Search for: random-forest
0.007 seconds
Total 21 records

    Prediction of shear strength parameters of hydrocarbon contaminated sand based on machine learning methods

    , Article Georisk ; 2020 Rezaee, M ; Mojtahedi, S. F. F ; Taherabadi, E ; Soleymani, K ; Pejman, M ; Sharif University of Technology
    Taylor and Francis Ltd  2020
    The objective of this paper is to predict the effect of hydrocarbon contamination on the shear strength parameters of sand by using various machine learning platforms. Multilayer perceptron, support vector machine, random forest, gradient boosting method, and multi-output support vector machine were methods used to predict the hydrocarbon contamination impacts on the internal friction angle and cohesion of contaminated sand. Random forest exhibited the best results for cohesion, whereas, for the friction angle, the gradient boosting method outperformed other approaches. Moreover, the multi-output support vector machine yielded better results than those pertaining to a single support vector... 

    Predicting Football Match Results by Machine Learning Methods

    , M.Sc. Thesis Sharif University of Technology Behradfar, Mohsen (Author) ; Rafiee, Majid (Supervisor)
    Nowadays, electrical energy is one of the crucial requirements in human beings’ lives. Since the majority of the world’s energy is met by fossil fuel, the problems such as global warming, the reduction of fossil fuel resources, and unpredictable oscillation of the prices of such fuels have led to a serious crisis for the people of the world. Furthermore, due to the increasing energy demand, the economic development of most countries has a strong correlation with fossil fuel prices. The abovementioned problems have made a lot of countries take alternative policies in terms of generating energy, one of which is using renewable energy resources. This approach is reported to be clean and... 

    A comparative study of various machine learning methods for performance prediction of an evaporative condenser

    , Article International Journal of Refrigeration ; Volume 126 , 2021 , Pages 280-290 ; 01407007 (ISSN) Behnam, P ; Faegh, M ; Shafii, M. B ; Khiadani, M ; Sharif University of Technology
    Elsevier Ltd  2021
    Evaporative condensers are regarded as highly-efficient and eco-friendly heat exchangers in refrigeration systems. Data-driven methods can play a key role in performance prediction of evaporative condensers, conducted without the complexity of theoretical analysis. In this study, four machine learning models including multi-layer perceptron artificial neural network (ANNMLP), support vector regression (SVR), decision tree (DT), and random forest (RF) models have been employed to predict heat transfer rate and overall heat transfer coefficient of a small-scale evaporative condenser functioning under a wide range of working conditions. A set of experimental tests were conducted, where inlet... 

    Font recognition for Persian optical character recognition system

    , Article Iranian Conference on Machine Vision and Image Processing, MVIPVolume 2017-November, 19 April 2018 ; Volume 2017 -November , April , 2018 , Pages 252-257 ; 21666776 (ISSN) ; 9781538644041 (ISBN) Eghbali, K ; Veisi, H ; Mirzaie, M ; Mohseni Behbahani, Y ; Sharif University of Technology
    IEEE Computer Society  2018
    Font recognition is one of the pre-processing steps in optical character recognition (OCR) systems that affects on their performance. In this paper two methods are proposed for Persian font recognition. In the first method, Gabor filter is used for feature extraction from the images, then principle component analysis (PCA) applied to reduce feature dimensions and finally, a multi-layer Perceptron (MLP) neural network is used for the classification. In the second techniques, random forest is utilized for recognizing fonts. For evaluation, a dataset includes 10 popular Persian fonts is used. The proposed Gabor-PCA-MLP method has achieved 98.70% of F-measure, and random forest resulted in of... 

    An Improved Clustering Method of Data Mining in Healthcare and Its Implementation

    , M.Sc. Thesis Sharif University of Technology Shourabizadeh, Hamed (Author) ; Akhavan Niaki, Taghi (Supervisor)
    In this study, a brief definition of data mining and its variants were mentioned. Then the methods and algorithms for clustering and their application in the field of healthcare is studied. Concidering the available data for anemia disease, including numeric and categorical attributes, the k-medoids clustering algorithm was selected. This algorithm is one of the simple, powerful and most widely used methods for clustering. The drawbacks of this algorithm are as follow: requires a user input on the number of clusters, depends on the initial data and traps in the local optima. In this thesis, an improved method of clustering-based on Random Forest and k-medoids algorithms has been developed.... 

