• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Journal of the Korean Dietetic Association
• /
• v.25 no.1
• /
• pp.44-58
• /
• 2019

Predicting the number of meals in a foodservice organization is an important decision-making process that is essential for successful food production, such as reducing the amount of residue, preventing menu quality deterioration, and preventing rising costs. Compared to other demand forecasts, the menu of dietary personnel includes diverse menus, and various dietary supplements include a range of side dishes. In addition to the menus, diverse subjects for prediction are very difficult problems. Therefore, the purpose of this study was to establish a method for predicting the number of meals including predictive modeling and considering various factors in addition to menus which are actually used in the field. For this purpose, 63 variables in eight categories such as the daily available number of people for the meals, the number of people in the time series, daily menu details, weekdays or seasons, days before or after holidays, weather and temperature, holidays or year-end, and events were identified as decision variables. An ensemble model using six prediction models was then constructed to predict the number of meals. As a result, the prediction error rate was reduced from 10%~11% to approximately 6~7%, which was expected to reduce the residual amount by approximately 40%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.510-512
• /
• 2022

Various studies are being conducted to improve the accuracy of fine dust, but there is a problem that deep learning models are not well learned due to various characteristics according to the concentration of fine dust. This paper proposes an EEMD-based fine dust concentration prediction model to decompose the characteristics of fine dust concentration and reflect the characteristics. After decomposing the fine dust concentration through EEMD, the final fine dust concentration value is derived by ensemble of the prediction results according to the characteristics derived from each. As a result of the model's performance evaluation, 91.7% of the fine dust concentration prediction accuracy was confirmed.

• International Journal of Contents
• /
• v.15 no.4
• /
• pp.8-15
• /
• 2019

This paper presents a solution of the 'Quick, Draw! Doodle Recognition Challenge' hosted by Google. Doodles are drawings comprised of concrete representational meaning or abstract lines creatively expressed by individuals. In this challenge, a doodle is presented as a sequence of sketches. From the view of at the sketch level, to learn the pattern of strokes representing a doodle, we propose a sequential model stacked with multiple convolution layers and Long Short-Term Memory (LSTM) cells following the attention mechanism [15]. From the view at the image level, we use multiple models pre-trained on ImageNet to recognize the doodle. Finally, an ensemble and a post-processing method using the minimum cost flow algorithm are introduced to combine multiple models in achieving better results. In this challenge, our solutions garnered 11th place among 1,316 teams. Our performance was 0.95037 MAP@3, only 0.4% lower than the winner. It demonstrates that our method is very competitive. The source code for this competition is published at: https://github.com/ngxbac/Kaggle-QuickDraw.

• KIPS Transactions on Computer and Communication Systems
• /
• v.8 no.2
• /
• pp.35-40
• /
• 2019

Traditional Malware Detection is susceptible for detecting malware which is modified by polymorphism or obfuscation technology. By learning patterns that are embedded in malware code, machine learning algorithms can detect similar behaviors and replace the current detection methods. Data must collected continuously in order to learn malicious code patterns that change over time. However, the process of storing and processing a large amount of malware files is accompanied by high space and time complexity. In this paper, an HDFS-based distributed processing system is designed to reduce space complexity and accelerate feature extraction time. Using a distributed processing system, we extract two API features based on filtering basis, 2-gram feature and APICFG feature and the generalization performance of ensemble learning models is compared. In experiments, the time complexity of the feature extraction was improved about 3.75 times faster than the processing time of a single computer, and the space complexity was about 5 times more efficient. The 2-gram feature was the best when comparing the classification performance by feature, but the learning time was long due to high dimensionality.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1773-1793
• /
• 2023

Automated machine learning, often referred to as "AutoML," is the process of automating the time-consuming and iterative procedures that are associated with the building of machine learning models. There have been significant contributions in this area across a number of different stages of accomplishing a data-mining task, including model selection, hyper-parameter optimization, and preprocessing method selection. Among them, preprocessing method selection is a relatively new and fast growing research area. The current work is focused on the recommendation of preprocessing methods, i.e., feature subset selection (FSS) algorithms. One limitation in the existing studies regarding FSS algorithm recommendation is the use of a single learner for meta-modeling, which restricts its capabilities in the metamodeling. Moreover, the meta-modeling in the existing studies is typically based on a single group of data characterization measures (DCMs). Nonetheless, there are a number of complementary DCM groups, and their combination will allow them to leverage their diversity, resulting in improved meta-modeling. This study aims to address these limitations by proposing an architecture for preprocess method selection that uses ensemble learning for meta-modeling, namely AutoFE-Sel. To evaluate the proposed method, we performed an extensive experimental evaluation involving 8 FSS algorithms, 3 groups of DCMs, and 125 datasets. Results show that the proposed method achieves better performance compared to three baseline methods. The proposed architecture can also be easily extended to other preprocessing method selections, e.g., noise-filter selection and imbalance handling method selection.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.3
• /
• pp.763-768
• /
• 2023

