• Journal of KIISE
• /
• v.43 no.1
• /
• pp.80-86
• /
• 2016

A speech-act is a behavior intended by users in an utterance. Speech-act classification is important in a dialogue system. The machine learning and rule-based methods have mainly been used for speech-act classification. In this paper, we propose a speech-act classification method based on the combination of support vector machine (SVM) and transformation-based learning (TBL). The user's utterance is first classified by SVM that is preferentially applied to categories with a low utterance rate in training data. Next, when an utterance has negative scores throughout the whole of the categories, the utterance is applied to the correction phase by rules. The results from our method were higher performance over the baseline system long with error-reduction.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.267-270
• /
• 2007

기계학습에서 분류기들의 집합으로 구성된 앙상블 분류기는 단일 분류기에 비해 정확도가 높다는 것이 입증되었다. 본 논문에서는 새로운 앙상블 학습으로서 데이터의 지역 기반 분류기들의 앙상블 학습을 제시하여 기존의 앙상블 학습과의 비교를 통해 성능을 검증하고자 한다. 지역 기반 분류기의 앙상블 학습은 데이터의 분포가 지역에 따라 다르다는 점에 착안하여 학습 데이터를 분할하고 해당하는 지역에 기반을 둔 분류기들을 만들어 나간다. 이렇게 만들어진 분류기들로부터 지역에 따라 가중치를 둔 투표를 하여 앙상블 방법을 이끌어낸다. 본 논문에서 제시한 앙상블 분류기의 성능평가를 위해 UCI Machine Learning Repository에 있는 11개의 데이터 셋을 이용하여 단일 분류기와 기존의 앙상블 분류기인 배깅과 부스팅등의 정확도를 비교하였다. 그 결과 기본 분류기로 나이브 베이즈와 SVM을 사용했을 때 새로운 앙상블 방법이 다른 방법보다 좋은 성능을 보이는 것을 알 수 있었다.

• The Journal of Information Systems
• /
• v.20 no.3
• /
• pp.237-255
• /
• 2011

지난 수십 년 간 연구자들은 효과적인 진료지원시스템을 개발하기 위해 다양한 도구와 방법론들을 제안하였고 지금도 새로운 방법론과 도구들을 계속적으로 개발하고 있다. 그 중에서 흉통으로 응급실에 내원한 노인환자에 대한 정확한 진단은 중요한 이슈 중의 하나였다. 따라서 많은 연구자들이 의사의 진단 능력을 향상시키기 위한 지능적인 의료의사결정과 시스템 개발에 투신하고 있지만 전통적인 의료시스템에 따른 대부분의 진료의사결정이 단일 분류기(classifier)에 기반하고 있어 만족스런 성능을 보여주지 못하고 있는 것이 현실이다. 따라서 이 논문은 앙상블 전략을 활용하여 의사들이 노인환자들의 흉통을 더 정확하고 빠르게 진단하는데 있어 도움을 줄 수 있게 하였다. 의사결정나무, 인공신경망, SVM 모델을 결합한 앙상블 기법을 실제 응급실에서 수집한 응급실 자료에 적용하였고, 그 결과 단일 분류기를 사용하는 것에 비해 월등히 향상된 진단 성과를 보이는 것을 관찰 할 수 있었다.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.4
• /
• pp.813-823
• /
• 2013

On-line monitoring system is important for high voltage vacuum circuit breakers (HVCBs) in operation condition assessment and fault diagnosis. A distributed multilayer system with client/server architecture is developed on rated voltage 10kV HVCB with spring operating mechanism. It can collect data when HVCB switches, calculate the necessary parameters, show the operation conditions and provide abundant information for fault diagnosis. Ensemble empirical mode decomposition (EEMD) is used to detect the singular point which is regarded as the contact moment. This method has been applied to on-line monitoring system successfully and its satisfactory effect has been proved through experiments. SVM and FCM are both effective methods for fault diagnosis. A combinative algorithm is designed to judge the faults of HVCB's operating mechanism. The system's precision and stability are confirmed by field tests.

