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.

향상된 다이내믹 프로그래밍 기반 RNA 이차구조 예측

( Oyun-erdene Namsrai ),정광수 ( Kwang Su Jung ),김선신 ( Sunshin Kim ),류근호 ( Keun Ho Ryu ) 한국정보처리학회 2005 한국정보처리학회 학술대회논문집 Vol.12 No.2

A ribonucleic acid (RNA) is one of the two types of nucleic acids found in living organisms. An RNA molecule represents a long chain of monomers called nucleotides. The sequence of nucleotides of an RNA molecule constitutes its primary structure, and the pattern of pairing between nucleotides determines the secondary structure of an RNA. Non-coding RNA genes produce transcripts that exert their function without ever producing proteins. Predicting the secondary structure of non-coding RNAs is very important for understanding their functions. We focus on Nussinov’s algorithm as useful techniques for predicting RNA secondary structures. We introduce a new traceback matrix and scoring table to improve above algorithm. And the improved prediction algorithm provides better levels of performance than the originals.

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.

Testosterone Deficiency with Erectile Dysfunction in Mongolian Men

Nansalmaa Naidan,Oyun-Erdene Rivaad,Namsrai Muukhai,Munkhtsetseg Janlav 대한남성과학회 2013 The World Journal of Men's Health Vol.31 No.2

Purpose: To detect the testosterone deficiency syndrome in Mongolian men over 40 years old with erectile dysfunction (ED).Materials and Methods: Total of 309 males over 40 years of age who received medical care at the ADAM Urology and Andrology Clinic from 2010 to 2011 were included in this study. An approval from the Ethics Committee of the Ministry of Health of Mongolia was obtained, and each study participant signed a consent form at the beginning of the study. The participants were assigned to either an ED group or a control group, depending on the results of the international index of erectile function (IIEF)-5 questionnaire. The ED group was further divided into three groups (moderate, severe, and very severe) based on the level of ED. The total testosterone (TT) levels were determined in the blood serum using a competitive enzyme-linked immunesorbent assay (ELISA) analytical system UBI MagiwelTM Testosterone Quantitative test, and free testosterone (FT) calculated as described by the Vermeulen calculation. Test samples were collected between 8:00 and 11:00 am in the mornings and testosterone deficiency syndrome was diagnosed based on the International Society for the Study of the Aging Male guidelines, particularly, if TT was ≤3.46 ng/ml or free testosterone FT was ≤0.072 ng/ml.Results: ED of moderate, severe, and very severe levels was diagnosed in 199 (64.41%) out of 309 participants. There was an inverse relationship between the main IIEF-5 score and age (r=−0.380, p<0.01). The average TT was 5.75±2.316 ng/ml and FT was 0.091±0.0084 ng/ml. Compared to the ED group, the control group had a higher TT level: 5.6440±1.177 ng/ml and 5.812±2.316 ng/ml, respectively. In the control group, the FT level was 0.061±0.0084 ng/ml, whereas it was 0.041±0.0076 ng/ml in the ED group.Conclusions: Our study showed that most of the aging males who came to the clinic had moderate to very severe ED (64.55%). The levels of TT (5.644±1.177 ng/ml) and FT (0.041±0.0036 ng/ml) were significantly lower in ED patients (p<0.05). The testosterone deficiency syndrome was detected in 24.27% of the ED group.

Self-training in significance space of support vectors for imbalanced biomedical event data

Munkhdalai, Tsendsuren,Namsrai, Oyun-Erdene,Ryu, Keun Ho BioMed Central 2015 BMC bioinformatics Vol.16 No.suppl7

<P><B>Background</B></P><P>Pairwise relationships extracted from biomedical literature are insufficient in formulating biomolecular interactions. Extraction of complex relations (namely, biomedical events) has become the main focus of the text-mining community. However, there are two critical issues that are seldom dealt with by existing systems. First, an annotated corpus for training a prediction model is highly imbalanced. Second, supervised models trained on only a single annotated corpus can limit system performance. Fortunately, there is a large pool of unlabeled data containing much of the domain background that one can exploit.</P><P><B>Results</B></P><P>In this study, we develop a new semi-supervised learning method to address the issues outlined above. The proposed algorithm efficiently exploits the unlabeled data to leverage system performance. We furthermore extend our algorithm to a two-phase learning framework. The first phase balances the training data for initial model induction. The second phase incorporates domain knowledge into the event extraction model. The effectiveness of our method is evaluated on the Genia event extraction corpus and a PubMed document pool. Our method can identify a small subset of the majority class, which is sufficient for building a well-generalized prediction model. It outperforms the traditional self-training algorithm in terms of f-measure. Our model, based on the training data and the unlabeled data pool, achieves comparable performance to the state-of-the-art systems that are trained on a larger annotated set consisting of training and evaluation data.</P>

Correlation based clustering of the Mongolian stock exchange for portfolio management

