Loading...
Search for:
feature-selection
0.005 seconds
Total 57 records
Scalable feature selection via distributed diversity maximization
, Article 31st AAAI Conference on Artificial Intelligence, AAAI 2017, 4 February 2017 through 10 February 2017 ; 2017 , Pages 2876-2883 ; Ghadiri, M ; Mirrokni, V ; Zadimoghaddam, M ; Sharif University of Technology
Abstract
Feature selection is a fundamental problem in machine learning and data mining. The majority of feature selection algorithms are designed for running on a single machine (centralized setting) and they are less applicable to very large datasets. Although there are some distributed methods to tackle this problem, most of them are distributing the data horizontally which are not suitable for datasets with a large number of features and few number of instances. Thus, in this paper, we introduce a novel vertically distributable feature selection method in order to speed up this process and be able to handle very large datasets in a scalable manner. In general, feature selection methods aim at...
A new prediction model based on cascade NN for wind power prediction
, Article Computational Economics ; March , 2018 , Pages 1-25 ; 09277099 (ISSN) ; Kiaian Mousavy, S. A ; Dashti, V ; Saeedi, M ; Yousefi, N ; Sharif University of Technology
Springer New York LLC
2018
Abstract
This paper presents a new prediction model based on empirical mode decomposition, feature selection and hybrid forecast engine. The whole structure of proposed model is based on nonstationarity and non-convex nature of wind power signal. The hybrid forecast engine consists of three main stages as; empirical mode decomposition, an intelligent algorithm and three stage neural network. All parameters of proposed neural network will be optimized by intelligent algorithm. Effectiveness of the proposed model is tested with real-world hourly data of wind farms in Canada, Spain and Texas. In order to demonstrate the validity of the proposed model, it is compared with several other wind speed and...
A new prediction model based on cascade NN for wind power prediction
, Article Computational Economics ; Volume 53, Issue 3 , 2019 , Pages 1219-1243 ; 09277099 (ISSN) ; Kiaian Mousavy, S. A ; Dashti, V ; Saeedi, M ; Yousefi, N ; Sharif University of Technology
Springer New York LLC
2019
Abstract
This paper presents a new prediction model based on empirical mode decomposition, feature selection and hybrid forecast engine. The whole structure of proposed model is based on nonstationarity and non-convex nature of wind power signal. The hybrid forecast engine consists of three main stages as; empirical mode decomposition, an intelligent algorithm and three stage neural network. All parameters of proposed neural network will be optimized by intelligent algorithm. Effectiveness of the proposed model is tested with real-world hourly data of wind farms in Canada, Spain and Texas. In order to demonstrate the validity of the proposed model, it is compared with several other wind speed and...
Data Stream Whole Clustering
, M.Sc. Thesis Sharif University of Technology ; Abolhassani, Hassan (Supervisor)
Abstract
Due to the application of data streams in various data sources such as Web click streams, Web pages, and data generated by sensors and satellites, data streams have attracted a huge attention recently. A data stream is an ordered sequence of points that must be accessed in order and can be read only once or a small number of times. For mining such data, the ability to process in one pass along with limited memory usage is very important. Data stream clustering also has received a huge attention in recent years and numerous algorithms are developed in this field. None of them has paid attention to the feature selection problem as an effective factor in clustering quality especially when the...
A clinical decision support system based on support vector machine and binary particle swarm optimisation for cardiovascular disease diagnosis
, Article International Journal of Data Mining and Bioinformatics ; Volume 15, Issue 4 , 2016 , Pages 312-327 ; 17485673 (ISSN) ; Shavandi, H ; Sadeghi, M ; Sharif University of Technology
Inderscience Enterprises Ltd
2016
Abstract
Cardiovascular diseases have been known as one of the main reasons of mortality all around the world. Nevertheless, this disease is preventable if it can be diagnosed in an early stage. Therefore, it is crucial to develop Clinical Decision Support Systems (CDSSs) that are able to help physicians diagnose the disease and its related risks. This study focuses on cardiovascular disease diagnosis in an Iranian community by developing a CDSS, based on Support Vector Machine (SVM) combined with Binary Particle Swarm Optimisation (BPSO). We used SVM as the classifier and benefited enormously from optimisation capabilities of BPSO in model development as well as feature selection. Finally,...
