• KIPS Transactions on Software and Data Engineering
• /
• v.7 no.9
• /
• pp.351-360
• /
• 2018

The LWR (Locally Weighted Regression) model, which is traditionally a lazy learning model, is designed to obtain the solution of the prediction according to the input variable, the query point, and it is a kind of the regression equation in the short interval obtained as a result of the learning that gives a higher weight value closer to the query point. We study on an incremental ensemble learning approach for LWR, a form of lazy learning and memory-based learning. The proposed incremental ensemble learning method of LWR is to sequentially generate and integrate LWR models over time using a genetic algorithm to obtain a solution of a specific query point. The weaknesses of existing LWR models are that multiple LWR models can be generated based on the indicator function and data sample selection, and the quality of the predictions can also vary depending on this model. However, no research has been conducted to solve the problem of selection or combination of multiple LWR models. In this study, after generating the initial LWR model according to the indicator function and the sample data set, we iterate evolution learning process to obtain the proper indicator function and assess the LWR models applied to the other sample data sets to overcome the data set bias. We adopt Eager learning method to generate and store LWR model gradually when data is generated for all sections. In order to obtain a prediction solution at a specific point in time, an LWR model is generated based on newly generated data within a predetermined interval and then combined with existing LWR models in a section using a genetic algorithm. The proposed method shows better results than the method of selecting multiple LWR models using the simple average method. The results of this study are compared with the predicted results using multiple regression analysis by applying the real data such as the amount of traffic per hour in a specific area and hourly sales of a resting place of the highway, etc.

• Proceedings of the Korean Society for Bioinformatics Conference
• /
• 2005.09a
• /
• pp.237-242
• /
• 2005

Predicting the cellular location of an unknown protein gives a valuable information for inferring the possible function of the protein. For more accurate prediction system, we need a good feature extraction method that transforms the raw sequence data into the numerical feature vector, minimizing information loss. In this paper, we propose new methods of extracting underlying features only from the sequence data by computing pairwise sequence alignment scores. In addition, we use composition based features to improve prediction accuracy. To construct an SVM ensemble from separately trained SVM classifiers, we propose specificity based weighted majority voting. The overall prediction accuracy evaluated by the 5-fold cross-validation reached 88.53% for the eukaryotic animal data set. By comparing the prediction accuracy of various feature extraction methods, we could get the biological insight on the location of targeting information. Our numerical experiments confirm that our new feature extraction methods are very useful for predicting subcellular localization of proteins.

• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.333-336
• /
• 2016

본 연구에서는 개체명 인식의 성능을 향상시키기 위해, 가중 투표 방법을 이용하여 개체명 인식 모델을 앙상블 하는 방법을 제안한다. 각 모델은 Conditional Random Fields의 변형 알고리즘을 사용하여 학습하고, 모델들의 가중치는 다목적 함수 최적화 기법인 NSGA-II 알고리즘으로 학습한다. 실험 결과 제안 시스템은 $F_1Score$ 기준으로 87.62%의 성능을 보여, 단독 모델 중 가장 높은 성능을 보인 방법보다 2.15%p 성능이 향상되었다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.333-336
• /
• 2016

본 연구에서는 개체명 인식의 성능을 향상시키기 위해, 가중 투표 방법을 이용하여 개체명 인식 모델을 앙상블 하는 방법을 제안한다. 각 모델은 Conditional Random Fields의 변형 알고리즘을 사용하여 학습하고, 모델들의 가중치는 다목적 함수 최적화 기법인 NSGA-II 알고리즘으로 학습한다. 실험 결과 제안 시스템은 $F_1Score$기준으로 87.62%의 성능을 보여, 단독 모델 중 가장 높은 성능을 보인 방법보다 2.15%p 성능이 향상되었다.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.19 no.6
• /
• pp.19-30
• /
• 2016

The projection of climate-related range shift is critical information for conservation planning of Korean fir (Abies koreana E. H. Wilson). We first modeled the distribution of Korean fir under current climate condition using five single-model species distribution models (SDMs) and the pre-evaluation weighted ensemble method and then predicted the distributions under future climate conditions projected with HadGEM2-AO under four $CO_2$ emission scenarios, the Representative Concentration Pathways (RCP) 2.6, 4.5, 6.0 and 8.5. We also investigated the predictive uncertainty stemming from five individual algorithms and four $CO_2$ emission scenarios for better interpretation of SDM projections. Five individual algorithms were Generalized linear model (GLM), Generalized additive model (GAM), Multivariate adaptive regression splines (MARS), Generalized boosted model (GBM) and Random forest (RF). The results showed high variations of model performances among individual SDMs and the wide range of diverging predictions of future distributions of Korean fir in response to RCPs. The ensemble model presented the highest predictive accuracy (TSS = 0.97, AUC = 0.99) and predicted that the climate habitat suitability of Korean fir would increase under climate changes. Accordingly, the fir distribution could expand under future climate conditions. Increasing precipitation may account for increases in the distribution of Korean fir. Increasing precipitation compensates the negative effects of increasing temperature. However, the future distribution of Korean fir is also affected by other ecological processes, such as interactions with co-existing species, adaptation and dispersal limitation, and other environmental factors, such as extreme weather events and land-use changes. Therefore, we need further ecological research and to develop mechanistic and process-based distribution models for improving the predictive accuracy.

