• Journal of the Korea Society of Computer and Information
• /
• v.23 no.10
• /
• pp.103-109
• /
• 2018

In this paper, we propose a dynamic document classification method which breaks away from existing document classification method with artificial categorization rules focusing on suppliers and has changing categorization rules according to users' needs or social trends. The core of this dynamic document classification method lies in the fact that it creates classification criteria real-time by using topic modeling techniques without standardized category rules, which does not force users to use unnecessary frames. In addition, it can also search the details through the relevance analysis by calculating the relationship between the words that is difficult to grasp by word frequency alone. Rather than for logical and systematic documents, this method proposed can be used more effectively for situation analysis and retrieving information of unstructured data which do not fit the category of existing classification such as VOC (Voice Of Customer), SNS and customer reviews of Internet shopping malls and it can react to users' needs flexibly. In addition, it has no process of selecting the classification rules by the suppliers and in case there is a misclassification, it requires no manual work, which reduces unnecessary workload.

• Journal of Institute of Control, Robotics and Systems
• /
• v.22 no.4
• /
• pp.253-265
• /
• 2016

We propose an obstacle classification method using multi-decision factors and decision sections based on Single 2D LiDAR. The existing obstacle classification method based on single 2D LiDAR has two specific advantages: accuracy and decreased calculation time. However, it was difficult to classify obstacle type, and therefore accurate path planning was not possible. To overcome this problem, a method of classifying obstacle type based on width data was proposed. However, width data was not sufficient to enable accurate obstacle classification. The proposed algorithm of this paper involves the comparison between decision factor and decision section to classify obstacle type. Decision factor and decision section was determined using width, standard deviation of distance, average normalized intensity, and standard deviation of normalized intensity data. Experiments using a real autonomous vehicle in a real environment showed that calculation time decreased in comparison with 2D LiDAR-based method, thus demonstrating the possibility of obstacle type classification using single 2D LiDAR.

• Journal of information and communication convergence engineering
• /
• v.19 no.2
• /
• pp.79-83
• /
• 2021

This paper proposes a novel image classification method based on few-shot learning, which is mainly used to solve model overfitting and non-convergence in image classification tasks of small datasets and improve the accuracy of classification. This method uses model structure optimization to extend the basic convolutional neural network (CNN) model and extracts more image features by adding convolutional layers, thereby improving the classification accuracy. We incorporated certain measures to improve the performance of the model. First, we used general methods such as setting a lower learning rate and shuffling to promote the rapid convergence of the model. Second, we used the data expansion technology to preprocess small datasets to increase the number of training data sets and suppress over-fitting. We applied the model to 10 monkey species and achieved outstanding performances. Experiments indicated that our proposed method achieved an accuracy of 87.92%, which is 26.1% higher than that of the traditional CNN method and 1.1% higher than that of the deep convolutional neural network ResNet50.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1D
• /
• pp.149-156
• /
• 2006

Existing method of landuse classification using aerial photographs or field survey requires relatively higher amount of time and cost due to necessary manual work. Especially in urban area where the pattern of landuse is densely aggregated, a landuse classification using satellite image is more complex. In this background, this study proposes a landuse classification method to utilize 1:1,000 digital topographic data and IKONOS satellite image. To prove the possibility of this method, the method was applied to Seoul metropolitan area. The results shows the total accuracy of approximately 95% and 14 landuse classes extracted. Based on the results from the pilot study, this method is applicable to landuse classification in urban area.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.6 no.3
• /
• pp.107-116
• /
• 2003

This study investigated how appropriate the classification methods based on conventional spectral characteristics are for high resolution imagery. A supervised classification mixing parametric and non-parametric rules, a method in which fuzzy theory is applied to such classification, and an unsupervised method were performed and compared to each other for accuracy. In addition, comparing the result screen-digitized through interpretation to the classification result using spectral characteristics, this study analyzed the conformity of both methods. Although the supervised classification to which fuzzy theory was applied showed the best performance, the application of conventional classification techniques to high resolution imagery had some limitations due to there being too much information unnecessary to classification, shadows, and a lack of spectral information. Consequently, more advanced techniques including integration with other advanced remote sensing technologies, such as lidar, and application of filtering or template techniques, are required to classify land cover/use or to extract useful information from high resolution imagery.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.311-319
• /
• 2009

