• Journal of Information Processing Systems
• /
• 제19권5호
• /
• pp.652-662
• /
• 2023

Identifying highly discriminating genes is a critical step in tumor recognition tasks based on microarray gene expression profile data and machine learning. Gene selection based on tree models has been the subject of several studies. However, these methods are based on a single-tree model, often not robust to ultra-highdimensional microarray datasets, resulting in the loss of useful information and unsatisfactory classification accuracy. Motivated by the limitations of single-tree-based gene selection, in this study, ensemble gene selection methods based on multiple-tree models were studied to improve the classification performance of tumor identification. Specifically, we selected the three most representative tree models: ID3, random forest, and gradient boosting decision tree. Each tree model selects top-n genes from the microarray dataset based on its intrinsic mechanism. Subsequently, three ensemble gene selection methods were investigated, namely multipletree model intersection, multiple-tree module union, and multiple-tree module cross-union, were investigated. Experimental results on five benchmark public microarray gene expression datasets proved that the multiple tree module union is significantly superior to gene selection based on a single tree model and other competitive gene selection methods in classification accuracy.

• 한국정보과학회논문지:데이타베이스
• /
• 제31권3호
• /
• pp.252-260
• /
• 2004

최근에 움직이는 객체의 미래 위치를 위한 TPR-tree가 제안되었으며, TPR-tree를 이용한 많은 연구들이 제안되었다. 그러나, TPR-tree가 시공간 데이타베이스에서 널리 사용됨에도 불구하고, TPR-tree를 위한 비용 모델은 제안되지 않았다. R-tree와 같은 공간 색인을 위한 비용 모델들은 움직이는 객체들의 미래 위치를 전혀 고려하지 않기 때문에, TPR-tree에 대한 시공간 질의를 위한 디스크 액세스 수를 정확하게 예측하지 못한다. 본 논문에서는 움직이는 객체들의 미래 위치를 고려한 TPR-tree를 위한 비용 모델을 처음으로 제안한다. 다양한 실험 결과, 제안된 TPR-tree의 비용 모델은 디스크 액세스 수를 정확하게 예측한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권7호
• /
• pp.3284-3306
• /
• 2018

This study proposes a method to develop an authoring system(GeoMaTree) for diverse trees that constitute a virtual landscape. The GeoMaTree system enables the simple, intuitive production of an efficient structure, and supports real-time processing. The core of the proposed system is a procedural modeling based on a mathematical model and an application that supports digital content creation on diverse platforms. The procedural modeling allows users to control the complex pattern of branch propagation through an intuitive process. The application is a multi-resolution 3D model that supports appropriate optimization for a tree structure. The application and a compatible function, with commercial tools for supporting the creation of realistic synthetic images and virtual landscapes, are implemented, and the proposed system is applied to a variety of 3D image content.

• 대한음성학회지:말소리
• /
• 제45호
• /
• pp.117-126
• /
• 2003

Decision tree-based state tying has been proposed in recent years as the most popular approach for clustering the states of context-dependent hidden Markov model-based speech recognition. The aims of state tying is to reduce the number of free parameters and predict state probability distributions of unseen models. But, when doing state tying, the size of a decision tree is very important for word independent recognition. In this paper, we try to construct optimized decision tree based on the average of feature vectors in state pool and the number of seen modes. We observed that the proposed optimal decision tree is effective in predicting the state probability distribution of unseen models.

• 상하수도학회지
• /
• 제20권6호
• /
• pp.789-797
• /
• 2006

This study was conducted to establish the refined model for prediction of flocculation/sedimentation efficiency in factual drinking water treatment plants using model tree technique. In order to carry out machine leaning for determining each linear model, five parameters; time, coagulant dose, raw water turbidity, SCD and conductivity, which were measured and collected from the field (K_DWTP), were selected and used. The existing analytical models developed by previous researchers were used only to examine closely the mechanism of flocculation rather than to apply it for practical purpose. The refined model established using model tree technique in this study could predict the factual sedimentation efficiency accurately (below 9% of average absolute error). Also, in aspect of engineering convenience, without any additional manipulation of parameters, it can be applied to practical works.