• KIISE Transactions on Computing Practices
• /
• v.21 no.1
• /
• pp.64-69
• /
• 2015

Place awareness is an essential for location-based services that are widely provided to smartphone users. However, traditional GPS-based methods are only valid outdoors where the GPS signal is strong and also require symbolic place information of the physical location. In this paper, environmental sounds and images are used to recognize important aspects of each place. The proposed method extracts feature vectors from visual, auditory and location data recorded by a smartphone with built-in camera, microphone and GPS sensors modules. The heterogeneous feature vectors were then learned by an ensemble learning method that learns each group of feature vectors for each classifier respectively and votes to produce the highest weighted result. The proposed method is evaluated for place recognition using a data group of 3000 samples in six places and the experimental results show a remarkably improved recognition accuracy when using all kinds of sensory data comparing to results using data from a single sensor or audio-visual integrated data only.

• Smart Structures and Systems
• /
• v.31 no.4
• /
• pp.335-349
• /
• 2023

Bridges constantly undergo deterioration and damage, the most common ones being concrete damage and exposed rebar. Periodic inspection of bridges to identify damages can aid in their quick remediation. Likewise, identifying components can provide context for damage assessment and help gauge a bridge's state of interaction with its surroundings. Current inspection techniques rely on manual site visits, which can be time-consuming and costly. More recently, robotic inspection assisted by autonomous data analytics based on Computer Vision (CV) and Artificial Intelligence (AI) has been viewed as a suitable alternative to manual inspection because of its efficiency and accuracy. To aid research in this avenue, this study performs a comparative assessment of different architectures, loss functions, and ensembling strategies for the autonomous segmentation of bridge components and damages. The experiments lead to several interesting discoveries. Nested Reg-UNet architecture is found to outperform five other state-of-the-art architectures in both damage and component segmentation tasks. The architecture is built by combining a Nested UNet style dense configuration with a pretrained RegNet encoder. In terms of the mean Intersection over Union (mIoU) metric, the Nested Reg-UNet architecture provides an improvement of 2.86% on the damage segmentation task and 1.66% on the component segmentation task compared to the state-of-the-art UNet architecture. Furthermore, it is demonstrated that incorporating the Lovasz-Softmax loss function to counter class imbalance can boost performance by 3.44% in the component segmentation task over the most employed alternative, weighted Cross Entropy (wCE). Finally, weighted softmax ensembling is found to be quite effective when used synchronously with the Nested Reg-UNet architecture by providing mIoU improvement of 0.74% in the component segmentation task and 1.14% in the damage segmentation task over a single-architecture baseline. Overall, the best mIoU of 92.50% for the component segmentation task and 84.19% for the damage segmentation task validate the feasibility of these techniques for autonomous bridge component and damage segmentation using RGB images.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.226-228
• /
• 2005

Predicting the cellular location of an unknown protein gives valuable information for inferring the possible function of the protein. For more accurate Prediction system, we need a good feature extraction method that transforms the raw sequence data into the numerical feature vector, minimizing information loss. In this paper we propose new methods of extracting underlying features only from the sequence data by computing pairwise sequence alignment scores. In addition, we use composition based features to improve prediction accuracy. To construct an SVM ensemble from separately trained SVM classifiers, we propose specificity based weighted majority voting . The overall prediction accuracy evaluated by the 5-fold cross-validation reached $88.53\%$ for the eukaryotic animal data set. By comparing the prediction accuracy of various feature extraction methods, we could get the biological insight on the location of targeting information. Our numerical experiments confirm that our new feature extraction methods are very useful forpredicting subcellular localization of proteins.

• Journal of Environmental Impact Assessment
• /
• v.27 no.3
• /
• pp.291-306
• /
• 2018

This study was designed to predict potential habitat of Japanese evergreen oak (Quercus acuta Thunb.) in Korean Peninsula considering its dispersal ability under climate change. We used a species distribution model (SDM) based on the current species distribution and climatic variables. To reduce the uncertainty of the SDM, we applied nine single-model algorithms and the pre-evaluation weighted ensemble method. Two representative concentration pathways (RCP 4.5 and 8.5) were used to simulate the distribution of Japanese evergreen oak in 2050 and 2070. The final future potential habitat was determined by considering whether it will be dispersed from the current habitat. The dispersal ability was determined using the Migclim by applying three coefficient values (${\theta}=-0.005$, ${\theta}=-0.001$ and ${\theta}=-0.0005$) to the dispersal-limited function and unlimited case. All the projections revealed potential habitat of Japanese evergreen oak will be increased in Korean Peninsula except the RCP 4.5 in 2050. However, the future potential habitat of Japanese evergreen oak was found to be limited considering the dispersal ability of this species. Therefore, estimation of dispersal ability is required to understand the effect of climate change and habitat distribution of the species.

• Journal of the Korean Society for Nondestructive Testing
• /
• v.30 no.4
• /
• pp.302-310
• /
• 2010

In this paper, we studied Bagging neural network for predicting defect size of steam generator(SG) tube in nuclear power plant. Bagging is a method for creating an ensemble of estimator based on bootstrap sampling. For predicting defect size of SG tube, we first generated eddy current testing signals for 4 defect patterns of SG tube with various widths and depths. Then, we constructed single neural network(SNN) and Bagging neural network(BNN) to estimate width and depth of each defect. The estimation performance of SNN and BNN were measured by means of peak error. According to our experiment result, average peak error of SNN and BNN for estimating defect depth were 0.117 and 0.089mm, respectively. Also, in the case of estimating defect width, average peak error of SNN and BNN were 0.494 and 0.306mm, respectively. This shows that the estimation performance of BNN is superior to that of SNN.