• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.2
• /
• pp.43-49
• /
• 2020

In this paper, we propose a stage-wise knowledge transfer method that uses block-wise retraining to transfer the useful knowledge of a pre-trained residual network (ResNet) in a teacher-student framework (TSF). First, multiple hint information transfer and block-wise supervised retraining of the information was alternatively performed between teacher and student ResNet models. Next, Softened output information-based knowledge transfer was additionally considered in the TSF. The results experimentally showed that the proposed method using multiple hint-based bottom-up knowledge transfer coupled with incremental block-wise retraining provided the improved student ResNet with higher accuracy than existing KD and hint-based knowledge transfer methods considered in this study.

• Proceedings of the IEEK Conference
• /
• 2000.07b
• /
• pp.1021-1024
• /
• 2000

In this paper, a neuro-fuzzy system for quantitative characterization of chaotic signals is proposed. The proposed system is differ from the previous methods in that the nonlinear functions of the nonlinear dynamical systems are calculated as the invariant factor. In the proposed neuro-fuzzy system, the nonlinear functions are determined by supervised learning. From the reconstructed nonlinear functions, the proposed system can generate extrapolated chaotic signals. This feature will help the study of nonlinear dynamical systems which require large number of chaotic data. To confirm the validity of the proposed system, nonlinear functions are reconstructed from 1-dimensional and 2-dimensional chaotic signals.

• Journal of Advanced Information Technology and Convergence
• /
• v.10 no.1
• /
• pp.45-56
• /
• 2020

Despite extensive research, performance enhancement of keyphrase (KP) extraction remains a challenging problem in modern informatics. Recently, deep learning-based supervised approaches have exhibited state-of-the-art accuracies with respect to this problem, and several of the previously proposed methods utilize Bidirectional Encoder Representations from Transformers (BERT)-based language models. However, few studies have investigated the effective application of BERT-based fine-tuning techniques to the problem of KP extraction. In this paper, we consider the aforementioned problem in the context of scientific articles by investigating the fine-tuning characteristics of two distinct BERT models - BERT (i.e., base BERT model by Google) and SciBERT (i.e., a BERT model trained on scientific text). Three different datasets (WWW, KDD, and Inspec) comprising data obtained from the computer science domain are used to compare the results obtained by fine-tuning BERT and SciBERT in terms of KP extraction.

• International Journal of Computer Science & Network Security
• /
• v.22 no.10
• /
• pp.11-16
• /
• 2022

Applying predictive analytics to predict software defects has improved the overall quality and decreased maintenance costs. Many supervised and unsupervised learning algorithms have been used for defect prediction on publicly available datasets. Most of these datasets suffer from an imbalance in the output classes. We study the impact of class imbalance in the defect datasets on the efficiency of the defect prediction model and propose a CPP method for handling imbalances in the dataset. The performance of the methods is evaluated using measures like Matthew's Correlation Coefficient (MCC), Recall, and Accuracy measures. The proposed sampling technique shows significant improvement in the efficiency of the classifier in predicting defects.

• The Bulletin of The Korean Astronomical Society
• /
• v.39 no.2
• /
• pp.89-89
• /
• 2014

While the current consensus view holds that galaxy mergers are commonplace, it is sometimes speculated that Globular Clusters (GCs) may also have undergone merging events, possibly resulting in massive objects with a strong metallicity spread such as Omega Centauri. Galaxies are mostly far, unresolved systems whose mergers are most likely wet, resulting in observational as well as modeling difficulties, but GCs are resolved into stars that can be used as discrete dynamical tracers, and their mergers might have been dry, therefore easily simulated with an N-body code. It is however difficult to determine the observational parameters best suited to reveal a history of merging based on the positions and kinematics of GC stars, if evidence of merging is at all observable. To overcome this difficulty, we investigate the applicability of supervised and unsupervised machine learning to the automatic reconstruction of the dynamical history of a stellar system. In particular we test whether statistical clustering methods can classify simulated systems into monolithic versus merger products. We run direct N-body simulations of two identical King-model clusters undergoing a head-on collision resulting in a merged system, and other simulations of isolated King models with the same total number of particles as the merged system. After several relaxation times elapse, we extract a sample of snapshots of the sky-projected positions of particles from each simulation at different dynamical times, and we run a variety of clustering and classification algorithms to classify the snapshots into two subsets in a relevant feature space.

