• Journal of Korea Multimedia Society
• /
• v.21 no.10
• /
• pp.1195-1202
• /
• 2018

Artificial neural network is widely used for its excellent performance and implementability. However, traditional neural network needs to learn the system from scratch, with the addition of new input data, the variation of the observation environment, or the change in the form of input/output data. To resolve such a problem, the technique of transfer learning has been proposed. Transfer learning constructs a newly developed target system partially updating existing system and hence provides much more efficient learning process. Until now, transfer learning is mainly studied in the field of image processing and is not yet widely employed in acoustic data processing. In this paper, focusing on the scalability of transfer learning, we apply the concept of transfer learning to the problem of guitar chord classification and evaluate its performance. For this purpose, we build a target system of convolutional neutral network (CNN) based 48 guitar chords classification system by applying the concept of transfer learning to a source system of CNN based 24 guitar chords classification system. We show that the system with transfer learning has performance similar to that of conventional system, but it requires only half the learning time.

• Journal of Internet Computing and Services
• /
• v.22 no.3
• /
• pp.9-16
• /
• 2021

This study aims to develop a human activity recognition (HAR) system as a Deep-Learning (DL) classification model, distinguishing various human activities. We solely rely on the signals from a wristband accelerometer worn by a person for the user's convenience. 3-axis sequential acceleration signal data are gathered within a predefined time-window-slice, and they are used as input to the classification system. We are particularly interested in developing a Deep-Learning model that can outperform conventional machine learning classification performance. A total of 13 activities based on the laboratory experiments' data are used for the initial performance comparison. We have improved classification performance using the Convolutional Neural Network (CNN) combined with an auto-encoder feature reduction and parameter tuning. With various publically available HAR datasets, we could also achieve significant improvement in HAR classification. Our CNN model is also compared against Recurrent-Neural-Network(RNN) with Long Short-Term Memory(LSTM) to demonstrate its superiority. Noticeably, our model could distinguish both general activities and near-identical activities such as sitting down on the chair and floor, with almost perfect classification accuracy.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.6
• /
• pp.1743-1758
• /
• 2023

Automatic modulation classification is a critical algorithm for non-cooperative communication systems. This paper addresses the challenging problem of closed-set and open-set signal modulation classification in complex channels. We propose a novel approach that incorporates a self-learning filter and center-loss in Deep Residual Shrinking Networks (DRSN) for closed-set modulation classification, and the Opendistance method for open-set modulation classification. Our approach achieves better performance than existing methods in both closed-set and open-set recognition. In closed-set recognition, the self-learning filter and center-loss combination improves recognition performance, with a maximum accuracy of over 92.18%. In open-set recognition, the use of a self-learning filter and center-loss provide an effective feature vector for open-set recognition, and the Opendistance method outperforms SoftMax and OpenMax in F1 scores and mean average accuracy under high openness. Overall, our proposed approach demonstrates promising results for automatic modulation classification, providing better performance in non-cooperative communication systems.

• International Journal of Computer Science & Network Security
• /
• v.24 no.2
• /
• pp.150-154
• /
• 2024

In recent years, there are extensive researches on the applications of machine learning to the automation and decision support for medical experts during disease detection. However, the performance of machine learning still needs improvement so that machine learning model produces result that is more accurate and reliable for disease detection. Selecting the hyper-parameter that could produce the possible maximum classification accuracy on medical dataset is the most challenging task in developing decision support systems with machine learning algorithms for medical dataset classification. Moreover, selecting the features that best characterizes a disease is another challenge in developing machine-learning model with better classification accuracy. In this study, we have proposed an optimized decision tree model for heart disease classification by using heart disease dataset collected from kaggle data repository. The proposed model is evaluated and experimental test reveals that the performance of decision tree improves when an optimal number of features are used for training. Overall, the accuracy of the proposed decision tree model is 98.2% for heart disease classification.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• Communications for Statistical Applications and Methods
• /
• v.26 no.6
• /
• pp.591-610
• /
• 2019

