• International Journal of Computer Science & Network Security
• /
• v.24 no.7
• /
• pp.11-23
• /
• 2024

Triage is a practice of accurately prioritizing patients in emergency department (ED) based on their medical condition to provide them with proper treatment service. The variation in triage assessment among medical staff can cause mis-triage which affect the patients negatively. Developing ED triage system based on machine learning (ML) techniques can lead to accurate and efficient triage outcomes. This study aspires to develop a triage system using machine learning techniques to predict ED triage levels using patients' information. We conducted a retrospective study using Security Forces Hospital ED data, from 2021 through 2023 during Hajj period in Saudia Arabi. Using demographics, vital signs, and chief complaints as predictors, two machine learning models were investigated, naming gradient boosted decision tree (XGB) and deep neural network (DNN). The models were trained to predict ED triage levels and their predictive performance was evaluated using area under the receiver operating characteristic curve (AUC) and confusion matrix. A total of 11,584 ED visits were collected and used in this study. XGB and DNN models exhibit high abilities in the predicting performance with AUC-ROC scores 0.85 and 0.82, respectively. Compared to the traditional approach, our proposed system demonstrated better performance and can be implemented in real-world clinical settings. Utilizing ML applications can power the triage decision-making, clinical care, and resource utilization.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.42 no.4
• /
• pp.747-749
• /
• 2017

In this paper, we propose a novel precoder selection technique for maximum-likelihood (ML) detected spatially multiplexed multiple-input multiple-output (MIMO) systems. Previous precoder selection techniques were designed without considering UEP, however the proposed technique is designed considering multi-antenna unequal error protection (UEP). Simulations demonstrate the improved multi-antenna UEP performance by the proposed technique.

• The Korean Journal of Ecology
• /
• v.28 no.6
• /
• pp.405-415
• /
• 2005

This paper considers the advantage of Machine Learning (ML) implemented to freshwater ecosystem research. Currently, many studies have been carried out to find the patterns of environmental impact on dynamics of communities in aquatic ecosystems. Ecological models popularly adapted by many researchers have been a means of information processing in dealing with dynamics in various ecosystems. The up-to-date trend in ecological modelling partially turns to the application of ML to explain specific ecological events in complex ecosystems and to overcome the necessity of complicated data manipulation. This paper briefly introduces ML techniques applied to freshwater ecosystems in Korea. The manuscript provides promising information for the ecologists who utilize ML for elucidating complex ecological patterns and undertaking modelling of spatial and temporal dynamics of communities.

• Advances in Traditional Medicine
• /
• v.8 no.2
• /
• pp.171-177
• /
• 2008

Epigallocatechin 3-gallate (EGCG), one of the major catechins of tea, was isolated from the decaffeinated, crude methanolic extract of Indian green tea (Camellia sinensis L. O. Kuntze) using chromatographic techniques. EGCG was then screened for antidiarrhoeal activity against 30 strains (clinical isolates) of V. cholerae, which is a well known Gram negative bacillus functioning as the pathogen of cholera. V. cholerae strains like V. cholerae 69, 71, 83, 214, 978, 1021, 1315, 1347, 1348, 569B and ATCC 14033 were inhibited by EGCG at a concentration of $25\;{\mu}g/ml$ whereas V. cholerae 10, 522, 976 were even more sensitive, being inhibited at $10\;{\mu}g/ml$ level. However, V. cholerae DN 16, DN 26, 30, 42, 56, 58, 113, 117, 564, 593, 972 and ATCC 14035 were inhibited at $50\;{\mu}g/ml$ level of EGCG. Only four strains were inhibited at $100\;{\mu}g/ml$. In this study the isolated compound was found to be bacteriostatic in its mechanism of action. In the in vivo experiment using the rabbit ileal loop model two different dosages of EGCG ($500\;{\mu}g/ml$ and $1,000\;{\mu}g/ml$) were able to protect the animals when they were challenged with V. cholerae 569B in the ileum.

• Journal of Multimedia Information System
• /
• v.6 no.4
• /
• pp.165-172
• /
• 2019

Machine-learning techniques have been actively employed to information security in recent years. Traditional rule-based security solutions are vulnerable to advanced attacks due to unpredictable behaviors and unknown vulnerabilities. By employing ML techniques, we are able to develop intrusion detection systems (IDS) based on anomaly detection instead of misuse detection. Moreover, threshold issues in anomaly detection can also be resolved through machine-learning. There are very few datasets for network intrusion detection compared to datasets for malicious code. KDD CUP 99 (KDD) is the most widely used dataset for the evaluation of IDS. Numerous studies on ML-based IDS have been using KDD or the upgraded versions of KDD. In this work, we develop an IDS model using CSE-CIC-IDS 2018, a dataset containing the most up-to-date common network attacks. We employ deep-learning techniques and develop a convolutional neural network (CNN) model for CSE-CIC-IDS 2018. We then evaluate its performance comparing with a recurrent neural network (RNN) model. Our experimental results show that the performance of our CNN model is higher than that of the RNN model when applied to CSE-CIC-IDS 2018 dataset. Furthermore, we suggest a way of improving the performance of our model.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.4C
• /
• pp.462-467
• /
• 2007

This paper proposes Minimum Distance Searching Algorithm (MDSA) which reduces the computational complexity (CC) of the ML, the kind of Spatial Multiplexing (SM) MIMO system. The MDSA searchs candidate symbols with a starting symbol, which is called reference bits. We used the linear receiver of MIMO techniques to find a starting symbol. The MDSA searchs the shortest path to a transmitted symbol using reference bits and Minimum Distance(MD) concept. The CC of MDSA is reduced to the 0.21% to the ML as the transmit antennas is 4 in 16QAM. The simulation result shows the BER of MDSA is nearly same to the BER of ML as the transmit antennas is 2 and the receive antennas is 3 in 16QAM and slightly degraded to the BER of ML as the transmit antennas is 4 and the receive antennas is 6 in QPSK.

