• Journal of KIISE:Software and Applications
• /
• v.28 no.9
• /
• pp.681-687
• /
• 2001

In order to raise a class discrimination power by combining multiple classifiers under the Bayesian decision theory, the upper bound of a Bayes error rate bounded by the conditional entropy of a class variable and decision variables obtained from training data samples should be minimized. Wang and Wong proposed a tree dependence first-order approximation scheme of a high order probability distribution composed of the class and multiple feature pattern variables for minimizing the upper bound of the Bayes error rate. This paper presents an extended high order product approximation scheme dealing with higher order dependency more than the first-order tree dependence, based on the minimization of the upper bound of the Bayes error rate. Multiple recognizers for unconstrained handwritten numerals from CENPARMI were combined by the proposed approximation scheme using the Bayesian formalism, and the high recognition rates were obtained by them.

• Journal of KIISE:Software and Applications
• /
• v.32 no.11
• /
• pp.1136-1146
• /
• 2005

Fingerprint classification is useful in an automated fingerprint identification system (AFIS) to reduce the matching time by categorizing fingerprints. Based on Henry system that classifies fingerprints into S classes, various techniques such as neural networks and support vector machines (SVMs) have been widely used to classify fingerprints. Especially, SVMs of high classification performance have been actively investigated. Since the SVM is binary classifier, we propose a novel classifier-combination model, multiple decision templates (MuDTs), to classily fingerprints. The method extracts several clusters of different characteristics from samples of a class and constructs a suitable combination model to overcome the restriction of the single model, which may be subject to the ambiguous images. With the experimental results of the proposed on the FingerCodes extracted from NIST Database4 for the five-class and four-class problems, we have achieved a classification accuracy of $90.4\%\;and\;94.9\%\;with\;1.8\%$ rejection, respectively.

• Journal of Navigation and Port Research
• /
• v.39 no.3
• /
• pp.165-172
• /
• 2015

The class of Dynamic Positioning System is divided in 3 classes depending on its redundancy and reliability according to IMO and classification society. There are 3 DP classes such as DP Class 1, 2 and 3 according to IMO MSC/Circ. 645. Higher DP class vessel has higher reliability, since redundancy concept is applied to the DP vessel depending on its DP class and can operate more safely. There are not enough information about DP class notation, which are needed when a company builds a new or buys second hand DP vessel or modifies DP classes, even the Korean shipyard is building a lot of DP vessels now. Also, the practical case of DP vessel modification, which had been done in Korea, to meet DP notation of IMO and classification society, will be helpful for DP vessel modification and sales industry development in Korea as a new business. As such this research identified what kind of requirements need to be taken into account to be from DP class 1 to DP class 2. The real DP class modification case is used to identify the requirements of DP class upgrade. Through the FMEA the redundancy concept on power system, thruster system and DP control system need to apply for DP class upgrade. The power system have to keep its DP function even if just a single fault happens on the generator or switchboard. Also, the PMS is required to monitor and control power system. Ship's Surge, Sway and Yaw movements can be controlled by the remaining thruster system after a single thruster fails. Lastly, multiple installation of PRS, sensors and DP control system are required to keep DP ability after a single fault on the DP control systems.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.2
• /
• pp.9-19
• /
• 2017

Useful continuous sensing applications are increasingly emerging as a new class of mobile applications. Meanwhile, open, multi-use sensor devices are newly adopted beyond smartphones, and provide huge opportunities to expand potential application categories. In this upcoming environment, uncoordinated use of sensor devices would cause severe imbalance in power consumption of devices, and thus result in early shutdown of some sensing applications depending on power-hungry devices. In this paper, we propose EnergyCordy, a novel inter-device energy use coordination system; with a system-wide holistic view, it coordinates the energy use of concurrent sensing applications over multiple sensor devices. As its key approach, we propose a relaxed sensor association; it decouples the energy use of an application from specific sensor devices leveraging multiple context inference alternatives, allowing flexible energy coordination at runtime. We demonstrated the effectiveness of EnergyCordy by developing multiple example applications over custom-designed wearable senor devices. We show that EnergyCordy effectively coordinates the power usage of concurrent sensing applications over multiple devices and prevent undesired early shutdown of applications.

• Communications for Statistical Applications and Methods
• /
• v.20 no.6
• /
• pp.439-454
• /
• 2013

This study presents a Bayesian multiple change-point detection approach to segment and classify the observations that no longer come from an initial population after a certain time. Inferences are based on the multiple change-points in a sequence of random variables where the probability distribution changes. Bayesian multiple change-point estimation is classifies each observation into a segment. We use a truncated Poisson distribution for the number of change-points and conjugate prior for the exponential family distributions. The Bayesian method can lead the unsupervised classification of discrete, continuous variables and multivariate vectors based on latent class models; therefore, the solution for change-points corresponds to the stochastic partitions of observed data. We demonstrate segmentation with real data.

