A Feature Selection-based Ensemble Method for Arrhythmia Classification

Namsrai, Erdenetuya,Munkhdalai, Tsendsuren,Li, Meijing,Shin, Jung-Hoon,Namsrai, Oyun-Erdene,Ryu, Keun Ho Korea Information Processing Society 2013 Journal of information processing systems Vol.9 No.1

In this paper, a novel method is proposed to build an ensemble of classifiers by using a feature selection schema. The feature selection schema identifies the best feature sets that affect the arrhythmia classification. Firstly, a number of feature subsets are extracted by applying the feature selection schema to the original dataset. Then classification models are built by using the each feature subset. Finally, we combine the classification models by adopting a voting approach to form a classification ensemble. The voting approach in our method involves both classification error rate and feature selection rate to calculate the score of the each classifier in the ensemble. In our method, the feature selection rate depends on the extracting order of the feature subsets. In the experiment, we applied our method to arrhythmia dataset and generated three top disjointed feature sets. We then built three classifiers based on the top-three feature subsets and formed the classifier ensemble by using the voting approach. Our method can improve the classification accuracy in high dimensional dataset. The performance of each classifier and the performance of their ensemble were higher than the performance of the classifier that was based on whole feature space of the dataset. The classification performance was improved and a more stable classification model could be constructed with the proposed approach.

A Feature Selection-based Ensemble Method for Arrhythmia Classification

Erdenetuya Namsrai,Tsendsuren Munkhdalai,Meijing Li,Jung Hoon Shin,Oyun Erdene Namsrai,Keun Ho Ryu 한국정보처리학회 2013 Journal of information processing systems Vol.9 No.1

In this paper a novel method is proposed to build an ensemble of classifiers by using a feature selection schema. The feature selection schema identifies the best feature sets that affect the arrhythmia classification. Firstly a number of feature subsets are extracted by applying the feature selection schema to the original dataset. Then classification models are built by using the each feature subset. Finally we combine the classification models by adopting a voting approach to form a classification ensemble. The voting approach in our method involves both classification error rate and feature selection rate to calculate the score of the each classifier in the ensemble. In our method the feature selection rate depends on the extracting order of the feature subsets. In the experiment we applied our method to arrhythmia dataset and generated three top disjointed feature sets. We then built three classifiers based on the top-three feature subsets and formed the classifier ensemble by using the voting approach. Our method can improve the classification accuracy in high dimensional dataset. The performance of each classifier and the performance of their ensemble were higher than the performance of the classifier that was based on whole feature space of the dataset. The classification performance was improved and a more stable classification model could be constructed with the proposed approach.

Surgical treatment of mucin-producing cholangiocarcinoma arising from intraductal papillary neoplasm of the intrahepatic bile duct

Namsrai Baterdene,Shin Hwang,Jong-Wook Lee,Min-Jae Jung,Heeji Shin,Hye Kyoung Seo,Myeong-Hwan Kim,Sung-Koo Lee 한국간담췌외과학회 2016 한국간담췌외과학회지 Vol.20 No.3

Intraductal papillary neoplasms of the bile duct (IPNB) leads to malignant transformation and mucin production. Herein, we presented two cases of mucin-producing IPNB with obstructive jaundice who underwent resection of the intrahepatic lesions and bypass hepaticojejunostomy. The first case was a 69 year-old male patient with 5-year follow up for gallstone disease. Imaging studies showed mucin-secreting IPNB mainly in the hepatic segment III bile duct (B3) and multiple intrahepatic duct stones for which, segment III resection, intrahepatic stone removal, end-to-side choledochojejunostomy and B3 hepaticojejunostomy were conducted. The second case was a 74 year-old female patient with 11-year follow up for gallstone disease. Imaging studies showed mucin-producing IPNB with dilatation of the segment IV duct (B4) and mural nodules for which, segment IV resection, partial resection of the diaphragm and central hepaticojejunostomy were conducted. Both patients recovered uneventfully from surgery. These cases highlight that in patients with IPNB, abundant production of highly viscous mucin inducing obstructive jaundice may be associated with malignant transformation.