In this paper, the existing statistical method (ARIMA) and machine learning method (Informer) were developed and compared to predict the distribution volume of pharmaceuticals. It was found that a machine learning-based model is advantageous for daily data prediction, and it is effective to use ARIMA for monthly prediction and switch to Informer as the data increases. The prediction error rate (RMSE) was reduced by 26.6% compared to the previous method, and the prediction accuracy was improved by 13%, resulting in a result of 86.2%. Through this thesis, we find that there is an advantage of obtaining the best results by ensembleing statistical methods and machine learning methods. In addition, machine learning-based AI models can derive the best results through deep learning operations even in irregular situations, and after commercialization, performance is expected to improve as the amount of data increases.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1191-1205
• /
• 2022

Particulate matter(PM) among air pollutants with complex and widespread causes is classified according to particle size. Among them, PM_2.5 is very small in size and can cause diseases in the human respiratory tract or cardiovascular system if inhaled by humans. In order to prepare for these risks, state-centered management and preventable monitoring and forecasting are important. This study tried to predict PM_2.5 in Seoul, where high concentrations of fine dust occur frequently, using two ensemble models, random forest (RF) and extreme gradient boosting (XGB) using 15 local data assimilation and prediction system (LDAPS) weather-related factors, aerosol optical depth (AOD) and 4 chemical factors as independent variables. Performance evaluation and factor importance evaluation of the two models used for prediction were performed, and seasonal model analysis was also performed. As a result of prediction accuracy, RF showed high prediction accuracy of R² = 0.85 and XGB R² = 0.91, and it was confirmed that XGB was a more suitable model for PM_2.5 prediction than RF. As a result of the seasonal model analysis, it can be said that the prediction performance was good compared to the observed values with high concentrations in spring. In this study, PM_2.5 of Seoul was predicted using various factors, and an ensemble-based PM_2.5 prediction model showing good performance was constructed.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.4
• /
• pp.21-29
• /
• 2023

In this paper, we propose an AI-based system for automatically classifying industry and occupation codes in the population census. The accurate classification of industry and occupation codes is crucial for informing policy decisions, allocating resources, and conducting research. However, this task has traditionally been performed by human coders, which is time-consuming, resource-intensive, and prone to errors. Our system represents a significant improvement over the existing rule-based system used by the statistics agency, which relies on user-entered data for code classification. In this paper, we trained and evaluated several models, and developed an ensemble model that achieved an 86.76% match accuracy in industry and 81.84% in occupation, outperforming the best individual model. Additionally, we propose process improvement work based on the classification probability results of the model. Our proposed method utilizes an ensemble model that combines transfer learning techniques with pre-trained models. In this paper, we demonstrate the potential for AI-based systems to improve the accuracy and efficiency of population census data classification. By automating this process with AI, we can achieve more accurate and consistent results while reducing the workload on agency staff.

• Journal of Service Research and Studies
• /
• v.14 no.1
• /
• pp.27-43
• /
• 2024

The market share of online platform services in the used car market continues to expand. And The used car online platform service provides service users with specifications of vehicles, accident history, inspection details, detailed options, and prices of used cars. SUV vehicle type's share in the domestic automobile market will be more than 50% in 2023, Sales of Hybrid vehicle type are doubled compared to last year. And these vehicle types are also gaining popularity in the used car market. Prior research has proposed a used car price prediction model by executing a Machine Learning model for all vehicles or vehicles by brand. On the other hand, the popularity of SUV and Hybrid vehicles in the domestic market continues to rise, but It was difficult to find a study that proposed a used car price prediction model for these vehicle type. This study selects a used car price prediction model by vehicle type using vehicle specifications and options for Sedans, SUV, and Hybrid vehicles produced by domestic brands. Accordingly, after selecting feature through the Lasso regression model, which is a feature selection, the ensemble model was sequentially executed with the same sampling, and the best model by vehicle type was selected. As a result, the best model for all models was selected as the CBR model, and the contribution and direction of the features were confirmed by visualizing Tree SHAP Value for the best model for each model. The implications of this study are expected to propose a used car price prediction model by vehicle type to sales officials using online platform services, confirm the attribution and direction of features, and help solve problems caused by asymmetry fo information between them.