    Developing an Artificial Intelligence Algorithm for Diagnosis and Prognosis of Failures

    , M.Sc. Thesis Sharif University of Technology Chenariyan Nakhaee, Muhammad (Author) ; Houshmand, Mahmood (Supervisor) ; Fattahi, Omid (Co-Advisor)
    Prognostics is necessary to ensure the reliability and safety of lithium-ion batteries for hybrid electric vehicles or satellites. This process can be achieved by capacity estimation, which is a direct fading indicator for assessing the state of health of a battery. However, the capacity of a lithium-ion battery onboard is difficult to monitor. This paper presents a data-driven approach for capacity estimation. First, new features are extracted from cyclic charge/discharge cycles and used as health indicators. Three algorithms are used to characterize the relationship between extracted features and battery capacity. Decision tree, random forest and boosting algorithms are trained using a... 

    Credit Risk Measurement of Loan Portfolio Based on the Classification of Debtors Using Machine Learning

    , M.Sc. Thesis Sharif University of Technology Ahmadnejad Saein, Mohammad Reza (Author) ; Zamani, Shiva (Supervisor) ; Haghpanah, Farshad (Co-Supervisor)
    Credit risk is the possibility of a loss resulting from a borrower's failure to repay a loan or meet contractual obligations. All banks and financial institutions need to manage credit risk of their lending portfolios to maximize risk-adjusted rate of return and obey regulatory rules. The most commonly used method for determining credit risk is to calculate the maximum loss within the “Value at Risk” framework.Previous studies proposed different models for Credit VaR calculation like Vasicek Model and Credit Risk Plus Model. Furthermore, due to the high growth of computing power and easy access to information, the application of data-driven models such as Machine Learning has been increasing... 

    Performance Evaluation of Machine Learning and Statistical Approaches for Wildfire Modeling and Prediction

    , M.Sc. Thesis Sharif University of Technology Mehrabi, Majid (Author) ; Moghim, Sanaz (Supervisor)
    Wildfires are complex phenomena with many indeterminate and highly unpredictable driving factors that have remained unresolved. During the last decade, machine learning methods have successfully excelled in wildfire prediction as an alternative to traditional field research methods by elucidating the relationship between historical wildfire events and various important variables. The main purpose of this research is to evaluate the random forest machine learning approach and the logistic regression statistical approach to prepare a wildfire susceptibility map using data related to historical wildfires and effective variables in the Okanogan region in Washington province of the United States... 

    Algorithms of Genome-Wide Association Studies

    , M.Sc. Thesis Sharif University of Technology Valishirin, Hossein (Author) ; Foroughmand Aarabi, Mohammad Hadi (Supervisor)
    The field of Genome-Wide Asocciation Studies (GWAS) plays a vital role in understanding the genetic basis of complex traits and diseases. In this thesis, the focus is on investigating the effectiveness of two approaches combining Differential Evolution (DE) with Random Forest (RF) and support vector machine (SVM) for feature selection in the context of GWAS. Arabidopsois Thaliana dataset is used as experimental dataset for comparative analysis. The main goal is to achieve more efficient feature selection while maintaining competitive accuracy compared to RF and SVM without using DE. This research includes conducting experiments using DE with RF and DE with SVM followed by a comprehensive... 

    A metabonomics investigation of multiple sclerosis by nuclear magnetic resonance

    , Article Magnetic Resonance in Chemistry ; Volume 51, Issue 2 , DEC , 2013 , Pages 102-109 ; 07491581 (ISSN) Mehrpour, M ; Kyani, A ; Tafazzoli, M ; Fathi, F ; Joghataie, M. T ; Sharif University of Technology
    Multiple sclerosis (MS) is a nervous system disease that affects the fatty myelin sheaths around the axons of the brain and spinal cord, leading to demyelination and a broad range of signs and symptoms. MS can be difficult to diagnose because its signs and symptoms may be similar to other medical problems. To find out which metabolites in serum are effective for the diagnosis of MS, we utilized metabolic profiling using proton nuclear magnetic resonance spectroscopy (1H-NMR). Random forest (RF) was used to classify the MS patients and healthy subjects. Atomic absorption spectroscopy was used to measure the serum levels of selenium. The results showed that the levels of selenium were lower in... 