Characterizing Pseudobase and Predicting RNA Secondary Structure with Simple H-Type Pseudoknots Based on Dynamic Programming

Namsrai, O.-E.,Ryu, K. H. SPRINGER-VERLAG 2007 Lecture Notes in Computer Science Vol.4632 No.-

Influences of lake altitude, surface area and trophic state on waterfowl diversity and abundance in Mongolia

Namsrai Jargal,Zoljargal Purewdorj,Onolragchaa Ganbold,Woon Kee Paek,Kwang Guk An 한국환경생물학회 2020 한국환경생물학회 학술대회 Vol.2020 No.07

RNA Secondary Structure Prediction with Simple Pseudoknots Based on Dynamic Programming

Namsrai, O.-E.,Jung, K. S.,Kim, S.,Ryu, K. H. SPRINGER-VERLAG 2006 ICHIT 2012 Vol.4115 No.-

Investigating EFL Students’ Speaking Self-Efficacy and Self-Regulated Learning Strategies

Namsrai, Enkhchimeg 한국ESP학회 2022 ESP REVIEW Vol.4 No.2

Learning strategies are crucial for university students learning English as a foreign language (EFL) or English for academic purposes (EAP). The employment of language learning strategies facilitates attainment and assists learners in different ways. This study investigated college students’ self-efficacy and self-regulated learning strategies in L2 acquisition. As speaking is the most difficult skill for Mongolian learners to acquire, they have reason to design strategies for speaking when learning English. 252 EFL college students in Mongolia were given the Questionnaire of English Speaking Self-Efficacy (QESSE) to estimate their self-efficacy, and the Questionnaire of English Speaking Self-Regulated Learning Strategies (QESSRLS) was administrated to evaluate their use of specific learning strategies. The results showed that the students reported overall a moderate level of self-efficacy and self-regulated learning (SRL) strategies in L2 speaking. This study discovered that the students were the most positive about student-centered methods. The dominant English classroom instruction pedagogy in Mongolia is still teacher-centered; students follow teachers’ words and commands. Students are not encouraged to develop their own learning strategies and focus on content knowledge. The current study allows EFL/EAP teachers to determine their students’ profiles, and design strategies to improve students’ L2 speaking outcomes. Moreover, it provides teachers with insights about how to organize suitable approaches for teaching EFL/EAP.

A Study on the Internal Resource Capacity, Competitive Strategy, and Performance of Korean Companies in Mongolia

Namsrai Altantuya,Eui-Burm Park,Jong-Wook Kwon 한국무역연구원 2014 무역연구 Vol.10 No.1

This study is to seek strategic measures to expand Korean companies' business operations in the Mongolian market by understanding factors of core capabilities and competitive strategy and analyzing the effects of such factors on their financial and strategic performance. For these purposes of empirical study, firstly the authors analyzed frequencies to find out the general characteristics of respondents and basic statistics for each question by using a statistical program SPSS(Statistical Package for the Social Science) PC +12.0.Then, factors were analyzed by applying principal component analysis in order to test validity. Finally, the Cronbach Alpha coefficient for reliability on internal consistency was calculated to verify credibility by regression analysis to test the hypotheses. The factors of companies' core capabilities were classified into capabilities of financial accounting, marketing, product service, and managers. Then, competitive strategy was classified into four different strategies such as overall cost, leadership, differentiation, and focus.

다목적 저수지 유입 영양염류, Chl-a 상태에 대한 장마기간의 장기적 및 계절 영향 평가

남스라이자갈 ( Namsrai Jargal ),우스만아티크 ( Usman Atique ),마문MD ( Md Mamun ),블랜디나제네스카코레 ( Blandina Genes Kakore ),안광국 ( Kwang-guk An ) 한국환경생태학회 2021 한국환경생태학회 학술대회지 Vol.2021 No.2

