• KIPS Transactions on Software and Data Engineering
• /
• v.3 no.8
• /
• pp.321-328
• /
• 2014

In this paper, we present a MapReduce-based mobility pattern mining system which can predict efficiently the next place of mobile users. It learns the mobility pattern model of each user, represented by Hidden Markov Models(HMM), from a large-scale trajectory dataset, and then predicts the next place for the user to visit by applying the learned models to the current trajectory. Our system consists of two parts: the back-end part, in which the mobility pattern models are learned for individual users, and the front-end part, where the next place for a certain user to visit is predicted based on the mobility pattern models. While the back-end part comprises of three distinct MapReduce modules for POI extraction, trajectory transformation, and mobility pattern model learning, the front-end part has two different modules for candidate route generation and next place prediction. Map and reduce functions of each module in our system were designed to utilize the underlying Hadoop infrastructure enough to maximize the parallel processing. We performed experiments to evaluate the performance of the proposed system by using a large-scale open benchmark dataset, GeoLife, and then could make sure of high performance of our system as results of the experiments.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.13 no.3
• /
• pp.14-28
• /
• 2010

The aim of this study is to investigate crime hotspot areas using the spatio-temporal cluster analysis which is possible to search simultaneously time range as well as space range as an alternative method of existing hotspot analysis only identifying crime occurrence distribution patterns in urban area. As for research method, first, crime data were collected from criminal registers provided by official police authority in M city, Gyeongnam and crime occurrence patterns were drafted on a map by using Geographic Information Systems(GIS). Second, by utilizing Ripley K-function and Space-Time Scan Statistics analysis, the spatio-temporal distribution of crime was examined. The results showed that the risk of crime was significantly clustered at relatively few places and the spatio-temporal clustered areas of crime were different from those predicted by existing spatial hotspot analysis such as kernel density analysis and k-means clustering analysis. Finally, it is expected that the results of this study can be not only utilized as a valuable reference data for establishing urban planning and crime prevention through environmental design(CPTED), but also made available for the allocation of police resources and the improvement of public security services.

• The Journal of the Korea Contents Association
• /
• v.14 no.9
• /
• pp.529-537
• /
• 2014

To establish a foundation to objectively interpret Chosun history, we construct people network of the Chosun dynasty. The network shows scale free network properties as if most social networks do. The people network is composed of 1,379 nodes and 3,874 links and its diameter is 14. To analysis of the network dynamics, whole network that is composed of 27 king networks were constructed by adding the first king, Taejo network to the second king, Jeongjong network and then continuously adding the next king networks. Interestingly, betweenness and closeness centralities were gradually decreased but stress centrality was drastically increased. These results indicate that information flow is gradually slowing and hub node position is more centrally oriented as growing the network. To elucidate key persons from the network, k-core and MCODE algorithms that can extract core or module structures from whole network were employed. It is a possible to obtain new insight and hidden information by analyzing network dynamics. Due to lack of the dynamic interacting data, there is a limit for network dynamic research. In spite of using concise data, this research provides us a possibility that annals of the Chosun dynasty are very useful historical data for analyzing network dynamics.

• Food Science of Animal Resources
• /
• v.30 no.2
• /
• pp.313-320
• /
• 2010

The purpose of this study was to evaluate the freshness of chicken meat during 19 d of storage at $4^{\circ}C$ using a portable electronic nose. The portable system consisted of six different metal oxide sensors and a moisture sensor. Determination of volatile compounds with gas chromatography-mass spectrometry, total bacterial count (TBC), and 2-thiobarbituric acid reactive substances (TBARS) monitored the quality change of the samples. These results were compared with the results measured by the electronic nose system. TBC and TBARS measurements could be separated into five groups and seven groups, respectively, among ten groups. According to principal component analysis and linear discriminant analysis with the signals of the portable electronic nose, the sample groups could be clearly separated into eight groups and nine groups, respectively, among ten groups. The portable electronic nose demonstrated potential for evaluating freshness of stored chicken.

