• Journal of Advanced Marine Engineering and Technology
• /
• v.26 no.6
• /
• pp.695-704
• /
• 2002

Generally, induction motor controller requires rotor speed sensor for commutation and current control, but it increases cost and size of the motor. So in these days, various researches including speed sensorless vector control have been reported and some of them have been put to practical use. In this paper a new speed estimation method using neural networks is proposed. The optimal neural network structure was tracked down by trial and error, and it was found that the 8-16-1 neural network has given correct results for the instantaneous rotor speed. Supervised learning methods, through which the neural network is trained to learn the input/output pattern presented, are typically used. The back-propagation technique is used to adjust the neural network weights during training. The rotor speed is calculated by weights and eight inputs to the neural network. Also, the proposed method has advantages such as the independency on machine parameters, the insensitivity to the load condition, and the stability in the low speed operation.

• The Journal of Bigdata
• /
• v.6 no.1
• /
• pp.23-35
• /
• 2021

In the era of the 4th industrial revolution, smart factories have received great attention, where production and manufacturing technology and ICT converge. With the development of IoT technology and big data, automation of production systems has become possible. In the advanced manufacturing industry, production systems are subject to unscheduled performance degradation and downtime, and there is a demand to reduce safety risks by detecting and reparing potential errors as soon as possible. This study designs a model based on supervised and unsupervised learning for detecting anomalies. The accuracy of XGBoost, LightGBM, and CNN models was compared as a supervised learning analysis method. Through the evaluation index based on the confusion matrix, it was confirmed that LightGBM is most predictive (97%). In addition, as an unsupervised learning analysis method, MD, AE, and LSTM-AE models were constructed. Comparing three unsupervised learning analysis methods, the LSTM-AE model detected 75% of anomalies and showed the best performance. This study aims to contribute to the advancement of the smart factory by combining supervised and unsupervised learning techniques to accurately diagnose equipment failures and predict when abnormal situations occur, thereby laying the foundation for preemptive responses to abnormal situations. do.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.5
• /
• pp.102-108
• /
• 1996

Adaptive resonance theory (ART2) characterized by its built-in mechanism of handling the stability-plasticity switching and by the adaptive learning without forgetting informations learned in the past, is based on an unsupervised template matching. We propose an improved tow-stage learning algorithm for aRT2: the original unsupervised learning followed by a new supervised learning. Each of the output nodes, after the unsupervised learning, is labeled according to the category informations to reinforce the template pattern associated with the target output node belonging to the same category some dominant classes from exhausting a finite number of template patterns in ART2 inefficiently. Experimental results on a set of 2545 FLIR images show that the ART2 trained by the two-stage learning algorithm yields better accuracy than the original ART2, regardless of th esize of the network and the methods of evaluating the accuracy. This improvement shows the effectiveness of the two-stage learning process.

• Journal of architectural history
• /
• v.31 no.1
• /
• pp.41-50
• /
• 2022

Traditional wooden buildings deform over time and are vulnerable to fire or earthquakes. Therefore, traditional wooden buildings require continuous management and repair, and securing architectural drawings is essential for repair and restoration. Unlike modernized CAD drawings, traditional wooden building drawings scan and store hand-drawn drawings, and in this process, many noise is included due to damage to the drawing itself. These drawings are digitized, but their utilization is poor due to noise. Difficulties in systematic management of traditional wooden buildings are increasing. Noise removal by existing algorithms has limited drawings that can be applied according to noise characteristics and the performance is not uniform. This study presents deep artificial neural network based noised reduction for architectural drawings. Front/side elevation drawings, floor plans, detail drawings of Korean wooden treasure buildings were considered. First, the noise properties of the architectural drawings were learned with both a cycle generative model and heuristic image fusion methods. Consequently, a noise reduction network was trained through supervised learning using training sets prepared using the noise models. The proposed method provided effective removal of noise without deteriorating fine lines in the architectural drawings and it showed good performance for various noise types.

• Journal of Broadcast Engineering
• /
• v.27 no.4
• /
• pp.511-518
• /
• 2022

Supervised learning based on deep learning has made a leap forward in various application fields. However, many supervised learning methods work under the common assumption that training and test data are extracted from the same distribution. If it deviates from this constraint, the deep learning network trained in the training domain is highly likely to deteriorate rapidly in the test domain due to the distribution difference between domains. Domain adaptation is a methodology of transfer learning that trains a deep learning network to make successful inferences in a label-poor test domain (i.e., target domain) based on learned knowledge of a labeled-rich training domain (i.e., source domain). In particular, the unsupervised domain adaptation technique deals with the domain adaptation problem by assuming that only image data without labels in the target domain can be accessed. In this paper, we explore the unsupervised domain adaptation techniques.

