• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.6
• /
• pp.749-754
• /
• 2011

In this study, we propose the design of optimized pRBFNNs-based face recognition system using two-dimensional Image and ASM algorithm. usually the existing 2 dimensional face recognition methods have the effects of the scale change of the image, position variation or the backgrounds of an image. In this paper, the face region information obtained from the detected face region is used for the compensation of these defects. In this paper, we use a CCD camera to obtain a picture frame directly. By using histogram equalization method, we can partially enhance the distorted image influenced by natural as well as artificial illumination. AdaBoost algorithm is used for the detection of face image between face and non-face image area. We can butt up personal profile by extracting the both face contour and shape using ASM(Active Shape Model) and then reduce dimension of image data using PCA. The proposed pRBFNNs consists of three functional modules such as the condition part, the conclusion part, and the inference part. In the condition part of fuzzy rules, input space is partitioned with Fuzzy C-Means clustering. In the conclusion part of rules, the connection weight of RBFNNs is represented as three kinds of polynomials such as constant, linear, and quadratic. The essential design parameters (including learning rate, momentum coefficient and fuzzification coefficient) of the networks are optimized by means of Differential Evolution. The proposed pRBFNNs are applied to real-time face image database and then demonstrated from viewpoint of the output performance and recognition rate.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.8
• /
• pp.49-59
• /
• 2022

In this paper, we designed a new enzyme function prediction model PSCREM based on a study that compared and evaluated CNN and LSTM/GRU models, which are the most widely used deep learning models in the field of predicting functions and structures using protein sequences in 2020, under the same conditions. Sequence evolution information was used to preserve detailed patterns which would miss in CNN convolution, and the relationship information between amino acids with functional significance was extracted through overlapping RNNs. It was referenced to feature map production. The RNN family of algorithms used in small CNN-RNN models are LSTM algorithms and GRU algorithms, which are usually stacked two to three times over 100 units, but in this paper, small RNNs consisting of 10 and 20 units are overlapped. The model used the PSSM profile, which is transformed from protein sequence data. The experiment proved 86.4% the performance for the problem of predicting the main classes of enzyme number, and it was confirmed that the performance was 84.4% accurate up to the sub-sub classes of enzyme number. Thus, PSCREM better identifies unique patterns related to protein function through overlapped RNN, and Overlapped RNN is proposed as a novel methodology for protein function and structure prediction extraction.

• Information Systems Review
• /
• v.22 no.1
• /
• pp.183-198
• /
• 2020

The development of IT increases the importance of understanding of IT-driven ecosystems. Platform business is the representative business model in the era of innovative IT-based businesses. However, it lacks the review research that entails ecosystem perspectives from traditional disciplines in which the perspective of ecosystem had been applied. Further most of platform research have focused on the comparison between ecosystems as a whole rather than exploration on complementors in the ecosystem who are selected and survive and, in turn, contributed to maintain the ecosystem to compete with other ecosystems. The current study listed highly cited papers from economics, sociological ecology, socio-technical ecology, organization studies, and marketing research which have cumulated research on ecosystems. And the three most critical features that determine the success of complementors, which are competition, relationality, and adaptability. Present study showed how the features were explained by each perspective from the different disciplines.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.25 no.1
• /
• pp.107-122
• /
• 2015

Many defensive mechanisms have been evolved as new attack methods are developed. However, APT attacks using e-mail are still hard to detect and prevent. Recently, many organizations in the government sector or private sector have been hacked by malicious e-mail based APT attacks. In this paper, first, we built hacking e-mail database based on the real e-mail data which were used in attacks on the Korean government organizations in recent years. Then, we extracted features from the hacking e-mails for profiling them. We design a case vector that can describe the specific characteristics of hacking e-mails well. Finally, based on case based reasoning, we made an algorithm for retrieving the most similar case from the hacking e-mail database when a new hacking e-mail is found. As a result, hacking e-mails have common characteristics in several features such as geo-location information, and these features can be used for classifying benign e-mails and malicious e-mails. Furthermore, this proposed case based reasoning algorithm can be useful for making a decision to analyze suspicious e-mails.

• KIPS Transactions on Computer and Communication Systems
• /
• v.2 no.9
• /
• pp.405-412
• /
• 2013

Current web has evolved to a mashed-up format according to the change of the implementation and usage patterns. Web services and user experiences have improved, however, security threats are also increased as the web contents that are not yet verified combine together. To mitigate the threats incurred as an adverse effect of the web development, we need to check security on the combined web contents. In this paper, we propose a scheduling method to detect malicious web pages not only inside but also outside through extended links for secure operation of a web site. The scheduling method considers several aspects of each page including connection popularity, suspiciousness, and check elapse time to make a decision on the order for security check on numerous web pages connected with links. We verified the effectiveness of the security check complying with the scheduling method that uses the priority given to each page.

• Korean Journal of Ecology and Environment
• /
• v.54 no.3
• /
• pp.229-239
• /
• 2021

Thamnaconus modestus, distributed in the Northwest Pacific, has high economic value and is used in various seafood. In this study, the morphological and genetic characteristics of T. modestus and T. septentrionalis were compared and analyzed. We observed the external and internal morphology of T. modestus, sketched skeletal elements, and analyzed phylogenetic evolutionary relationships using the cytochrome c oxidase subunit I (COI) gene on mitochondrial DNA compared to T. septentrionalis. The T. modestus observed in this study had blackish-brown patterns irregularly scattered on the gray-brown body, and the fins were blue-green. Genetic analysis results based on the COI sequences of T. modestus showed seven types of base sequence variation; however, the homology was more than 98.8%. In addition, as a result of comparison of the COI nucleotide sequences and phylogenetic analysis in Tetraodontiformes, two T. septentrionalis sequences (JN813099, MW485059) were similar to T. modestus with 99% homology, and the other two T. septentrionalis sequences (EF607583, KP267619) were similar to those of species belonging to another genus Thamnaconus with 95% homology with T. modestus. It was not easy to classify the species based on morphological characteristics, and phylogenetic analysis between T. modestus and T. septentrionalis confirmed the difference in classification. These results provide the external and internal morphology of T. modestus and will be used as important information for the taxonomic study of T. modestus and T. septentrionalis.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.24 no.5
• /
• pp.897-909
• /
• 2014

With the development of the Internet, the number of cyber threats is continuously increasing and their techniques are also evolving for the purpose of attacking our crucial systems. Since attackers are able to easily make exploit codes, i.e., malware, using dedicated generation tools, the number of malware is rapidly increasing. However, it is not easy to analyze all of malware due to an extremely large number of malware. Because of this, many researchers have proposed the malware classification methods that aim to identify unforeseen malware from the well-known malware. The existing malware classification methods used malicious information obtained from the static and the dynamic malware analysis as the criterion of calculating the similarity between malwares. Also, most of them used API functions and their sequences that are divided into a certain length. Thus, the accuracy of the malware classification heavily depends on the length of divided API sequences. In this paper, we propose an extraction method of optimized API sequence length and combination that can be used for improving the performance of the malware classification.