• Journal of Intelligence and Information Systems
• /
• v.24 no.3
• /
• pp.21-44
• /
• 2018

In recent years, the rapid development of internet technology and the popularization of smart devices have resulted in massive amounts of text data. Those text data were produced and distributed through various media platforms such as World Wide Web, Internet news feeds, microblog, and social media. However, this enormous amount of easily obtained information is lack of organization. Therefore, this problem has raised the interest of many researchers in order to manage this huge amount of information. Further, this problem also required professionals that are capable of classifying relevant information and hence text classification is introduced. Text classification is a challenging task in modern data analysis, which it needs to assign a text document into one or more predefined categories or classes. In text classification field, there are different kinds of techniques available such as K-Nearest Neighbor, Naïve Bayes Algorithm, Support Vector Machine, Decision Tree, and Artificial Neural Network. However, while dealing with huge amount of text data, model performance and accuracy becomes a challenge. According to the type of words used in the corpus and type of features created for classification, the performance of a text classification model can be varied. Most of the attempts are been made based on proposing a new algorithm or modifying an existing algorithm. This kind of research can be said already reached their certain limitations for further improvements. In this study, aside from proposing a new algorithm or modifying the algorithm, we focus on searching a way to modify the use of data. It is widely known that classifier performance is influenced by the quality of training data upon which this classifier is built. The real world datasets in most of the time contain noise, or in other words noisy data, these can actually affect the decision made by the classifiers built from these data. In this study, we consider that the data from different domains, which is heterogeneous data might have the characteristics of noise which can be utilized in the classification process. In order to build the classifier, machine learning algorithm is performed based on the assumption that the characteristics of training data and target data are the same or very similar to each other. However, in the case of unstructured data such as text, the features are determined according to the vocabularies included in the document. If the viewpoints of the learning data and target data are different, the features may be appearing different between these two data. In this study, we attempt to improve the classification accuracy by strengthening the robustness of the document classifier through artificially injecting the noise into the process of constructing the document classifier. With data coming from various kind of sources, these data are likely formatted differently. These cause difficulties for traditional machine learning algorithms because they are not developed to recognize different type of data representation at one time and to put them together in same generalization. Therefore, in order to utilize heterogeneous data in the learning process of document classifier, we apply semi-supervised learning in our study. However, unlabeled data might have the possibility to degrade the performance of the document classifier. Therefore, we further proposed a method called Rule Selection-Based Ensemble Semi-Supervised Learning Algorithm (RSESLA) to select only the documents that contributing to the accuracy improvement of the classifier. RSESLA creates multiple views by manipulating the features using different types of classification models and different types of heterogeneous data. The most confident classification rules will be selected and applied for the final decision making. In this paper, three different types of real-world data sources were used, which are news, twitter and blogs.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.5
• /
• pp.949-959
• /
• 2012

This article is to find the right size of decision trees that performs better for boosting algorithm. First we defined the tree size D as the depth of a decision tree. Then we compared the performance of boosting algorithm with different tree sizes in the experiment. Although it is an usual practice to set the tree size in boosting algorithm to be small, we figured out that the choice of D has a significant influence on the performance of boosting algorithm. Furthermore, we found out that the tree size D need to be sufficiently large for some dataset. The experiment result shows that there exists an optimal D for each dataset and choosing the right size D is important in improving the performance of boosting. We also tried to find the model for estimating the right size D suitable for boosting algorithm, using variables that can explain the nature of a given dataset. The suggested model reveals that the optimal tree size D for a given dataset can be estimated by the error rate of stump tree, the number of classes, the depth of a single tree, and the gini impurity.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.53-67
• /
• 2020

This study employees a supervised learning prediction model to detect nonconformity in advance of processed food manufacturing and processing businesses. The study was conducted according to the standard procedure of machine learning, such as definition of objective function, data preprocessing and feature engineering and model selection and evaluation. The dependent variable was set as the number of supervised inspection detections over the past five years from 2014 to 2018, and the objective function was to maximize the probability of detecting the nonconforming companies. The data was preprocessed by reflecting not only basic attributes such as revenues, operating duration, number of employees, but also the inspections track records and extraneous climate data. After applying the feature variable extraction method, the machine learning algorithm was applied to the data by deriving the company's risk, item risk, environmental risk, and past violation history as feature variables that affect the determination of nonconformity. The f1-score of the decision tree, one of ensemble models, was much higher than those of other models. Based on the results of this study, it is expected that the official food control for food safety management will be enhanced and geared into the data-evidence based management as well as scientific administrative system.

