• Journal of Korean Society of Water and Wastewater
• /
• v.35 no.4
• /
• pp.259-275
• /
• 2021

This study explored the usefulness and implications of the Bayesian hyperparameter optimization in developing species distribution models (SDMs). A variety of machine learning (ML) algorithms, namely, support vector machine (SVM), random forest (RF), boosted regression tree (BRT), XGBoost (XGB), and Multilayer perceptron (MLP) were used for predicting the occurrence of four benthic macroinvertebrate species. The Bayesian optimization method successfully tuned model hyperparameters, with all ML models resulting an area under the curve (AUC) > 0.7. Also, hyperparameter search ranges that generally clustered around the optimal values suggest the efficiency of the Bayesian optimization in finding optimal sets of hyperparameters. Tree based ensemble algorithms (BRT, RF, and XGB) tended to show higher performances than SVM and MLP. Important hyperparameters and optimal values differed by species and ML model, indicating the necessity of hyperparameter tuning for improving individual model performances. The optimization results demonstrate that for all macroinvertebrate species SVM and RF required fewer numbers of trials until obtaining optimal hyperparameter sets, leading to reduced computational cost compared to other ML algorithms. The results of this study suggest that the Bayesian optimization is an efficient method for hyperparameter optimization of machine learning algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.2A
• /
• pp.112-120
• /
• 2010

In this paper, we propose semi-ML adaptive parallel QRDM (APQRDM) and iterative QRDM (AIQRDM) algorithms based on set grouping. Using the set grouping, the tree-search stage of QRDM algorithm is divided into partial detection phases (PDP). Therefore, when the treesearch stage of QRDM is divided into 4 PDPs, the APQRDM latency is one fourth of that of the QRDM, and the hardware requirements of AIQRDM is approximately one fourth of that of QRDM. Moreover, simulation results show that in $4{\times}4$ system and at Eb/N0 of 12 dB, APQRDM decreases the average computational complexity to approximately 43% of that of the conventional QRDM. Also, at Eb/N0 of 0dB, AIQRDM reduces the computational complexity to about 54% and the average number of metric comparisons to approximately 10% of those required by the conventional QRDM and AQRDM.

• Smart Media Journal
• /
• v.13 no.6
• /
• pp.9-15
• /
• 2024

In recent trends, there has been an increase in 'Qshing' attacks, a hybrid form of phishing that exploits fake QR (Quick Response) codes impersonating government agencies to steal personal and financial information. Particularly, this attack method is characterized by its stealthiness, as victims can be redirected to phishing pages or led to download malicious software simply by scanning a QR code, making it difficult for them to realize they have been targeted. In this paper, we have developed a classification technique utilizing machine learning algorithms to identify the maliciousness of URLs embedded in QR codes, and we have explored ways to integrate this with existing QR code readers. To this end, we constructed a dataset from 128,587 malicious URLs and 428,102 benign URLs, extracting 35 different features such as protocol and parameters, and used AutoML to identify the optimal algorithm and hyperparameters, achieving an accuracy of approximately 87.37%. Following this, we designed the integration of the trained classification model with existing QR code readers to implement a service capable of countering Qshing attacks. In conclusion, our findings confirm that deriving an optimized algorithm for classifying malicious URLs in QR codes and integrating it with existing QR code readers presents a viable solution to combat Qshing attacks.

• Journal of Korea Multimedia Society
• /
• v.13 no.3
• /
• pp.359-368
• /
• 2010

Algorithm visualization is useful for program testing, debugging and evaluating, as well as visual aids in education. When teaching algorithms and data structures, showing exact behaviors by graphics or animation is more suitable than just explaining them. Current systems for algorithm animation are limited to a couple of specific applications and need a special environment. In the proposed system, programmer writes source program in C and animator embeds visualization scripts in the appropriate location of the program. Then user can get an animation code in form of SVG and see a graphical representation on the web browser. Generated SVG animation code is platform independent and can also interact with other XML applications or HTML.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.8A
• /
• pp.768-776
• /
• 2010

In this paper, we consider a linear precoding for the effective spectrum sharing in multiple-input multiple-output (MIMO) cognitive radio system where a secondary user coexists with primary users. The secondary user employs the orthogonal space time block coding (OSTBC) at the transmitter. Assuming a flat fading channel and a maximum-likelihood receiver, the optimum precoder forces transmission referred to as eigen-beamforming. In this paper, to eliminate the interference, ZF criterion based eigen-beamforming is not only used but also the precoding weight is chosen to cancel the remaining interference. This weight is computed by vector's likelihood. Simulation results show stronger interference suppression capability, better SER performance, and higher capacity than the algorithm in [4].