• The KIPS Transactions:PartA
• /
• v.12A no.5 s.95
• /
• pp.451-460
• /
• 2005

We present a methodology to design energy-efficient data paths using FPGAs. Our methodology integrates domain specific modeling, coarse-grained performance evaluation, design space exploration, and low-level simulation to understand the tradeoffs between energy, latency, and area. The domain specific modeling technique defines a high-level model by identifying various components and parameters specific to a domain that affect the system-wide energy dissipation. A domain is a family of architectures and corresponding algorithms for a given application kernel. The high-level model also consists of functions for estimating energy, latency, and area that facilitate tradeoff analysis. Design space exploration(DSE) analyzes the design space defined by the domain and selects a set of designs. Low-level simulations are used for accurate performance estimation for the designs selected by the DSE and also for final design selection We illustrate our methodology using a family of architectures and algorithms for matrix multiplication. The designs identified by our methodology demonstrate tradeoffs among energy, latency, and area. We compare our designs with a vendor specified matrix multiplication kernel to demonstrate the effectiveness of our methodology. To illustrate the effectiveness of our methodology, we used average power density(E/AT), energy/(area x latency), as themetric for comparison. For various problem sizes, designs obtained using our methodology are on average $25\%$ superior with respect to the E/AT performance metric, compared with the state-of-the-art designs by Xilinx. We also discuss the implementation of our methodology using the MILAN framework.

• Journal of Internet Computing and Services
• /
• v.15 no.2
• /
• pp.19-31
• /
• 2014

In this paper, a parallel backoff delay procedure on multiple IEEE 802.15.4 channels and a channel searching method considering the frequency spectrum of WiFi traffic are studied for IEEE 802.15.4 nodes to avoid the interference from WiFi traffic. In order to search the channels being occupied by WiFi traffic, we analyzed the methods measuring the powers of adjacent channels simultaneously, checking the duration of measured power levels greater than a threshold, and finding the same periodicity of sampled RSSI data as the beacon frame by signal processing. In an wireless channel overlapped with IEEE 802.11 network, the operation of CSMA-CA algorithm for IEEE 802.15.4 nodes is explained. A method to execute a parallel backoff procedure on multiples IEEE 802.15.4 channels by an IEEE 802.15.4 device is proposed with the description of its algorithm. When we analyze the data measured by the experimental system implemented with the proposed method, it is observed that medium access delay times increase at the same time in the associated IEEE 802.15.4 channels that are adjacent each other during the generation of WiFi traffic. A channel evaluation function to decide the interference from other traffic on an IEEE 802.15.4 channel is defined. A channel searching method considering the channel evaluations on the adjacent channels together is proposed in order to search the IEEE 802.15.4 channels interfered by WiFi, and the experimental results show that it correctly finds the channels interfered by WiFi traffic.

• Journal of Convergence for Information Technology
• /
• v.11 no.5
• /
• pp.30-37
• /
• 2021

Various devices are connected to the Internet, and attacks using the Internet are occurring. Among such attacks, there are attacks that use malicious URLs to make users access to wrong phishing sites or distribute malicious viruses. Therefore, how to detect such malicious URL attacks is one of the important security issues. Among recent deep learning technologies, neural networks are showing good performance in image recognition, speech recognition, and pattern recognition. This neural network can be applied to research that analyzes and detects patterns of malicious URL characteristics. In this paper, performance analysis according to various parameters was performed on a method of detecting malicious URLs using neural networks. In this paper, malicious URL detection performance was analyzed while changing the activation function, learning rate, and neural network structure. The experimental data was crawled by Alexa top 1 million and Whois to build the data, and the machine learning library used TensorFlow. As a result of the experiment, when the number of layers is 4, the learning rate is 0.005, and the number of nodes in each layer is 100, the accuracy of 97.8% and the f1 score of 92.94% are obtained.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.12
• /
• pp.519-524
• /
• 2023

