• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.39C no.3
• /
• pp.255-263
• /
• 2014

Today, a number of CCTV on the highway is to observe the flow of traffics. There have been a number of studies where traffic data (e.g., the speed of vehicles and the amount of traffic on the road) are transferred back to the centralized server so that an appropriate action can be taken. This paper introduces a system that detects the changes of traffic flows caused by an accident or unexpected stopping (i.e., vehicle remains idle) by monitoring each lane separately. The traffic flows of each lane are level spacing curve that shows Wigner distribution for location vector. Applying calogero-moser system and Hamiltonian system, probability equation for each level-spacing curve is derived. The high level of modification of the signal means that the lane is in accident situation. This is different from previous studies in that it does more than looking for the signal from only one lane, now it is able to detect an accident in entire flow of traffic. In process of monitoring traffic flow of each lane, when camera recognizes a shadow of vehicle as a vehicle, it will affect the accident detecting capability. To prevent this from happening, the study introduces how to get rid of such shadow. The system using Basian network method is being compared for capability evaluation of the system of the study. As a result, the system of the study appeared to be better in performance in detecting the modification of traffic flow caused by idle vehicle.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.6
• /
• pp.163-172
• /
• 2012

Collaborative filtering which is used explicit method in a existing recommedation system, can not only reflect exact attributes of item but also still has the problem of sparsity and scalability, though it has been practically used to improve these defects. This paper proposes the personalized recommendation system using RFM method and k-means clustering in u-commerce which is required by real time accessablity and agility. In this paper, using a implicit method which is is not used complicated query processing of the request and the response for rating, it is necessary for us to keep the analysis of RFM method and k-means clustering to be able to reflect attributes of the item in order to find the items with high purchasablity. The proposed makes the task of clustering to apply the variable of featured vector for the customer's information and calculating of the preference by each item category based on purchase history data, is able to recommend the items with efficiency. To estimate the performance, the proposed system is compared with existing system. As a result, it can be improved and evaluated according to the criteria of logicality through the experiment with dataset, collected in a cosmetic internet shopping mall.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.5
• /
• pp.57-70
• /
• 2009

A CAD design drawing based on vector data that is very important art work in industrial fields has been considered to content that the copyright protection is urgently needed. This paper presents a watermarking scheme based on k-means++ for CAD design drawing. One CAD design drawing consists of several layers and each layer consists of various geometric objects such as LINE, POLYLINE, CIRCLE, ARC, 3DFACE and POLYGON. POLYLINE with LINE, 3DFACE and ARC that are fundamental objects make up the majority in CAD design drawing. Therefore, the proposed scheme selects the target object with high distribution among POLYLINE, 3DFACE and ARC objects in CAD design drawing and then selects layers that include the most target object. Then we cluster the target objects in the selected layers by using k-means++ and embed the watermark into the geometric distribution of each group. The geometric distribution is the normalized length distribution in POLYLINE object, the normalized area distribution in 3DFACE object and the angle distribution in ARC object. Experimental results verified that the proposed scheme has the robustness against file format converting, layer attack as well as various geometric editing provided in CAD editing tools.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.4
• /
• pp.45-54
• /
• 2012

Location-based services with GPS positioning technology as a key technology, but recognizing the current location through satellite communication is not possible in an indoor location-aware technology, low-power short-range communication is primarily made of the study. Especially, as Chirp Spread Spectrum(CSS) based location-aware approach for low-power physical layer IEEE802.15.4a is selected as a standard, Ranging distance estimation techniques and data transfer speed enhancements have been more developed. It is known that the distance measured by CSS ranging has quite a lot of noise as well as its bias. However, the noise problem can be adjusted by modeling the non-zero mean noise value by a scaling factor which corresponds to the change of magnitude of a measured distance vector. In this paper, we propose a localization system using the CSS signal to measure distance for a mobile node taken a measurement of the exact coordinates. By applying the extended kalman filter and least mean squares method, the localization system is faster, more stable. Finally, we evaluate the reliability and accuracy of the proposed algorithm's performance by the experiment for the realization of localization system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.10
• /
• pp.1422-1432
• /
• 1993

Nonlinear mapping function of the HCNN( Hidden Control Neural Network ) can change over time to model the temporal variability of a speech signal by combining the nonlinear prediction of conventional neural networks with the segmentation capability of HMM. We have two things in this paper. first, we showed that the performance of the HCNN is better than that of HMM. Second, the HCNN with its prediction error measure given by weighted distance is proposed to use suitable distance measure for the HCNN, and then we showed that the superiority of the proposed system for speaker-independent speech recognition tasks. Weighted distance considers the differences between the variances of each component of the feature vector extraced from the speech data. Speaker-independent Korean digit recognition experiment showed that the recognition rate of 95%was obtained for the HCNN with Euclidean distance. This result is 1.28% higher than HMM, and shows that the HCNN which models the dynamical system is superior to HMM which is based on the statistical restrictions. And we obtained 97.35% for the HCNN with weighted distance, which is 2.35% better than the HCNN with Euclidean distance. The reason why the HCNN with weighted distance shows better performance is as follows : it reduces the variations of the recognition error rate over different speakers by increasing the recognition rate for the speakers who have many misclassified utterances. So we can conclude that the HCNN with weighted distance is more suit-able for speaker-independent speech recognition tasks.

