• Korean Journal of Remote Sensing
• /
• v.34 no.1
• /
• pp.45-74
• /
• 2018

Landslide is a natural hazard that threats lives and properties in many areas around the world. Landslides are difficult to recognize, particularly in rainforest regions. Thus, an accurate, detailed, and updated inventory map is required for landslide susceptibility, hazard, and risk analyses. The inconsistency in the results obtained using different features selection techniques in the literature has highlighted the importance of evaluating these techniques. Thus, in this study, six techniques of features selection were evaluated. Very-high-resolution LiDAR point clouds and orthophotos were acquired simultaneously in a rainforest area of Cameron Highlands, Malaysia by airborne laser scanning (LiDAR). A fuzzy-based segmentation parameter (FbSP optimizer) was used to optimize the segmentation parameters. Training samples were evaluated using a stratified random sampling method and set to 70% training samples. Two machine-learning algorithms, namely, Support Vector Machine (SVM) and Random Forest (RF), were used to evaluate the performance of each features selection algorithm. The overall accuracies of the SVM and RF models revealed that three of the six algorithms exhibited higher ranks in landslide detection. Results indicated that the classification accuracies of the RF classifier were higher than the SVM classifier using either all features or only the optimal features. The proposed techniques performed well in detecting the landslides in a rainforest area of Malaysia, and these techniques can be easily extended to similar regions.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.1
• /
• pp.61.1-61.1
• /
• 2013

In this study we apply Support Vector Machine (SVM) to the prediction of geo-effective halo coronal mass ejections (CMEs). The SVM, which is one of machine learning algorithms, is used for the purpose of classification and regression analysis. We use halo and partial halo CMEs from January 1996 to April 2010 in the SOHO/LASCO CME Catalog for training and prediction. And we also use their associated X-ray flare classes to identify front-side halo CMEs (stronger than B1 class), and the Dst index to determine geo-effective halo CMEs (stronger than -50 nT). The combinations of the speed and the angular width of CMEs, and their associated X-ray classes are used for input features of the SVM. We make an attempt to find the best model by using cross-validation which is processed by changing kernel functions of the SVM and their parameters. As a result we obtain statistical parameters for the best model by using the speed of CME and its associated X-ray flare class as input features of the SVM: Accuracy=0.66, PODy=0.76, PODn=0.49, FAR=0.72, Bias=1.06, CSI=0.59, TSS=0.25. The performance of the statistical parameters by applying the SVM is much better than those from the simple classifications based on constant classifiers.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.3
• /
• pp.569-577
• /
• 2012

The analysis and management of business failure has been recognized to be important in the area of financial management in the evaluation of firms' performance and the assessment of their viability. To this end, effective failure-prediction models are needed. This paper describes a new approach to prediction of business failure using the total margin algorithm which is a kind of support vector machine. It will be shown that the proposed method can evaluate the risk of failure better than existing methods through some real data.

• Proceedings of the KIEE Conference
• /
• 1993.07a
• /
• pp.227-230
• /
• 1993

Recently, the development of novel control methodology enables us to improve the performance of AC-machine drives by using pulse width modulation (PWM) technique. Usually, the dynamic characteristic of induction motor (IM) has been represented by the 5-th order nonlinear differential equation. This dynamics, however, can be reduced to 3-rd order dynamics by applying direct control of IM input current. This methodology concludes that it is much easier to control IM by means of the field-oriented methods employing the current controller. Therefore a precise current control is crucial to achieve a high control performance both in dynamic and steady state operations. This paper presents an adaptive fuzzy current controller with artificial neural network (ANN) for field-oriented controlled IM. This new control structure is able to adaptively minimize a current ripple while maintaining constant switching frequency. Especially the proposed controller employs neuro-computing philosophy as well as adaptive learning pattern recognizing principles with respect to variations of the system parameters. The proposed approach is applied to the IM drive system, and its performance is tested through various simulations. Simulation results show that the proposed system, compared among several known classical methods, has a superb performance.

• The Journal of Korea Robotics Society
• /
• v.10 no.2
• /
• pp.70-78
• /
• 2015

The purpose of this study was to investigate how young children recognize the image of robots, and how they understand the relationship between themselves and robots based on school experience. 20 children from kindergarten A had no direct experience with educational robots, whereas 20 children from kindergarten B had experience in using educational robots in their classroom. Total 40 children from age group 5 class participated in this study. We collected data using interview and drawing test. The findings of the study are as follows: First, participating children recognized robots as having both the character of a machine and a human. But children with previous robot experience provided description of robots as a machine-tool. Both groups were not able to explain the structure of robots in details. Second, participating children understood that they can develop a range of social relationships with robots, including simple help to family replacement. There were mixed views on robots among the children with previous experience, but children with no experience described robots as taking the role of peers or family members. These findings could contribute to the development of robots and related programs in the field of early childhood education.

