A HIERARCHICAL METHOD FOR FINDING OPTIMAL ARCHITECTURE AND WEIGHTS USING EVOLUTIONARY LEAST SQUARE BASED LEARNING

2003 ◽  
Vol 13 (01) ◽  
pp. 13-24 ◽  
Author(s):  
RANADHIR GHOSH ◽  
BRIJESH VERMA
Keyword(s):  
Neural Network ◽  
Solution Space ◽  
Least Square ◽  
Optimum Number ◽  
Data Sets ◽  
Data Set ◽  
Architecture Model ◽  
Starting Point ◽  
Xor Problem ◽  
Hidden Neurons

In this paper, we present a novel approach of implementing a combination methodology to find appropriate neural network architecture and weights using an evolutionary least square based algorithm (GALS).1 This paper focuses on aspects such as the heuristics of updating weights using an evolutionary least square based algorithm, finding the number of hidden neurons for a two layer feed forward neural network, the stopping criterion for the algorithm and finally some comparisons of the results with other existing methods for searching optimal or near optimal solution in the multidimensional complex search space comprising the architecture and the weight variables. We explain how the weight updating algorithm using evolutionary least square based approach can be combined with the growing architecture model to find the optimum number of hidden neurons. We also discuss the issues of finding a probabilistic solution space as a starting point for the least square method and address the problems involving fitness breaking. We apply the proposed approach to XOR problem, 10 bit odd parity problem and many real-world benchmark data sets such as handwriting data set from CEDAR, breast cancer and heart disease data sets from UCI ML repository. The comparative results based on classification accuracy and the time complexity are discussed.

A Comparison Study of Mahalanobis-Taguchi System and Neural Network for Multivariate Pattern Recognition

2005 ◽  
Author(s):  
Jungeui Hong ◽  
Elizabeth A. Cudney ◽  
Genichi Taguchi ◽  
Rajesh Jugulum ◽  
Kioumars Paryani ◽  
...  
Keyword(s):  
Neural Network ◽  
Small Data ◽  
Data Sets ◽  
Comparison Study ◽  
Data Set ◽  
Set Size ◽  
Breast Cancer Study ◽  
Discriminant Ability ◽  
Small Data Sets ◽  
Multivariate Pattern

The Mahalanobis-Taguchi System is a diagnosis and predictive method for analyzing patterns in multivariate cases. The goal of this study is to compare the ability of the Mahalanobis-Taguchi System and a neural network to discriminate using small data sets. We examine the discriminant ability as a function of data set size using an application area where reliable data is publicly available. The study uses the Wisconsin Breast Cancer study with nine attributes and one class.

scholarly journals Organic aerosol source apportionment in London 2013 with ME-2: exploring the solution space with annual and seasonal analysis

2016 ◽  
Vol 16 (24) ◽  
pp. 15545-15559 ◽  
Author(s):  
Ernesto Reyes-Villegas ◽  
David C. Green ◽  
Max Priestman ◽  
Francesco Canonaco ◽  
Hugh Coe ◽  
...  
Keyword(s):  
Source Apportionment ◽  
Statistical Tests ◽  
Solution Space ◽  
Organic Aerosol ◽  
Data Sets ◽  
Environmental Implications ◽  
Data Set ◽  
Diesel Vehicles ◽  
Heavy Duty Diesel ◽  
Paul Scherrer

Abstract. The multilinear engine (ME-2) factorization tool is being widely used following the recent development of the Source Finder (SoFi) interface at the Paul Scherrer Institute. However, the success of this tool, when using the a value approach, largely depends on the inputs (i.e. target profiles) applied as well as the experience of the user. A strategy to explore the solution space is proposed, in which the solution that best describes the organic aerosol (OA) sources is determined according to the systematic application of predefined statistical tests. This includes trilinear regression, which proves to be a useful tool for comparing different ME-2 solutions. Aerosol Chemical Speciation Monitor (ACSM) measurements were carried out at the urban background site of North Kensington, London from March to December 2013, where for the first time the behaviour of OA sources and their possible environmental implications were studied using an ACSM. Five OA sources were identified: biomass burning OA (BBOA), hydrocarbon-like OA (HOA), cooking OA (COA), semivolatile oxygenated OA (SVOOA) and low-volatility oxygenated OA (LVOOA). ME-2 analysis of the seasonal data sets (spring, summer and autumn) showed a higher variability in the OA sources that was not detected in the combined March–December data set; this variability was explored with the triangle plots f44 : f43 f44 : f60, in which a high variation of SVOOA relative to LVOOA was observed in the f44 : f43 analysis. Hence, it was possible to conclude that, when performing source apportionment to long-term measurements, important information may be lost and this analysis should be done to short periods of time, such as seasonally. Further analysis on the atmospheric implications of these OA sources was carried out, identifying evidence of the possible contribution of heavy-duty diesel vehicles to air pollution during weekdays compared to those fuelled by petrol.

