• Proceedings of the KSR Conference
• /
• 2004.10a
• /
• pp.240-245
• /
• 2004

It is necessary to analyze the railway accidents and incidents for the purpose of understanding present safety state and enhancing its system. Korea National Railroad has its accident/incident reporting codes, but it is relatively not sufficient for detail classification and investigation of accident and incident compared with foreign countries. This paper suggests how to classify the railway accident/incident, and describes the analysis result for domestic recent railway accidents and incidents according to the new suggested classifying system.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1995.10b
• /
• pp.52-58
• /
• 1995

A shared-weight neural network that performed classification based on the features extracted with the fuzzy morphological operation is introduced. Learning rules for the structuring elements, degree of membership, and weighting factors are also precisely described. In application to handwritten digit recognition problem, the fuzzy morphological shared-weight neural network produced the results which are comparable to the state-of-art for this problem.

• Bulletin of the Korean Mathematical Society
• /
• v.56 no.6
• /
• pp.1525-1537
• /
• 2019

Let ${\tilde{M}}$ be a Riemannian manifold equipped with a concircular vector field ${\tilde{X}}$ and M a submanifold (with its induced metric) of ${\tilde{M}}$. Denote by X the restriction of ${\tilde{X}}$ on M and by $X^T$ the tangential component of X, called the canonical field of M. In this article we study submanifolds of ${\tilde{M}}$ whose canonical field $X^T$ is also concircular. Several characterizations and classification results in this respect are obtained.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.2
• /
• pp.841-854
• /
• 2020

The paper explains the method to process, analyze and predict traffic patterns in Los Angeles county using Big Data and Machine Learning. The dataset is used from a popular navigating platform in the USA, which tracks information on the road using connected users' devices and also collects reports shared by the users through the app. The dataset mainly consists of information about traffic jams and traffic incidents reported by users, such as road closure, hazards, accidents. The major contribution of this paper is to give a clear view of how the large-scale road traffic data can be stored and processed using the Big Data system - Hadoop and its ecosystem (Hive). In addition, analysis is explained with the help of visuals using Business Intelligence and prediction with classification machine learning model on the sampled traffic data is presented using Azure ML. The process of modeling, as well as results, are interpreted using metrics: accuracy, precision and recall.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.1
• /
• pp.35-43
• /
• 2015

Over the past couple of years, tremendous progress has been made in applying deep learning (DL) techniques to computer vision. Especially, deep convolutional neural networks (DCNNs) have achieved state-of-the-art performance on standard recognition datasets and tasks such as ImageNet Large-Scale Visual Recognition Challenge (ILSVRC). Among them, GoogLeNet network which is a radically redesigned DCNN based on the Hebbian principle and scale invariance set the new state of the art for classification and detection in the ILSVRC 2014. Since there exist various deep learning techniques, this review paper is focusing on techniques directly related to DCNNs, especially those needed to understand the architecture and techniques employed in GoogLeNet network.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.22 no.2
• /
• pp.105-122
• /
• 1994

The landscape plannings were composed of the fundamental provisions which are important factors for the city and local plannings in Germany. The State Government run institute like LOLF in NRW have been established to solve the serious ecological problems. The institute have completed the basic system covering various ecological problems, such as environmental affection analysis and ecological risk, to contribute the application of the various land-using plants. The ecological landscape plannings are regarded as major fundamental plans of environmental preservation for the plannings development. The problems about the local landscape plannings needed to be preserved ecologically are introduced into the Federal Nature Preservation Law and State Congress Law in order to carry out the environmental protection policy effectively. The local landscape plannings are divided into two-step models by the special classification system for the planning. One is build-up of an independent design separated from city plans functionally, and the other is a systematic regulation for cooperation between the local plans and ecological conservation functions. The local plannings including lands to be preserved as the Nature Preservatin Zone is classified a rest area total sectioned plains.

• Journal of Institute of Control, Robotics and Systems
• /
• v.6 no.6
• /
• pp.486-491
• /
• 2000

The dicing process cuts a semiconductor wafer to lengthwise and crosswise direction by using a rotating circular diamond blade. But inferior goods are made under the influence of several parameters in dicing such as blade, wafer, cutting water and cutting conditions. This paper describes a monitoring algorithm using neural network in order to find out an instant of vibration signal change when bad dicing appears. The algorithm is composed of two steps: feature extraction and decision. In the feature extraction, five features processed from vibration signal which is acquired by accelerometer attached on blade head are proposed. In the decision, back-propagation neural network is adopted to classify the dicing process into normal and abnormal dicing, and normal and damaged blade. Experiments have been performed for GaAs semiconductor wafer in the case of normal/abnormal dicing and normal/damaged blade. Based upon observation of the experimental results, the proposed scheme shown has a good accuracy of classification performance by which the inferior goods decreased from 35.2% to 6.5%.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 1996.04a
• /
• pp.169-175
• /
• 1996

We provide a fudamental set of information on technologies in the area of mechanical engineering to draw action plans for intermationalization of the National R&D activites. First, we design and use a modified Delphi method to evaluate levels of our technological capabilities and developed countries' as well. We investigate technology acquistion methodolgies, technology characteristcs and various aspects of international cooperation in terms of technology. Secondly, we analyze final responses of participants (i.e,. the third round results of Delphi method) tosee the correlation between various factors in developing mechanical engineering technologies through international cooperation. The technology classification used in this research is developed by STEPI (Science and Technology Policy Institute).

• 한국신재생에너지학회:학술대회논문집
• /
• 2008.10a
• /
• pp.357-360
• /
• 2008

There are many control strategies for HEV in today. Expanding motor-driving and operating at good-efficient point in engine is the key of the HEV control to increase fuel economy. There are two types of HEV supervisory control. One is rule-based control and the other is optimization control. MAX-SOC control, thermostat control, baseline status control and state-machine control are in deterministic RBC. It is simple, but powerful and easy to apply in real-time circumstance. In this study, we analysis these four control strategies in RBC (Rule-based control) and identify the each advantage and disadvantage.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.3 no.1
• /
• pp.65-75
• /
• 1993

Fuzzy data is a phenomenon often occurring in real life. There is the inherent vagueness of classification terms referring to a continuous scale, the uncertainty of linguistic terms such as "I almost agree" or the vagueness of terms and concepts due to the statistical variability in communication [20] and many more. Previously, such fuzzy data was approximated by non-fuzzy (crisp) data, which obviously did not lead to a correct and precise representation of the real world. Fuzzy set theory has been developed to represent and manipulate fuzzy data [18]. Explicitly managing the degree of fuzziness in databases allows the system to distinguish between what is known, what is not known and what is partially known. Systems in the literature whose specific objective is to handle imprecision in databases present various approaches. This paper is concerned with the different ways uncertainty and imprecision are handled in database design. It outlines the major areas of fuzzification in (relational) database systems.