Deep Learning is one of the machine learning methods to find features from a huge data using non-linear transformation. It is now commonly used for supervised learning in many fields. In particular, Convolutional Neural Network (CNN) is the best technique for the image classification since 2012. For users who consider deep learning models for real-world applications, Keras is a popular API for neural networks written in Python and also can be used in R. We try examine the parameter estimation procedures of Deep Neural Network and structures of CNN models from basics to advanced techniques. We also try to figure out some crucial steps in CNN that can improve image classification performance in the CIFAR10 dataset using Keras. We found that several stacks of convolutional layers and batch normalization could improve prediction performance. We also compared image classification performances with other machine learning methods, including K-Nearest Neighbors (K-NN), Random Forest, and XGBoost, in both MNIST and CIFAR10 dataset.

• Journal of Computing Science and Engineering
• /
• v.5 no.2
• /
• pp.150-160
• /
• 2011

Automatic text classification has a long history and many studies have been conducted in this field. In particular, many machine learning algorithms and information retrieval techniques have been applied to text classification tasks. Even though much technical progress has been made in text classification, there is still room for improvement in text classification. In this paper, we will discuss remaining issues in improving text classification. In this paper, three improvement issues are presented including automatic training data generation, noisy data treatment and term weighting and indexing, and four actual studies and their empirical results for those issues are introduced. First, the semi-supervised learning technique is applied to text classification to efficiently create training data. For effective noisy data treatment, a noisy data reduction method and a robust text classifier from noisy data are developed as a solution. Finally, the term weighting and indexing technique is revised by reflecting the importance of sentences into term weight calculation using summarization techniques.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.33-40
• /
• 2023

Machine-learning techniques are discovering effective performance on data analytics. Classification and regression are supported for prediction on different kinds of data. There are various breeds of classification techniques are using based on nature of data. Threshold determination is essential to making better model for unlabelled data. In this paper, threshold value applied as range, based on min-max normalization technique for creating labels and multiclass classification performed on rainfall data. Binary classification is applied on autism data and classification techniques applied on child abuse data. Performance of each technique analysed with the evaluation metrics.

• Journal of the Korean Geographical Society
• /
• v.43 no.5
• /
• pp.761-774
• /
• 2008

Integration of GIS data and human expert knowledge into digital image processing has long been acknowledged as a necessity to improve remote sensing image analysis. We propose inductive machine learning algorithm for GIS data integration and rule-based classification method for land cover classification. Proposed method is tested with a land cover classification of a Landsat ETM+ multispectral image and GIS data layers including elevation, aspect, slope, distance to water bodies, distance to road network, and population density. Decision trees and production rules for land cover classification are generated by C5.0 inductive machine learning algorithm with 350 stratified random point samples. Production rules are used for land cover classification integrated with unsupervised ISODATA classification. Result shows that GIS data layers such as elevation, distance to water bodies and population density can be effectively integrated for rule-based image classification. Intuitive production rules generated by inductive machine learning are easy to understand. Proposed method demonstrates how various GIS data layers can be integrated with remotely sensed imagery in a framework of knowledge base construction to improve land cover classification.

• Journal of the Semiconductor & Display Technology
• /
• v.19 no.4
• /
• pp.121-125
• /
• 2020

With miniaturization of semiconductor, the manufacturing process become more complex, and undetected small changes in the state of the equipment have unexpectedly changed the process results. Fault detection classification (FDC) system that conducts more active data analysis is feasible to achieve more precise manufacturing process control with advanced machine learning method. However, applying machine learning, especially in supervised learning criteria, requires an arduous data labeling process for the construction of machine learning data. In this paper, we propose a semi-supervised learning to minimize the data labeling work for the data preprocessing. We employed equipment status variable identification (SVID) data and optical emission spectroscopy data (OES) in silicon etch with SF6/O2/Ar gas mixture, and the result shows as high as 95.2% of labeling accuracy with the suggested semi-supervised learning algorithm.