• Journal of the Korea Safety Management & Science
• /
• v.8 no.3
• /
• pp.1-9
• /
• 2006

Recently data mining techniques have been used for analysis and classification of data related to industrial accidents. The main objective of this study is to compare algorithms for data analysis of industrial accidents and this paper provides an optimal predicting model of 5 kinds of algorithms including CHAID, CART, C4.5, LR (Logistic Regression) and NN (Neural Network) with ROC chart, lift chart and response threshold. Also, this paper provides an approximation model for an optimal predicting model based on NN. The approximation model provided in this study can be utilized for easy interpretation of data analysis using NN. This study uses selected ten independent variables to group injured people according to a dependent variable in a way that reduces variation. In order to find an optimal predicting model among 5 algorithms, a retrospective analysis was performed in 67,278 subjects. The sample for this work chosen from data related to industrial accidents during three years ($2002\;{\sim}\;2004$) in korea. According to the result analysis, NN has excellent performance for data analysis and classification of industrial accidents.

• Asian-Australasian Journal of Animal Sciences
• /
• v.31 no.11
• /
• pp.1700-1713
• /
• 2018

Objective: The aim of this study was to verify the usefulness of artificial neural networks (ANN), multivariate adaptive regression splines (MARS), naïve Bayes classifier (NBC), general discriminant analysis (GDA), and logistic regression (LR) for dystocia detection in Polish Holstein-Friesian Black-and-White heifers and cows and to indicate the most influential predictors of calving difficulty. Methods: A total of 1,342 and 1,699 calving records including six categorical and four continuous predictors were used. Calving category (difficult vs easy or difficult, moderate and easy) was the dependent variable. Results: The maximum sensitivity, specificity and accuracy achieved for heifers on the independent test set were 0.855 (for ANN), 0.969 (for NBC), and 0.813 (for GDA), respectively, whereas the values for cows were 0.600 (for ANN), 1.000 and 0.965 (for NBC, GDA, and LR), respectively. With the three categories of calving difficulty, the maximum overall accuracy for heifers and cows was 0.589 (for MARS) and 0.649 (for ANN), respectively. The most influential predictors for heifers were an average calving difficulty score for the dam's sire, calving age and the mean yield of the farm, where the heifer was kept, whereas for cows, these additionally included: calf sex, the difficulty of the preceding calving, and the mean daily milk yield for the preceding lactation. Conclusion: The potential application of the investigated models in dairy cattle farming requires, however, their further improvement in order to reduce the rate of dystocia misdiagnosis and to increase detection reliability.

• Proceedings of the Safety Management and Science Conference
• /
• 2006.04a
• /
• pp.51-56
• /
• 2006

Recently data mining techniques have been used for analysis and classification of data related to industrial accidents. The main objective of this study is to compare performance of algorithms for data analysis of industrial accidents and this paper provides a comparative analysis of 5 kinds of algorithms including CHAID, CART, C4.5, LR (Logistic Regression) and NN (Neural Network) with ROC chart, lift chart and response threshold. In this study, data on 67,278 accidents were analyzed to create risk groups for a number of complications, including the risk of disease and accident. The sample for this work chosen from data related to manufacturing industries during three years $(2002\sim2004)$ in korea. According to the result analysis, NN has excellent performance for data analysis and classification of industrial accidents.

• Proceedings of the Korea Contents Association Conference
• /
• 2018.05a
• /
• pp.7-8
• /
• 2018

This study proposes an analysis of the Word2vec-based machine learning classifiers for the sake of opinion mining tasks. As a bench-marking method, BOW (Bag-of-Words) was adopted. On the basis of utilizing the Word2vec and BOW as feature extraction methods, we applied Laptop and Restaurant dataset to LR, DT, SVM, RF classifiers. The results showed that the Word2vec feature extraction yields more improved performance.

• Journal of the Korean Geotechnical Society
• /
• v.39 no.6
• /
• pp.5-12
• /
• 2023

Proactive assessment of landslide susceptibility is necessary for minimizing casualties. This study proposes a methodology for classifying the landslide safety factor using a classification algorithm based on machine learning techniques. The high-risk area model is adopted to perform the classification and eight geotechnical parameters are adopted as inputs. Four classification algorithms-namely decision tree, k-nearest neighbor, logistic regression, and random forest-are employed for comparing classification accuracy for the safety factors ranging between 1.2 and 2.0. Notably, a high accuracy is demonstrated in the safety factor range of 1.2~1.7, but a relatively low accuracy is obtained in the range of 1.8~2.0. To overcome this issue, the synthetic minority over-sampling technique (SMOTE) is adopted to generate additional data. The application of SMOTE improves the average accuracy by ~250% in the safety factor range of 1.8~2.0. The results demonstrate that SMOTE algorithm improves the accuracy of classification algorithms when applied to geotechnical data.

• Journal of Korean Society for Geospatial Information Science
• /
• v.25 no.1
• /
• pp.19-27
• /
• 2017

This research was aimed to analyze landslide susceptibility and compare the prediction accuracy using ensemble frequency ratio (FR) and logistic regression at the Inje area, Korea. The landslide locations were identified with the before and after aerial photographs of landslide occurrence that were randomly selected for training (70%) and validation (30%). The total twelve landslide-related factors were elevation, slope, aspect, distance to drainage, topographic wetness index, stream power index, soil texture, soil sickness, timber age, timber diameter, timber density, and timber type. The spatial relationship between landslide occurrence and landslide-related factors was analyzed using FR and ensemble model. The produced LSI maps were validated and compared using relative operating characteristics (ROC) curve. The prediction accuracy of produced ensemble LSI map was about 2% higher than FR LSI map. The LSI map produced in this research could be used to establish land use planning and mitigate the damages caused by disaster.