• Atmosphere
• /
• v.30 no.4
• /
• pp.439-454
• /
• 2020

This study investigates the change in the fine particulate matter (PM2.5) concentration and World Health Organization (WHO) air quality index (AQI) in East Asia (EA) under Shared Socioeconomic Pathways (SSPs). AQI is an indicator of increasing levels about health concern, divided into six categories based on PM2.5 annual concentrations. Here, we utilized the ensemble results of UKESM1, the climate model operated in Met Office, UK, for the analysis of long-term variation during the historical (1950~2014) and future (2015~2100) period. The results show that the spatial distributions of simulated PM2.5 concentrations in present-day (1995~2014) are comparable to observations. It is found that most regions in EA exceeded the WHO air quality guideline except for Japan, Mongolia regions, and the far seas during the historical period. In future scenarios containing strong air quality (SSP1-2.6, SSP5-8.5) and medium air quality (SSP2-4.5) controls, PM2.5 concentrations are substantially reduced, resulting in significant improvement in AQI until the mid-21st century. On the other hand, the mild air pollution controls in SSP3-7.0 tend to lead poor AQI in China and Korea. This study also examines impact of increased in PM2.5 concentrations on downward shortwave energy at the surface. As a result, strong air pollution controls can improve air quality through reduced PM2.5 concentrations, but lead to an additional warming in both the near and mid-term future climate over EA.

• Tunnel and Underground Space
• /
• v.32 no.6
• /
• pp.502-517
• /
• 2022

As the use of TBM increases, research has recently increased to to analyze TBM data with machine learning techniques to predict the exchange cycle of disc cutters, and predict the advance rate of TBM. In this study, a regression prediction of disc cutte wear of slurry shield TBM site was made by combining machine learning based on the machine data and the geotechnical data obtained during the excavation. The data were divided into 7:3 for training and testing the prediction of disc cutter wear, and the hyper-parameters are optimized by cross-validated grid-search over a parameter grid. As a result, gradient boosting based on the ensemble model showed good performance with a determination coefficient of 0.852 and a root-mean-square-error of 3.111 and especially excellent results in fit times along with learning performance. Based on the results, it is judged that the suitability of the prediction model using data including mechanical data and geotechnical information is high. In addition, research is needed to increase the diversity of ground conditions and the amount of disc cutter data.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.119-137
• /
• 2022

Recent recommendation system studies apply various deep learning models to represent user and item interactions better. One of the noteworthy studies is ONCF(Outer product-based Neural Collaborative Filtering) which builds a two-dimensional interaction map via outer product and employs CNN (Convolutional Neural Networks) to learn high-order correlations from the map. However, ONCF has limitations in recommendation performance due to the problems with CNN and the absence of side information. ONCF using CNN has an inductive bias problem that causes poor performances for data with a distribution that does not appear in the training data. This paper proposes to employ a Vision Transformer (ViT) instead of the vanilla CNN used in ONCF. The reason is that ViT showed better results than state-of-the-art CNN in many image classification cases. In addition, we propose a new architecture to reflect side information that ONCF did not consider. Unlike previous studies that reflect side information in a neural network using simple input combination methods, this study uses an independent auxiliary classifier to reflect side information more effectively in the recommender system. ONCF used a single latent vector for user and item, but in this study, a channel is constructed using multiple vectors to enable the model to learn more diverse expressions and to obtain an ensemble effect. The experiments showed our deep learning model improved performance in recommendation compared to ONCF.