The ensemble approach in comparison with the diverse feature selection techniques for estimating NPPs parameters using the different learning algorithms of the feed-forward neural network
, Article Nuclear Engineering and Technology ; Volume 53, Issue 12 , 2021 , Pages 3944-3951 ; 17385733 (ISSN) ; Sharif University of Technology
Korean Nuclear Society
2021
Abstract
Several reasons such as no free lunch theorem indicate that there is not a universal Feature selection (FS) technique that outperforms other ones. Moreover, some approaches such as using synthetic dataset, in presence of large number of FS techniques, are very tedious and time consuming task. In this study to tackle the issue of dependency of estimation accuracy on the selected FS technique, a methodology based on the heterogeneous ensemble is proposed. The performance of the major learning algorithms of neural network (i.e. the FFNN-BR, the FFNN-LM) in combination with the diverse FS techniques (i.e. the NCA, the F-test, the Kendall's tau, the Pearson, the Spearman, and the Relief) and...
Prediction of gas chromatographic retention indices of a diverse set of toxicologically relevant compounds
, Article Journal of Chromatography A ; Volume 1028, Issue 2 , 2004 , Pages 287-295 ; 00219673 (ISSN) ; Karlovits, M ; Demuth, W ; Stimpfl, T ; Vycudilik, W ; Jalali Heravi, M ; Varmuza, K ; Sharif University of Technology
Elsevier
2004
Abstract
For a set of 846 organic compounds, relevant in forensic analytical chemistry, with highly diverse chemical structures, the gas chromatographic Kovats retention indices have been quantitatively modeled by using a large set of molecular descriptors generated by software Dragon. Best and very similar performances for prediction have been obtained by a partial least squares regression (PLS) model using all considered 529 descriptors, and a multiple linear regression (MLR) model using only 15 descriptors obtained by a stepwise feature selection. The standard deviations of the prediction errors (SEP), were estimated in four experiments with differently distributed training and prediction sets....
A novel method based on empirical mode decomposition for P300-Based detection of deception
, Article IEEE Transactions on Information Forensics and Security ; Volume 11, Issue 11 , 2016 , Pages 2584-2593 ; 15566013 (ISSN) ; Moradi, M. H ; Janghorbani, A ; Sharif University of Technology
Institute of Electrical and Electronics Engineers Inc
Abstract
Conventional polygraphy has several alternatives and one of them is P300-based guilty knowledge test. The purpose of this paper is to apply a new method called empirical mode decomposition (EMD) to extract features from electroencephalogram (EEG) signal. EMD is an appropriate tool to deal with the nonlinear and nonstationary nature of EEG. In the previous studies on the same data set, some morphological, frequency, and wavelet features were extracted only from Pz channel, and used for the detection of guilty and innocent subjects. In this paper, an EMD-based feature extraction was done on EEG recorded signal. Features were extracted from all three recorded channels (Pz, Cz, and Fz) for...
Using empirical covariance matrix in enhancing prediction accuracy of linear models with missing information
, Article 2017 12th International Conference on Sampling Theory and Applications, SampTA 2017, 3 July 2017 through 7 July 2017 ; 2017 , Pages 446-450 ; 9781538615652 (ISBN) ; Shahsavari, S ; Esmaeili, A ; Marvasti, F ; Sharif University of Technology
Abstract
Inference and Estimation in Missing Information (MI) scenarios are important topics in Statistical Learning Theory and Machine Learning (ML). In ML literature, attempts have been made to enhance prediction through precise feature selection methods. In sparse linear models, LASSO is well-known in extracting the desired support of the signal and resisting against noisy systems. When sparse models are also suffering from MI, the sparse recovery and inference of the missing models are taken into account simultaneously. In this paper, we will introduce an approach which enjoys sparse regression and covariance matrix estimation to improve matrix completion accuracy, and as a result enhancing...