• The KIPS Transactions:PartB
• /
• v.11B no.2
• /
• pp.187-198
• /
• 2004

The World-Wide Web is the largest distributed Information space and has grown to encompass diverse information resources. However, although Web is growing exponentially, the individual's capacity to read and digest contents is essentially fixed. From the view point of Web users, they can be confused by explosion of Web information, by constantly changing Web environments, and by lack of understanding needs of Web users. In these Web environments, mining traversal patterns is an important problem in Web mining with a host of application domains including system design and Information services. Conventional traversal pattern mining systems use the inter-pages association in sessions with only a very restricted mechanism (based on vector or matrix) for generating frequent k-Pagesets. We develop a family of novel algorithms (termed WebPR - Web Page Recommend) for mining frequent traversal patterns and then pageset to recommend. Our algorithms provide Web users with new page views, which Include pagesets to recommend, so that users can effectively traverse its Web site. The main distinguishing factors are both a point consistently spanning schemes applying inter-pages association for mining frequent traversal patterns and a point proposing the most efficient tree model. Our experimentation with two real data sets, including Lady Asiana and KBS media server site, clearly validates that our method outperforms conventional methods.

• Journal of KIISE:Software and Applications
• /
• v.31 no.5
• /
• pp.595-604
• /
• 2004

Automatic analysis of concepts or semantic relations from text documents enables not only an efficient acquisition of relevant information, but also a comparison of documents in the concept level. We present a multiple cause model-based approach to text analysis, where latent topics are automatically extracted from document sets and similarity between documents is measured by semantic kernels constructed from the extracted topics. In our approach, a document is assumed to be generated by various combinations of underlying topics. A topic is defined by a set of words that are related to the same topic or cooccur frequently within a document. In a network representing a multiple-cause model, each topic is identified by a group of words having high connection weights from a latent node. In order to facilitate teaming and inferences in multiple-cause models, some approximation methods are required and we utilize an approximation by Helmholtz machines. In an experiment on TDT-2 data set, we extract sets of meaningful words where each set contains some theme-specific terms. Using semantic kernels constructed from latent topics extracted by multiple cause models, we also achieve significant improvements over the basic vector space model in terms of retrieval effectiveness.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.1
• /
• pp.169-177
• /
• 2020

In South Korea, the results of R&D in science and technology are submitted to the National Science and Technology Information Service (NTIS) in reports that have Korea national science and technology standard classification codes (K-NSCC). However, considering there are more than 2000 sub-categories, it is non-trivial to choose correct classification codes without a clear understanding of the K-NSCC. In addition, there are few cases of automatic document classification research based on the K-NSCC, and there are no training data in the public domain. To the best of our knowledge, this study is the first attempt to build a highly performing K-NSCC classification system based on NTIS report meta-information from the last five years (2013-2017). To this end, about 210 mid-level categories were selected, and we conducted preprocessing considering the characteristics of research report metadata. More specifically, we propose a convolutional neural network (CNN) technique using only task names and keywords, which are the most influential fields. The proposed model is compared with several machine learning methods (e.g., the linear support vector classifier, CNN, gated recurrent unit, etc.) that show good performance in text classification, and that have a performance advantage of 1% to 7% based on a top-three F1 score.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.10
• /
• pp.2004-2010
• /
• 2009

The characteristic of the backbond for distribution service in WMNs(Wireless Mesh Networks) is that WMNs has multiple links connected to mesh points and dynamic routing protocol such as AODV to establish routing paths. When the terminal is communicating with the service through new AP, mobile nodes can resume communication by setting only the link between new AP and mobile node in the case of existing WLANs, but WMNs needs path establishment process in multihop networks. Our goal in this paper is to support the seamless communication service by eliminating path establishment delay in WMNs. We present the method that eliminates the handover latency by predicting the location of handover using GPS information and making the paths to their destination in advance. We implement mesh nodes using embedded board that contains proposed handover method and evaluate performance of handover latency. Our experiment shows that handover delay time is decreased from 2.47 to 0.05 seconds and data loss rate is decreased from 20~35% in the existing method to 0~10% level.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.5
• /
• pp.3378-3384
• /
• 2015

The problem of fast search for raman spectrum has attracted much attention recently. By far the most simple and widely used method is to calculate and compare the Euclidean distance between the given spectrum and the spectra in a database. But it is non-trivial problem because of the inherent high dimensionality of the data. One of the most serious problems is the high computational complexity of searching for the closet codeword. To overcome this problem, The fast codeword search algorithm based on the mean pyramids of codewords is currently used in image coding applications. In this paper, we present three new methods for the fast algorithm to search for the closet codeword. the proposed algorithm uses two significant features of a vector, mean values and variance, to reject many unlikely codewords and save a great deal of computation time. The Experiment results show about 42.8-55.2% performance improvement for the 1DMPS+PDS. The results obtained confirm the effectiveness of the proposed algorithm.