• The Journal of Bigdata
• /
• v.7 no.2
• /
• pp.11-29
• /
• 2022

Pre-trained language models are the most important and widely used tools in natural language processing tasks. Since those have been pre-trained for a large amount of corpus, high performance can be expected even with fine-tuning learning using a small number of data. Since the elements necessary for implementation, such as a pre-trained tokenizer and a deep learning model including pre-trained weights, are distributed together, the cost and period of natural language processing has been greatly reduced. Transformer variants are the most representative pre-trained language models that provide these advantages. Those are being actively used in other fields such as computer vision and audio applications. In order to make it easier for researchers to understand the pre-trained language model and apply it to natural language processing tasks, this paper describes the definition of the language model and the pre-learning language model, and discusses the development process of the pre-trained language model and especially representative Transformer variants.

• Transactions of Materials Processing
• /
• v.16 no.7
• /
• pp.502-508
• /
• 2007

Effect of the process parameters of the cogging process on the void crushing has been studied in order to identify the most effective factor. The process parameters considered in this study are die width ratio, reduction ratio and pre-cooling time before cogging process. Void crushing analysis with the selected process parameters was carried out using FE analysis. The results of FE analysis were evaluated by Taguchi method. It was found that the efficiency of void crushing increases with an increase in the values of all selected process parameters and the principal factor controlling the void crushing was identified as the reduction ratio.

• Plant Journal
• /
• v.8 no.4
• /
• pp.40-44
• /
• 2012

The pre-treatment of natural gas to remove $H_2S$ and $CO_2$ before liquefaction in natural gas processing is required, and amine-based absorption processes are widely used in gas processing. The current study aims to model amine-based absorption process and to find cost-effective design through systematic analysis of energy systems, together with column design. Different design options for absorber and stripper are investigated in a holistic manner, and heat integration technique has been applied to investigate how design of columns is interacted with energy efficiency for the pre-treatment process considered. Case study has been presented to demonstrate the applicability of heat integration method for improving energy efficiency in practice.

• International Journal of Computer Science & Network Security
• /
• v.21 no.3
• /
• pp.177-184
• /
• 2021

All over the world, people are affected by many chronic diseases and medical practitioners are working hard to find out the symptoms and remedies for the diseases. Many researchers focus on the feature detection of the disease and trying to get a better health recommendation system. It is necessary to detect the features automatically to provide the most relevant solution for the disease. This research gives the framework of Health Recommendation System (HRS) for identification of relevant and non-redundant features in the dataset for prediction and recommendation of diseases. This system consists of three phases such as Pre-processing, Feature Selection and Performance evaluation. It supports for handling of missing and noisy data using the proposed Imputation of missing data and noise detection based Pre-processing algorithm (IMDNDP). The selection of features from the pre-processed dataset is performed by proposed ensemble-based feature selection using an expert's knowledge (EFS-EK). It is very difficult to detect and monitor the diseases manually and also needs the expertise in the field so that process becomes time consuming. Finally, the prediction and recommendation can be done using Support Vector Machine (SVM) and rule-based approaches.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.14 no.1
• /
• pp.26-33
• /
• 2001

The proposed system was composed of pre-processor which was executing binary/high-pass filtering and post-processor which ranged from statistic data to prediction. In post-processor work, step one was filter process of image, step two was image recognition, and step three was destruction degree/time prediction. After these processing, we could predict image of the last destruction timestamp. This research was produced variation value according to growth of tree pattern. This result showed improved correction, when this research was applied image Processing. Pre-processing step of original image had good result binary work after high pas- filter execution. In the case of using partial discharge of the image, our research could predict the last destruction timestamp. By means of experimental data, this prediction system was acquired $\pm$3.2% error range.

• The Transactions of the Korea Information Processing Society
• /
• v.3 no.7
• /
• pp.1868-1879
• /
• 1996

There are limitations to process high bandwidth traffic in B-ISDN with mesh- topology single bus architecture of current mobile communication system. And, it is impossible to import ATM switch using fixed length packet rather than variable length packet. Some implementations are able to process variable length packet, but there are some problems such as pre-processing for synchronization and bit delay. In this paper, we design a concentrator that can manipulate variable length packet without additional pre-process. There is on bit delay for packet starting signal in input interface, So it is more efficient to process packets, such that the concentrator can reduce he processing time as $\ulcornerlog2N\lrcorne＋1$ bit-time rather than N bit-time delay in ordinary concentrator. It is expected that the mobile communication system with partial mesh topology bus adopting the knockout switch architecture can process high bandwidth traffic in B-ISDN.

• Journal of Environmental Impact Assessment
• /
• v.19 no.3
• /
• pp.315-321
• /
• 2010

ArcView and water network models have been integrated to develop the water network modeling system based upon GIS. To develop this system, pre, main, and post processing systems are required. GIS programming technique was adopted by using the ArcView's script language Avenue. The input data of models have been prepared by using the AutoCAD Map3D through the conversion of modeling input data to GIS data for A city. The modeling has been implemented by using EPANET, WaterCAD, InfoWorks. To develop the post processing system, the modeling results of the water network models have been analyzed by using GIS. During the application process of the developed system to B city with 300,000 population, main problems were found in the constructed GIS DB of that city. Thus, pilot study area of B city has been constructed, and pre-, main, and post-processing techniques were invented based upon GIS. Finally, the problems related to waterworks GIS projects in Korea were discussed and solutions were suggested.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.11a
• /
• pp.120-121
• /
• 2021

This study examines the performance of the pre-treatment process system to use CGS, a by-product generated in IGCC, as a concrete fine aggregate for construction materials, on the quality of CGS fine aggregate. As a result of the analysis, it is judged that the quality of fine aggregates of CGS can be improved at both density, absorption rate, and 0.08mm body passage amount after the hydroelectric screening process using water as a medium during the pretreatment process. It is believed that it can be used as basic data for national standard certification of CGS fine aggregates in the future.

• Proceedings of the Korean Powder Metallurgy Institute Conference
• /
• 2006.09b
• /
• pp.826-827
• /
• 2006

Copper infiltration is demonstrated as a viable solution to achieve higher mechanical properties by filling the interconnected porosities of a ferrous structure with copper infiltrant. This paper will present the results of a design of experiments study based on the selected processing variables in the copper infiltration process. The variables are the following: Infiltrating temperatures, infiltrating time at pre-heat zone and hot zone, the green density of iron part, the migration of copper into the iron part at different processing conditions. The results show the flexibility of the infiltration process to attain certain mechanical properties by changing the processing conditions.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.50 no.12
• /
• pp.590-596
• /
• 2001

The pre-emphasis is an essential process for speech signal processing. Widely used two methods are the typical method using a fixed value near unity and te optimal method using the autocorrelation ratio of the signal. This study proposes a new pre-emphasis method using the short-term energy difference of speech signal, which can effectively compensate the glottal source characteristics and lip radiation characteristics. Using the proposed pre-emphasis, speech analysis, such as spectrum estimation, formant detection, is performed and the results are compared with those of the conventional two pre-emphasis methods. The speech analysis with 5 single vowels showed that the proposed method enhanced the spectral shapes and gave nearly constant formant frequencies and could escape the overlapping of adjacent two formants. comparison with FFT spectra had verified the above results and showed the accuracy of the proposed method. The computational complexity of the proposed method reduced to about 50% of the optimal method.