• Journal of Information Processing Systems
• /
• v.13 no.5
• /
• pp.1203-1212
• /
• 2017

Intrusion detection systems (IDSs) are crucial in this overwhelming increase of attacks on the computing infrastructure. It intelligently detects malicious and predicts future attack patterns based on the classification analysis using machine learning and data mining techniques. This paper is devoted to thoroughly evaluate classifier ensembles for IDSs in IEEE 802.11 wireless network. Two ensemble techniques, i.e. voting and stacking are employed to combine the three base classifiers, i.e. decision tree (DT), random forest (RF), and support vector machine (SVM). We use area under ROC curve (AUC) value as a performance metric. Finally, we conduct two statistical significance tests to evaluate the performance differences among classifiers.

• Journal of Wetlands Research
• /
• v.25 no.4
• /
• pp.306-314
• /
• 2023

Stream sediments are an important component of water quality management because they are receptors of various pollutants such as heavy metals and organic matters emitted from upland sources and can be secondary pollution sources, adversely affecting water environment. To effectively manage the stream sediments, identification of primary sources of sediment contamination and source-associated control strategies will be required. We evaluated the performance of machine learning models in identifying primary sources of sediment contamination based on the physico-chemical properties of stream sediments. A total of 356 stream sediment data sets of 18 quality parameters including 10 heavy metal species(Cd, Cu, Pb, Ni, As, Zn, Cr, Hg, Li, and Al), 3 soil parameters(clay, silt, and sand fractions), and 5 water quality parameters(water content, loss on ignition, total organic carbon, total nitrogen, and total phosphorous) were collected near abandoned metal mines and industrial complexes across the four major river basins in Korea. Two machine learning algorithms, linear discriminant analysis (LDA) and support vector machine (SVM) classifiers were used to classify the sediments into four cases of different combinations of the sampling period and locations (i.e., mine in dry season, mine in wet season, industrial complex in dry season, and　industrial complex in wet season). Both models showed good performance in the classification, with SVM outperformed LDA; the accuracy values of LDA and SVM were 79.5% and 88.1%, respectively. An SVM ensemble model was used for multi-label classification of the multiple contamination sources inlcuding landuses in the upland areas within 1 km radius from the sampling sites. The results showed that the multi-label classifier was comparable performance with sinlgle-label SVM in classifying mines and industrial complexes, but was less accurate in classifying dominant land uses (50~60%). The poor performance of the multi-label SVM is likely due to the overfitting caused by small data sets compared to the complexity of the model. A larger data set might increase the performance of the machine learning models in identifying contamination sources.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.21 no.1
• /
• pp.103-118
• /
• 2018

Ecosystems in subalpine regions are recognized as areas vulnerable to climatic changes because rainfall and the possibility of flora migration are very low due to the characteristics of topography in the regions. In this context, habitat niche was formulated for representative species of arbors in subalpine regions in order to understand the effects of climatic changes on alpine arbor ecosystems. The current potential habitats were modeled as future change areas according to the climatic change scenarios. Based on the growth conditions and environmental characteristics of the habitats, the study was conducted to identify direct and indirect causes affecting the habitat reduction of Abies nephrolepis. Diverse model algorithms for explanation of the relationship between the emergence of biological species and habitat environments were reviewed to construct the environmental data suitable for the six models(GLM, GAM, RF, MaxEnt, ANN, and SVM). Weights determined through TSS were applied to the six models for ensemble in an attempt to minimize the uncertainty of the models. Based on the current climate determined by averaging the climates over the past 30years(1981~2010) and the HadGEM-RA model was applied to fabricate bioclimatic variables for scenarios RCP 4.5 and 8.5 on the near and far future. The results of models of the alpine region tree species studied were put together and evaluated and the results indicated that a total of eight national parks such as Mt. Seorak, Odaesan, and Hallasan would be mainly affected by climatic changes. Changes in the Baekdudaegan reserves were analyzed and in the results, A. nephrolepis was predicted to be affected the most in the RCP8.5. The results of analysis as such are expected to be finally utilizable in the survey of biological species in the Korean peninsula, restoration and conservation strategies considering climatic changes as the analysis identified the degrees of impacts of climatic changes on subalpine region trees in Korean peninsula with very high conservation values.

