• International journal of advanced smart convergence
• /
• v.12 no.1
• /
• pp.53-58
• /
• 2023

With the fast development of artificial intelligence day by day, users are demanding explanations about the results of algorithms and want to know what parameters influence the results. In this paper, we propose a model for bankruptcy prediction with interpretability using the SHAP framework. SHAP (SHAPley Additive exPlanations) is framework that gives a visualized result that can be used for explanation and interpretation of machine learning models. As a result, we can describe which features are important for the result of our deep learning model. SHAP framework Force plot result gives us top features which are mainly reflecting overall model score. Even though Fully Connected Neural Networks are a "black box" model, Shapley values help us to alleviate the "black box" problem. FCNNs perform well with complex dataset with more than 60 financial ratios. Combined with SHAP framework, we create an effective model with understandable interpretation. Bankruptcy is a rare event, then we avoid imbalanced dataset problem with the help of SMOTE. SMOTE is one of the oversampling technique that resulting synthetic samples are generated for the minority class. It uses K-nearest neighbors algorithm for line connecting method in order to producing examples. We expect our model results assist financial analysts who are interested in forecasting bankruptcy prediction of companies in detail.

• Steel and Composite Structures
• /
• v.49 no.1
• /
• pp.109-123
• /
• 2023

Headed studs welded to steel beams and embedded within the concrete of deck slabs are vital components of modern composite floor systems, where safety and economy depend on the accurate predictions of the stud shear resistance. The multitude of existing deck profiles and the complex behavior of studs in deck slab ribs makes developing accurate and reliable mechanical or empirical design models challenging. The paper addresses this issue by presenting a machine learning (ML) model developed from the natural gradient boosting (NGBoost) algorithm capable of producing probabilistic predictions and a database of 464 push-out tests, which is considerably larger than the databases used for developing existing design models. The proposed model outperforms models based on other ML algorithms and existing descriptive equations, including those in EC4 and AISC 360, while offering probabilistic predictions unavailable from other models and producing higher shear resistances for many cases. The present study also showed that the stud shear resistance is insensitive to the concrete elastic modulus, stud welding type, location of slab reinforcement, and other parameters considered important by existing models. The NGBoost model was interpreted by evaluating the feature importance and dependence determined with the SHapley Additive exPlanations (SHAP) method. The model was calibrated via reliability analyses in accordance with the Eurocodes to ensure that its predictions meet the required reliability level and facilitate its use in design. An interactive open-source web application was created and deployed to the cloud to allow for convenient and rapid stud shear resistance predictions with the developed model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.2
• /
• pp.230-235
• /
• 2005

In this paper, we propose an algorithm that improves a tone quality of a noisy audio signal in order to enhance a performance of perceptual filter using intelligent estimation of noise pattern from a band degraded by additive noise. The proposed method doesn't use the estimated noise which is obtained from silent range. Instead new estimated noise according to the power of signal and effect of noise variation is considered for each frame. So the noisy audio signal is enhanced by the method which controls a estimation of noise Pattern effectively in a noise corruption band. To show the performance of the proposed algorithm, various input signals which had a different signal-to-noise ratio(SNR) such as $5\cal{dB},\;10\cal{dB},\;15\cal{dB}\;and\;20\cal{dB}$ were used to test the proposed algorithm. we carry out SSNR and NMR of objective measurement and MOS test of subjective measurement. An approximate improvement of $7.4\cal{dB},\;6.8\cal{dB},\;5.7\cal{dB},\;5.1\cal{dB}$ in SSNR and $15.7\cal{dB},\;15.5\cal{dB},\;15.2\cal{dB},\;14.8\cal{dB}$ in NMR is achieved with the input signals, respectively. And we confirm the enhancement of tone quality in terms of mean opinion score(MOS) test which is result of subjective measurement.

