• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.8
• /
• pp.2993-3010
• /
• 2021

The job shop scheduling problem (JSSP) plays a critical role in smart manufacturing, an effective JSSP scheduler could save time cost and increase productivity. Conventional methods are very time-consumption and cannot deal with complicated JSSP instances as it uses one optimal algorithm to solve JSSP. This paper proposes an effective scheduler based on deep learning technology named self-supervised long-short term memory (SS-LSTM) to handle complex JSSP accurately. First, using the optimal method to generate sufficient training samples in small-scale JSSP. SS-LSTM is then applied to extract rich feature representations from generated training samples and decide the next action. In the proposed SS-LSTM, two channels are employed to reflect the full production statues. Specifically, the detailed-level channel records 18 detailed product information while the system-level channel reflects the type of whole system states identified by the k-means algorithm. Moreover, adopting a self-supervised mechanism with LSTM autoencoder to keep high feature extraction capacity simultaneously ensuring the reliable feature representative ability. The authors implemented, trained, and compared the proposed method with the other leading learning-based methods on some complicated JSSP instances. The experimental results have confirmed the effectiveness and priority of the proposed method for solving complex JSSP instances in terms of make-span.

• Journal of Korean Institute of Industrial Engineers
• /
• v.37 no.1
• /
• pp.64-73
• /
• 2011

Oil price prediction is an important issue for the regulators of the government and the related industries. When employing the time series techniques for prediction, however, it becomes difficult and challenging since the behavior of the series of oil prices is dominated by quantitatively unexplained irregular external factors, e.g., supply- or demand-side shocks, political conflicts specific to events in the Middle East, and direct or indirect influences from other global economical indices, etc. Identifying and quantifying the relationship between oil price and those external factors may provide more relevant prediction than attempting to unclose the underlying structure of the series itself. Technically, this implies the prediction is to be based on the vectoral data on the degrees of the relationship rather than the series data. This paper proposes a novel method for time series prediction of using Semi-Supervised Learning that was originally designed only for the vector types of data. First, several time series of oil prices and other economical indices are transformed into the multiple dimensional vectors by the various types of technical indicators and the diverse combination of the indicator-specific hyper-parameters. Then, to avoid the curse of dimensionality and redundancy among the dimensions, the wellknown feature extraction techniques, PCA and NLPCA, are employed. With the extracted features, a timepointspecific similarity matrix of oil prices and other economical indices is built and finally, Semi-Supervised Learning generates one-timepoint-ahead prediction. The series of crude oil prices of West Texas Intermediate (WTI) was used to verify the proposed method, and the experiments showed promising results : 0.86 of the average AUC.

• Journal of KIISE
• /
• v.43 no.2
• /
• pp.269-274
• /
• 2016

Named entity recognition is a process which extracts named entities in sentences and determines categories of the named entities. Previous studies on named entity recognition have primarily been used for supervised learning. For supervised learning, a large training corpus manually annotated with named entity categories is needed, and it is a time-consuming and labor-intensive job to manually construct a large training corpus. We propose a semi-supervised learning method to minimize the cost needed for training corpus construction and to rapidly enhance the performance of named entity recognition. The proposed method uses distance supervision for the construction of the initial training corpus. It can then effectively remove noise sentences in the initial training corpus through the use of an active bagging method, an ensemble method of bagging and active learning. In the experiments, the proposed method improved the F1-score of named entity recognition from 67.36% to 76.42% after active bagging for 15 times.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.5
• /
• pp.187-197
• /
• 2020

Recently, various methods of text embedding using deep learning algorithms have been proposed. Especially, the way of using pre-trained language model which uses tremendous amount of text data in training is mainly applied for embedding new text data. However, traditional pre-trained language model has some limitations that it is hard to understand unique context of new text data when the text has too many tokens. In this paper, we propose self-supervised learning-based fine tuning method for pre-trained language model to infer vectors of long-text. Also, we applied our method to news articles and classified them into categories and compared classification accuracy with traditional models. As a result, it was confirmed that the vector generated by the proposed model more accurately expresses the inherent characteristics of the document than the vectors generated by the traditional models.