• Journal of Korean Home Economics Education Association
• /
• v.17 no.2
• /
• pp.1-10
• /
• 2005

The purpose of the study was to examine the effects of the lesson using Multiple Intelligence on self-esteem in the unit of Resources management and environment in Technology${\cdot}$Home Economics of middle school. In this study, the lesson using Multiple Intelligence theory for six hours was developed and applied to a class of 37 students of middle school as an experimental group. Lecture-style lesson was applied to the other class of 37 students as a control group. Significant differences between pre-test and post-test scores of self-esteem were found in the experimental group. Post-test scores of self-esteem were higher than pre-test scores of self-esteem in the experimental group. There were no significant differences between pre-test and post-test scores of self-esteem in the control group. It was found that the lesson using Multiple Intelligence theory was effective in improving self-esteem of middle school students.

• Journal of Internet Computing and Services
• /
• v.15 no.1
• /
• pp.63-71
• /
• 2014

Currently many non-traditional application areas such as artificial intelligence and web databases require advanced modeling power than the existing relational data model. In those application areas, object-oriented database (OODB) is better data model since an OODB can providemodeling power as grouping similar objects into class, and organizing all classes into a hierarchy where a subclass inherits all definitions from its superclasses. The purpose of this paper is to develop an OODB concurrency control scheme dealing with multiple inheritance. The proposed scheme, called Multiple Inheritance Implicit Locking (MIIL), is based on so-called implicit locking. In the proposed scheme, we eliminate redundant locks that are necessary in the existing implicit locking scheme. Intention locks are required as the existing implicit locking scheme. In this paper, it is shown that MIIL has less locking overhead than implicit locking does. We use only OODB inheritance hierarchies, single inheritance and multiple inheritance so that no additional overhead is necessary for reducing locking overhead.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.257-263
• /
• 2021

This study analyzed the factors affecting online class satisfaction and tried to use it as basic data for more effective class management. The survey was collected on 208 students majoring in dental laboratory technology in Jeonbuk and Chungbuk. The data were analyzed by reliability analysis, descriptive stastistics, compare means(t-test, one-way ANOVA), Pearson's correlation coefficient and stepwies multiple regression analysis(SPSS program). The analysis results showed that expected benefit of time was 3.87, expected benefit of learning was 3.30, technology familiarity was 3.40, and class satisfaction was 3.21. The most influential factor in class satisfaction was technology familiarity. In order to increase the online class satisfaction, it is necessary to build a learning environment to improve the ability to learning tools used in classes and technology familiarity with the online class system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.958-964
• /
• 2016

In this paper, we propose an i-vector based language recognition system to identify the spoken language of the speaker, which uses multiple discriminative deep neural network (DNN) models analogous to the multi-class support vector machine (SVM) classification system. The proposed model was trained and tested using the i-vectors included in the NIST 2015 i-vector Machine Learning Challenge database, and shown to outperform the conventional language recognition methods such as cosine distance, SVM and softmax NN classifier in open-set experiments.

• Restorative Dentistry and Endodontics
• /
• v.25 no.1
• /
• pp.56-62
• /
• 2000

The purpose of this study was to assess the effects of restorative materials on the marginal leakage of wedge-shaped class V cavity. The study was performed in vitro in 25 defect-free permanent, extracted teeth. Wedge-shaped class V cavities were prepared and then the teeth were randomly selected and restored according to the following. Group A : restoration with Tetric Ceram(composite resin) Group B : restoration with Tetric flow(flowable resin) Group C : restoration with Compoglass after acid etching(compomer) Group D : restoration with Compoglass(compomer) Group E : restoration with Fuji II LC improved(resin-modified GIC) After thermocycling, the specimens were immersed in 5% basic fuchsin solution for 6 hours and sectioned longitudinally through the center of the restoration. The degree of marginal leakage was measured as the extent of dye penetration under the stereomicroscope. The data were analysed using one-way ANOVA. When significant differences found, multiple comparisons were made using Duncan's Multiple Range Test. The results were as follows: 1. The occlusal margins of all groups except for Fuji II LC improved showed lesser leakage than gingival margins and there was statistically significant difference(p<0.05). 2. At the occlusal margins, group A, B showed same marginal leakage scores, and others were decreased as group C, D, E in that order. There were statistically significant difference between group A, Band group D, E, group C and group E(p<0.05). 3. At the gingival margins, group B, C showed same marginal leakage scores, and others were decreased as group A, D, E in that order. But there was statistically significant difference between group B, C and group E(p<0.05). 4. In the Compoglass restoration, acid-etching technique was beneficial for marginal sealing ability at all of margins. But there was no statistically significant difference (p>0.05). In the restorations for wedge-shaped class V cavities, resin restoration with acid etching technique is recommended.