• Journal of Digital Convergence
• /
• v.17 no.2
• /
• pp.163-170
• /
• 2019

Over the past decade, the development of the Web explosively increased the data. Feature selection step is an important step in extracting valuable data from a large amount of data. This study proposes a novel opinion mining model based on combining feature selection (FS) methods with Word embedding to vector (Word2vec) and BOW (Bag-of-words). FS methods adopted for this study are CFS (Correlation based FS) and IG (Information Gain). To select an optimal FS method, a number of classifiers ranging from LR (logistic regression), NN (neural network), NBN (naive Bayesian network) to RF (random forest), RS (random subspace), ST (stacking). Empirical results with electronics and kitchen datasets showed that LR and ST classifiers combined with IG applied to BOW features yield best performance in opinion mining. Results with laptop and restaurant datasets revealed that the RF classifier using IG applied to Word2vec features represents best performance in opinion mining.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.22
• /
• pp.9599-9602
• /
• 2014

Background: Postoperative adjuvant radiotherapy (RT) in the management of early stage endometrial cancer (EC) is still controversial. Here we report our institutional experience with patients who received postoperative RT for stage I-II EC over a period of 35 years and assess potential predictors of local recurrence (LR), distant metastasis (DM), and overall survival (OS). Materials and Methods: A total of 188 patients undergoing postoperative RT for stage IA-II EC between 1977 and 2012 were evaluated. Some 96 received median 46 Gy whole pelvic radiotherapy (WPRT) (range: 40-60 Gy), 37 were given WPRT with vaginal cuff therapy (VCT), and 55 received only VCT either with brachytherapy (BT) or stereotactic body radiotherapy (SBRT). Chemotherapy was given to 5 patients with uterine papillary serous carcinoma (UPSC). Logistic regression analysis was used to assess the effect of clinicopathological factors on LR, DM, and OS. Results: Median follow-up time was 11 years (range: 1-35 years). At the time of analysis, 34 patients were not alive. Of the 15 patients with LR, 7 (46.7%) recurred in the vaginal stump, 5 (33.3%) in the pelvic region, and 3 (20%) in the paraaortic nodal region, while 12 had distant metastasis. UPSC histology (p=0.027), sole VCT (p=0.041), high histologic grade (p=0.034), and age ${\geq}71$ (p=0.04) were poor prognostic factors on univariate analysis. Conclusions: In our patients receiving radiotherapy for early-stage EC, grade III disease and age ${\geq}71$ were associated with shorter OS whereas UPSC histology was an independent predictor for both LR and DM.

• Journal of Korea Water Resources Association
• /
• v.54 no.9
• /
• pp.667-680
• /
• 2021

We investigated changes in frost days and frost-free periods and to comparatively assess frost event prediction models developed using logistic regression (LR), random forest (RF), and long short-term memory (LSTM) networks. The meteorological variables for the model development were collected from the Suwon, Cheongju, and Gwangju stations for the period of 1973-2019 for spring (March - May) and fall (September - November). The developed models were then evaluated by Precision, Recall, and f-1 score and graphical evaluation methods such as AUC and reliability diagram. The results showed that significant decreases (significance level of 0.01) in the frequencies of frost days were at the three stations in both spring and fall. Overall, the evaluation metrics showed that the performance of RF was highest, while that of LSTM was lowest. Despite higher AUC values (above 0.9) were found at the three stations, reliability diagrams showed inconsistent reliability. A further study is suggested on the improvement of the predictability of both frost events and the first and last frost days by the frost event prediction models and reliability of the models. It would be beneficial to replicate this study at more stations in other regions.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.248-248
• /
• 2023

In recent years, the number of landslides in Korea has been increasing due to extreme weather events such as localized heavy rainfall and typhoons. Landslides often occur with debris flows, land subsidence, and earthquakes. They cause significant damage to life and property. 64% of Korea's land area is made up of mountains, the government wanted to predict landslides to reduce damage. In response, the Korea Forest Service has established a 'Landslide Information System' to predict the likelihood of landslides. This system selects a total of 13 landslide factors based on past landslide events. Using the LR technique (Logistic Regression) to predict the possibility of a landslide occurrence and the accuracy is known to be 0.75. However, most of the data used for learning in the current system is on landslides that occurred from 2005 to 2011, and it does not reflect recent typhoons or heavy rain. Therefore, in this study, we will apply a total of six machine learning techniques (KNN, LR, SVM, XGB, RF, GNB) to predict the occurrence of landslides based on the data of Inje, Gangwon-do, which was recently produced by the National Institute of Forest. To predict the occurrence of landslides, it is necessary to process converting landslide events and factors data into a suitable form for machine learning techniques through ArcGIS and Python. In addition, there is a large difference in the number of data between areas where landslides occurred or not. Therefore, the prediction was performed after correcting the unbalanced data using Tomek Links and Near Miss techniques. Moreover, to control unbalanced data, a model that reflects soil properties will use to remove absolute safe areas.