• ETRI Journal
• /
• v.43 no.6
• /
• pp.1058-1080
• /
• 2021

Crime is not a completely random event but rather shows a pattern in space and time. Capturing the dynamic nature of crime patterns is a challenging task. Crime prediction models that rely only on neighborhood influence and demographic features might not be able to capture the dynamics of crime patterns, as demographic data collection does not occur frequently and is static. This work proposes a novel approach for crime count and hotspot prediction to capture the dynamic nature of crime patterns using taxi data along with historical crime and demographic data. The proposed approach predicts crime events in spatial units and classifies each of them into a hotspot category based on the number of crime events. Four models are proposed, which consider different covariates to select a set of independent variables. The experimental results show that the proposed combined subset model (CSM), in which static and dynamic aspects of crime are combined by employing the taxi dataset, is more accurate than the other models presented in this study.

• Journal of Information Processing Systems
• /
• v.17 no.2
• /
• pp.399-410
• /
• 2021

Aiming at the problem that complex external variables in natural scenes have a greater impact on facial expression recognition results, a facial expression recognition method based on two-stream convolutional neural network is proposed. The model introduces exponentially enhanced shared input weights before each level of convolution input, and uses soft attention mechanism modules on the space-time features of the combination of static and dynamic streams. This enables the network to autonomously find areas that are more relevant to the expression category and pay more attention to these areas. Through these means, the information of irrelevant interference areas is suppressed. In order to solve the problem of poor local robustness caused by lighting and expression changes, this paper also performs lighting preprocessing with the lighting preprocessing chain algorithm to eliminate most of the lighting effects. Experimental results on AFEW6.0 and Multi-PIE datasets show that the recognition rates of this method are 95.05% and 61.40%, respectively, which are better than other comparison methods.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.123-124
• /
• 2019

This paper proposes an improved deep learning method based on small data sets for animal image classification. Firstly, we use a CNN to build a training model for small data sets, and use data augmentation to expand the data samples of the training set. Secondly, using the pre-trained network on large-scale datasets, such as VGG16, the bottleneck features in the small dataset are extracted and to be stored in two NumPy files as new training datasets and test datasets. Finally, training a fully connected network with the new datasets. In this paper, we use Kaggle famous Dogs vs Cats dataset as the experimental dataset, which is a two-category classification dataset.

• ETRI Journal
• /
• v.43 no.4
• /
• pp.746-757
• /
• 2021

Side-channel attacks pose an inevitable challenge to the implementation of cryptographic algorithms, and it is important to mitigate them. This work identifies a novel data encoding technique based on 1-of-4 codes to resist differential power analysis attacks, which is the most investigated category of side-channel attacks. The four code words of the 1-of-4 codes, namely (0001, 0010, 1000, and 0100), are split into two sets: set-0 and set-1. Using a select signal, the data processed in hardware is switched between the two encoding sets alternately such that the Hamming weight and Hamming distance are equalized. As a case study, the proposed technique is validated for the NIST standard AES-128 cipher. The proposed technique resists differential power analysis performed using statistical methods, namely correlation, mutual information, difference of means, and Welch's t-test based on the Hamming weight and distance models. The experimental results show that the proposed countermeasure has an area overhead of 2.3× with no performance degradation comparatively.

• International Journal of Computer Science & Network Security
• /
• v.22 no.1
• /
• pp.183-188
• /
• 2022

This paper explores the opportunities for scholarly publishing as part of the educational activities of the system of extracurricular research and experimental education, UNESCO Category 2 Centre, "Junior Academy of Sciences of Ukraine" (hereinafter referred to as the JASU), using Open Journal Systems (hereinafter referred to as OJS) publishing platform. The purpose of the study was to identify and select a publishing model and substantiate the procedures for the launch of an OJS-based open-access student scholarly e-journal with a view of improving scholarly communication between gifted children from Ukraine and other countries.

• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.631-640
• /
• 2023

Short-text similarity calculation is one of the hot issues in natural language processing research. The conventional keyword-overlap similarity algorithms merely consider the lexical item information and neglect the effect of the word order. And some of its optimized algorithms combine the word order, but the weights are hard to be determined. In the paper, viewing the keyword-overlap similarity algorithm, the short English text similarity algorithm based on lexical chunk theory (LC-SETSA) is proposed, which introduces the lexical chunk theory existing in cognitive psychology category into the short English text similarity calculation for the first time. The lexical chunks are applied to segment short English texts, and the segmentation results demonstrate the semantic connotation and the fixed word order of the lexical chunks, and then the overlap similarity of the lexical chunks is calculated accordingly. Finally, the comparative experiments are carried out, and the experimental results prove that the proposed algorithm of the paper is feasible, stable, and effective to a large extent.

• The Journal of Internal Korean Medicine
• /
• v.22 no.4
• /
• pp.601-611
• /
• 2001

Objectives : We aimed to identify the dose-dependent inhibitory effects of Cheonsagunja-tang(喘四君子湯), leech(Hirudo medicinalis/水蛭) roasted with Ephedrae Herba(麻黃) on the mRNA expression of IL-6, IL-16, GM-CSF involved in the asthma model. Methods : In the study BEAS-2B cell lines, human epithelial cells were used. These cells were stimulated with tumor necrosis factor(TNF)-${\alpha}$ for artificial inflammatory expression. ${\beta}$-actin messenger RNA(mRNA) was used by internal standard. After 24 hours of the Cheonsagunja-tang, leech-treatment, total cellular RNAs were collected treating RNA zol directly on the living cells. Then the transcriptional activities of IL-6, 16 and GM-CSF were measured by RT-PCR with electrophoresis. Results: In the Cheonsagunja-tang study, the mRNA expression of IL-6 showed 30% transcriptional inhibitory effect compared to the control group in the $100{\mu}l/ml$ category(p<0.005). In the IL-16, there was 26%, 31% and 31% transcriptional inhibitory effect compared to the control groups in the $4{\mu}l/ml$, $20{\mu}l/ml$ and $100{\mu}l/ml$ categories, respectively(p<0.05). In the GM-CSF, the experimental group had 56% transcriptional inhibitory effect compared to the control group in the $100{\mu}l/ml$ category(p<0.001). In other concentrations, there was no inhibitory effect. In the leech study, the mRNA expression of IL-6 showed 37% transcriptional inhibitory effect compared to the control group in the $100{\mu}l/ml$ category(p<0.001). In the IL-16, there was 63% and 67% transcriptional inhibitory effect compared to the control groups in the $20{\mu}/ml$ and $100{\mu}/ml$ categories, respectively(p<0.001). In the GM-CSF, there was 64% and 68% transcriptional inhibitory effect compared to the control groups in the $20{\mu}l/ml$ and $100{\mu}l/ml$ categories, respectively(p<0.001). In other concentrations, there was no inhibitory effect. Conclusions : This study shows that Cheonsagunja-tang and leech roasted with Ephedrae Herba have dose-dependent inhibitory effects on the mRNA expression of IL-6, IL-16 and GM-CSF in BEAS-2B cell lines, human epithelial cells.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.4
• /
• pp.471-478
• /
• 2020

This study aimed to suggest implications for the future development of agricultural R&D through a priority analysis of excellent agricultural R&D performance considering productivity, marketability, and effects on region, which are important factors of agricultural R&D evaluation. The subjects of analysis were R&D of Rural Development Administration (RDA), which was selected as national R&D excellent Performance from 2006 to 2018, and the priority was analyzed using Analytic Hierarchy Process (AHP). The results are as follows. First, the relative importance of the criteria decreased in the order of marketability(0.423) > regional impact(0.360) > productivity(0.216). Second, as a result of analyzing the relative importance of the sub-criteria, the possibility of commercialization(0.515) was high in the category of marketability. The possibility of technology expansion(0.538) was high in the category of effects on region and the quality improvement(0.464) was high in the category of productivity. Third, the excellent agricultural R&D performance that had high relative importance was the technologies associated with improving the agricultural competitiveness of horticultural crops.

