• Proceedings of the KSRS Conference
• /
• 2007.10a
• /
• pp.594-597
• /
• 2007

We present in this paper a novel mid and long term power load prediction method using temporal pattern mining from AMR (Automatic Meter Reading) data. Since the power load patterns have time-varying characteristic and very different patterns according to the hour, time, day and week and so on, it gives rise to the uninformative results if only traditional data mining is used. Also, research on data mining for analyzing electric load patterns focused on cluster analysis and classification methods. However despite the usefulness of rules that include temporal dimension and the fact that the AMR data has temporal attribute, the above methods were limited in static pattern extraction and did not consider temporal attributes. Therefore, we propose a new classification method for predicting power load patterns. The main tasks include clustering method and temporal classification method. Cluster analysis is used to create load pattern classes and the representative load profiles for each class. Next, the classification method uses representative load profiles to build a classifier able to assign different load patterns to the existing classes. The proposed classification method is the Calendar-based temporal mining and it discovers electric load patterns in multiple time granularities. Lastly, we show that the proposed method used AMR data and discovered more interest patterns.

• Journal of KIISE:Software and Applications
• /
• v.35 no.11
• /
• pp.681-691
• /
• 2008

In this paper, we have evaluated and compared each feature and feature combinations necessary for statistical Korean dialogue act classification. We have implemented a Korean dialogue act classification system by using the Support Vector Machine method. The experimental results show that the POS bigram does not work well and the morpheme-POS pair and other features can be complementary to each other. In addition, a small number of features, which are selected by a feature selection technique such as chi-square, are enough to show steady performance of dialogue act classification. We also found that the last eojeol plays an important role in classifying an entire sentence, and that Korean characteristics such as free order and frequent subject ellipsis can affect the performance of dialogue act classification.

• Journal of the Institute of Convergence Signal Processing
• /
• v.8 no.3
• /
• pp.185-191
• /
• 2007

Remote sensing images are multispectral image data collected from several band divided by wavelength ranges. The classification of remote sensing images is the method of classifying what has similar spectral characteristics together among each pixel composing an image as the important algorithm in this field. This paper presents a pattern classification method of remote sensing images by applying a possibilistic fuzzy c-means (PFCM) algorithm. The PFCM algorithm is a hybridization of a FCM algorithm, which adopts membership degree depending on the distance between data and the center of a certain cluster, combined with a PCM algorithm, which considers class typicality of the pattern sets. In this proposed method, we select the training data for each class and perform supervised classification using the PFCM algorithm with spectral signatures of the training data. The application of the PFCM algorithm is tested and verified by using Landsat TM and IKONOS remote sensing satellite images. As a result, the overall accuracy showed a better results than the FCM, PCM algorithm or conventional maximum likelihood classification(MLC) algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.2
• /
• pp.17-26
• /
• 2018

In this paper, we propose a gait type classification method based on pressure sensor which reflects various terrain and velocity variations. In order to obtain stable gait classification performance, we divide the whole gait data into several steps by detecting the swing phase, and normalize each step. Then, we extract robust features for both topographic variation and speed variation by using the Null-LDA(Null-Space Linear Discriminant Analysis) method. The experimental results show that the proposed method gives a good performance of gait type classification even though there is a change in the gait velocity and the terrain.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.10
• /
• pp.31-38
• /
• 2015

In this paper, we propose the ensemble-based classification model which extracts just new data patterns from the streaming-data by using clustering and generates new classification models to be added to the ensemble in order to reduce the number of data labeling while it keeps the accuracy of the existing system. The proposed technique performs clustering of similar patterned data from streaming data. It performs the data labeling to each cluster at the point when a certain amount of data has been gathered. The proposed technique applies the K-NN technique to the classification model unit in order to keep the accuracy of the existing system while it uses a small amount of data. The proposed technique is efficient as using about 3% less data comparing with the existing technique as shown the simulation results for benchmarks, thereby using clustering.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.267-270
• /
• 2019

Through this research, the rating data of shops were analyzed. The model was designed for discrete multiple classification as to the corresponding data, and the following experiments were initiated to observe the learned machine. By comparing each benchmarks in the experiments, which contains different setting variables for the machine model, the hit ratio was measured which indicates how much it is matched with the expected label. By analyzing those results from each benchmarks, the model was redesigned one time during the research and the effects of each setting variables on this machine were clarified. Furthermore, the research result left the future works, which are related with how the learning could be improved and what should be designed in the further research.

