Sharif Digital Repository

This paper addresses the midterm preventive maintenance outage scheduling problem of thermal generating units which is becoming increasingly important due to the aging of power generation fleet. In this context, a novel midterm preventive maintenance outage scheduler is proposed based on decision tree and mixed integer linear formation which explicitly considers the thermal units aging momentum in terms of failure rate. This allows the system operators to determine the thermal units' maintenance outage window based on the cost/benefit analysis of preventive maintenance tasks while optimizing the time interval between consecutive maintenance tasks. Additionally, the division of the year-long... 

In this paper, the joint problem of project selection and project scheduling under uncertain environment is formulated, analyzed, and solved by multistage stochastic programs. First of all, a general mathematical formulation which can address several versions of the problem is presented. A multi-period project selection and scheduling problem is introduced and modeled by multistage stochastic programs, which are effective for solving long-term planning problems under uncertainty. A set of scenarios and corresponding probabilities is applied to model the multivariate random data process (costs or revenues, available budget, chance of success). Then, due to computational complexity, a scenario... 

Decision tree-clustered context-dependent hidden semi-Markov models (HSMMs) are typically used in statistical parametric speech synthesis to represent probability densities of acoustic features given contextual factors. This paper addresses three major limitations of this decision tree-based structure: (i) The decision tree structure lacks adequate context generalization. (ii) It is unable to express complex context dependencies. (iii) Parameters generated from this structure represent sudden transitions between adjacent states. In order to alleviate the above limitations, many former papers applied multiple decision trees with an additive assumption over those trees. Similarly, the current... 

Background and aims: Among different types of occupational accidents in the construction industry, falls and falling objects accidents (group I) account for 44% of construction accidents. Hit by vehicle, electric shock, collapse in the excavation and fire or explosion accidents (group II), while are only 7% frequent, make up about 26% of all fatalities and total disabling accidents. The aim of this study is to investigate these two groups of accidents and to discuss the obtained results in order to identify the potential hazards of construction industry. Methods: Data mining methods are employed to analyze data in this research. Hence, 21864 data records which were provided by Social... 

Crohn's disease and celiac disease belong to a group of autoimmune conditions that affect the digestive system, specifically the small intestine. They both attack the digestive tract and share many symptoms. Thus, the discovery of proper methods would be a major step toward differentiating celiac disease from Crohn's disease. The aim of this study was to search for the metabolic biomarkers to differentiate between these two diseases. Proton nuclear magnetic resonance spectroscopy (1H NMR) was employed as the metabolic profiling method to look for serum metabolites that differentiate between celiac disease and Crohn's disease. Classification of celiac disease and Crohn's disease was done... 

Internet traffic classification is important in many aspects of network management such as data exploitation detection, malicious user identification, and restricting application traffic. Previously, features such as port and protocol numbers have been used to classify traffic, but these features can now be changed easily, making their use in traffic classification inadequate. Consequently, traffic classification based on machine learning (ML) is now employed. The number of features used in an ML algorithm has a significant impact on performance, in particular accuracy. In this paper, a minimum best feature set is chosen using a supervised method to obtain uncorrelated features. Outlier... 

Trip destination and mode choice are highly influenced by travelers' perceptions and behaviors; selecting a destination and a vehicle for a trip are two interdependent problems. This paper presents and applies a disaggregate joint model for traveler destination and mode choice. The choice model uses fuzzy set and probability theory to deal with the uncertainty embedded in travelers' perceptions and behaviors. The model is structured as a decision tree in which the fuzzy and non-fuzzy classification of influential variables regarding destination selection and mode choice expand the tree. The most influential explanatory variables among all the variables categorized for travelers' household,... 

Speaker adaptive speech synthesis based on Hidden Semi-Markov Model (HSMM) has been demonstrated to be dramatically effective in the presence of confined amount of speech data. However, we could intensify this effectiveness by training the average voice model appropriately. Hence, this study presents a new method for training the average voice model. This method guarantees that data from every speaker contributes to all the leaves of decision tree. We considered this fact that small training data and highly diverse contexts of training speakers are considered as disadvantages which degrade the quality of average voice model impressively, and further influence the adapted model and synthetic... 