• Journal of Sensor Science and Technology
• /
• v.29 no.3
• /
• pp.194-200
• /
• 2020

Human detection is an important aspect in many video-based sensing and monitoring systems. Studies have been actively conducted for the automatic detection of humans in camera images, and various methods have been proposed. However, there are still problems in terms of performance and computational cost. In this paper, we describe a method for efficient human detection in the field of view of a camera, which may be static or moving, through multiple processing steps. A detection line is designated at the position where a human appears first in a sensing area, and only the one-dimensional gray pixel values of the line are monitored. If any noticeable change occurs in the detection line, corner detection and optical flow computation are performed in the vicinity of the detection line to confirm the change. When significant changes are observed in the corner numbers and optical flow vectors, the final determination of human presence in the monitoring area is performed using the Histograms of Oriented Gradients method and a Support Vector Machine. The proposed method requires processing only specific small areas of two consecutive gray images. Furthermore, this method enables operation not only in a static condition with a fixed camera, but also in a dynamic condition such as an operation using a camera attached to a moving vehicle.

• Journal of Korean Institute of Industrial Engineers
• /
• v.40 no.4
• /
• pp.404-414
• /
• 2014

Yield prediction is important to manage semiconductor quality. Many researches with machine learning algorithms such as SVM (support vector machine) are conducted to predict yield precisely. However, yield prediction using SVM is hard because extremely imbalanced and big data are generated by final test procedure in semiconductor manufacturing process. Using SVM algorithm with imbalanced data sometimes cause unnecessary support vectors from major class because of unselected support vectors from minor class. So, decision boundary at target class can be overwhelmed by effect of observations in major class. For this reason, we propose a under-sampling method with minor class based SVM (MCSVM) which overcomes the limitations of ordinary SVM algorithm. MCSVM constructs the model that fixes some of data from minor class as support vectors, and they can be good samples representing the nature of target class. Several experimental studies with using the data sets from UCI and real manufacturing process represent that our proposed method performs better than existing sampling methods.

• Journal of Information Science Theory and Practice
• /
• v.2 no.1
• /
• pp.6-21
• /
• 2014

This paper proposes a novel knowledge extraction system, TAKES (Two-step Approach for Knowledge Extraction System), which integrates advanced techniques from Information Retrieval (IR), Information Extraction (IE), and Natural Language Processing (NLP). In particular, TAKES adopts a novel keyphrase extraction-based query expansion technique to collect promising documents. It also uses a Conditional Random Field-based machine learning technique to extract important biological entities and relations. TAKES is applied to biological knowledge extraction, particularly retrieving promising documents that contain Protein-Protein Interaction (PPI) and extracting PPI pairs. TAKES consists of two major components: DocSpotter, which is used to query and retrieve promising documents for extraction, and a Conditional Random Field (CRF)-based entity extraction component known as FCRF. The present paper investigated research problems addressing the issues with a knowledge extraction system and conducted a series of experiments to test our hypotheses. The findings from the experiments are as follows: First, the author verified, using three different test collections to measure the performance of our query expansion technique, that DocSpotter is robust and highly accurate when compared to Okapi BM25 and SLIPPER. Second, the author verified that our relation extraction algorithm, FCRF, is highly accurate in terms of F-Measure compared to four other competitive extraction algorithms: Support Vector Machine, Maximum Entropy, Single POS HMM, and Rapier.

• Journal of Korea Multimedia Society
• /
• v.19 no.8
• /
• pp.1516-1529
• /
• 2016

While global manufacturing is becoming more competitive due to variety of customer demand, increase in production cost and uncertainty in resource availability, the future ability of manufacturing industries depends upon the implementation of Smart Factory. With the convergence of new information and communication technology, Smart Factory enables manufacturers to respond quickly to customer demand and minimize resource usage while maximizing productivity performance. This paper presents the development of a big data analytics platform architecture for Smart Factory. As this platform represents a conceptual software structure needed to implement data-driven decision-making mechanism in shop floors, it enables the creation and use of diagnosis, prediction and optimization models through the use of data analytics and big data. The completion of implementing the platform will help manufacturers: 1) acquire an advanced technology towards manufacturing intelligence, 2) implement a cost-effective analytics environment through the use of standardized data interfaces and open-source solutions, 3) obtain a technical reference for time-efficiently implementing an analytics modeling environment, and 4) eventually improve productivity performance in manufacturing systems. This paper also presents a technical architecture for big data infrastructure, which we are implementing, and a case study to demonstrate energy-predictive analytics in a machine tool system.

• Design & Manufacturing
• /
• v.13 no.3
• /
• pp.53-58
• /
• 2019

This paper presents Artificial Neural Network(ANN) method to predict maximum injection pressure of injection molding machine and weights of injection molding products. 5 hidden layers with 10 neurons is used in the ANN. The ANN was conducted with 5 Input parameters and 2 response data. The input parameters, i.e., melt temperature, mold temperature, fill time, packing pressure, and packing time were selected. The combination of the orthogonal array L27 data set and 23 randomly generated data set were applied in order to train and test for ANN. According to the experimental result, error of the ANN for weights was $0.49{\pm}0.23%$. In case of maximum injection pressure, error of the ANN was $1.40{\pm}1.19%$. This value showed that ANN can be successfully predict the injection pressure and the weights of injection molding products.