• Journal of radiological science and technology
• /
• v.44 no.6
• /
• pp.655-661
• /
• 2021

Positron emission tomography (PET) images is affected by acquisition time, short acquisition times results in low gamma counts leading to degradation of image quality by statistical noise. Noise2Void(N2V) is self supervised denoising model that is convolutional neural network (CNN) based deep learning. The purpose of this study is to evaluate denoising performance of N2V for PET image with a short acquisition time. The phantom was scanned as a list mode for 10 min using Biograph mCT40 of PET/CT (Siemens Healthcare, Erlangen, Germany). We compared PET images using NEMA image-quality phantom for standard acquisition time (10 min), short acquisition time (2min) and simulated PET image (S2 min). To evaluate performance of N2V, the peak signal to noise ratio (PSNR), normalized root mean square error (NRMSE), structural similarity index (SSIM) and radio-activity recovery coefficient (RC) were used. The PSNR, NRMSE and SSIM for 2 min and S2 min PET images compared to 10min PET image were 30.983, 33.936, 9.954, 7.609 and 0.916, 0.934 respectively. The RC for spheres with S2 min PET image also met European Association of Nuclear Medicine Research Ltd. (EARL) FDG PET accreditation program. We confirmed generated S2 min PET image from N2V deep learning showed improvement results compared to 2 min PET image and The PET images on visual analysis were also comparable between 10 min and S2 min PET images. In conclusion, noisy PET image by means of short acquisition time using N2V denoising network model can be improved image quality without underestimation of radioactivity.

• Journal of Korea Foundry Society
• /
• v.42 no.6
• /
• pp.369-376
• /
• 2022

The die-casting process is an important process for various industries, but there are limitations in the profitability and productivity of related companies due to the high defect rate. In order to overcome this, this study has developed die-casting fault detection modules based on industrial AI technologies. The developed module is constructed from three-stage models depending on the characteristics of the dataset. The first-stage model conducts fault detection based on supervised learning from the dataset without labels. The second-stage model realizes one-class classification based on semi-supervised learning, where the dataset only has production success labels. The third-stage model corresponds to fault detection based on supervised learning, where the dataset includes a small amount of production failure cases. The developed fault detection module exhibited outstanding performance with roughly 96% accuracy for actual process data.

• 한국정보교육학회:학술대회논문집
• /
• 2021.08a
• /
• pp.213-221
• /
• 2021

With the advent of the 4th industrial revolution, the need for artificial intelligence education has increased. The online learning environment caused by COVID-19 made it possible to use variety of artificial intelligence platforms. In this study, an aritificial intelligence class plan was developed and proposed to achieve the goal of artificial intelligence education using an AI platform. The AI platform used is AI for Oceans, With the theme of creating a program for the environment, designed a 6-hour project class using Novel Engineering-based on STEAM model. Students experience AI for Oceans enough time and learn supervised learning by experience. Based on understanding of supervised learning, students design their own programs for the environment using Entry's AI blocks. In this study, for AI convergence education, this lesson was developed and presented with the goal of acquiring the creative problem solving ability and integrated thinking ability by using the principles of artificial intelligence to solve problems.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.380-382
• /
• 2022

Extracting informative representation of molecules using graph neural networks(GNNs) is crucial in AI-driven drug discovery. Recently, the graph research community has been trying to replicate the success of self supervised in natural language processing, with several successes claimed. However, we find the benefit brought by self-supervised learning on applying varitional auto-encoders can be potentially effective on molecular data.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.6 no.1
• /
• pp.7-17
• /
• 1997

During the past decade, there were many well-established theories for the adaptive control of linear systems, but there exists relatively little general theory for the adaptive control of nonlinear systems. Adaptive control technique is essential for providing a stable and robust performance for application of industrial robot control. Neural network computing methods provide one approach to the development of adaptive and learning behavior in robotic system for manufacturing. Computational neural networks have been demonstrated which exhibit capabilities for supervised learning, matching, and generalization for problems on an experimental scale. Supervised learning could improve the efficiency of training and development of robotic systems. In this paper, a new scheme of adaptive-neuro control system to implement real-time control of robot manipulator using digital signal processors is proposed. Digital signal processors, DSPs, are micro-processors that are developed particularly for fast numerical computations involving sums and products of variables. The proposed neuro control algorithm is one of learning a model based error back-propagation scheme using Lyapunov stability analysis method. The proposed adaptive-neuro control scheme is illustrated to be an efficient control scheme for implementation of real-time control for SCARA robot with four-axes by experiment.

• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.457-462
• /
• 2006

In many applications, an enormous amount of unlabeled data is available with little cost. Therefore, it is natural to ask whether we can take advantage of these unlabeled data in classification learning. In this paper, we analyzed the role of unlabeled data in the context of naive Bayesian learning. Experimental results show that including unlabeled data as part of training data can significantly improve the performance of classification accuracy. The effect of using unlabeled data is especially important in case labeled data are sparse.