• The Journal of Korean Medicine
• /
• v.23 no.2
• /
• pp.211-224
• /
• 2002

Background and Objective : In order to investigate the effects of Gilgyunglwedok-tang (GRT) on antitumor activity and antimetastatic activity, studies were done experimentally. Materials and Methods : Experimental studies were perfonned for the cytotoxic effect on BALB/c mouse lung fibroblast cells, the proliferating effect of splenic lymphocyte, the expression of CD3e/CD4, CD3e/CD8, and B220 in peripheral blood mononuclear cells (PBMCs), the cytotoxic effect on A549, SK-OV-3, SK-MEL-2, MCF-7 cells, the inhibitory effect on the activity of DNA topoisomerase I, the T/C% in ICR mice bearing S-180, the inhibitory effect of Cell adhesive of A549 Cells and SK-OY-3 Cells to complex extracellular matrix, the inhibitory effect on lung colonies, the change of lung tissue, the antiangiogenic activity, and the effect on MMP-2 and MMP-9 gene expression in the RT1080 cell line. Results and Conclusion : The results were obtained as follows : 1. In the cytotoxic effect on BALB/C mouse lung fibroblast Cell, GHT didn't show the significant cytotoxic effect on BALB/C mouse lung fibroblast cell compared to the control group. 2. In thymidine uptake assay, GHT showed the significant proliferating effect of splenic lymphocyte in proportion to the concentration. 3. In the expression of CD3e/CD4, CD3e/CD8, and B220 in peripheral blood mononuclea cells (PBMCs) of mice, GRT had no significant change to the normal group in CD4. However, GRT showed an increase to the normal group in CD8 and GHT in the only $1\mu\textrm{g}/ml$ category showed an increase to the normal group in B220. 4. In the cytotoxic effect of GRT on A549, SK-OY-3, SK-MEL-2 and MCF-7 cells, there was no significant cytotoxic effect compared to the control group. 5. In the inhibitory effect on the activity of DNA topoisomerase I, GHT in the $10\mu\textrm{g}/ml$ category showed the inhibitory effect on the activity of DNA topoisomerase I in proportion to the concentration. 6. In the T/C% in ICRmice bearing S-180, GHTtreated group showed 123.7% of T/C% compared to the control group. 7. In the inhibitory effect of cell adhesive of A549 Cells and SK-OV-3 Cells to complex extracellular matrix, GRT in the only $100\mu\textrm{g}/ml$ category showed the significant inhibitory effect compared to the control group. 8. In the inhibitory effect on lung colonies, GHT showed the significant inhibitory effect on lung colonies compared to the control group. 9. In the change of lung tissue, GHT showed a significant decrease of lung cancer growth, interalveolar fibrosis and hyaline material compared to the control group. In the development of lymphocyte around lung cancer cells and lung parenchymal, GHT showed the significant inducement efficacy compared to the control group. 10. In CAM assay, the antiangiogenic activity of GHT showed 30%. 11. In the effect on MMP-2 and MMP-9 gene expression in the RT1080 cell line, GHT had no significant inhibitory effect on MMP-2 and MMP-9 gene expression compared to the control group. According to the above results, it could be suggested that GHT has an antitumor activity and antimetastatic activity.

• Journal of the Korean Society for information Management
• /
• v.23 no.4 s.62
• /
• pp.277-294
• /
• 2006

This paper examines a level of categorization performance in a real-life collection of abstract articles in the fields of science and technology, and tests the optimal size of documents per category in a training set using a kNN classifier. The corpus is built by choosing categories that hold more than 2,556 documents first, and then 2,556 documents per category are randomly selected. It is further divided into eight subsets of different size of training documents : each set is randomly selected to build training documents ranging from 20 documents (Tr-20) to 2,000 documents (Tr-2000) per category. The categorization performances of the 8 subsets are compared. The average performance of the eight subsets is 30% in $F_1$ measure which is relatively poor compared to the findings of previous studies. The experimental results suggest that among the eight subsets the Tr-100 appears to be the most optimal size for training a km classifier In addition, the correctness of subject categories assigned to the training sets is probed by manually reclassifying the training sets in order to support the above conclusion by establishing a relation between and the correctness and categorization performance.