Enkhtuul Bukhsuren,Batnyam Battulga,Oyun-Erdene Namsrai ASCONS 2018 INTERNATIONAL JOURNAL OF EMERGING MULTIDISCIPLINAR Vol.2 No.2

This research paper analyzed the stock prices of Mongolia Stock Exchange TOP 20 index from 1 January 2012 to 31 December 2016, and estimated the return rate of these stocks. A hierarchical clustering was created from the correlation matrix of stock returns. From this clustering five stocks were selected for the portfolio construction and the rate of return was maximized using the Modern Portfolio Theory developed by Harry Markowitz. The weight of each stock in the portfolio was calculated for maximization the return rate of the portfolio, and 12 portfolios were constructed from these five stocks. An investor can select appropriate one of these portfolios in accordance with his or her risk and return characteristics.

Decision Support System for Mongolian Portfolio Selection

Enkhtuul Bukhsuren,Uyanga Sambuu,Oyun-Erdene Namsrai,Batnasan Namsrai,류근호 한국정보처리학회 2022 Journal of information processing systems Vol.18 No.5

Investors aim to increase their profitability by investing in the stock market. An adroit strategy for minimizingrelated risk lies through diversifying portfolio operationalization. In this paper, we propose a six-step stocksportfolio selection model. This model is based on data mining clustering techniques that reflect the ensuingimpact of the political, economic, legal, and corporate governance in Mongolia. As a dataset, we have selectedstock exchange trading price, financial statements, and operational reports of top-20 highly capitalized stocksthat were traded at the Mongolian Stock Exchange from 2013 to 2017. In order to cluster the stock returns andrisks, we have used k-means clustering techniques. We have combined both k-means clustering withMarkowitz's portfolio theory to create an optimal and efficient portfolio. We constructed an efficient frontier,creating 15 portfolios, and computed the weight of stocks in each portfolio. From these portfolio options, theinvestor is given a choice to choose any one option.

An Active Co-Training Algorithm for Biomedical Named-Entity Recognition

Munkhdalai, Tsendsuren,Li, Meijing,Yun, Unil,Namsrai, Oyun-Erdene,Ryu, Keun Ho Korea Information Processing Society 2012 Journal of information processing systems Vol.8 No.4

Exploiting unlabeled text data with a relatively small labeled corpus has been an active and challenging research topic in text mining, due to the recent growth of the amount of biomedical literature. Biomedical named-entity recognition is an essential prerequisite task before effective text mining of biomedical literature can begin. This paper proposes an Active Co-Training (ACT) algorithm for biomedical named-entity recognition. ACT is a semi-supervised learning method in which two classifiers based on two different feature sets iteratively learn from informative examples that have been queried from the unlabeled data. We design a new classification problem to measure the informativeness of an example in unlabeled data. In this classification problem, the examples are classified based on a joint view of a feature set to be informative/non-informative to both classifiers. To form the training data for the classification problem, we adopt a query-by-committee method. Therefore, in the ACT, both classifiers are considered to be one committee, which is used on the labeled data to give the informativeness label to each example. The ACT method outperforms the traditional co-training algorithm in terms of f-measure as well as the number of training iterations performed to build a good classification model. The proposed method tends to efficiently exploit a large amount of unlabeled data by selecting a small number of examples having not only useful information but also a comprehensive pattern.

An Active Co-Training Algorithm for Biomedical Named-Entity Recognition

Tsendsuren Munkhdalai,Mei Jing Li,Un Il Yun,Oyun Erdene Namsrai,Keun Ho Ryu 한국정보처리학회 2012 Journal of information processing systems Vol.8 No.4

A Dependency Graph-Based Keyphrase Extraction Method Using Anti-patterns

Batsuren, Khuyagbaatar,Batbaatar, Erdenebileg,Munkhdalai, Tsendsuren,Li, Meijing,Namsrai, Oyun-Erdene,Ryu, Keun Ho Korea Information Processing Society 2018 Journal of information processing systems Vol.14 No.5

Keyphrase extraction is one of fundamental natural language processing (NLP) tools to improve many text-mining applications such as document summarization and clustering. In this paper, we propose to use two novel techniques on the top of the state-of-the-art keyphrase extraction methods. First is the anti-patterns that aim to recognize non-keyphrase candidates. The state-of-the-art methods often used the rich feature set to identify keyphrases while those rich feature set cover only some of all keyphrases because keyphrases share very few similar patterns and stylistic features while non-keyphrase candidates often share many similar patterns and stylistic features. Second one is to use the dependency graph instead of the word co-occurrence graph that could not connect two words that are syntactically related and placed far from each other in a sentence while the dependency graph can do so. In experiments, we have compared the performances with different settings of the graphs (co-occurrence and dependency), and with the existing method results. Finally, we discovered that the combination method of dependency graph and anti-patterns outperform the state-of-the-art performances.