A robust L1 data depth was used in clustering and classification, so called DDclus and DDclass by Jornsten (2004). SVM-based classification works well in most of the situation but show some weakness in the presence of outliers. Proper gene selection is important in classification since there are so many redundant genes. Either by selecting appropriate genes or by gene clustering combined with classification method enhance the overall performance of classification. The performance of depth based method are evaluated among several SVM-based classification methods.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.28 no.3
• /
• pp.217-224
• /
• 2017

In this paper, we propose a fusion classification method combining HRRP and JEM classifier with complementary properties for the classification of aircraft. The fusion method is based on the confidence of a classifier for a classification result to improve performance compared with single classifier in various situations. The confidence is defined as the posterior probability estimated from the classification performance of a classifier and it depends on the aspect angle and the certainty for a classification result. Through the classification test using simulation data, we can verify that the proposed fusion method shows good performance by fusing the classifiers effectively.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.2
• /
• pp.807-824
• /
• 2016

In this paper, a method of mood classification based on user brainwaves is proposed for real-time application in commercial services. Unlike conventional mood analyzing systems, the proposed method focuses on classifying real-time user moods by analyzing the user's brainwaves. Applying brainwave-related research in commercial services requires two elements - robust performance and comfortable fit of. This paper proposes a filter based on Regularized Common Spatial Patterns (RCSP) and presents its use in the implementation of mood classification for a music service via a wireless consumer electroencephalography (EEG) device that has only 14 pins. Despite the use of fewer pins, the proposed system demonstrates approximately 10% point higher accuracy in mood classification, using the same dataset, compared to one of the best EEG-based mood-classification systems using a skullcap with 32 pins (EU FP7 PetaMedia project). This paper confirms the commercial viability of brainwave-based mood-classification technology. To analyze the improvements of the system, the changes of feature variations after applying RCSP filters and performance variations between users are also investigated. Furthermore, as a prototype service, this paper introduces a mood-based music list management system called MyMusicShuffler based on the proposed mood-classification method.

• IE interfaces
• /
• v.25 no.2
• /
• pp.170-177
• /
• 2012

Process control is essential to operate the semiconductor process efficiently. This paper consider fault classification of semiconductor based cyclic signal for process control. In general, process signal usually take the different pattern depending on some different cause of fault. If faults can be classified by cause of faults, it could improve the process control through a definite and rapid diagnosis. One of the most important thing is a finding definite diagnosis in fault classification, even-though it is classified several times. This paper proposes the method that one-class classifier classify fault causes as each classes. Hotelling T2 chart, kNNDD(k-Nearest Neighbor Data Description), Distance based Novelty Detection are used to perform the one-class classifier. PCA(Principal Component Analysis) is also used to reduce the data dimension because the length of process signal is too long generally. In experiment, it generates the data based real signal patterns from semiconductor process. The objective of this experiment is to compare between the proposed method and SVM(Support Vector Machine). Most of the experiments' results show that proposed method using Distance based Novelty Detection has a good performance in classification and diagnosis problems.

• Journal of Information Processing Systems
• /
• v.19 no.2
• /
• pp.258-266
• /
• 2023

Aiming at the problems of poor customer satisfaction and poor accuracy of customer classification, this paper proposes a customer classification model based on speech recognition. First, this paper analyzes the temporal data characteristics of customer demand data, identifies the influencing factors of customer demand behavior, and determines the process of feature extraction of customer voice signals. Then, the emotional association rules of customer demands are designed, and the classification model of customer demands is constructed through cluster analysis. Next, the Euclidean distance method is used to preprocess customer behavior data. The fuzzy clustering characteristics of customer demands are obtained by the fuzzy clustering method. Finally, on the basis of naive Bayesian algorithm, a customer demand classification model based on speech recognition is completed. Experimental results show that the proposed method improves the accuracy of the customer demand classification to more than 80%, and improves customer satisfaction to more than 90%. It solves the problems of poor customer satisfaction and low customer classification accuracy of the existing classification methods, which have practical application value.