• Journal of rehabilitation welfare engineering & assistive technology
• /
• v.11 no.2
• /
• pp.143-146
• /
• 2017

In this paper, a novel approach is proposed to improve the performance of speech/music classification using the recurrent neural network (RNN) in the enhanced voice services (EVS) of 3GPP for hearing aids. Feature vectors applied to the RNN are selected from the relevant parameters of the EVS for efficient speech/music classification. The performance of the proposed algorithm is evaluated under various conditions and large speech/music data. The proposed algorithm yields better results compared with the conventional scheme implemented in the EVS.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.6
• /
• pp.1557-1563
• /
• 2015

Multiclass classification is typically performed using the voting scheme method based on combining binary classifications. In this paper we propose multiclass classification method for large data, which can be regarded as the revised one-vs-all method. The multiclass classification is performed by using the hat matrix of least squares support vector machine (LS-SVM) ensemble, which is obtained by aggregating individual LS-SVM trained on each subset of whole large data. The cross validation function is defined to select the optimal values of hyperparameters which affect the performance of multiclass LS-SVM proposed. We obtain the generalized cross validation function to reduce computational burden of cross validation function. Experimental results are then presented which indicate the performance of the proposed method.

• International Journal of Contents
• /
• v.14 no.4
• /
• pp.24-29
• /
• 2018

EEG-based brain-computer interfaces has focused on explicitly expressed intentions to assist physically impaired patients. For EEG-based-computer interfaces to function effectively, it should be able to understand users' implicit information. Since it is hard to gather EEG signals of human brains, we do not have enough training data which are essential for proper classification performance of implicit intention. In this paper, we improve the subject independent classification of implicit intention through the generation of additional training data. In the first stage, we perform the PCA (principal component analysis) of training data in a bid to remove redundant components in the components within the input data. After the dimension reduction by PCA, we train ICA (independent component analysis) network whose outputs are statistically independent. We can get additional training data by adding Gaussian noises to ICA outputs and projecting them to input data domain. Through simulations with EEG data provided by CNSL, KAIST, we improve the classification performance from 65.05% to 66.69% with Gamma components. The proposed sample generation method can be applied to any machine learning problem with fewer samples.

• Journal of Korea Multimedia Society
• /
• v.25 no.1
• /
• pp.29-38
• /
• 2022

Although image compression is one of the essential technologies to transmit image data on a variety of surveillance and mobile healthcare applications, it causes unnecessary compression artifacts such as blocking and ringing artifacts by the lossy compression in the limited network bandwidth. Recently, image restoration methods using convolutional neural network (CNN) show the significant improvement of image quality from the compressed images. In this paper, we propose Image Denoising Convolutional Neural Networks (IDCNN) to reduce the compression artifacts for the purpose of improving the performance of object classification. In order to evaluate the classification accuracy, we used the ImageNet test dataset consisting of 50,000 natural images and measured the classification performance in terms of Top-1 and Top-5 accuracy. Experimental results show that the proposed IDCNN can improve Top-1 and Top-5 accuracy as high as 2.46% and 2.42%, respectively.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.20 no.2
• /
• pp.1-10
• /
• 2024

For the purpose of predicting credit card customer churn accurately through data analysis, a model can be constructed with various machine learning algorithms, including decision tree. And feature importance has been utilized in selecting better input features that can improve performance of data analysis models for several application areas. In this paper, a method of utilizing feature importance calculated from the MDI method and its effects are investigated in the credit card customer churn prediction problem with classification trees. Compared with several random feature selections from case data, a set of input features selected from higher value of feature importance shows higher predictive power. It can be an efficient method for classifying and choosing input features necessary for improving prediction performance. The method organized in this paper can be an alternative to the selection of input features using feature importance in composing and using classification trees, including credit card customer churn prediction.

