2,159 Hits in 3.7 sec


Zuhair Hussein Ali, College of Education/ AL-Mustansiriyah University, Baghdad, Iraq, Ameen A. Noor
2019 Iraqi Journal for Computers and Informatics  
In this paper new methods based on Radial Basis Function (RBF) and Fuzzy Radial Basis Function (FRBF) are used to solve the problem of text classification, where a set of features extracted for each sentence  ...  Reuters 21578 dataset utilized for the purpose of text classification. The results showed the effectiveness of FRBF is better than RBF  ...  Text GCN basically builds on GCN which is semi supervised learning consists of a multilayer neural network that work on graph directly.  ... 
doi:10.25195/2017/4513 fatcat:2wmrnyuuevbazlg37ncpwt327e

Semi-supervised Persian font recognition

Maryam Bahojb Imani, Mohamad Reza Keyvanpour, Reza Azmi
2011 Procedia Computer Science  
So many different semi-supervised learning methods have been studied recently.  ...  Classical supervised methods need lot of labeled data to train a classifier. Since it is very costly and time consuming to label large amounts of data, it is useful to use data sets without labels.  ...  Semi-supervised learning (SSL) is applied for classification, clustering and regression. In this paper, we used it in classification tasks.  ... 
doi:10.1016/j.procs.2010.12.057 fatcat:hsjonhfxqrfl3lpi5ihbnmxu3e

Active Learning with the Probabilistic RBF Classifier [chapter]

Constantinos Constantinopoulos, Aristidis Likas
2006 Lecture Notes in Computer Science  
In this work we present an active learning methodology for training the probabilistic RBF (PRBF) network.  ...  It is a special case of the RBF network, and constitutes a generalization of the Gaussian mixture model.  ...  Thus this problem is closely related to semi-supervised learning. Algorithms for semi-supervised learning have been proposed for Gaussian mixtures in [8, 9] , as well as for the RBF network [10] .  ... 
doi:10.1007/11840817_38 fatcat:qijzqdrelzay3b7fmd2bc7ikmi

Semi-Supervised Target-Dependent Sentiment Classification for Micro-Blogs

Shadi I. Abudalfa, Moataz A. Ahmed
2019 Journal of Computer Science and Technology  
Semi-supervised learning techniques have been known in the literature to improve classification accuracy in comparison with supervised learning techniques; however, they use same number of labeled samples  ...  In this work, we propose a new semi-supervised learning technique that uses less number of labeled microblogs than that used with supervised learning techniques.  ...  Fabian Gieseke for answering some enquiries about his developed tool which has been used also in this research work.  ... 
doi:10.24215/16666038.19.e06 fatcat:sx4wk3rhvnas5arzbmnc24apcy

Learning with Local and Global Consistency

Dengyong Zhou, Olivier Bousquet, Thomas Navin Lal, Jason Weston, Bernhard Schölkopf
2003 Neural Information Processing Systems  
Our method yields encouraging experimental results on a number of classification problems and demonstrates effective use of unlabeled data.  ...  We consider the general problem of learning from labeled and unlabeled data, which is often called semi-supervised learning or transductive inference.  ...  Special thanks go to Xiaojin Zhu, Zoubin Ghahramani, and John Lafferty who communicated with us on the important post-processing step class mass normalization used in their method and also provided us  ... 
dblp:conf/nips/ZhouBLWS03 fatcat:kh6vjlszazhf3lud26to3wt34a

Using Semi-Supervised Learning for the Creation of Medical Systematic Review: An Exploratory Analysis

Prem Timsina, Jun Liu, Omar El-Gayar, Yanyan Shang
2016 2016 49th Hawaii International Conference on System Sciences (HICSS)  
In this research, we explore semi-supervised learning based classifiers to identify articles that can be included when creating medical systematic reviews (SRs).  ...  Specifically, we perform comparative study of various semi-supervised learning algorithm, and identify the best technique that is suited for SRs creation.  ...  Existing research on semi-supervised learning also demonstrate promise for text classification with few labeled examples.  ... 
doi:10.1109/hicss.2016.151 dblp:conf/hicss/TimsinaLES16 fatcat:dechwix62jhjvcb4uijn4sphk4

Classification of Functional and Non-functional Requirements in Agile by Cluster Neuro-Genetic Approach

Daminderjit Sunner, Harpreet Bajaj
2016 International Journal of Software Engineering and Its Applications  
This paper proposes a supervised learning based automated (neural network with the genetic algorithm) approach for successfully classifying functional and non-functional requirements from multiple requirements  ...  Machine) with RBF (Radial Basis Function) kernel and neural network with a genetic algorithm are implemented. These two classifier models are compared and analyzed by precision, recall, and accuracy.  ...  After this, supervised learning is applied to implement two classifier models: SVM with RBF kernel: SVM can only perform linear classification.  ... 
doi:10.14257/ijseia.2016.10.10.13 fatcat:pxmzbnywzjd23dddfqjuytp42i