Unsupervised feature selection for phoneme sound classification using particle swarm optimization
, Article 5th Iranian Joint Congress on Fuzzy and Intelligent Systems - 16th Conference on Fuzzy Systems and 14th Conference on Intelligent Systems, CFIS 2017, 7 March 2017 through 9 March 2017 ; 2017 , Pages 86-90 ; 9781509040087 (ISBN) ; Bagheri Shourak, S ; Faraji, M. M ; Sharif University of Technology
Abstract
This paper proposes a new method based on Particle Swarm Optimization (PSO) for feature selection in phonemes sound classification. Inspired of biologist's studies, each particle is represented by filterbank which is motivated by human hearing. Thus, we propose a technique in which PSO is used to extract audio features similar to human's ear in order to achieve better classification. We use PSO technique for optimizing particle's filterbank in order to classify sound signals accurately. Then, feature extraction is done by using particle's information. Moreover, a classification method based on nearest neighbor is used. Furthermore, by using a defined fitness function in this paper, the...
Investigating the performance of the supervised learning algorithms for estimating NPPs parameters in combination with the different feature selection techniques
, Article Annals of Nuclear Energy ; Volume 158 , 2021 ; 03064549 (ISSN) ; Sharif University of Technology
Elsevier Ltd
2021
Abstract
Several reasons such as no free lunch theorem indicates that any learning algorithm in combination with a specific feature selection (FS) technique may give more accurate estimation than other learning algorithms. Therefore, there is not a universal approach that outperforms other algorithms. Moreover, due to the large number of FS techniques, some recommended solutions such as using synthetic dataset or combining different FS techniques are very tedious and time consuming. In this study to tackle the issue of more accurate estimation of NPPs parameters, the performance of the major supervised learning algorithms in combination with the different FS techniques which are appropriate for...
Author Identification Using Statistical Methods
, M.Sc. Thesis Sharif University of Technology ; Beigy, Hamid (Supervisor)
Abstract
With the increasing use of the Internet, we are witnessing the exchange of gigabytes of text in cyberspace. Cyberspace makes it possible for individuals to hide their true identity and enter this space with an spurious one. Abuses that occur in online communities due to the use of unknown identities, reduce confidence in this type of communication and create many challenges in this area. Hence the importance of maintaining the security of the space, controling the user-generated content and identifying the authors of texts increases day by day. In this Research we have presented an approach to author identification. This approach is based on modeling the style of the authors on the basis of...
Selection of efficient features for discrimination of hand movements from MEG using a BCI competition IV data set
, Article Frontiers in Neuroscience ; Issue APR , 2012 ; 16624548 (ISSN) ; Shamsollahi, M. B ; Sharif University of Technology
2012
Abstract
The aim of a brain-computer interface (BCI) system is to establish a new communication system that translates human intentions, reflected by measures of brain signals such as magnetoencephalogram (MEG), into a control signal for an output device. In this paper, an algorithm is proposed for discriminating MEG signals, which were recorded during hand movements in four directions. These signals were presented as data set 3 of BCI competition IV. The proposed algorithm has four main stages: pre-processing, primary feature extraction, the selection of efficient features, and classification. The classification stage was a combination of linear SVM and linear discriminant analysis classifiers. The...
Developing an approach to evaluate stocks by forecasting effective features with data mining methods
, Article Expert Systems with Applications ; Volume 42, Issue 3 , February , 2014 , Pages 1325-1339 ; 09574174 (ISSN) ; Modarres, M ; Sharif University of Technology
Elsevier Ltd
2014
Abstract
In this research, a novel approach is developed to predict stocks return and risks. In this three stage method, through a comprehensive investigation all possible features which can be effective on stocks risk and return are identified. Then, in the next stage risk and return are predicted by applying data mining techniques for the given features. Finally, we develop a hybrid algorithm, on the basis of filter and function-based clustering; the important features in risk and return prediction are selected then risk and return re-predicted. The results show that the proposed hybrid model is a proper tool for effective feature selection and these features are good indicators for the prediction...
An Investigation of Data Mining Methods in E-Learning
, M.Sc. Thesis Sharif University of Technology ; Habibi, Jafar (Supervisor)
Abstract
In the pas few years, the use of web-based education systems have grown exponentially spurred by the fact that neither students nor teachers are bound to a specific location and that this form of computer-based education is virtually independent of any specific hardware platforms. These systems can offer a great variety of channels and workspaces to facilitate information sharing and communication between participants in a course, let educators distribute information to students, produce content material, prepare assignments and tests, engage in discussions, manage distance classes and enable collaborative learning with virtual classroom sessions, forums, chats, file storage areas, news...