• Journal of the Institute of Electronics Engineers of Korea TC
• /
• v.49 no.7
• /
• pp.48-55
• /
• 2012

Because of increasing complexity in maximum-likelihood (ML) decoding of four of higher antenna scenario, Partial Interference Cancellation (PIC) group decoding could be the perfect solution to reduce the decoding complexity occurs in ML decoding. In this paper, we separate the symbols the users in the layered basis and find the equivalent channel matrix. Based on the equivalent channel matrix we provide the grouping scheme. In our paper, we construct a block wise transmission technique which will achieve the desired code rate and reduce the complexity and provide less transmission time. Finally we show the different grouping performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.12
• /
• pp.2525-2532
• /
• 2009

In this paper, a fast sphere decoder is proposed for the joint detection of phase-shift keying (PSK) signals in uncoded Vertical Bell Laboratories Layered Space Time (V-BLAST) systems. The proposed decoder, PSD, consists of preprocessing stage and search stage. The search stage of PSD relies on the depth-first branch-and-bound (BB) algorithm with "best-first" orders stored in lookup tables. Simulation results show that the PSD is able to provide the system with the maximum likelihood (ML) performance at low complexity.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2005.11a
• /
• pp.488-496
• /
• 2005

현재의 웹이 HTML을 바탕으로 인간 사용자와의 인터페이스에 초점을 맞추고 있는데 비하여, 차세대 웹은 XML 및 XML 기반 각종 표준들을 바탕으로 소프트웨어 에이전트와의 인터페이스에 초점을 맞추어 나가고 있다. 차세대 웹에서 소프트웨어 에이전트의 두뇌 역할을 수행하기 위하여 추론엔진은 차세대 웹의 표준 언어인 시맨틱 웹(Semantic Web)을 충실히 이해할 수 있어야 한다. 이를 위한 기초 작업의 일환으로 OWL(Web Ontology Language)과 RuleML(Rule Markup Language)이 W3C에 제안된 바 있다. 본 연구에서는 SWRL을 규칙 표현 방법으로 사용하고, OWL을 사실 표현 방법으로 사용하는 역방향 추론엔진인 SMART-B(SeMantic web Agent Reasoning Tools - Backward chaining inference engine)을 개발하고자 한다. 이를 위하여 SWRL 기반 역방향 추론을 위한 요구 기능을 분석하고, 기존 역방향 추론 알고리즘에 차세대 시맨틱 웹을 요구 기능을 반영한 역방향 추론 알고리즘을 설계하였다. 또한, 유비쿼터스 환경에서의 각종 플랫폼의 독립성과 이식성을 확보하고 기기 간의 성능 차이를 극복할 수 있도록 사실 베이스 및 규칙 베이스의 관리도구와 역방향 추론 엔진 등을 Java 프로그래밍 언어를 이용하여 단위 컴포넌트의 형태로 개발 중에 있다.

• Journal of IKEEE
• /
• v.25 no.1
• /
• pp.25-31
• /
• 2021

The conventional credit card number recognition system generally needs a card to be placed in a designated location before its processing, which is not an ideal user experience especially for people with visual impairment. To improve the user experience, this paper proposes a novel algorithm that can automatically detect the location of a credit card number based on the fact that a group of sixteen digits has a fixed aspect ratio. The proposed algorithm first performs morphological operations to obtain multiple candidates of the credit card number with >4:1 aspect ratio, then recognizes the card number by testing each candidate via OCR and BIN matching techniques. Implemented with OpenCV and Firebase ML, the proposed scheme achieves 77.75% accuracy in the credit card number recognition task.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.226-228
• /
• 2021

We propose a fall recognition system using the Pose Detection of Google ML kit using video data. Using the Pose detection algorithm, 33 three-dimensional feature points extracted from the body are used to recognize the fall. The algorithm that recognizes the fall by analyzing the extracted feature points uses k-NN. While passing through the normalization process in order not to be influenced in the size of the human body within the size of image and image, analyzing the relative movement of the feature points and the fall recognizes, thirteen of the thriteen test videos recognized the fall, showing an 100% success rate.