In hashing-based image retrieval, the hash code of a manipulated image is different from the original image, making it difficult to search for the same image. This paper proposes and evaluates a self-supervised deephashing model that generates perceptual hash codes from feature information such as texture, shape, and color of images. The comparison models are autoencoder-based variational inference models, but the encoder is designed with a fully connected layer, convolutional neural network, and transformer modules. The proposed model is a variational inference model that includes a SimAM module of extracting geometric patterns and positional relationships within images. The SimAM module can learn latent vectors highlighting objects or local regions through an energy function using the activation values of neurons and surrounding neurons. The proposed method is a representation learning model that can generate low-dimensional latent vectors from high-dimensional input images, and the latent vectors are binarized into distinguishable hash code. From the experimental results on public datasets such as CIFAR-10, ImageNet, and NUS-WIDE, the proposed model is superior to the comparative model and analyzed to have equivalent performance to the supervised learning-based deephashing model. The proposed model can be used in application systems that require low-dimensional representation of images, such as image search or copyright image determination.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.14 no.2
• /
• pp.122-127
• /
• 2010

Purpose: The Xpress3.$cardiac^{TM}$ which is a kind of wide beam reconstruction (WBR) method developed by UltraSPECT (Haifa, Israel) enables the acquisition of at quarter time while maintaining image quality. The purpose of this study is to investigate the usefulness of WBR method for decreasing scan times and to compare to it with filtered back projection (FBP), which is the method routinely used. Materials and Methods: Phantom and clinical studies were performed. The anthropomorphic torso phantom was made on an equality with counts from patient's body. The Tl-201 concentrations in the compartments were 74 kBq (2 ${\mu}Ci$)/cc in myocardium, 11.1 kBq (0.3 ${\mu}Ci$)/cc in soft tissue, and 2.59 kBq (0.07 ${\mu}Ci$)/cc in lung. The non-gated Tl-201 myocardial perfusion SPECT data were acquired with the phantom. The former study was scanned for 50 seconds per frame with FBP method, and the latter study was acquired for 13 seconds per frame with WBR method. Using the Xeleris ver. 2.0551, full width at half maximum (FWHM) and average image contrast were compared. In clinical studies, we analyzed the 30 patients who were examined by Tl-201 gated myocardial perfusion SPECT in department of nuclear medicine at Asan Medical Center from January to April 2010. The patients were imaged at full time (50 second per frame) with FBP algorithm and again quarter-time (13 second per frame) with the WBR algorithm. Using the 4D MSPECT (4DM), Quantitative Perfusion SPECT (QPS), and Quantitative Gated SPECT (QGS) software, the summed stress score (SSS), summed rest score (SRS), summed difference score, end-diastolic volume (EDV), end-systolic volume (ESV) and ejection fraction (EF) were analyzed for their correlations and statistical comparison by paired t-test. Results: As a result of the phantom study, the WBR method improved FWHM more than about 30% compared with FBP method (WBR data 5.47 mm, FBP data 7.07 mm). And the WBR method's average image contrast was also higher than FBP method's. However, in result of quantitative indices, SSS, SDS, SRS, EDV, ESV, EF, there were statistically significant differences from WBR and FBP(p<0.01). In the correlation of SSS, SDS, SRS, there were significant differences for WBR and FBP (0.18, 0.34, 0.08). But EDV, ESV, EF showed good correlation with WBR and FBP (0.88, 0.89, 0.71). Conclusion: From phantom study results, we confirmed that the WBR method reduces an acquisition time while improving an image quality compared with FBP method. However, we should consider significant differences in quantitative indices. And it needs to take an evaluation test to apply clinical study to find a cause of differences out between phantom and clinical results.

• The KIPS Transactions:PartD
• /
• v.14D no.6
• /
• pp.675-688
• /
• 2007

Memory access errors are frequently occurred in C programs. A number of tools and research works have been trying to detect the errors automatically. However, they have one or more of the following problems: inability to detect all memory errors, changing the memory allocation mechanism, incompatibility with libraries, and excessive performance overhead. In this paper, we suggest a new method to solve these problems, and then present a result of comparison to the previous research works through the experiments. Our approach consists of two phases. First is to transform source code at compile time through inserting instrumentation into the source code. And second is to detect memory errors at run time with a bitmap that maintains information about memory allocation. Our approach has improved the error detection abilities against the binary code analysis based ones by using the source code analysis technique, and enhanced performance in terms of both space and time, too. In addition, our approach has no problem with respect to compatibility with shared libraries as well as does not need to modify memory allocation mechanism.