    Persian handwritten digit recognition by random forest and convolutional neural networks

    , Article 9th Iranian Conference on Machine Vision and Image Processing,18 November 2015 through 19 November 2015 ; Volume 2016-February , 2015 , Pages 37-40 ; 21666776 (ISSN) ; 9781467385398 (ISBN) Zamani, Y ; Souri, Y ; Rashidi, H ; Kasaei, S ; Sharif University of Technology
    IEEE Computer Society 
    Persian handwritten digit recognition has attracted some interests in the research community by introduction of large Hoda dataset. In this paper, the well-known random forest (RF) and convolutional neural network (CNN) algorithms are investigated for Persian handwritten digit recognition on the Hoda dataset. Using the Hoda dataset as a standard testbed, we have performed some experiments with different preprocessing steps, feature types, and baselines. It is then shown that RFs and CNNs perform competitively with the state-of-the-art methods on this dataset, while CNNs being the fastest if appropriate hardware is available  

    Detecting malicious applications using system services request behavior

    , Article 16th EAI International Conference on Mobile and Ubiquitous Systems: Computing, Networking and Services, MobiQuitous 2019, 12 November 2019 through 14 November 2019 ; 2019 , Pages 200-209 ; 9781450372831 (ISBN) Salehi, M ; Amini, M ; Crispo, B ; Sharif University of Technology
    Association for Computing Machinery  2019
    Widespread growth in Android malware stimulates security researchers to propose different methods for analyzing and detecting malicious behaviors in applications. Nevertheless, current solutions are ill-suited to extract the fine-grained behavior of Android applications accurately and efficiently. In this paper, we propose ServiceMonitor, a lightweight host-based detection system that dynamically detects malicious applications directly on mobile devices. ServiceMonitor reconstructs the fine-grained behavior of applications based on their interaction with system services (i.e. SMS manager, camera, wifi networking, etc). ServiceMonitor monitors the way applications request system services in... 

    Predicting scientific research trends based on link prediction in keyword networks

    , Article Journal of Informetrics ; Volume 14, Issue 4 , 2020 Behrouzi, S ; Shafaeipour Sarmoor, Z ; Hajsadeghi, K ; Kavousi, K ; Sharif University of Technology
    Elsevier Ltd  2020
    The rapid development of scientific fields in this modern era has raised the concern for prospective scholars to find a proper research field to conduct their future studies. Thus, having a vision of future could be helpful to pick the right path for doing research and ensuring that it is worth investing in. In this study, we use article keywords of computer science journals and conferences, assigned by INSPEC controlled indexing, to construct a temporal scientific knowledge network. By observing keyword networks snapshots over time, we can utilize the link prediction methods to foresee the future structures of these networks. We use two different approaches for this link prediction problem.... 

    Capital Market Forecasting with Machine Learning Model and Comparing it with Forecasting Using System Dynamic

    , M.Sc. Thesis Sharif University of Technology Kazem Dehbashi, Sina (Author) ; Kianfar, Farhad (Supervisor)
    Prediction is an important issue in many areas. Proper planning for the future requires careful forecasting, so providing accurate methods, especially in the financial field, is invaluable. In this study, the main problem is predicting the price of global gold. Factors for gold prices include oil, gas, silver, soybeans, copper, the s & p500 index, the Dow Jones index, the British and Japanese stock market indices, the dollar index, the multi-currency exchange rate (pound-euro-yuan-yen) with the dollar to The title of the influential factors in this research is considered. The time frame of this research is daily, in other words, the data is collected on a daily basis and the goal is to... 

    1H NMR based metabolic profiling in Crohn's disease by random forest methodology

    , Article Magnetic Resonance in Chemistry ; Vol. 52, issue. 7 , July , 2014 , p. 370-376 Fathi, F ; Majari-Kasmaee, L ; Mani-Varnosfaderani, A ; Kyani, A ; Rostami-Nejad, M ; Sohrabzadeh, K ; Naderi, N ; Zali, M. R ; Rezaei-Tavirani, M ; Tafazzoli, M ; Arefi-Oskouie, A ; Sharif University of Technology
    The present study was designed to search for metabolic biomarkers and their correlation with serum zinc in Crohn's disease patients. Crohn's disease (CD) is a form of inflammatory bowel disease that may affect any part of the gastrointestinal tract and can be difficult to diagnose using the clinical tests. Thus, introduction of a novel diagnostic method would be a major step towards CD treatment.Proton nuclear magnetic resonance spectroscopy ( 1H NMR) was employed for metabolic profiling to find out which metabolites in the serum have meaningful significance in the diagnosis of CD. CD and healthy subjects were correctly classified using random forest methodology. The classification model for... 

    NMR based metabonomics study on celiac disease in the blood serum

    , Article Gastroenterology and Hepatology from Bed to Bench ; Volume 6, Issue 4 , 2013 , Pages 190-194 ; 20082258 (ISSN) Fathi, F ; Ektefa, F ; Arefi Oskouie, A ; Rostami, K ; Rezaei Tavirani, M ; Mohammad Alizadeh, A. H ; Tafazzoli, M ; Rostami Nejad, M ; Sharif University of Technology
    Aim: The aim of this study is to look for the proper methods that would be a major step towards untreated CD diagnosis and seek the metabolic biomarkers causes of CD and compare them to control group. Background: Celiac disease (CD) is a common autoimmune disorder that is not easily diagnosed using the clinical tests. Patients and methods: Thirty cases and 30 controls were entered into this study. Metabolic profiling was obtained using proton nuclear magnetic resonance spectroscopy (1HNMR) to seek metabolites that are helpful for the detection of CD. Classification of CD and healthy subject was done using random forest (RF). Results: The obtained classification model showed an 89% correct... 