• Asia Marketing Journal
• /
• v.11 no.1
• /
• pp.39-72
• /
• 2009

With the growing importance of accountability, it is getting necessary to test the impact of marketing expenditure on marketing performance. Including recent ROM, we can find a few researches about marketing accountability. But there are a few problems about definitions and metric of marketing expenditure and marketing performance. Therefore, by defining and analyzing the impact of marketing expenditure on marketing performance, we are going to set the classification scheme of marketing expenditure and marketing performance. Based on research findings, new definitions and metrics are proposed as follows. First, we suggest the classification scheme of marketing expenditure. Marketing expenditure is defined as expense accounts in the balance sheet for doing marketing tasks. Marketing expenditures includes many accounts, for example, marketing research, advertising, sales promotion, foreign market development, physical distribution, after services. Among these marketing investment, advertising expenses have a positive effect on marketing performance. Second, we suggest the classification scheme of marketing performance. Already, marketing performance has been defined as financial metrics, customer metrics, market metrics, and corporate social responsibility. But, in this study, we find that the process model is not relevant for explaining association between the performance metrics. The process model is a virtuous cycle: "customer metrics→market metrics→financial metrics→firm valuation metrics." But, in this study, it is not supported or a little significant association between these metrics. Based on these results, we suggest the balance model or flower model as the classification scheme of marketing performance.

• Journal of Korea Multimedia Society
• /
• v.19 no.2
• /
• pp.209-214
• /
• 2016

The purpose of this study was to classify healty persons and Parkinson disease patients from the vocal characteristics of healty persons and the of Parkinson disease patients using Machine Learning algorithms. So, we compared the most widely used algorithms for Machine Learning such as J48 algorithm and REPTree algorithm. In order to evaluate the classification performance of the two algorithms, the results were compared with depending on vocal characteristics. The classification performance of depending on vocal characteristics show 88.72% and 84.62%. The test results showed that the J48 algorithms was superior to REPTree algorithms.

• JSTS:Journal of Semiconductor Technology and Science
• /
• v.12 no.2
• /
• pp.162-167
• /
• 2012

This paper describes the design of a high-performance unified SVM classifier circuit. The proposed circuit supports both linear and non-linear SVM classifications. In order to ensure efficient classification, a 48x96 or 64x64 sliding window with 20 window strides is used. We reduced the circuit size by sharing most of the resources required for both types of classification. We described the proposed unified SVM classifier circuit using the Verilog HDL and synthesized the gate-level circuit using 65nm standard cell library. The synthesized circuit consists of 661,261 gates, operates at the maximum operating frequency of 152 MHz and processes up to 33.8 640x480 image frames per second.

• MALSORI
• /
• no.64
• /
• pp.77-88
• /
• 2007

Nowadays, mel-frequency cesptral coefficients (MFCCs) and Gaussian mixture models (GMMs) are used for the pathological voice detection. This paper suggests a method to improve the performance of the pathological/normal voice classification based on the MFCC-based GMM. We analyze the characteristics of the mel frequency-based filterbank energies using the fisher discriminant ratio (FDR). And the feature vectors through the linear discriminant analysis (LDA) transformation of the filterbank energies (FBE) and the MFCCs are implemented. An accuracy is measured by the GMM classifier. This paper shows that the FBE LDA-based GMM is a sufficiently distinct method for the pathological/normal voice classification, with a 96.6% classification performance rate. The proposed method shows better performance than the MFCC-based GMM with noticeable improvement of 54.05% in terms of error reduction.

• Proceedings of the Safety Management and Science Conference
• /
• 2006.04a
• /
• pp.51-56
• /
• 2006

Recently data mining techniques have been used for analysis and classification of data related to industrial accidents. The main objective of this study is to compare performance of algorithms for data analysis of industrial accidents and this paper provides a comparative analysis of 5 kinds of algorithms including CHAID, CART, C4.5, LR (Logistic Regression) and NN (Neural Network) with ROC chart, lift chart and response threshold. In this study, data on 67,278 accidents were analyzed to create risk groups for a number of complications, including the risk of disease and accident. The sample for this work chosen from data related to manufacturing industries during three years $(2002\sim2004)$ in korea. According to the result analysis, NN has excellent performance for data analysis and classification of industrial accidents.