• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.624-626
• /
• 2023

Credit scoring is a technique used by financial institutions to assess the creditworthiness of potential borrowers. This involves evaluating a borrower's credit history to predict the likelihood of defaulting on a loan. This paper presents an ensemble of two Transformer based models within a framework for discriminating the default risk of loan applications in the field of credit scoring. The first model is FinBERT, a pretrained NLP model to analyze sentiment of financial text. The second model is FT-Transformer, a simple adaptation of the Transformer architecture for the tabular domain. Both models are trained on the same underlying data set, with the only difference being the representation of the data. This multi-modal approach allows us to leverage the unique capabilities of each model and potentially uncover insights that may not be apparent when using a single model alone. We compare our model with two famous ensemble-based models, Random Forest and Extreme Gradient Boosting.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.27 no.1
• /
• pp.115-127
• /
• 2024

This research aimed to construct models with various structures based on the Transformer module and to perform land cover classification, thereby examining the applicability of the Transformer module. For the classification of land cover, the Unet model, which has a CNN structure, was selected as the base model, and a total of four deep learning models were constructed by combining both the encoder and decoder parts with the Transformer module. During the training process of the deep learning models, the training was repeated 10 times under the same conditions to evaluate the generalization performance. The evaluation of the classification accuracy of the deep learning models showed that the Model D, which utilized the Transformer module in both the encoder and decoder structures, achieved the highest overall accuracy with an average of approximately 89.4% and a Kappa coefficient average of about 73.2%. In terms of training time, models based on CNN were the most efficient. however, the use of Transformer-based models resulted in an average improvement of 0.5% in classification accuracy based on the Kappa coefficient. It is considered necessary to refine the model by considering various variables such as adjusting hyperparameters and image patch sizes during the integration process with CNN models. A common issue identified in all models during the land cover classification process was the difficulty in detecting small-scale objects. To improve this misclassification phenomenon, it is deemed necessary to explore the use of high-resolution input data and integrate multidimensional data that includes terrain and texture information.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.9
• /
• pp.17-25
• /
• 2023

In this paper, we propose a model that can perform human pose estimation through a MobileViT-based model with fewer parameters and faster estimation. The based model demonstrates lightweight performance through a structure that combines features of convolutional neural networks with features of Vision Transformer. Transformer, which is a major mechanism in this study, has become more influential as its based models perform better than convolutional neural network-based models in the field of computer vision. Similarly, in the field of human pose estimation, Vision Transformer-based ViTPose maintains the best performance in all human pose estimation benchmarks such as COCO, OCHuman, and MPII. However, because Vision Transformer has a heavy model structure with a large number of parameters and requires a relatively large amount of computation, it costs users a lot to train the model. Accordingly, the based model overcame the insufficient Inductive Bias calculation problem, which requires a large amount of computation by Vision Transformer, with Local Representation through a convolutional neural network structure. Finally, the proposed model obtained a mean average precision of 0.694 on the MS COCO benchmark with 3.28 GFLOPs and 9.72 million parameters, which are 1/5 and 1/9 the number compared to ViTPose, respectively.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.22 no.8
• /
• pp.515-522
• /
• 2010

In order to improve the efficiency of a transformer cooling system, the intelligent algorithm was developed. The intelligent algorithm is composed of a setpoint algorithm and a control algorithm. The setpoint algorithm was developed by the neural network, and the control algorithm was developed by the fuzzy logic. These algorithms were used for the control of a blower and an oil pump of the transformer cooling system. In order to analyse performances of these algorithms, the dynamic model of a transformer cooling system was used. Based on various performance tests, energy savings and stable controls of a transformer cooling system were observed. Therefore, control algorithms developed for this study may be effectively used for the control of a transformer cooling system.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.51 no.12
• /
• pp.624-629
• /
• 2002

This paper presents the first Simulation model using EMTDC program to analyze the electrified train voltage distribution characteristics in ac auto-transformer 1110 railroads. In general, all of the electrified train supply system has the characteristics that the train supply line is a naturally non-symmetrical and unbalanced system. Also, it is needed to model the Scott transformer which invert the balanced 3-phase quantity into 2-phase. Therefore, the general simulation methodology using previous simplified equivalent circuit or RMS based program can't obtain the accurate results to reflect the real-time operation because these methodology is basically assumed on completely 3-phase balanced system. To overcome these defects, in this paper, the EMTDC simulation model to analysis the completely electrified railroad system with Scott transformer and AC auto-transformer is presented. Also, the correctness of EMTDC modeling is confirmed by the old basic concepts and we think that this EMTDC model has the future powerful capability for application of railroad system analysis.