• Atmosphere
• /
• v.34 no.2
• /
• pp.123-138
• /
• 2024

Using 18 multi-model-based a Shared Socioeconomic Pathway (SSP) and Representative Concentration Pathways (RCP) climate change scenarios, future changes in temperature and warmth index on the Korean Peninsula in the 21st century (2011~2100) were analyzed. In the analysis of the current climate (1981~2010), the ensemble averaged model results were found to reproduce the observed average values and spatial patterns of temperature and warmth index similarly well. In the future climate projections, temperature and warmth index are expected to rise in the 21st century compared to the current climate. They go further into the future and the higher carbon scenario (SSP5-8.5), the larger the increase. In the 21st century, in the low-carbon scenario (SSP1-2.6), temperature and warmth index are expected to rise by about 2.5℃ and 24.6%, respectively, compared to the present, while in the high-carbon scenario, they are expected to rise by about 6.2℃ and 63.9%, respectively. It was analyzed that reducing carbon emissions could contribute to reducing the increase in temperature and warmth index. The increase in the warmth index due to climate change can be positively analyzed to indicate that the effective heat required for plant growth on the Korean Peninsula will be stably secured. However, it is necessary to comprehensively consider negative aspects such as changes in growth conditions during the plant growth period, increase in extreme weather such as abnormally high temperatures, and decrease in plant diversity. This study can be used as basic scientific information for adapting to climate change and preparing response measures.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.37 no.1
• /
• pp.9-17
• /
• 2017

Characteristics of aerodynamic damping ratios of a helical $180^{\circ}$ model which shows better aerodynamic behavior in both along-wind and across-wind responses on a super tall building was investigated by an aeroelastic model test. The aerodynamic damping ratio was evaluated from the wind-induced responses of the model by using Random Decrement (RD) technique. Further, various triggering levels in evaluation of aerodynamic damping ratios using RD technique were also examined. As a result, it was found that when at least 2000 segments were used for evaluating aerodynamic damping ratio for ensemble averaging, the aerodynamic damping ratio can be obtained more consistently with lower irregular fluctuations. This is good agreement with those of previous studies. Another notable observation was that for square and helical $180^{\circ}$ models, the aerodynamic damping ratios in along-wind direction showed similar linear trends with reduced wind speeds regarding of building shapes. On the other hand, for the helical $180^{\circ}$ model, the aerodynamic damping ratio in across-wind direction showed quite different trends with those of the square model. In addition, the aerodynamic damping ratios of the helical $180^{\circ}$ model showed very similar trends with respect to the change of wind direction, and showed gradually increasing trends having small fluctuations with reduced wind speeds. Another observation was that in definition of triggering levels in RD technique on aerodynamic damping ratios, it may be possible to adopt the triggering levels of "standard deviation" or "${\sqrt{2}}$ times of the standard deviation" of the response time history if RD functions have a large number of triggering points. Further, these triggering levels may result in similar values and distributions with reduced wind speeds and either may be acceptable.

• Economic and Environmental Geology
• /
• v.46 no.4
• /
• pp.313-327
• /
• 2013

In the study, three-dimensional geostatistical simulations on McMurray Formation which is the largest oil sand reservoir in Athabasca area, Canada were performed, and the optimal site for steam assisted gravity drainage (SAGD) was selected based on the predictions. In the selection, the factors related to the vertical extendibility of steam chamber were considered as the criteria for an optimal site. For the predictions, 110 borehole data acquired from the study area were analyzed in the Markovian transition probability (TP) framework and three-dimensional distributions of the composing media were predicted stochastically through an existing TP based geostatistical model. The potential of a specific medium at a position within the prediction domain was estimated from the ensemble probability based on the multiple realizations. From the ensemble map, the cumulative thickness of the permeable media (i.e. Breccia and Sand) was analyzed and the locations with the highest potential for SAGD applications were delineated. As a supportive criterion for an optimal SAGD site, mean vertical extension of a unit permeable media was also delineated through transition rate based computations. The mean vertical extension of a permeable media show rough agreement with the cumulative thickness in their general distribution. However, the distributions show distinctive disagreement at a few locations where the cumulative thickness was higher due to highly alternating juxtaposition of the permeable and the less permeable media. This observation implies that the cumulative thickness alone may not be a sufficient criterion for an optimal SAGD site and the mean vertical extension of the permeable media needs to be jointly considered for the sound selections.