• 디지털산업정보학회논문지
• /
• 제6권3호
• /
• pp.49-57
• /
• 2010

We extended a general attack tree to apply cyber attack model for network vulnerability analysis. We defined an extended cyber attack tree (E-CAT) which extends the general attack tree by associating each node of the tree with a transition of attack that could have contributed to the cyber attack. The E-CAT resolved the limitation that a general attack tree can not express complex and sophisticate attacks. Firstly, the Boolean expression can simply express attack scenario with symbols and codes. Secondary, An Attack Generation Probability is used to select attack method in an attack tree. A CONDITION-composition can express new and modified attack transition which a aeneral attack tree can not express. The E-CAT is possible to have attack's flexibility and improve attack success rate when it is applied to cyber attack model.

• International Journal of Advanced Culture Technology
• /
• 제11권3호
• /
• pp.310-314
• /
• 2023

The purpose of this study is to predict the remaining capacity of lithium-ion batteries and evaluate their performance using five artificial intelligence models, including linear regression analysis, decision tree, random forest, neural network, and ensemble model. We is in the study, measured Excel data from the CS2 lithium-ion battery was used, and the prediction accuracy of the model was measured using evaluation indicators such as mean square error, mean absolute error, coefficient of determination, and root mean square error. As a result of this study, the Root Mean Square Error(RMSE) of the linear regression model was 0.045, the decision tree model was 0.038, the random forest model was 0.034, the neural network model was 0.032, and the ensemble model was 0.030. The ensemble model had the best prediction performance, with the neural network model taking second place. The decision tree model and random forest model also performed quite well, and the linear regression model showed poor prediction performance compared to other models. Therefore, through this study, ensemble models and neural network models are most suitable for predicting the remaining capacity of lithium-ion batteries, and decision tree and random forest models also showed good performance. Linear regression models showed relatively poor predictive performance. Therefore, it was concluded that it is appropriate to prioritize ensemble models and neural network models in order to improve the efficiency of battery management and energy systems.

• 한국해안·해양공학회논문집
• /
• 제23권1호
• /
• pp.109-117
• /
• 2011

국내 경사식 방파제의 대표적인 피복재인 Tetrapod는 대부분 경험식을 사용하여 중량을 산정한다. 경험식은 수리 실험의 결과를 곡선맞춤(curve-fitting)하여 제안되기 때문에 실험 오차에 따른 불확실성이 내포되어 있다. 이런 불확실성을 최소화하기 위해 인경신경망과 M5' model tree를 사용하여 피복재 안정수를 예측하였다. 각 모형의 불확실성의 정도는 예측된 안정수와 수리실험의 안정수 사이의 일치지수(index of agreement)를 사용하여 평가하였다. 일치지수가 가장 큰 인공신경망은 우수한 예측 능력을 가지고 있지만 일반 설계자들이 쉽게 사용할 수 없는 큰 단점이 있다. 반면에 M5' model tree는 인공신경망보다는 예측 능력이 조금 떨어지지만 기존의 경험식보다는 예측능력이 우수하고 또한 일반 설계자들이 쉽게 사용할 수 있는 공식의 형태로 주어지는 장점이 있다.

• 한국경영과학회지
• /
• 제40권2호
• /
• pp.19-29
• /
• 2015

The purpose of this study is to suggest the grid search method for selecting an optimal decision tree model. It chooses optimal values for the maximum depth of tree and the minimum number of observations that must exist in a node in order for a split to be attempted. Therefore, the grid search method guarantees building a decision tree model that shows more precise and stable classifying performance. Through empirical analysis using data of job satisfaction of workplace reserve force commanders, we show that the grid search method helps us generate an optimal decision tree model that gives us hints for the improvement direction of labor conditions of Korean workplace reserve force commanders.

• 사상체질의학회지
• /
• 제22권3호
• /
• pp.87-97
• /
• 2010

1. Objectives: In SCM, a personal Sasang constitution must be determined accurately before any Sasang treatment. The purpose of this study is to develop an objective method for classification of Sasang constitution. 2. Methods: We collected samples from 5 centers where SCM is practiced, and applied two-stage decision tree analysis on these samples. We recruited samples from 5 centers. The collected data were from subjects whose response to herbal medicine was confirmed according to Sasang constitution. 3. Results: The two-stage decision tree model shows higher classification power than a simple decision tree model. This study also suggests that gender must be considered in the first stage to improve the accuracy of classification. 4. Conclusions: We identified important factors for classifying Sasang constitutions through two-stage decision tree analysis. The two-stage decision tree model shows higher classification power than a simple decision tree model.