• Proceedings of the Korea Information Processing Society Conference
• /
• 2017.04a
• /
• pp.181-182
• /
• 2017

관성 센서 기반 행위인지는 스마트폰과 웨어러블 밴드 등의 출현으로 보다 간편한 방법으로 행위인지가 가능해졌다. 현재 대부분의 행위인지 서비스나 연구들은 단일 행위의 결론만을 도출하고 있으나, 이러한 방식은 한 행위에서 한 가지 동작밖에 취할 수 없는 경우에는 문제가 없지만 두 가지 이상의 동작이 합쳐진 경우에 어떤 행위를 최종 결론으로 도출해야 하는지에 대한 문제점을 내포한다. 따라서 본 논문에서는 이러한 문제점을 해결하기 위해 세 개의 센서 기기 (스마트폰, 스마트워치, 웨어러블 센서)를 이용한 멀티 레이블 행위인지를 제안한다. 스마트폰은 신체 전반적인 움직임 탐지를 위하여 소지위치가 정해지지 않은 비고정식 센서의 보조적인 역할을 수행한다. 스마트워치는 사용자가 주로 사용하는 손의 손목, 그리고 웨어러블 센서는 사용자의 허벅지에 부착되어 각각 상하체의 움직임을 파악한다. 이후 각 기기에서 도출된 결론에 Majority Weighted Voting 기법을 적용하여 단일 혹은 멀티 레이블의 최종 행위를 도출한다.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 2001.10a
• /
• pp.106-110
• /
• 2001

The Analytic Hierarchy Process (AHP) is well suited to group decision making and offers numerous benefits as a synthesizing mechanism in group decisions. To date, the majority of AHP applications have been in group settings. In general, aggregation methods employed in AHP can be largely classified into two methods: geometric mean method and (weighted) arithmetic mean method. In a situation where there do not exist clear guidelines for selection between them, two methods do not always guarantee the same group decision result. Thus we suggest a simulation approach for building group consensus as a complementary tool, even when just group judgments are required. Without any efforts to make point estimates from individual diverse preference judgments, a simulation approach suggests the process how the individual preference judgments are aggregated into consensus, displaying possible disagreements as is natural in group members' different viewpoints.

• Journal of Information Processing Systems
• /
• v.16 no.2
• /
• pp.360-376
• /
• 2020

Many heuristic attribute reduction algorithms have been proposed to find a single reduct that functions as the entire set of original attributes without loss of classification capability; however, the proposed reducts are not always perfect for these multiclass datasets. In this study, based on a probabilistic rough set model, we propose the class-oriented attribute reduction (COAR) algorithm, which separately finds a reduct for each target class. Thus, there is a strong dependence between a reduct and its target class. Consequently, we propose a type of ensemble constructed on a group of classifiers based on class-oriented reducts with a customized weighted majority voting strategy. We evaluated the performance of our proposed algorithm based on five real multiclass datasets. Experimental results confirm the superiority of the proposed method in terms of four general evaluation metrics.

• Journal of the military operations research society of Korea
• /
• v.28 no.1
• /
• pp.83-96
• /
• 2002

The Analytic Hierarchy Process (AHP) is well suited to group decision making and offers numerous benefits as a synthesizing mechanism in group decisions. To date, the majority of AHP applications have been in group settings. One reason for this may be that groups often have an advantage over individual when there exists a significant difference between the importance of quality in the decision and the importance of time in which to obtain the decision. Another reason may be the best alternative is selected by comparing alternative solutions, testing against selected criteria, a task ideally suited for AHP. In general, aggregation methods employed in group AHP can be largely classified into two methods: geometric mean method and (weighted) arithmetic mean method. In a situation where there do not exist clear guidelines for selection between them, two methods do not always guarantee the same group decision result. We propose a simulation approach for building group consensus without efforts to make point estimates from individual diverse preference judgments, displaying possible disagreements as is natural in group members'different viewpoints.

• Journal of Information Processing Systems
• /
• v.17 no.1
• /
• pp.89-106
• /
• 2021

Different samples can have different effects on learning support vector machine (SVM) classifiers. To rebalance an imbalanced dataset, it is reasonable to reduce non-informative samples and add informative samples for learning classifiers. Safe sample screening can identify a part of non-informative samples and retain informative samples. This study developed a resampling algorithm for Rebalancing imbalanced data using Safe Sample Screening (Re-SSS), which is composed of selecting Informative Samples (Re-SSS-IS) and rebalancing via a Weighted SMOTE (Re-SSS-WSMOTE). The Re-SSS-IS selects informative samples from the majority class, and determines a suitable regularization parameter for SVM, while the Re-SSS-WSMOTE generates informative minority samples. Both Re-SSS-IS and Re-SSS-WSMOTE are based on safe sampling screening. The experimental results show that Re-SSS can effectively improve the classification performance of imbalanced classification problems.

• Communications for Statistical Applications and Methods
• /
• v.30 no.2
• /
• pp.191-213
• /
• 2023

Unit distributions are frequently used in probability theory and statistics to depict meaningful variables having values between zero and one. Using convenient transformation, the unit inverse exponentiated weibull (UIEW) distribution, which is equally useful for modelling data on the unit interval, is proposed in this study. Quantile function, moments, incomplete moments, uncertainty measures, stochastic ordering, and stress-strength reliability are among the statistical properties provided for this distribution. To estimate the parameters associated to the recommended distribution, well-known estimation techniques including maximum likelihood, maximum product of spacings, least squares, weighted least squares, Cramer von Mises, Anderson-Darling, and Bayesian are utilised. Using simulated data, we compare how well the various estimators perform. According to the simulated outputs, the maximum product of spacing estimates has lower values of accuracy measures than alternative estimates in majority of situations. For two real datasets, the proposed model outperforms the beta, Kumaraswamy, unit Gompartz, unit Lomax and complementary unit weibull distributions based on various comparative indicators.