Semi-supervised and active learning with the probabilistic RBF classifier

Constantinos Constantinopoulos, Aristidis Likas
2008 Neurocomputing  
The probabilistic RBF network (PRBF) is a special case of the RBF network and constitutes a generalization of the Gaussian mixture model.  ...  In this paper we propose a semi-supervised learning method for PRBF, using labeled and unlabeled observations concurrently, that is based on the expectation-maximization (EM) algorithm.  ...  Algorithms for semi-supervised learning have been proposed for Gaussian mixtures in [10, 18] , as well as for the RBF network [16] .  ... 
doi:10.1016/j.neucom.2007.11.039 fatcat:spky4pe4f5f5xi43nw3p6klvzq

Graph topology inference benchmarks for machine learning [article]

Carlos Lassance and Vincent Gripon and Gonzalo Mateos
2020 arXiv   pre-print
As a tool used to express relationships between objects, graphs can be deployed to various ends: I) clustering of vertices, II) semi-supervised classification of vertices, III) supervised classification  ...  In this work, we introduce several ease-to-use and publicly released benchmarks specifically designed to reveal the relative merits and limitations of graph inference methods.  ...  It also contains 5 standard splits that are not used here (as we do unsupervised and semi-supervised classification).  ... 
arXiv:2007.08216v1 fatcat:cztqzvnw7rds5o4rjwlp6nvkia

Text Categorization Comparison between Simple BPNN and Combinatorial Method of LSI and BPNN

Hemlata Tekwani, Mahak Motwani
2014 International Journal of Computer Applications  
This paper proposed a text categorization comparison between simple BPNN and Combinatorial method of LSI and BPNN.  ...  Hence, this new method greatly reduces the dimension and better classification results can be achieved.  ...  The BP network and RBF network are widely used among artificial neural networks for automatic text classification.  ... 
doi:10.5120/17138-7723 fatcat:q3cnicxrsbhapmh7qlihvii7re

Kernel Methods for Deep Learning

Youngmin Cho, Lawrence K. Saul
2009 Neural Information Processing Systems  
These kernel functions can be used in shallow architectures, such as support vector machines (SVMs), or in deep kernel-based architectures that we call multilayer kernel machines (MKMs).  ...  A similar validation was provided by recent work on kernel methods for semi-supervised embedding [7] . We hope that our results inspire more work on kernel methods for deep learning.  ...  The use of LMNN is inspired by the supervised fine-tuning of weights in the training of deep architectures [18] .  ... 
dblp:conf/nips/ChoS09 fatcat:5b2gmqcjarebdgimc34lxzvjaq

Cold Start Active Learning Strategies in the Context of Imbalanced Classification [article]

Etienne Brangbour and Pierrick Bruneau and Thomas Tamisier and Stéphane Marchand-Maillet
2022 arXiv   pre-print
We present novel active learning strategies dedicated to providing a solution to the cold start stage, i.e. initializing the classification of a large set of data with no attached labels.  ...  Specifically, our active learning iterations address label scarcity and imbalance using element scores, combining information extracted from a clustering structure to a label propagation model.  ...  Alternatively, using a semi-supervised model will allow to gracefully integrate label feedback.  ... 
arXiv:2201.10227v1 fatcat:fdfvqgp245hz3eu3dtujepbfsa

Feature and kernel learning

Verónica Bolón-Canedo, Michele Donini, Fabio Aiolli
2015 The European Symposium on Artificial Neural Networks  
This tutorial is concerned with the use of data to learn features and kernels automatically.  ...  A nonparametric spectral transformation τ : λ s → µ s optimized to the specific task is performed using the semi-supervised information contained in L.  ...  This can be made by changing the spectral representation of L rescaling the eigenvalues in according to the semi-supervised information, for examples using linear programming [48] .  ... 
dblp:conf/esann/Bolon-CanedoDA15 fatcat:udrtxou2wvbbjdqkowahd4hx44

A Comparative Assessment of Data Mining Algorithms to Predict Fraudulent Firms

Harshit Monish, Avinash Chandra Pandey
2020 2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence)  
We have implemented Decision Trees, Linear Support Vector Machines, RBF Kernel Support Vector Machines, K-Nearest Neighbor, Artificial Neural Network and logistic regression classification models.  ...  Fig. 7 .Fig. 8 . 78 Accuracy Time consumption plot method for short text classification based on semi-supervised learning," in 2015 4th International Conference on Advanced Information Technology and Sensor  ...  EXPERIMENTAL RESULTS AND CONCLUSION In this paper, supervised learning models and techniques have been compared for the problem statement of finding the fraudulent company using classification models i.e  ... 
doi:10.1109/confluence47617.2020.9057968 fatcat:yxruvis7qzfqjgww6zz2gmxuae

Handwritten Devanagari Lipi using Support Vector Machine

Shailendra KumarShrivastava, Pratibha Chaurasia
2012 International Journal of Computer Applications  
The energy features of segment characters are used for the classification. The more no. of segmentation improves the recognition rate.  ...  Different technique and features are used for the faithful recognition characters. In this paper we have proposed a SVM (support vector machine) based technique for Devanagari character recognition.  ...  Supervised learning, Un-supervised learning, Semi-supervised learning and reinforcement learning are various types of Machine Learning.  ... 
doi:10.5120/6220-8785 fatcat:h6vok6webzgplepkhw4kstfo4e
« Previous Showing results 1 — 15 out of 2,159 results