Trajectory Estimation of a Vehicle Using Stereo Cameras
, M.Sc. Thesis Sharif University of Technology ; Moghadasi, Reza (Supervisor)
Abstract
Visual odometry(VO) is the process of estimating the egomotion of an agent(e.g., vehicle, human, and robot) using the input of a single or multiple cameras attached to it. Application domains include robotics, wearable computing, augmented reality, and automotive. The term was chosen for its similarity to wheel odometry, which incrementally estimates the motion of a vehicle by integrating the number of turns of its wheels over time. Likewise, VO operates by incrementally estimating the pose of the vehicle through examination of the changes that movements induces on the images of its onboard cameras. For the VO to work effectively, there should be sufficient illumination in the environment...
Small-scale building load forecast based on hybrid forecast engine
, Article Neural Processing Letters ; 2017 , Pages 1-23 ; 13704621 (ISSN) ; Talebpour, F ; Safaee, E ; Ghadimi, N ; Abedinia, O ; Sharif University of Technology
Abstract
Electricity load forecasting plays an important role for optimal power system operation. Accordingly, short term load forecast (STLF) is also becoming an important task by researchers to tackle the mentioned problem. As a consequence of the highly non-smooth and volatile trend of the load time series specially in building levels, its STLF is even a more complex procedure than that of a power system. For this purpose, in this paper we proposed a new prediction model based on a new feature selection algorithm and hybrid forecast engine of enhanced version of empirical mode decomposition named sliding window EMD bundled with an intelligent algorithm. The proposed forecast engine is combined...
Small-Scale building load forecast based on hybrid forecast engine
, Article Neural Processing Letters ; Volume 48, Issue 1 , 2018 , Pages 329-351 ; 13704621 (ISSN) ; Talebpour, F ; Safaee, E ; Ghadimi, N ; Abedinia, O ; Sharif University of Technology
Springer New York LLC
2018
Abstract
Electricity load forecasting plays an important role for optimal power system operation. Accordingly, short term load forecast (STLF) is also becoming an important task by researchers to tackle the mentioned problem. As a consequence of the highly non-smooth and volatile trend of the load time series specially in building levels, its STLF is even a more complex procedure than that of a power system. For this purpose, in this paper we proposed a new prediction model based on a new feature selection algorithm and hybrid forecast engine of enhanced version of empirical mode decomposition named sliding window EMD bundled with an intelligent algorithm. The proposed forecast engine is combined...
Development of a new features selection algorithm for estimation of NPPs operating parameters
, Article Annals of Nuclear Energy ; Volume 146 , October , 2020 ; Ghanbari, M ; Ghofrani, M. B ; Sharif University of Technology
Elsevier Ltd
2020
Abstract
One of the most important challenges in target parameters estimation via model-free methods is selection of the most effective input parameters namely features selection (FS). Indeed, irrelevant features can degrade the estimation performance. In the current study, the challenge of choosing among the several plant parameters is tackled by means of the innovative FS algorithm named ranking of features with minimum deviation from the target parameter (RFMD). The selected features accompanied with the stable and the fast learning algorithm of multilayer perceptron (MLP) neural network (i.e. Levenberg-Marquardt algorithm) which is a combination of gradient descent and Gauss-newton learning...
Multiple partial discharge sources separation using a method based on laplacian score and correlation coefficient techniques
, Article Electric Power Systems Research ; Volume 210 , 2022 ; 03787796 (ISSN) ; Vakilian, M ; Firuzi, K ; Sharif University of Technology
Elsevier Ltd
2022
Abstract
Partial discharge (PD) activity can be destructive to the transformer insulation, and ultimately may result in total breakdown of the insulation. Partial discharge sources identification in a power transformer enables the operator to evaluate the transformer insulation condition during its lifetime. In order to identify the PD source; in the case of presence of multiple sources; the first step is to capture the PD signals and to extract their specific features. In this contribution, the frequency domain analysis, the time domain analysis and the wavelet transform are employed for feature extraction purpose. In practice, there might be plenty of features, and in each scenario, only some of...