• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.543-552
• /
• 2022

When frost occurs, crops are directly damaged. When crops come into contact with low temperatures, tissues freeze, which hardens and destroys the cell membranes or chloroplasts, or dry cells to death. In July 2020, a sudden sub-zero weather and frost hit the Minas Gerais state of Brazil, the world's largest coffee producer, damaging about 30% of local coffee trees. As a result, coffee prices have risen significantly due to the damage, and farmers with severe damage can produce coffee only after three years for crops to recover, which is expected to cause long-term damage. In this paper, we tried to predict frost using frost generation data and weather observation data provided by the Korea Meteorological Administration to prevent severe frost. A model was constructed by reflecting weather factors such as wind speed, temperature, humidity, precipitation, and cloudiness. Using XGB(eXtreme Gradient Boosting), SVM(Support Vector Machine), Random Forest, and MLP(Multi Layer perceptron) models, various hyper parameters were applied as training data to select the best model for each model. Finally, the results were evaluated as accuracy(acc) and CSI(Critical Success Index) in test data. XGB was the best model compared to other models with 90.4% ac and 64.4% CSI, followed by SVM with 89.7% ac and 61.2% CSI. Random Forest and MLP showed similar performance with about 89% ac and about 60% CSI.

• The Korean Journal of Applied Statistics
• /
• v.35 no.4
• /
• pp.457-468
• /
• 2022

In aged society, it is important to prevent older people from being disability needing long-term care. The purpose of this study is to develop a prediction model to discover high-risk groups who are likely to be beneficiaries of Long-Term Care Insurance. This study is a retrospective study using database of National Health Insurance Service (NHIS) collected in the past of the study subjects. The study subjects are 7,724,101, the population over 65 years of age registered for medical insurance. To develop the prediction model, we used logistic regression, decision tree, random forest, and multi-layer perceptron neural network. Finally, random forest was selected as the prediction model based on the performances of models obtained through internal and external validation. Random forest could predict about 90% of the older people in need of long-term care using DB without any information from the assessment of eligibility for long-term care. The findings might be useful in evidencebased health management for prevention services and can contribute to preemptively discovering those who need preventive services in older people.

• Journal of Industrial Convergence
• /
• v.20 no.2
• /
• pp.1-9
• /
• 2022

As the need for improvement of transparency and fairness in the selection of national R&D projects has been continuously raised, we analyzed the impact on the evaluation selection results by evaluation indexes for The land transportation technology commercialization support project and searched for ways to improve indexes using the analysis results. As for the research data, it were applied as selection results of new R&D projects and evaluation indexes in two fields(SME innovation and start-up) in 2021. Logistic regression analysis is used for the influence of each evaluation indexes on the evaluation result, and for the regression model, evaluation indexes with low influence are removed in advance through artificial neural network multiple perceptron analysis to improve the reliability of the analysis results. As a result of the analysis, in the field of SME innovation, the influence of the evaluation index on the workforce planning was the lowest and the influence of the appropriateness of commercialization promotion plan was the highest. In the start-up field, the influence of the evaluation indexes for technology development suitability, marketability, and suitability for carrying out the project were estimated to be similar to each other, and the influence of the technology evaluation index was found to be the lowest. The analysis results of this thesis suggest the need for continuous improvement of selection and evaluation indexes, and by using the analysis results to select a fair R&D institution according to the selection of appropriate indexes, it will be possible to contribute to deriving excellent research results and fostering excellent companies in the field of land transportation.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• Korean Journal of Construction Engineering and Management
• /
• v.23 no.3
• /
• pp.45-55
• /
• 2022

BIM models allow building spaces to be instantiated and recognized as unique objects independently of model elements. These instantiated spaces provide the required semantics that can be leveraged for building code checking, energy analysis, and evacuation route analysis. However, theses spaces or rooms need to be designated manually, which in practice, lead to errors and omissions. Thus, most BIM models today does not guarantee the semantic integrity of space designations, limiting their potential applicability. Recent studies have explored ways to automate space allocation in BIM models using artificial intelligence algorithms, but they are limited in their scope and relatively low classification accuracy. This study explored the use of Graph Convolutional Networks, an algorithm exclusively tailored for graph data structures. The goal was to utilize not only geometry information but also the semantic relational data between spaces and elements in the BIM model. Results of the study confirmed that the accuracy was improved by about 8% compared to algorithms that only used geometric distinctions of the individual spaces.

• Knowledge Management Research
• /
• v.23 no.2
• /
• pp.277-299
• /
• 2022

The massive card transaction data generated in the tourism industry has become an important resource that implies tourist consumption behaviors and patterns. Based on the transaction data, developing a smart service system becomes one of major goals in both tourism businesses and knowledge management system developer communities. However, the lack of rating scores, which is the basis of traditional recommendation techniques, makes it hard for system designers to evaluate a learning process. In addition, other auxiliary factors such as temporal, spatial, and demographic information are needed to increase the performance of a recommendation system; but, gathering those are not easy in the card transaction context. In this paper, we introduce CTDDTR, a novel approach using card transaction data to recommend tourism services. It consists of two main components: i) Temporal preference Embedding (TE) represents tourist groups and services into vectors through Doc2Vec. And ii) Deep tourism Recommendation (DR) integrates the vectors and the auxiliary factors from a tourism RDF (resource description framework) through MLP (multi-layer perceptron) to provide services to tourist groups. In addition, we adopt RFM analysis from the field of knowledge management to generate explicit feedback (i.e., rating scores) used in the DR part. To evaluate CTDDTR, the card transactions data that happened over eight years on Jeju island is used. Experimental results demonstrate that the proposed method is more positive in effectiveness and efficacies.