• The Journal of the Korea Contents Association
• /
• v.15 no.5
• /
• pp.436-443
• /
• 2015

The purpose of this study was to develop a self-diagnostic linearity quality control techniques of computed tomography (CT) by using measured CT number values from the various concentrations of iodine contrast media (CM) is diluted with distilled water under each condition of the tube voltage. The equipment was used for four-channel MDCT, the iodine concentration were using 300 mgI/ml, 350 mgI/ml, 370 mgI/ml and 400mgI/ml. Dilution of CM in distilled water was increased by each 5% until the maximum CT number values were measured. We applied the tube voltages for 90 kVp, 120 kVp, 140 kVp. As a result, we was obtained to the nearest linearity as 0.993 of correlation coefficient between the iodinated CM from 5% to 25% in 400 mgI/ml and the CT number values by 90 kVp. In conclusion, the proposed self-diagnostic linearity quality assurance technique by using iodine CM can be utilized to replace the AAPM CT performance phantom.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.6
• /
• pp.1039-1044
• /
• 2008

Owing to rapid advancements of the mobile computing technologies and the performance of mobile device recently, the data synchronization techniques between severs and mobile clients are getting more and more important. OMA also proposes and recommends standard synchronization methods to use SyncML. However, the feasible data in the method are limited to normal document data, scheduler data, etc. This paper a standard framework based on SyncML. We call it SCGFG. The SCGFG is able to synchronize not only the above data but also GIS data which is very useful in mobile applications. It applies GML, international GIS standard, to the synchronization. By means of using XML, it is also able to resolve the serious problem that is the increase of data volume occurred by SyncML and GML. efficiently. It is highly expected to be useful in the smart synchronization of GIS data among several servers and mobile clients.

• The Journal of Information Systems
• /
• v.26 no.4
• /
• pp.17-37
• /
• 2017

Purpose Customer Loyalty is the most important factor of customer relationship management (CRM). Especially in retailing industry, where customers have many options of where to spend their money. Classifying loyal customers through customers' data can help retailing companies build more efficient marketing strategies and gain competitive advantages. This study aims to construct classification models of distinguishing the loyal customers within a Korean retailing company using data mining techniques with R language. Design/methodology/approach In order to classify retailing customers, we used combination of support vector machines (SVMs) and other classification algorithms of machine learning (ML) with the support of recursive feature elimination (RFE). In particular, we first clean the dataset to remove outlier and impute the missing value. Then we used a RFE framework for electing most significant predictors. Finally, we construct models with classification algorithms, tune the best parameters and compare the performances among them. Findings The results reveal that ML classification techniques can work well with CRM data in Korean retailing industry. Moreover, customer loyalty is impacted by not only unique factor such as net promoter score but also other purchase habits such as expensive goods preferring or multi-branch visiting and so on. We also prove that with retailing customer's dataset the model constructed by SVMs algorithm has given better performance than others. We expect that the models in this study can be used by other retailing companies to classify their customers, then they can focus on giving services to these potential vip group. We also hope that the results of this ML algorithm using R language could be useful to other researchers for selecting appropriate ML algorithms.

• Journal of Embryo Transfer
• /
• v.25 no.4
• /
• pp.291-295
• /
• 2010

Four estrus-induced Himalayan tahrs (Hemitragus jemlahicus) were inseminated with frozen-thawed semen by laparoscopic or transcervical insemination techniques with no regard to the site of ovulation in non-breeding season. In June and July, 2009, estrus was synchronized by Eazi-Breed $CIDR^{(R)}$ (Controlled internal drug release; Pfizer Animal Health, New Zealand) insertion for 16 days and PG 600 (PMSG 400IU, hCG 200 IU; Intervet, Netherlands) injection (IM) a day before removing $CIDR^{(R)}$. Forty eight hours later, laparoscopic or transcervical insemination was done to each of two tahrs under anesthetic condition inducted by ketamine (1.5 mg/kg) and medetomidine (0.09 mg/kg). For examination of estradiol and progesterone, blood was collected right before $CIDR^{(R)}$ insertion, PG 600 injection, $CIDR^{(R)}$ removal and insemination. Estradiol levels of four tahrs (No. 1, 2, 3, 4) before $CIDR^{(R)}$ insertion and insemination were 13.3, 8.8, 14.3, 12 pg/ml and 23.5, 25.5, 21.1, 11.5 pg/ml, respectively. Progesterone levels of four tahrs (No. 1, 2, 3, 4) before $CIDR^{(R)}$ insertion and insemination were 1.8, 0.05, 0.63, 0.61 ng/ml and 1.03, 0.37, 1.48, 2.12 ng/ml. Except for No. 4 tahr, cervices showed cervical mucus and opened enough to penetrate with embryo transfer gun sheet usually used for cows. Therefore, No.4 was laparoscopically inseminated together with No. 1. In conclusion, none of four Himalayan tahrs was pregnant. However, we proved that estrus could be induced by CIDR and PG 600 injection in non-breeding season, and laparoscopic or transcervical insemination with frozen-thawed semen could be one of assisted reproductive techniques in Himalayan Tahr.