• The Journal of Society for e-Business Studies
• /
• v.26 no.3
• /
• pp.97-117
• /
• 2021

Predicting term deposit subscriptions is one of representative financial marketing in banks, and banks can build a prediction model using various customer information. In order to improve the classification accuracy for term deposit subscriptions, many studies have been conducted based on machine learning techniques. However, even if these models can achieve satisfactory performance, utilizing them is not an easy task in the industry when their decision-making process is not adequately explained. To address this issue, this paper proposes an explainable scheme for term deposit subscription forecasting. For this, we first construct several classification models using decision tree-based ensemble learning methods, which yield excellent performance in tabular data, such as random forest, gradient boosting machine (GBM), extreme gradient boosting (XGB), and light gradient boosting machine (LightGBM). We then analyze their classification performance in depth through 10-fold cross-validation. After that, we provide the rationale for interpreting the influence of customer information and the decision-making process by applying Shapley additive explanation (SHAP), an explainable artificial intelligence technique, to the best classification model. To verify the practicality and validity of our scheme, experiments were conducted with the bank marketing dataset provided by Kaggle; we applied the SHAP to the GBM and LightGBM models, respectively, according to different dataset configurations and then performed their analysis and visualization for explainable term deposit subscriptions.

• The Korean Journal of Archival Studies
• /
• no.37
• /
• pp.189-237
• /
• 2013

Since central governments began to establish and use the Standard Records Management System(RMS) in 2007, more and more local governments and other public organizations have constructed RMS. RMS is the essential tool for records management in electronic environments, but it is not known how well the functions of RMS reflect standards and practice related records management or how many records managers use RMS in performing their works. This paper deals with analyzing the evaluation of 'classification & disposal schedule management' function in RMS. 'Classification & disposal schedule management' function has 4 subfunctions of review of classification & preservation period, management of the schedule items, assignment of classification scheme and reclassification. Classification and disposal schedule is at the heart of intellectual control of records and core area of records management. So it is important to analyze whether this function plays well a role in RMS or not. This research carried out evaluation of function and use about classification & disposal schedule management in RMS. Functional evaluation is to compare and analyze how well RMS meets the functional requirements which home and foreign standards give. Use evaluation is to investigate how records managers use RMS in accomplishing their task of managing classification & disposal schedule and to look into what is the problem with the use. This paper could get the implications through the survey of records managers who are working at central governments, regional local governments and basic local governments. And these implications are considered in institutional, functional, use and administrative aspect. It is important to communicate with stakeholders so that 'classification & disposal schedule management' function, further, all functions of the RMS in practice of records management could be used smoothly. Users of RMS have to raise demands or call for technical solutions of the problems which come up in use, while RMS developers and administrators must make more of an effort to satisfy their demands, reflect them on the RMS and enhance the system.

• The KIPS Transactions:PartA
• /
• v.13A no.5 s.102
• /
• pp.387-398
• /
• 2006

In this paper, we propose a new page allocation scheme for flash memory file system. The proposed scheme allocates pages by exploiting the concept of Purity, which is defined as the fraction of blocks where valid Pages and invalid Pages are coexisted. The Pity determines the cost of block cleaning, that is, the portion of pages to be copied and blocks to be erased for block cleaning. To enhance the purity, the scheme classifies hot-modified data and cold-modified data and allocates them into different blocks. The hot/cold classification is based on both static properties such as attribute of data and dynamic properties such as the frequency of modifications. We have implemented the proposed scheme in YAFFS and evaluated its performance on the embedded board equipped with 400MHz XScale CPU, 64MB SDRAM, and 64MB NAND flash memory. Performance measurements have shown that the proposed scheme can reduce block cleaning time by up to 15.4 seconds with an average of 7.8 seconds compared to the typical YAFFS. Also, the enhancement becomes bigger as the utilization of flash memory increases.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.2
• /
• pp.187-196
• /
• 2006

A new GA-based methodology using information granules is suggested for the construction of fuzzy classifiers. The proposed scheme consists of three steps: selection of information granules, construction of the associated fuzzy sets, and tuning of the fuzzy rules. First, the genetic algorithm (GA) is applied to the development of the adequate information granules. The fuzzy sets are then constructed from the analysis of the developed information granules. An interpretable fuzzy classifier is designed by using the constructed fuzzy sets. Finally, the GA is utilized for tuning of the fuzzy rules, which can enhance the classification performance on the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example, the classification of the Iris data, is provided.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.2
• /
• pp.209-215
• /
• 2003