• Journal of Multimedia Information System
• /
• v.3 no.2
• /
• pp.35-42
• /
• 2016

Electrocardiogram (ECG) signal gives a clear indication whether the heart is at a healthy status or not as the early notification of a cardiac problem in the heart could save the patient's life. Several methods were launched to clarify how to diagnose the abnormality over the ECG signal waves. However, some of them face the problem of lack of accuracy at diagnosis phase of their work. In this research, we present an accurate and successive method for the diagnosis of abnormality through Discrete Wavelet Transform (DWT), QRS complex detection and Support Vector Machines (SVM) classification with overall accuracy rate 95.26%. DWT Refers to sampling any kind of discrete wavelet transform, while SVM is known as a model with related learning algorithm, which is based on supervised learning that perform regression analysis and classification over the data sample. We have tested the ECG signals for 10 patients from different file formats collected from PhysioNet database to observe accuracy level for each patient who needs ECG data to be processed. The results will be presented, in terms of accuracy that ranged from 92.1% to 97.6% and diagnosis status that is classified as either normal or abnormal factors.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.135-148
• /
• 2012

This paper proposes a model for recognizing sentential paraphrases through Predicate-Argument Tuple (PAT)-based approximate alignment between two texts. We cast the paraphrase recognition problem as a binary classification by defining and applying various alignment features which could effectively express the semantic relatedness between two sentences. Experiment confirmed the potential of our approach and error analysis revealed various paraphrase patterns not being solved by our system, which can help us devise methods for further performance improvement.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.47 no.4
• /
• pp.666-683
• /
• 2023

This research aims to develop based clothing design process tailored to the industry-university collaborative setting and apply it in a fashion design class. into three distinct phases: designing and organizing our fashion design class, conducting our class at a university, and gathering student feedback. First, we conducted a literature review on employing new technologies in traditional clothing design processes. We consulted with industry professionals from the Samsung C&T Fashion Group to develop an AI-based clothing design process. We then developed in-class learning activities that leveraged fashion brand product databases, a supervised learning AI model, and operating an AI-based Creativity Support Tool (CST). Next, we setup an industry-university fashion design class at a university in South Korea. Finally, we obtained feedback from undergraduate students who participated in the class. The survey results showed a satisfaction level of 4.7 out of 5. The evaluations confirmed that the instructional methods, communication, faculty, and student interactions within the class were both adequate and appropriate. These research findings highlighted that our AI-based clothing design process applied within the fashion design class led to valuable data-driven convergent thinking and technical experience beyond that of traditional clothing design processes.

• Journal of the Korea Convergence Society
• /
• v.8 no.1
• /
• pp.25-34
• /
• 2017

With the recent spread of smartphones and the introduction of web services, online users can access large-scale content regardless of time or place. However, users have had trouble finding the content they wanted among large-scale content. To solve this problem, user modeling and content recommendation system have been actively studied in various fields. However, in spite of active changes in senior groups according to the changes in information environment, research on user modeling and content recommendation system focused on senior groups are insufficient. In this paper, we propose a method of modeling smart senior based on their preference, and further develop a smart senior classification model using machine learning methods. As a result, we can not only grasp the preferences of smart seniors, but also develop a smart senior classification model, which is the foundation for the research of a recommendation system which will provide the activities and contents most suitable for senior groups.

• Healthcare Informatics Research
• /
• v.24 no.4
• /
• pp.309-316
• /
• 2018

Objectives: Both the valence and arousal components of affect are important considerations when managing mental healthcare because they are associated with affective and physiological responses. Research on arousal and valence analysis, which uses images, texts, and physiological signals that employ deep learning, is actively underway; research investigating how to improve the recognition rate is needed. The goal of this research was to design a deep learning framework and model to classify arousal and valence, indicating positive and negative degrees of emotion as high or low. Methods: The proposed arousal and valence classification model to analyze the affective state was tested using data from 40 channels provided by a dataset for emotion analysis using electrocardiography (EEG), physiological, and video signals (the DEAP dataset). Experiments were based on 10 selected featured central and peripheral nervous system data points, using long short-term memory (LSTM) as a deep learning method. Results: The arousal and valence were classified and visualized on a two-dimensional coordinate plane. Profiles were designed depending on the number of hidden layers, nodes, and hyperparameters according to the error rate. The experimental results show an arousal and valence classification model accuracy of 74.65 and 78%, respectively. The proposed model performed better than previous other models. Conclusions: The proposed model appears to be effective in analyzing arousal and valence; specifically, it is expected that affective analysis using physiological signals based on LSTM will be possible without manual feature extraction. In a future study, the classification model will be adopted in mental healthcare management systems.