Due to rapid eutrophication, sustainable water quality management and supply are essential in drinking water sources and aquatic biota in large reservoirs. We evaluated the potentially crucial factors influencing the algal chlorophyll (Chl-a), nutrients, and the links between the rainfall and other vital elements in a large multipurpose reservoir (Yongdam Reservoir) during 2013-2019. We developed the empirical models on algal Chl-a, total phosphorus (TP), total nitrogen (TN), and TN:TP’s ambient ratios considering the maneuvering influence of Asian monsoons. The intensive rainfall during the monsoon months strongly impacted the nutrient regime and other vital factors. The seasonal patterns of algal Chl-a varied in response to the nutrient contents (TN, TP), suspended solids, and ambient N:P ratios along the longitudinal gradient. The conditional plot analysis, empirical modeling, and observations supported an overall P-limitation scenario, as was evident from the magnitude of N:P ratios (R2 = 0.36, F = 24.9, p < 0.001). Furthermore, the reservoir’s trophic status alluded to the larger particles and blue-green algae during the monsoon and postmonsoon months. The correlation analysis, Mann- Kendall trend test, and principal component analysis illustrated compelling links between CHL-a, TP, and rainfall regime. The outcomes suggested the reservoir was primarily controlled by phosphorus limitation, with an increasing CHL-a tendency along with nitrogen dilution. However, a slight decline in phosphorus was also detected. The Yongdam Reservoir is under the threat of recurrent eutrophication events that could jeopardize this vital drinking water facility due to increasing agricultural and anthropic activities.

인공저수지의 영양염류-N:P비율 상태의 계절적 변동과 토지이용도에 따른 영양염류-Chl-a 상관관계

남스라이자갈 ( Namsrai Jargal ),이상재 ( Sang-jae Lee ),안광 ( Kwang-guk An ) 한국환경생태학회 2021 한국환경생태학회 학술대회지 Vol.2021 No.2

Water quality degradation and eutrophication are strongly related to land-use practices around freshwater systems. Here, we investigated an impact of predominant land use on seasonal dynamics of organic matters, nutrients, the ratios of N:P and CHL-a:TP, and algal biomass in Korean reservoirs (n=9) based on long-term monitoring of 2006-2020. It was also examined relationships between nutrients and the algal biomass in reservoirs altered by predominant land-use pattern. The algal biomass was assessed by chlorophyll-a (CHL-a) concentration. Compared to forest-dominated reservoirs (FDRs), agricultural dominant reservoirs (ADRs) and estuarine reservoirs (ERs) manifested relatively high concentrations of suspended solids (TSS), organic matters, phosphorus contents, and the algal biomass, especially the ERs. According to total phosphorus (TP) averages in each type of reservoir, a risk of blue-green algal dominance was high in the ERs (~80%), moderate in the ADRs (~40%), and low in the FDRs (< 10%) for our study. The intense rainfall periods during July - August were a primary environmental factor for seasonal increases of nutrients into the reservoirs. Particularly, the escalation in TP was observably in line with rainfall runoff into the ADRs. But total nitrogen (TN) in the ERs visibly reduced during the intense rainfall months due to ionic dilution by rainfall-water. The seasonal dynamic of predicted CHL-a differed through the three types of reservoirs, showing a seasonal distinct pattern of algal bloom. Correspondingly, the seasonal dynamic in CHL-a:TP ratio varied by the land-use patterns. It was relatively higher during winter in the ERs, and yet showed higher values during summer and fall in the ADRs. The seasonal reduction in N:P ratio was strongly connected to the increase of TP, as well as an intensity of anthropogenic activity. Furthermore, the empirical regression model showed that the algal biomass was significant positive relation with both TN (R2=0.44, p<0.001) and TP (R2=0.78, p<0.001), whereas showing negatively associated to N:P ratio (R2=0.49, p<0.001). However, the empirical nutrients-CHL-a models in reservoirs altered by the predominant land-use patterns. TP was most important predictor of CHL-a level in both the FDRs (R2=0.60, p<0.001) and the ADRs (R2=0.87, p<0.001), supporting P-limitation for algal growth. But the power of TP-CHL-a relation observably reduced for the ERs (R2=0.25, p<0.05) compared to the FDRs and the ADRs. This result was mainly driven by hypereutrophic nutrient conditions, and thus light availability was likely a more important predictor of CHL-a level in the ERs.