• Journal of IKEEE
• /
• v.28 no.2
• /
• pp.145-149
• /
• 2024

Recently, research on temporal action localization has been actively conducted. In this paper, unlike existing methods, we propose two approaches for learning bidirectional information by creating reverse playback videos for fully supervised temporal action localization. One approach involves creating training data by combining reverse playback videos and forward playback videos, while the other approach involves training separate models on videos with different playback directions. Experiments were conducted on the THUMOS-14 dataset using TALLFormer. When using both reverse and forward playback videos as training data, the performance was 5.1% lower than that of the existing method. On the other hand, using a model ensemble shows a 1.9% improvement in performance.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.2
• /
• pp.159-167
• /
• 1996

In this paper, a new competitive learning algorithm called dynamic competitive learning (DCL) is presented. DCL is a supervised learning mehtod that dynamically generates output neuraons and nitializes weight vectors from training patterns. It introduces a new parameter called LOG (limit of garde) to decide whether or not an output neuron is created. In other words, if there exist some neurons in the province of LOG that classify the input vector correctly, then DCL adjusts the weight vector for the neuraon which has the minimum grade. Otherwise, it produces a new output neuron using the given input vector. It is largely learning is not limited only to the winner and the output neurons are dynamically generated int he trining process. In addition, the proposed algorithm has a small number of parameters. Which are easy to be determined and applied to the real problems. Experimental results for patterns recognition of remote sensing data and handwritten numeral data indicate the superiority of dCL in comparison to the conventional competitive learning methods.

• Asia pacific journal of information systems
• /
• v.32 no.2
• /
• pp.226-248
• /
• 2022

The existing model for sentiment analysis of product reviews learned from past data and new data was labeled based on training. But new data was never used by the existing system for making a decision. The proposed Aspect-based multi-agent Deep Reinforcement learning Sentiment Analysis (ADRSA) model learned from its very first data without the help of any training dataset and labeled a sentence with aspect category and sentiment polarity. It keeps on learning from the new data and updates its knowledge for improving its intelligence. The decision of the proposed system changed over time based on the new data. So, the accuracy of the sentiment analysis using deep reinforcement learning was improved over supervised learning and unsupervised learning methods. Hence, the sentiments of premium customers on a particular site can be explored to other customers effectively. A dynamic environment with a strong knowledge base can help the system to remember the sentences and usage State Action Reward State Action (SARSA) algorithm with Bidirectional Encoder Representations from Transformers (BERT) model improved the performance of the proposed system in terms of accuracy when compared to the state of art methods.

• Communications for Statistical Applications and Methods
• /
• v.12 no.3
• /
• pp.659-672
• /
• 2005

This paper evaluates discretization of continuous variables to select relevant variables for supervised learning using mutual information. Three discretization methods, MDL, Histogram and 4-Intervals are considered. The process of discretization and variable subset selection is evaluated according to the classification accuracies with the 6 real data sets of UCI databases. Results show that 4-Interval discretization method based on quantiles, is robust and efficient for variable selection process. We also visually evaluate the appropriateness of the selected subset of variables.

• International Journal of Contents
• /
• v.14 no.1
• /
• pp.1-6
• /
• 2018

Text, as one of the most influential inventions of humanity, has played an important role in human life since ancient times. The rich and precise information embodied in text is very useful in a wide range of vision-based applications such as the text data extracted from images that can provide information for automatic annotation, indexing, language translation, and the assistance systems for impaired persons. Therefore, natural-scene text detection with active research topics regarding computer vision and document analysis is very important. Previous methods have poor performances due to numerous false-positive and true-negative regions. In this paper, a fully-convolutional-network (FCN)-based method that uses supervised architecture is used to localize textual regions. The model was trained directly using images wherein pixel values were used as inputs and binary ground truth was used as label. The method was evaluated using ICDAR-2013 dataset and proved to be comparable to other feature-based methods. It could expedite research on text detection using deep-learning based approach in the future.