A new GA-based methodology using information granules is suggested for the construction of fuzzy classifiers. The proposed scheme consists of three steps: selection of information granules, construction of the associated fuzzy sets, and tuning of the fuzzy rules. First, the genetic algorithm (GA) is applied to the development of the adequate information granules. The fuzzy sets are then constructed from the analysis of the developed information granules. An interpretable fuzzy classifier is designed by using the constructed fuzzy sets. Finally, the GA are utilized for tuning of the fuzzy rules, which can enhance the classification performance on the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example, the classification of the Iris data, is provided.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.1
• /
• pp.107-113
• /
• 2003

A new GA-based methodology using information granules is suggested for the construction of fuzzy classifiers. The proposed scheme consists of three steps: selection of information granules, construction of the associated fuzzy sets, and tuning of the fuzzy rules. First, the genetic algorithm (GA) is applied to the development of the adequate information granules. The fuzzy sets are then constructed from the analysis of the developed information granules. An interpretable fuzzy classifier is designed by using the constructed fuzzy sets. Finally, the GA are utilized for tuning of the fuzzy rules, which can enhance the classification performance on the misclassified data (e.g., data with the strange pattern or on the boundaries of the classes). To show the effectiveness of the proposed method, an example, the classification of the Iris data, is provided.

• International Journal of Control, Automation, and Systems
• /
• v.4 no.3
• /
• pp.365-371
• /
• 2006

This paper presents an algorithm based on a combination of Discrete Wavelet Transforms and neural networks for detection and classification of internal faults in a two-winding three-phase transformer. Fault conditions of the transformer are simulated using ATP/EMTP in order to obtain current signals. The training process for the neural network and fault diagnosis decision are implemented using toolboxes on MATLAB/Simulink. Various cases and fault types based on Thailand electricity transmission and distribution systems are studied to verify the validity of the algorithm. It is found that the proposed method gives a satisfactory accuracy, and will be particularly useful in a development of a modern differential relay for a transformer protection scheme.

• Korean Journal of Veterinary Research
• /
• v.23 no.2
• /
• pp.165-172
• /
• 1983

The distribution of slide coagulase-negative staphylococci isolated from bovine mastitic milk samples was investigated mainly according to Kloos and Schleifer's classification scheme, and toxigenic and enzymatic characteristics of these strains were also examined. One-hundred-and-twenty-one strains of coagulase-negative staphylococci isolated were classified into 8 species. Of these species, Staphylococcus epidermidis, Staph. xylosus, Staph. haemolyticus and Staph. simulans were more frequently found in bovine mastitic milk samples, and toxin and enzyme production of these species were observed in relatively high frequency. Staph. hyicus subsp. hyicus was isolated from the 4 quarters with clinical mastitis included in this investigation. By the use of Baird-Parker and Pelzer's classification system, 44.6% and 18.2% of the strains could not be classified in any subgroup, respectively.

• 제어로봇시스템학회:학술대회논문집
• /
• 2004.08a
• /
• pp.1413-1415
• /
• 2004

This paper proposes a SVM(Support Vector Machine) training algorithm to control a service robot with voice command. The service robot with a stereo vision system and dual manipulators of four degrees of freedom implements a User-Dependent Voice Control System. The training of SVM algorithm that is one of the statistical learning theories leads to a QP(quadratic programming) problem. In this paper, we present an efficient SVM speech recognition scheme especially based on less learning data comparing with conventional approaches. SVM discriminator decides rejection or acceptance of user's extracted voice features by the MFCC(Mel Frequency Cepstrum Coefficient). Among several SVM kernels, the exponential RBF function gives the best classification and the accurate user recognition. The numerical simulation and the experiment verified the usefulness of the proposed algorithm.

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.2
• /
• pp.71.2-71.2
• /
• 2019

In order to gather clues to surface mineralogy of asteroids, we classify their taxonomy based on their reflected spectra. It is remarkable that a large number of asteroids plotted in the proper orbital element space with distinct colors according to their taxonomic types reveal the dynamical evolution and the structure in the near-Earth space, the main-belt and beyond. Although we have ~1×10⁶ known objects, no more than ~3×10³ of them are properly classified taxonomically as visible-near infrared spectroscopy is costly. On the other hand, multi-wavelength broadband photometry in the visible region provides a rather inexpensive alternative tool for approximate taxonomy. Thus we have conducted multi-band observations systematically using Korea Microlensing Telescope Network (KMTNet) with BVRI and griz filters since back in 2015. We then applied aperture photometry with elliptical apertures to fit the trails of objects during the exposures, and classified them with the principle component indices of Ivezic et al. (2001). We will make use of our new, three dimensional asteroid classification scheme for the next step.