• The Journal of Korean Medicine
• /
• v.35 no.3
• /
• pp.93-102
• /
• 2014

Objectives: This study aimed at revising the Korean Out-patient Groups for Korean Medicine (KOPG-OM, version 1.0) based on clinical similarity and resource use, by using the accumulated claims data, and evaluating the validity of the revised classification system. Methods: A clinical specialist panel involving 19 specialists from 8 Korean medicine (KM) specialty areas reviewed the classification tree, diagnosis groups and procedure groups in terms of clinical similarity. Several models of outpatient grouping were formulated, with the validity of each tested based on the $R^2$ coefficient of determination for the treatment costs of all visits. To add age splits, the variances of treatment costs by age groups were also analyzed. These statistical analyses were performed using KM claims data of National Health Insurance from 2010 to 2012. Results: The classification tree designed via panel discussions was used to allocate outpatient cases to 26 diagnosis groups, with cases involving procedures such as acupuncture, moxibustion and cupping, then allocated to 9 procedure groups in each diagnosis group. The cases without procedures were categorized into the visit index - medication group. This process resulted in 298 outpatient groups. The $R^2$ values for treatment costs of all visits ranged from 0.38 to 0.69 depending on the providers' types. Conclusions: The revised model of KOPG-KM has a higher validity for outpatient classification than the current system and can provide better management of the costs of outpatient care in KM.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.6
• /
• pp.67-75
• /
• 1993

ECG pattern was classified using a back-propagation neural network. An improved feature extractor of ECG is proposed for better classification capability. It is consisted of preprocessing ECG signal by an FIR filter faster than conventional one by a factor of 5. QRS complex recognition by moving-window integration, and peak extraction by quadratic approximation. Since the FIR filter had a periodic frequency spectrum, only one-fifth of usual processing time was required. Also, segmentation of ECG signal followed by quadratic approximation of each segment enabled accurate detection of both P and T waves. When improtant features were extracted and fed into back-propagation neural network for pattern classification, the required number of nodes in hidden and input layers was reduced compared to using raw data as an input, also reducing the necessary time for study. Accurate pattern classification was possible by an appropriate feature selection.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.6_2
• /
• pp.1055-1062
• /
• 2023

Globally, the need for remanufacturing or reusing ships and various mechanical parts continues to increase due to environmental problems including global warming. Research on remanufacturing is being carried out in many areas. However, research on inspection and classification to identify the performance or degree of wear of mechanical parts is insufficient. In particular, studies on the inspection-ability and classification-ability of mechanical parts equipped with various materials and complex forms are highly required. Remanufacturing must be considered from the stage of design to extend the life cycle of mechanical parts. Particularly, it is very important to perform research for evaluating the degree of ease to inspect and classify various sorts of wear or deterioration of parts caused by long-term use easily. In this study, the degree of ease in inspecting or classifying mechanical parts for remanufacturing is defined as inspection-ability and classification-ability. In fact, to remanufacture old parts, inspection-ability and classification-ability should be reflected from the stage of design. The purpose of this study is to evaluate the inspection-ability and classification-ability of ships and various mechanical parts. This researcher has presented the quantitative evaluation procedure of inspection-ability and classification-ability, derived the factors and ranges that influence each of the details of easiness, assigned scores according to the ranges of the factors, and calculated weights. Lastly, this study presents the procedure of scoring to evaluate the overall weights of inspection-ability and classification-ability and also inspection-ability and classification-ability quantitatively.

• Journal of Microbiology
• /
• v.33 no.2
• /
• pp.95-102
• /
• 1995

A probability identification matrix of acidophilic Streptomyces was constructed. The phenetic data of the strains were derived from numerical classification described by Seong et al. The minimum number of diagnostic characters was determined using computer programs for calculation of different separation indices. The resulting matrix consisted of 25 clusters versus 53 characters. Theoretical evaluation of this matrix was achieved by estimating the chuster overlap and the identification scores for the Hypothetical Median Organisms (HMO) and for the representatives of each cluster. Cluster overlap was found to be relatively small. Identification scores for the HMO and the randomly selected representatives of each cluster were satisfactory. The matrix was assessed practically by applying the matrix to the identification of unknown isolates. Of the unknown isolates, 71.9% were clearly identified to one of eight clusters. The numerical classification data was also used to design a selective isolation medium for antibiotic-producing organisms. Four chemical substances including 2 antibiotics were determined by the DLACHAR program as diagnostic for the isolation of target organisms which have antimicrobial activity against Micrococcus luteus. It was possible to detect the increased rate of selective isolation on the synthesized medium. Theresults show that the numerical phenetic data can be applied to a variety of purposes, such as construction of identification matrix and selective isolation medium for acidophilic antinomycetes.