• Journal of the Korean Geosynthetics Society
• /
• v.19 no.1
• /
• pp.45-53
• /
• 2020

There is an increasing tendency to try to make predictive analysis using measurement data based on machine learning techniques in the railway industries. In this paper, it was predicted that Track quality index (TQI) using vehicle acceleration data based on the machine learning method. The XGB (XGBoost) was the most accurate with 85% in the all data sets. Unlike the SVM model with a single algorithm, the RF and XGB model with a ensemble system were considered to be good at the prediction performance. In the case of the Surface TQI, it is shown that the acceleration of the z axis is highly related to the vertical direction and is in good agreement with the previous studies. Therefore, it is appropriate to apply the model with the ensemble algorithm to predict the track quality index using the vehicle vibration acceleration data because the accuracy may vary depending on the applied model in the machine learning methods.

• International Journal of Computer Science & Network Security
• /
• v.21 no.1
• /
• pp.27-33
• /
• 2021

The whole world now is dealing with Coronavirus, and it has turned to be one of the most widespread and long-lived pandemics of our times. Reports reveal that the infectious disease has taken toll of the almost 80% of the world's population. Amidst a lot of research going on with regards to the prediction on growth and transmission through Symptomatic carriers of the virus, it can't be ignored that pre-symptomatic and asymptomatic carriers also play a crucial role in spreading the reach of the virus. Classification Algorithm has been widely used to classify different types of COVID-19 carriers ranging from simple feature-based classification to Convolutional Neural Networks (CNNs). This research paper aims to present a novel technique using a Random Forest Machine learning algorithm with hyper-parameter tuning to classify different types COVID-19-carriers such that these carriers can be accurately characterized and hence dealt timely to contain the spread of the virus. The main idea for selecting Random Forest is that it works on the powerful concept of "the wisdom of crowd" which produces ensemble prediction. The results are quite convincing and the model records an accuracy score of 99.72 %. The results have been compared with the same dataset being subjected to K-Nearest Neighbour, logistic regression, support vector machine (SVM), and Decision Tree algorithms where the accuracy score has been recorded as 78.58%, 70.11%, 70.385,99% respectively, thus establishing the concreteness and suitability of our approach.

• Geomechanics and Engineering
• /
• v.34 no.6
• /
• pp.697-726
• /
• 2023

Brittleness as an important property of rock plays a crucial role both in the failure process of intact rock and rock mass response to excavation in engineering geological and geotechnical projects. Generally, rock brittleness indices are calculated from the mechanical properties of rocks such as uniaxial compressive strength, tensile strength and modulus of elasticity. These properties are generally determined from complicated, expensive and time-consuming tests in laboratory. For this reason, in the present research, an attempt has been made to predict the rock brittleness indices from simple, inexpensive, and quick laboratory test results namely dry unit weight, porosity, slake-durability index, P-wave velocity, Schmidt rebound hardness, and point load strength index using multiple linear regression, exponential regression, support vector machine (SVM) with various kernels, generating fuzzy inference system, and regression tree ensemble (RTE) with boosting framework. So, this could be considered as an innovation for the present research. For this purpose, the number of 39 rock samples including five igneous, twenty-six sedimentary, and eight metamorphic were collected from different regions of Iran. Mineralogical, physical and mechanical properties as well as five well known rock brittleness indices (i.e., B₁, B₂, B₃, B₄, and B₅) were measured for the selected rock samples before application of the above-mentioned machine learning techniques. The performance of the developed models was evaluated based on several statistical metrics such as mean square error, relative absolute error, root relative absolute error, determination coefficients, variance account for, mean absolute percentage error and standard deviation of the error. The comparison of the obtained results revealed that among the studied methods, SVM is the most suitable one for predicting B₁, B₂ and B₅, while RTE predicts B₃ and B₄ better than other methods.