• Journal of KIISE:Databases
• /
• v.31 no.6
• /
• pp.596-608
• /
• 2004

A lot of work has been done to reduce disk access time in I/O intensive systems, which store and handle massive amount of data, by distributing data across multiple disks and accessing them in parallel. Most of the previous work has focused on an efficient mapping from a grid cell to a disk number on the assumption that data space is regular grid-like partitioned. Although we can achieve good performance for low-dimensional data by grid-like partitioning, its performance becomes degenerate as grows the dimension of data even with a good disk allocation scheme. This comes from the fact that they partition entire data space equally regardless of distribution ratio of data objects. Most of the data in high-dimensional space exist around the surface of space. For that reason, we propose a new declustering algorithm based on the partitioning scheme which partition data space from the surface. With an unbalanced partitioning scheme, several experimental results show that we can remarkably reduce the number of data blocks touched by a query as grows the dimension of data and a query size. In this paper, we propose disk allocation schemes based on the layout of the resultant data blocks after partitioning. To show the performance of the proposed algorithm, we have performed several experiments with different dimensional data and for a wide range of number of disks. Our proposed disk allocation method gives a performance within 10 additive disk accesses compared with strictly optimal allocation scheme. We compared our algorithm with Kronecker sequence based declustering algorithm, which is reported to be the best among the grid partition and mapping function based declustering algorithms. We can improve declustering performance up to 14 times as grows dimension of data.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.15 no.3
• /
• pp.60-67
• /
• 2016

In cumulative transfer functions, as number of transfer increase, the impact of individual transfer to transfer cost increase linearly or non linearly. This function can effectively explain various passengers's travel behavior who choose their travel routes in integrated transit line networks including bus and railway modes. Using the function, it is possible to simulate general situations such that even though more travel times are expected, less number of transfer routes are preferred. However, because travel cost with cumulative transfer function is known as non additive cost function types in route search algorithms, finding an optimal route in integrated transit networks is confronted by the insolvable enumeration of all routes in many cases. This research proposes a methodology for finding an optimal path considering cumulative transfer function. For this purpose, the reversal phenomenon of optimal path generated in route search process is explained. Also a heuristic methodology for selecting an optimal route among multiple routes predefined by the K path algorithm. The incoming link based entire path deletion method is adopted for finding K ranking path thanks to the merit of security of route optimality condition. Through case studies the proposed methodology is discussed in terms of the applicability of real situations.

• KIPS Transactions on Computer and Communication Systems
• /
• v.6 no.7
• /
• pp.297-306
• /
• 2017

Citizens want more accurate forecast information using Bus Information System. However, most bus information systems that use an average based short-term prediction algorithm include many errors because they do not consider the effects of the traffic flow, signal period, and halting time. In this paper, we try to improve the precision of forecast information by analyzing the influencing factors of the error, thereby making the convenience of the citizens. We analyzed the influence factors of the error using BIS data. It is shown in the analyzed data that the effects of the time characteristics and geographical conditions are mixed, and that effects on halting time and passes speed is different. Therefore, the halt time is constructed using Generalized Additive Model with explanatory variable such as hour, GPS coordinate and number of routes, and we used Hidden Markov Model to construct a pattern considering the influence of traffic flow on the unit section. As a result of the pattern construction, accurate real-time forecasting and long-term prediction of route travel time were possible. Finally, it is shown that this model is suitable for travel time prediction through statistical test between observed data and predicted data. As a result of this paper, we can provide more precise forecast information to the citizens, and we think that long-term forecasting can play an important role in decision making such as route scheduling.

• School Mathematics
• /
• v.5 no.4
• /
• pp.421-440
• /
• 2003

It is difficult for the learner to understand completely the ratio concept which forms a basis of proportional reasoning. And proportional reasoning is, on the one hand, the capstone of children's elementary school arithmetic and, the other hand, it is the cornerstone of all that is to follow. But school mathematics has centered on the teachings of algorithm without dealing with its essence and meaning. The purpose of this study is to analyze the essence of ratio concept from multidimensional viewpoint. In addition, this study will show the direction for improvement of ratio concept. For this purpose, I tried to analyze the historical development of ratio concept. Most mathematicians today consider ratio as fraction and, in effect, identify ratios with what mathematicians called the denominations of ratios. But Euclid did not. In line with Euclid's theory, ratio should not have been represented in the same way as fraction, and proportion should not have been represented as equation, but in line with the other's theory they might be. The two theories of ratios were running alongside each other, but the differences between them were not always clearly stated. Ratio can be interpreted as a function of an ordered pair of numbers or magnitude values. A ratio is a numerical expression of how much there is of one quantity in relation to another quantity. So ratio can be interpreted as a binary vector which differentiates between the absolute aspect of a vector -its size- and the comparative aspect-its slope. Analysis on ratio concept shows that its basic structure implies 'proportionality' and it is formalized through transmission from the understanding of the invariance of internal ratio to the understanding of constancy of external ratio. In the study, a fittingness(or comparison) and a covariation were examined as the intuitive origins of proportion and proportional reasoning. These form the basis of the protoquantitative knowledge. The development of sequences of proportional reasoning was examined. The first attempts at quantifying the relationships are usually additive reasoning. Additive reasoning appears as a precursor to proportional reasoning. Preproportions are followed by logical proportions which refer to the understanding of the logical relationships between the four terms of a proportion. Even though developmental psychologists often speak of proportional reasoning as though it were a global ability, other psychologists insist that the evolution of proportional reasoning is characterized by a gradual increase in local competence.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.6
• /
• pp.446-453
• /
• 2004

