirable trade-off between time and memory usage, and $H^*$-cubing is better adapted to ad-hoc OLAP querieswith respect to the factors such as time and memory space.

• Smart Media Journal
• /
• v.11 no.10
• /
• pp.65-75
• /
• 2022

Deep learning is used as a creative tool that could overcome the limitations of existing analysis models and generate various types of results such as text, image, and music. In this paper, we propose a method necessary to preprocess audio data using the Niko's MIDI Pack sound source file as a data set and to generate music using Bi-LSTM. Based on the generated root note, the hidden layers are composed of multi-layers to create a new note suitable for the musical composition, and an attention mechanism is applied to the output gate of the decoder to apply the weight of the factors that affect the data input from the encoder. Setting variables such as loss function and optimization method are applied as parameters for improving the LSTM model. The proposed model is a multi-channel Bi-LSTM with attention that applies notes pitch generated from separating treble clef and bass clef, length of notes, rests, length of rests, and chords to improve the efficiency and prediction of MIDI deep learning process. The results of the learning generate a sound that matches the development of music scale distinct from noise, and we are aiming to contribute to generating a harmonistic stable music.

• Journal of KIISE:Software and Applications
• /
• v.29 no.4
• /
• pp.270-277
• /
• 2002

Text filtering is a task of deciding whether a document has relevance to a specified topic. As Internet and Web becomes wide-spread and the number of documents delivered by e-mail explosively grows the importance of text filtering increases as well. The aim of this paper is to improve the accuracy of text filtering systems by using machine learning techniques. We apply AdaBoost algorithms to the filtering task. An AdaBoost algorithm generates and combines a series of simple hypotheses. Each of the hypotheses decides the relevance of a document to a topic on the basis of whether or not the document includes a certain word. We begin with an existing AdaBoost algorithm which uses weak hypotheses with their output of 1 or -1. Then we extend the algorithm to use weak hypotheses with real-valued outputs which was proposed recently to improve error reduction rates and final filtering performance. Next, we attempt to achieve further improvement in the AdaBoost's performance by first setting weights randomly according to the continuous Poisson distribution, executing AdaBoost, repeating these steps several times, and then combining all the hypotheses learned. This has the effect of mitigating the ovefitting problem which may occur when learning from a small number of data. Experiments have been performed on the real document collections used in TREC-8, a well-established text retrieval contest. This dataset includes Financial Times articles from 1992 to 1994. The experimental results show that AdaBoost with real-valued hypotheses outperforms AdaBoost with binary-valued hypotheses, and that AdaBoost iterated with random weights further improves filtering accuracy. Comparison results of all the participants of the TREC-8 filtering task are also provided.

• Journal of KIISE
• /
• v.44 no.10
• /
• pp.1112-1123
• /
• 2017

As content providers further offload content-centric services to the cloud, data retrieval over the cloud typically results in many redundant items because there is a prevalent near-duplication of content on the Internet. Simply fetching all data from the cloud severely degrades efficiency in terms of resource utilization and bandwidth, and data can be encrypted by multiple content providers under different keys to preserve privacy. Thus, locating near-duplicate data in a privacy-preserving way is highly dependent on the ability to deduplicate redundant search results and returns best matches without decrypting data. To this end, we propose an efficient near-duplicate detection scheme for encrypted data in the cloud. Our scheme has the following benefits. First, a single query is enough to locate near-duplicate data even if they are encrypted under different keys of multiple content providers. Second, storage, computation and communication costs are alleviated compared to existing schemes, while achieving the same level of search accuracy. Third, scalability is significantly improved as a result of a novel and efficient two-round detection to locate near-duplicate candidates over large quantities of data in the cloud. An experimental analysis with real-world data demonstrates the applicability of the proposed scheme to a practical cloud system. Last, the proposed scheme is an average of 70.6% faster than an existing scheme.

• The Korean Journal of Archival Studies
• /
• no.37
• /
• pp.273-305
• /
• 2013

In order to actively use of records and information in current digital record management systems, we have to check whether a system is designed to fully support the user of records and whether the good use of the system is being. This study analyzed the status of the use of the search function of Records Management System(RMS) for public agencies, and evaluated them. In order to investigate the status of the use of the search function, it surveyed records managers of public agencies using the RMS. The result showed that records managers unsatisfied with the usability and the search performance of the RMS. To evaluate the search function, it identified the functional requirement of the system and develops a checklist that can be used for evaluation. Two assessments were conducted. Firstly, as pre-evaluation, it assessed the degree of implementation of the current RMS according to the checklist as an inspection chart using document examination method. Secondly, it assessed the degree of implementation using a survey of records managers of public agencies that use the RMS. Assessment results show the improvement of the basic features that are essential to the system is required. In particular, the search function is required to improve user-friendliness for the user. For the advance of RMS, this study discusses the necessity for improvement of the search functions, the build of continuous maintenance and management system, and the user education.

• Korean Journal of Plant Resources
• /
• v.35 no.2
• /
• pp.346-361
• /
• 2022

Food poisoning accidents caused by poisonous plants occur every year. As certain poisonous plants are mistaken for edible plants causing food poisoning, accurate species identification of poisonous plants is required. DNA barcodes suitable for species identification of poisonous plants and database that can be used for accurate species identification are necessary for their use in forensic cases. In this study, species identification of 19 poisonous plants native to Jeju Island using seven DNA barcodes (trnH-psbA, trnL-trnF, trnL intron, rbcL, matK, ITS1-ITS4, 18S rRNA) was performed to construct a database containing sequence information and DNA barcode universality. trnL-trnF barcode and ITS1-ITS4 barcode were the easiest markers for PCR amplification and sequence retrieval, and the combination of the two barcodes enabled single species identification in 18 out of 19 plants. Therefore, when an investigation of unknown poisonous plants is requested, combination of trnL-trnF and ITS1-ITS4 barcodes is considered as a primary marker for species identification. The database of recommended DNA barcodes for each poisonous plant presented in this study will be helpful in plants poisoning cases.