• Journal of the Korean Society for Railway
• /
• v.5 no.4
• /
• pp.253-259
• /
• 2002

This paper presents the first simulation model using EMTDC program to analyze the electrified train voltage distribution characteristics in ac auto-transformer fed railroads. In general, all of the electrified train supply system has the characteristics that the train supply line is a naturally non-symmetrical and unbalanced system. Also, it is needed to model the Scott transformer which invert the balanced 3-phase quantity into 2-phase. Therefore, the general simulation methodology using previous simplified equivalent circuit or RMS based program can't obtain the accurate results to reflect the real-time operation because these methodology is basically assumed on completely 3-phase balanced system. To overcome these defects, in this paper, the EMTDC simulation model to analysis the completely electrified railroad system with Scott transformer and AC auto-transformer is presented. Also, the correctness of EMTDC modeling is confirmed by the old basic concepts and we think that this EMTDC model has the future powerful capability for application of railroad system analysis.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.20 no.7
• /
• pp.48-57
• /
• 2021

The main research content of this paper is to evaluate the structural integrity and the cooling performance of 4250 kVA power transformer with ONAN(Oil Natural and Air Natural) mode. The dynamic analysis is used to verify the structural safety of the transformer by seismic loading. The transformer structure is simplified and NX software is used to build a three-dimensional model, and ANSYS commercial software is used to calculate the stress and deformation by applying corresponding load. The analysis result was evaluated whether it satisfies the design requirements according to the IEEE Std 693 standard. In terms of thermal analysis to evaluate the cooling performance, the thermal physical model is used to calculate the heat exchange between the radiator and the tank in the steady state, and the result is input into the Fluent software to calculate the internal temperature field of the transformer tank, which reduces the calculation cost of thermal fluid. Comparing the simulated hot spot temperature and top oil temperature of the transformer with the calculation results of the IEC60076 classic model, it is found that the error is only 1.9%.

• Proceedings of the KIEE Conference
• /
• 2002.07b
• /
• pp.607-612
• /
• 2002

This paper presents the first domestic model product of hybrid Insulation low-loss type pole transformer, with increased overload ability, increased life, reduced loss, focusing improved short-circuit characteristics. The capacity of this transformer is 100 KVZ, rates first voltage 13200 V and secondary voltage 230/115 type. The volume of model transformer that was produced in high temperature using hybrid insulation can be reduced to 10-20% in comparison with that of present transformer with the same capacity. In this paper, we did short-circuit test and tested general characteristics of model transformer. In addition, we suggested design application new concept and methods thorough this experiment.

• Proceedings of the KIEE Conference
• /
• 2006.11a
• /
• pp.151-153
• /
• 2006

This paper describes the transient phenomena considering transformer model in combined distribution systems with power cables. To evaluate the overvoltage, the change of transformer model and system structure are considered. Transformer parameters calculated by BCTRAN of EMTP are considered to evaluate surge generation in underground distribution systems when underground distribution system has various parameters. It is evaluated that overvoltage value increases and decreases according to the structure of system. However it is confirmed that there is not much effect to overvoltage as transformer is operated in distribution line.

• Smart Media Journal
• /
• v.13 no.4
• /
• pp.9-15
• /
• 2024

Tree-based algorithms have been the dominant methods used build a prediction model for tabular data. This also includes personal credit data. However, they are limited to compatibility with categorical and numerical data only, and also do not capture information of the relationship between other features. In this work, we proposed an ensemble model using the Transformer architecture that includes text features and harness the self-attention mechanism to tackle the feature relationships limitation. We describe a text formatter module, that converts the original tabular data into sentence data that is fed into FinBERT along with other text features. Furthermore, we employed FT-Transformer that train with the original tabular data. We evaluate this multi-modal approach with two popular tree-based algorithms known as, Random Forest and Extreme Gradient Boosting, XGBoost and TabTransformer. Our proposed method shows superior Default Recall, F1 score and AUC results across two public data sets. Our results are significant for financial institutions to reduce the risk of financial loss regarding defaulters.