• Journal of Information Processing Systems
• /
• v.19 no.5
• /
• pp.688-701
• /
• 2023

The conventional methods of network intrusion detection system (NIDS) cannot measure the trend of intrusiondetection targets effectively, which lead to low detection accuracy. In this study, a NIDS method which based on a deep neural network in a big-data environment is proposed. Firstly, the entire framework of the NIDS model is constructed in two stages. Feature reduction and anomaly probability output are used at the core of the two stages. Subsequently, a convolutional neural network, which encompasses a down sampling layer and a characteristic extractor consist of a convolution layer, the correlation of inputs is realized by introducing bidirectional long short-term memory. Finally, after the convolution layer, a pooling layer is added to sample the required features according to different sampling rules, which promotes the overall performance of the NIDS model. The proposed NIDS method and three other methods are compared, and it is broken down under the conditions of the two databases through simulation experiments. The results demonstrate that the proposed model is superior to the other three methods of NIDS in two databases, in terms of precision, accuracy, F1- score, and recall, which are 91.64%, 93.35%, 92.25%, and 91.87%, respectively. The proposed algorithm is significant for improving the accuracy of NIDS.

• Journal of Information Processing Systems
• /
• v.19 no.6
• /
• pp.842-857
• /
• 2023

Currently, most sentiment classification models on microblogging platforms analyze sentence parts of speech and emoticons without comprehending users' emotional inclinations and grasping moral nuances. This study proposes a hybrid sentiment analysis model. Given the distinct nature of microblog comments, the model employs a combined stop-word list and word2vec for word vectorization. To mitigate local information loss, the TextCNN model, devoid of pooling layers, is employed for local feature extraction, while BiLSTM is utilized for contextual feature extraction in deep learning. Subsequently, microblog comment sentiments are categorized using a classification layer. Given the binary classification task at the output layer and the numerous hidden layers within BiLSTM, the Tanh activation function is adopted in this model. Experimental findings demonstrate that the enhanced TextCNN-BiLSTM model attains a precision of 94.75%. This represents a 1.21%, 1.25%, and 1.25% enhancement in precision, recall, and F1 values, respectively, in comparison to the individual deep learning models TextCNN. Furthermore, it outperforms BiLSTM by 0.78%, 0.9%, and 0.9% in precision, recall, and F1 values.

• Journal of the korean society of oceanography
• /
• v.33 no.3
• /
• pp.113-122
• /
• 1998

The requisite numbers of sample replicates for the population study of soft-bottom benthos were estimated from survey data on the Songdo tidal flat and subtidal zone in Youngil Bay, Korea. Large numbers of samples were taken; two-hundred-fifty 0.02 m$^2$ box corers and fifty 0.1m$^2$ van Veen grabs were taken on the Songdo tidal flat and in Youngil Bay, respectively. The effect of sampler size on sampling efforts was investigated by pooling the unit samples in pairs, fours, eights, etc. The requisite number of sample replicates (n$_r$) was determined by sample variance (s$^2$) and mean (m) function (n$_r$:s$^2$/P$^2$m$^2$), at P=0.2 level, in which s$^2$ and m were calculated from the counts of individuals collected. For example, seven samples of 0.02 m$^2$ corer for the intertidal and two samples of 0.1 m$^2$ van Veen grab for subtidal fauna were required to estimate the total density of community. The smaller sampler size was more efficient than larger ones when sampling costs were compared on the basis of the total sampling area. The requisite number of sample replicates was also predicted ($\^{n}$n$_r$) by substituting $\^{s}$$^2$ obtained from the regression of s$^2$ against m using the Taylor's power law ($\^{s}$$^2$:am$^b$). The regression line of survey data on s$^2$ and m plotted on log scale was well fitted to the Taylor's power law (r$^2$${\geq}$0.95, p<;0.001) over the whole range of m. The exponent b was, however, varied when it was estimated from m which was categorized into classes by its scale. The fitted exponent b was large when both density class and the sampler size were large. The number of sample replicates, therefore, could be more significantly estimated, if regression coefficients (a and b) would be calculated from sample variance and mean categorized into density classes.

• Journal of Advanced Marine Engineering and Technology
• /
• v.33 no.8
• /
• pp.1137-1143
• /
• 2009

Experiments were conducted to evaluate pool boiling heat transfer performance between plain and micro finned surfaces with FC-72, which is chemically and electrically stable. Three kinds of micro fins with the dimension of $100{\mu}m\;{\times}\;10{\mu}m$, $150{\mu}m\;{\times}\;10{\mu}m$ and $200{\mu}m\;{\times}\;10{\mu}m$ (width $\times$ height) were fabricated on the surface of a silicon chip. The experiments were carried out on the liquid subcooling of 5, 10 and 15 K under the atmospheric condition. The micro finned surface with a larger fin width of $200{\mu}m$ provided a better pool boiling heat transfer performance. Also, the micro finned surfaces showed a sharp increase in heat flux with increasing wall superheat and a larger heat transfer enhancement compared to a plain surface.