• Journal of Korean Neurosurgical Society
• /
• v.39 no.3
• /
• pp.198-203
• /
• 2006

Objective : To evaluate the natural histories and growth rates of meningiomas, the authors perform this retrospective observational study and attempt to identify those factors predicting tumor growth. Methods : Between 1993 and 2004, a total of 83 patients were diagnosed by computed tomography[CT] scans or magnetic resonance[MR] imaging as having an intracranial meningioma, and were treated by observation only using regular clinical and radiological examinations. Twenty-six of these 83 patients, with available data were included in this study. Follow up periods ranged from 9 to 137 months [mean, 55.6 mo.; median, 60 mo.]. The tumor volumes, absolute growth rates, and tumor doubling times were calculated. Results : Patient age and sex distributions were comparable to those of other studies, but exceptionally 16 meningiomas [62%] were located at the skull base in the present study. During follow-up monitoring, the majority of meningiomas grew, though 77% showed low absolute annual growth rates [$<1cm^3/yr$]. The tumor doubling times ranged from 2.87 to 201.72 years [mean, 42.91 yr]. Based on Imaging analysis, peritumoral edema and the absence of calcification were probable factors predicting tumor growth. Tumor-related symptoms seemed to be slightly related to tumor growth. Other factors, e.g., gender, age, tumor location, and T2-weighted signal Intensities on MR imaging, were not significantly related to tumor growth. Conclusion : This study shows that the majority of meningiomas are slow growing. However, variations in tumor growth are unexplained, thus individualized optimal treatment strategies should be provided in each meningioma.

• Korean Journal of Head & Neck Oncology
• /
• v.15 no.2
• /
• pp.166-171
• /
• 1999

Background and Objectives: The aim of our study was to describe the appearance of recurrent and residual lesions in the head and neck tumors, and to evaluate the usefullness of CT and MRI. Materials and Methods: CT(n=42) and MRI(n=4) of 44 patients with recurrent head and neck tumors were reviewed retrospectively. Primary tumor sites were larynx/hypopharynx in 15, oral cavity/floor of mouth in 13, base of tongue/tonsil in 5, nasopharynx in 4, palate in 2, and others in 5 patients. Therapeutic modalities included sugery and radiotherapy in 23, radiotherapy in 11, surgery in 5, chemotherapy and radiotherapy in 4, and chemotherapy in 1 patient. Results: The patterns of tumor recurrence were nodal recurrence(n=17), primary tumor bed recurrence combined with nodal recurrence(n=12), primary tumor bed recurrence(n=10) and residual primary tumors(n=5). The most common appearance of residual/recurrent primary tumor on CT was focal or diffuse heterogenous mass with or without surrounding fat or muscle infiltration(25/27). On MRI, the recurrent lesions showed intermediate signal intensity on T1 weighted image and high signal intensity on T2 weighted image with heterogenous enhancement in the most cases(n=3). 38 out of 44 nodal recurrences(86%) which had been pathologically or clinically proved were more than 1 cm in diameter or contained central low density on CT scan. Conclusion: Although CT and MRI findings of recurrent and residual tumors of the head and neck were nonspecific, in the majority the lesions manifested as a mass at primary tumor bed and/or nodal disease including contralateral side of the neck. And CT and MRI are valuable for revealing above lesions.

• Geophysics and Geophysical Exploration
• /
• v.23 no.3
• /
• pp.131-140
• /
• 2020

In training a classification model using machine learning, the acquisition of training data is a very important stage, because the amount and quality of the training data greatly influence the model performance. However, when the cost of obtaining data is so high that it is difficult to build ideal training data, the number of samples for each class may be acquired very differently, and a serious data-imbalance problem can occur. If such a problem occurs in the training data, all classes are not trained equally, and classes containing relatively few data will have significantly lower recall values. Additionally, the reliability of evaluation indices such as accuracy and precision will be reduced. Therefore, this study sought to overcome the problem of data imbalance in two stages. First, we introduced weighted accuracy and weighted precision as new evaluation indices that can take into account a data-imbalance ratio by modifying conventional measures of accuracy and precision. Next, oversampling was performed to balance weighted precision and recall among classes. We verified the algorithm by applying it to the problem of facies classification. As a result, the imbalance between majority and minority classes was greatly mitigated, and the boundaries between classes could be more clearly identified.

• The Bulletin of The Korean Astronomical Society
• /
• v.40 no.2
• /
• pp.41.3-41.3
• /
• 2015

As part of the Herschel key program "Water in Star-forming Regions with Herschel (WISH)", PACS imaging spectroscopy data have been taken toward ten massive young stellar objects (YSOs): four high mass protostellar objects (HMPOs), two hot molecular cores (HMCs), and four ultracompact HII regions (UCHIIs). The spectra cover a broad range of wavelengths (55 to 210 micron) presenting various atomic and molecular lines as well as excellent dust thermal continua. By fitting the continua utilizing a modified black-body formula we estimate mass-weighted temperature and column density distributions of warm dust and find that UCHII regions are warmer and HMCs are more deeply embedded than the other types. We also estimate rotational temperature and column density distributions of warm CO gas using the rotational diagram analysis. In addition, based on the comparison of high J CO line fluxes to the RATRAN estimates of central heating envelope models, we find that majority of warm CO is originated from bipolar outflow shocks.