• Journal of The Korean Society of Grassland and Forage Science
• /
• v.43 no.1
• /
• pp.1-10
• /
• 2023

This study aimed to confirm the importance ratio of climate and management variables on production of orchardgrass in Korea (1982-2014). For the climate, the mean temperature in January (MTJ, ℃), lowest temperature in January (LTJ, ℃), growing days 0 to 5 (GD 1, day), growing days 5 to 25 (GD 2, day), Summer depression days (SSD, day), rainfall days (RD, day), accumulated rainfall (AR, mm), and sunshine duration (SD, hr) were considered. For the management, the establishment period (EP, 0-6 years) and number of cutting (NC, 2nd-5th) were measured. The importance ratio on production of orchardgrass was estimated using the neural network model with the perceptron method. It was performed by SPSS 26.0 (IBM Corp., Chicago). As a result, EP was the most important variable (100%), followed by RD (82.0%), AR (79.1%), NC (69.2%), LTJ (66.2%), GD 2 (63.3%), GD 1 (61.6%), SD (58.1%), SSD (50.8%) and MTJ (41.8%). It implies that EP, RD, AR, and NC were more important than others. Since the annual rainfall in Korea is exceed the required amount for the growth and development of orchardgrass, the damage caused by heavy rainfall exceeding the appropriate level could be reduced through drainage management. It means that, when cultivating orchardgrass, factors that can be controlled were relatively important. Although it is difficult to interpret the specific effect of climates on production due to neural networking modeling, in the future, this study is expected to be useful in production prediction and damage estimation by climate change by selecting major factors.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.2
• /
• pp.1-9
• /
• 2023

Research using AI and big data is also being actively conducted in the health and medical fields such as disease diagnosis and treatment. Most of the existing research data used cohort data from research institutes or some patient data. In this paper, the difference in the prediction rate of survival and the factors affecting survival between breast cancer patients in their 40~50s and other age groups was revealed using health insurance review claim data held by the HIRA. As a result, the accuracy of predicting patients' survival was 0.93 on average in their 40~50s, higher than 0.86 in their 60~80s. In terms of that factor, the number of treatments was high for those in their 40~50s, and age was high for those in their 60~80s. Performance comparison with previous studies, the average precision was 0.90, which was higher than 0.81 of the existing paper. As a result of performance comparison by applied algorithm, the overall average precision of Decision Tree, Random Forest, and Gradient Boosting was 0.90, and the recall was 1.0, and the precision of multi-layer perceptrons was 0.89, and the recall was 1.0. I hope that more research will be conducted using machine learning automation(Auto ML) tools for non-professionals to enhance the use of the value for health insurance review claim data held by the HIRA.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.2
• /
• pp.343-350
• /
• 2022

It is worth verifying the effectiveness of data integration between data with different features. This study investigated whether the data integration affects the accuracy of deep neural network (DNN), and which integration method shows the best improvement. This study used two different public datasets. One public dataset was taken in an actual farm in India. And another was taken in a laboratory environment in Korea. Leaf images were selected from two different public datasets to have five classes which includes normal and four different types of plant diseases. DNN used pre-trained VGG16 as a feature extractor and multi-layer perceptron as a classifier. Data were integrated into three different ways to be used for the training process. DNN was trained in a supervised manner via the integrated data. The trained DNN was evaluated by using a test dataset taken in an actual farm. DNN shows the best accuracy for the test dataset when DNN was first trained by images taken in the laboratory environment and then trained by images taken in the actual farm. The results show that data integration between plant images taken in a different environment helps improve the performance of deep neural networks. And the results also confirmed that independent use of plant images taken in different environments during the training process is more effective in improving the performance of DNN.

• The Journal of Engineering Geology
• /
• v.33 no.4
• /
• pp.673-687
• /
• 2023

Exploratory data analysis is the process of observing and understanding data collected from various sources to identify their distributions and correlations through their structures and characterization. This process can be used to identify correlations among conditioning factors and select the most effective factors for analysis. This can help the assessment of landslide susceptibility, because landslides are usually triggered by multiple factors, and the impacts of these factors vary by region. This study compared two stages of exploratory data analysis to examine the impact of the data exploration procedure on the landslide prediction model's performance with respect to factor selection. Deep-learning-based landslide susceptibility analysis used either a combinations of selected factors or all 23 factors. During the data exploration phase, we used a Pearson correlation coefficient heat map and a histogram of random forest feature importance. We then assessed the accuracy of our deep-learning-based analysis of landslide susceptibility using a confusion matrix. Finally, a landslide susceptibility map was generated using the landslide susceptibility index derived from the proposed analysis. The analysis revealed that using all 23 factors resulted in low accuracy (55.90%), but using the 13 factors selected in one step of exploration improved the accuracy to 81.25%. This was further improved to 92.80% using only the nine conditioning factors selected during both steps of the data exploration. Therefore, exploratory data analysis selected the conditioning factors most suitable for landslide susceptibility analysis and thereby improving the performance of the analysis.