• Asian Pacific Journal of Cancer Prevention
• /
• v.14 no.6
• /
• pp.3719-3724
• /
• 2013

Background: RASSF1A has been reported to be a candidate tumor suppressor in non-small cell lung cancer (NSCLC). However, the association between RASSF1A promoter methylation and NSCLC remains unclear, particularly in regarding links to clinicopathologic features. Methods: Eligible studies were identified through searching PubMed, EMBASE, Cochrane Library and China National Knowledge Infrastructure (CNKI) databases. Studies were pooled and odds ratios (ORs) with corresponding confidence intervals (CIs) were calculated. Funnel plots were also performed to evaluate publication bias. Results: Nineteen studies involving 2,063 cases of NSCLC and 1,184 controls were included in this meta-analysis. A significant association was observed between RASSF1A methylation and NSCLC in the complete data set (OR = 19.42, 95% CI: 14.04-26.85, P < 0.001). Pooling the control tissue subgroups (heterogeneous/autologous) gave pooled ORs of 32.4 (95% CI, 12.4-84.5) and 17.7 (95% CI, 12.5-25.0) respectively. Racial subgroup (Caucasian/Asian) analysis gave pooled ORs of 26.6 (95% CI, 10.9-64.9) and 20.9 (95% CI, 14.4-30.4) respectively. The OR for RASSF1A methylation in poorly-differentiated vs. moderately/well-differentiated NSCLC tissues was 1.88 (95% CI, 1.32-2.68, P<0.001), whereas there were no significant differences in RASSF1A methylation in relation to gender, pathology, TNM stage and smoking behavior among NSCLC cases. Conclusion: This meta-analysis suggests a significant association between RASSF1A methylation and NSCLC, confirming the role of RASSF1A as a tumor suppressor gene. Large-scale and well-designed case-control studies are needed to validate the associations identified in the present meta-analysis.

• Journal of KIISE
• /
• v.44 no.9
• /
• pp.992-1004
• /
• 2017

To assess image quality accurately, an image quality assessment (IQA) metric is required to reflect the human visual system (HVS) properly. In other words, the structure, color, and contrast ratio of the image should be evaluated in consideration of various factors. In addition, as mobile embedded devices such as smartphone become popular, a fast computing speed is important. In this paper, the proposed IQA metric combines color similarity, gradient similarity, and phase similarity synergistically to satisfy the HVS and is designed by using optimized pooling and quantization for fast computation. The proposed IQA metric is compared against existing 13 methods using 4 kinds of evaluation methods. The experimental results show that the proposed IQA metric ranks the first on 3 evaluation methods and the first on the remaining method, next to VSI which is the most remarkable IQA metric. Its computing speed is on average about 20% faster than VSI's. In addition, we find that the proposed IQA metric has a bigger amount of correlation with the HVS than existing IQA metrics.

• The Korean Journal of Applied Statistics
• /
• v.31 no.1
• /
• pp.67-75
• /
• 2018

Multidimensional scaling (MDS) is an exploratory analysis of multivariate data to represent the dissimilarity among objects in the geometric low-dimensional space. However, a general MDS map only shows the information of objects without any information about variables. In this study, we used MDS based on the algorithm of Torgerson (Theory and Methods of Scaling, Wiley, 1958) to visualize some clusters of objects in categorical data. For this, we convert given data into a multiple indicator matrix. Additionally, we added the information of levels for each categorical variable on the MDS map by applying the partition method of Shin et al. (Korean Journal of Applied Statistics, 28, 1171-1180, 2015). Therefore, we can find information on the similarity among objects as well as find associations among categorical variables using the proposed MDS map.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.14 no.6
• /
• pp.1207-1214
• /
• 2019

Technologies of the 4th industrial revolution have been unknowingly seeping into our lives. Many IoT based home security systems are using the convolutional neural network(CNN) as good biometrics to recognize a face and protect home and family from intruders since CNN has demonstrated its excellent ability in image recognition. In this paper, three layouts of CNN for 2D and 2.5D image of small dataset with various input image size and filter size are explored. The simulation results show that the layout of CNN with 50*50 input size of 2.5D image, 2 convolution and max pooling layer, and 3*3 filter size for small dataset of 2.5D image is optimal for a home security system with recognition accuracy of 0.966. In addition, the longest CPU time consumption for one input image is 0.057S. The proposed layout of CNN for a face recognition is suitable to control the actuators in the home security system because a home security system requires good face recognition and short recognition time.

• Journal of KIISE:Information Networking
• /
• v.30 no.2
• /
• pp.244-251
• /
• 2003

As the Internet and wireless mobile communication is being widespread, the WAP (Wireless Application Protocol) that merges these two technologies has emerged. Also, the research of the WAP gateway that enables efficient processing of multiple user demands concurrently and prompt response to requests from various wireless devices has been performed in many working groups. The WAP stack is the most fundamental technology among these researches. In this paper, we implement the WAP container utilizing the JAVA's multithreading and applying to the connection pooling technique to manage the available resource in the container efficiently. Also, we design and implement the WTP(Wireless Transaction Protocol) layer and the UDP(User Datagram Protocol) layer based on the priority Policy. Our design and implementation showed the shorter waiting time than that of the existing FCFS(First-Come, First-Served) system in the transaction layer and its efficiency was proved through simulation.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.15 no.4
• /
• pp.17-25
• /
• 2020

Venture debt is a prominent funding tool to promote scale-up of ventures. In the growth stage, venture firms that need large-scale funding can accelerate their growth by leveraging venture debt without diluting their shares, while venture capitals can quickly recollect their investments by accelerating the growth of the ventures they invest. By supplying venture debt, banks can diversify their asset primarily concentrated on loans, and improve the return on assets. As in the case of Silicon Valley Bank, a leading venture lender, closer cooperation between the two agents is essential to supply venture debt. One is the venture capital, an equity capital supplier, and the other is the bank, a debt capital supplier. To this end, we propose "credit risk sharing venture loans" and "venture loan pooling". The former encourages banks' participation in the venture debt market where the manager of Korean Fund of Funds, KVIC and policy guarantee schemes such as KODIT and KIBO screen or partially absorbe the risks inherent in venture loans. The latter reduces the burden of banking on individual venture loans through securitization.