Background: The abundance of gene expression microarray data has led to the development of machine learning algorithms applicable for tackling disease diagnosis, disease prognosis, and treatment selection problems. However, these algorithms often produce classifiers with weaknesses in terms of accuracy, robustness, and interpretability. This paper introduces fuzzy support vector machine which is a learning algorithm based on combination of fuzzy classifiers and kernel machines for microarray classification.Results: Experimental results on public leukemia, prostate, and colon cancer datasets show that fuzzy support vector machine applied in combination with filter or wrapper feature selection... 

Parkinson's disease (PD) is a neurodegenerative disease, which is not easily diagnosed using clinical tests and the discovery of proper methods would be a major step towards a successful diagnosis. In the present study, we employed metabolic profiling using proton nuclear magnetic resonance spectroscopy to find metabolites in serum, which are helpful for the diagnosis of PD. Classification of PD and healthy subject was done using random forest. Serum levels of selenium measured by atomic absorption spectrometry in PD group were lower than the serum selenium levels in the control group. The metabolites causing selenium changes in PD patients were identified using random forest, and a model... 

Data streams are endless flow of data produced in high speed, large size and usually non-stationary environments. The main property of these streams is the occurrence of concept drifts. Using decision trees is shown to be a powerful approach for accurate and fast learning of data streams. In this paper, we present an incremental regression tree that can predict the target variable of newly incoming instances. The tree is updated in the case of occurring concept drifts either by altering its structure or updating its embedded models. Experimental results show the effectiveness of our algorithm in speed and accuracy aspects in comparison to the best state-of-the-art methods  

Online voltage security assessment is necessary to choose appropriate remedial and corrective actions in order to prevent a large-scale blackout. This paper presents a new online voltage security assessment method based on wide-area measurements and decision-tree algorithm. For the predicted load and generation variation scenarios (i.e., a day ahead), the database is obtained using continuation power flow. Moreover, each operating point is labeled as 'secure' or 'insecure' from voltage stability points of view based on WECC voltage security criteria. Decision trees are trained on the subset of the existing data by applying two famous splitting rules and various predictors. Bagging and... 

In the last few years there has been growing interest in the use of functional Magnetic Resonance Imaging (fMRI) for brain mapping. To decode brain patterns in fMRI data, we need reliable and accurate classifiers. Towards this goal, we compared performance of eleven popular pattern recognition methods. Before performing pattern recognition, applying the dimensionality reduction methods can improve the classification performance; therefore, seven methods in region of interest (RDI) have been compared to answer the following question: which dimensionality reduction procedure performs best? In both tasks, in addition to measuring prediction accuracy, we estimated standard deviation of... 

Reviewing the various Partial Discharges (PD data mining researches which have been reported so far, this study compares the performance of different feature spaces and different classifiers employed for PD classification in insulation condition monitoring of power transformers. In this process, first a knowledge basis is developed through construction of 4 different types of PD models in the high voltage laboratory. Background noise is considered as one class in this knowledge basis. The high frequency time domain current signals of high voltage equipment are captured over one power frequency cycle. The single PD activities within this captured signal are extracted by application of a... 

Under smart grid environment, islanding detection plays an important role in reliable operation of distributed generation (DG) units. In this paper an intelligent-based islanding detection algorithm for PV and DFIG units is proposed. Decision tree algorithm is used to classify islanding detection instances. This algorithm is rapid, simple, intelligible and easy to interpret. The error rate of this method is reduced by Adaptive Boosting (AdaBoost) technique. The proposed method is tested on a distribution system including PV, DFIG and synchronous generator. Probable events in the system are simulated under diverse operating states to generate classification data set. First and second order... 

This paper introduces a novel approach for power system fault diagnosis based on synchronized phasor measurements during the fault. The synchronized measurements are obtained in real time from Phasor Measurement Units (PMUs) and compared with offline thresholds determined by decision trees (DTs) to diagnose the fault. The DTs have already been trained offline using detailed power system analysis for different fault cases. While the traditional methods for fault diagnosis use the status of protective relays (PRs) and circuit breakers (CBs) to infer the fault section in the power system, the proposed method uses the available signals following the fault and thus can be trusted even in case of... 

In this paper, we introduce a new hybrid learning algorithm, called DTGP, to construct cost-sensitive classifiers. This algorithm uses a decision tree as its basic classifier and the constructed decision tree will be pruned by a genetic programming algorithm using a fitness function that is sensitive to misclassification costs. The proposed learning algorithm has been examined through six cost-sensitive problems. The experimental results show that the proposed learning algorithm outperforms in comparison to some other known learning algorithms like C4.5 or naïve Bayesian