    Semantic segmentation of RGB-D images using 3D and local neighbouring features

    , Article 2015 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2015, 23 November 2015 through 25 November 2015 ; 2015 ; 9781467367950 (ISBN) Fooladgar, F ; Kasaei, S ; Sharif University of Technology
    Institute of Electrical and Electronics Engineers Inc  2015
    3D scene understanding is one of the most important problems in the field of computer vision. Although, in the past decades, considerable attention has been devoted on the 2D scene understanding problem, now with the development of the depth sensors (like Microsoft Kinect), the 3D scene understanding has become a very challenging task. Traditionally, the scene understanding problem was considered as the semantic labeling of each image pixel. Semantic labeling of RGB-D images has not attained a comparable success, as the RGB semantic labeling, due to the lack of a challenging dataset. With the introduction of an RGB-D dataset, called NYU-V2, it became possible to propose a novel method to... 

    Automated detection of autism spectrum disorder using a convolutional neural network

    , Article Frontiers in Neuroscience ; Volume 13 , 2020 Sherkatghanad, Z ; Akhondzadeh, M ; Salari, S ; Zomorodi Moghadam, M ; Abdar, M ; Acharya, U. R ; Khosrowabadi, R ; Salari, V ; Sharif University of Technology
    Frontiers Media S.A  2020
    Background: Convolutional neural networks (CNN) have enabled significant progress in speech recognition, image classification, automotive software engineering, and neuroscience. This impressive progress is largely due to a combination of algorithmic breakthroughs, computation resource improvements, and access to a large amount of data. Method: In this paper, we focus on the automated detection of autism spectrum disorder (ASD) using CNN with a brain imaging dataset. We detected ASD patients using most common resting-state functional magnetic resonance imaging (fMRI) data from a multi-site dataset named the Autism Brain Imaging Exchange (ABIDE). The proposed approach was able to classify ASD... 

    Metabolomics analysis of the saliva in patients with chronic hepatitis b using nuclear magnetic resonance: A pilot study

    , Article Iranian Journal of Basic Medical Sciences ; Volume 22, Issue 9 , 2019 , Pages 1044-1049 ; 20083866 (ISSN) Gilany, K ; Mohamadkhani, A ; Chashmniam, S ; Shahnazari, P ; Amini, M ; Arjmand, B ; Malekzadeh, R ; Nobakht Motlagh Ghoochani, B. F ; Sharif University of Technology
    Mashhad University of Medical Sciences  2019
    Objective(s): Hepatitis B virus infection causes chronic disease such as cirrhosis and hepatocellular carcinoma. The metabolomics investigations have been demonstrated to be related to pathophysiologic mechanisms in many disorders such as hepatitis B infection. The aim of this study was to investigate the saliva metabolic profile of patients with chronic hepatitis B infection and to identify underlying mechanisms as well as potential biomarkers associated with the disease. Materials and Methods: Saliva from 16 healthy subjects and 20 patients with chronic hepatitis B virus were analyzed by nuclear magnetic resonance (NMR). Then, multivariate statistical analysis was performed to identify... 

    The 2017 and 2018 Iranian Brain-Computer interface competitions

    , Article Journal of Medical Signals and Sensors ; Volume 10, Issue 3 , 2020 , Pages 208-216 Aghdam, N ; Moradi, M ; Shamsollahi, M ; Nasrabadi, A ; Setarehdan, S ; Shalchyan, V ; Faradji, F ; Makkiabadi, B ; Sharif University of Technology
    Isfahan University of Medical Sciences(IUMS)  2020
    This article summarizes the first and second Iranian brain-computer interface competitions held in 2017 and 2018 by the National Brain Mapping Lab. Two 64-channel electroencephalography (EEG) datasets were contributed, including motor imagery as well as motor execution by three limbs. The competitors were asked to classify the type of motor imagination or execution based on EEG signals in the first competition and the type of executed motion as well as the movement onset in the second competition. Here, we provide an overview of the datasets, the tasks, the evaluation criteria, and the methods proposed by the top-ranked teams. We also report the results achieved with the submitted algorithms...