In this paper, a new adaptive perceptual filter using variable threshold to enhance audio signals degraded by additively nonstationary noise is proposed. The adaptive perceptual filter updates variable threshold each time according to the power of signal and the effect of noise variation. So the noisy audio signal is enhanced by the method which controls a residual noise effectively. The proposed algorithm uses the perceptual filter which transforms a time domain signal into frequency domain and calculates an intensity energy and an excitation energy in bark domain. In this method. the stage updated the response of filter is decided by threshold. The proposed algorithm using vairable threshold effectively controls a residual noise using the energy difference of audio signals degraded by the additive nonstationary noise. The proposed method is tested with the noisy audio signals degraded by nonstationary noise at various signal -to-noise ratios (SNR). We carry out NMR and MOS test when the input SNR is 15dB. 20dB. 25dB and 30dB. An approximate improvement of 17.4dB. 15.3dB, 12.8dB. 9.8dB in NMR and enhancement of 2.9, 2.5, 2.3, 1.7 in MOS test is achieved with the input signals. respectively.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.22 no.6
• /
• pp.1231-1243
• /
• 1997

Digital satellite communication channels have nonlinearities with memory due to saturation characteristics of traveling wave tube amplifier in the satellite and transmitter/receiver linear filters. In this paper, we propose a network structure and a learning algorithm for complex pi-sigma network (CPSK) and exploit CPSN in the problem of equalization of nonlinear satellite channels. The proposed CPSN is a complex-valued extension of real-valued pi-sigma network that is a higher-order feedforward network with fast learning while greatly reducing network complexity by utilizing efficient form of polynomials for many input variables. The performance of the proposed CPSN is demonstrated by computer simulations on the equalization of complex-valued QPSK input symbols distorted by a nonlinear channel modeled as a Volterra series and additive noise. The results indicate that the CPSN shows good equalization performance, fast convergence, and less computations as compared to conventional higher-order models such as Volterra filters.

• Asian-Australasian Journal of Animal Sciences
• /
• v.13 no.10
• /
• pp.1347-1352
• /
• 2000

The objectives of the study were to examine non-genetic factors that influence somatic cell counts in dairy cattle and to estimate the genetic parameters of somatic cell counts. A total of 34, 097-test day somatic cell count records were obtained from the Zimbabwe Dairy Services Association (ZDSA). The data were from 5, 615 Holstein daughters of 390 sires and 2, 541 dams tested between May 1994 and December 1998. First lactation cows contributed 22, 147 records to the data set, while 11, 950 records were from second and later parity cows. The model for analysis included fixed effects of month of calving, year of calving, stage of lactation, calving interval and test date. Milk yield and age on test day were fitted in the model as covariates. The additive genetic effects pertaining to cows, sires and dams and the residual error were the random effects. The Average Information Restricted Maximum Likelihood algorithm was used for analysis. The heritability of somatic cell scores was low at $0.027{\pm}0.013$ for parity one cows and $0.087{\pm}0.031$ for parity two and above. Repeatability estimates were $0.22{\pm}0.01$ and $0.30{\pm}0.01$ for the two lactation groups, respectively. Genetic and phenotypic correlations between the somatic cell scores and test day milk production were small and negative. It seems that there is no genetic link between somatic cell counts and milk yield in Holstein cattle in Zimbabwe. The results also seem to indicate that somatic cell count is a trait that is mainly governed by environmental factors.