scholarly journals Modeling of daily runoff from a small agricultural watershed using artificial neural network with resampling techniques

2014 ◽  
Vol 17 (1) ◽  
pp. 56-74 ◽  
Author(s):  
Gurjeet Singh ◽  
Rabindra K. Panda ◽  
Marc Lamers
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Agricultural Watershed ◽  
Training Data ◽  
Optimum Number ◽  
Training Process ◽  
Data Set ◽  
Reported Study ◽  
Artificial Neural ◽  
Artificial Neural Network Ann

The reported study was undertaken in a small agricultural watershed, namely, Kapgari in Eastern India having a drainage area of 973 ha. The watershed was subdivided into three sub-watersheds on the basis of drainage network and land topography. An attempt was made to relate the continuously monitored runoff data from the sub-watersheds and the whole-watershed with the rainfall and temperature data using the artificial neural network (ANN) technique. The reported study also evaluated the bias in the prediction of daily runoff with shorter length of training data set using different resampling techniques with the ANN modeling. A 10-fold cross-validation (CV) technique was used to find the optimum number of hidden neurons in the hidden layer and to avoid neural network over-fitting during the training process for shorter length of data. The results illustrated that the ANN models developed with shorter length of training data set avoid neural network over-fitting during the training process, using a 10-fold CV method. Moreover, the biasness was investigated using the bootstrap resampling technique based ANN (BANN) for short length of training data set. In comparison with the 10-fold CV technique, the BANN is more efficient in solving the problems of the over-fitting and under-fitting during training of models for shorter length of data set.

The Application of Probabilistic Neural Network in Speech Recognition Based on Partition Clustering

2012 ◽  
Vol 263-266 ◽  
pp. 2173-2178
Author(s):  
Xin Guang Li ◽  
Min Feng Yao ◽  
Li Rui Jian ◽  
Zhen Jiang Li
Keyword(s):  
Neural Network ◽  
Speech Recognition ◽  
Clustering Algorithm ◽  
Probabilistic Neural Network ◽  
Back Propagation ◽  
Back Propagation Neural Network ◽  
Data Sets ◽  
Data Set ◽  
Proposed Model ◽  
Partition Clustering

A probabilistic neural network (PNN) speech recognition model based on the partition clustering algorithm is proposed in this paper. The most important advantage of PNN is that training is easy and instantaneous. Therefore, PNN is capable of dealing with real time speech recognition. Besides, in order to increase the performance of PNN, the selection of data set is one of the most important issues. In this paper, using the partition clustering algorithm to select data is proposed. The proposed model is tested on two data sets from the field of spoken Arabic numbers, with promising results. The performance of the proposed model is compared to single back propagation neural network and integrated back propagation neural network. The final comparison result shows that the proposed model performs better than the other two neural networks, and has an accuracy rate of 92.41%.

scholarly journals Ambulatory and Laboratory Stress Detection Based on Raw Electrocardiogram Signals Using a Convolutional Neural Network

Sensors ◽  
2019 ◽  
Vol 19 (20) ◽  
pp. 4408 ◽  
Author(s):  
Hyun-Myung Cho ◽  
Heesu Park ◽  
Suh-Yeon Dong ◽  
Inchan Youn
Keyword(s):  
Neural Network ◽  
Transfer Learning ◽  
Mental Arithmetic ◽  
Small Data ◽  
Data Sets ◽  
Learning Method ◽  
Data Set ◽  
Electrocardiogram Signals ◽  
Proposed Model ◽  
Small Data Set

The goals of this study are the suggestion of a better classification method for detecting stressed states based on raw electrocardiogram (ECG) data and a method for training a deep neural network (DNN) with a smaller data set. We suggest an end-to-end architecture to detect stress using raw ECGs. The architecture consists of successive stages that contain convolutional layers. In this study, two kinds of data sets are used to train and validate the model: A driving data set and a mental arithmetic data set, which smaller than the driving data set. We apply a transfer learning method to train a model with a small data set. The proposed model shows better performance, based on receiver operating curves, than conventional methods. Compared with other DNN methods using raw ECGs, the proposed model improves the accuracy from 87.39% to 90.19%. The transfer learning method improves accuracy by 12.01% and 10.06% when 10 s and 60 s of ECG signals, respectively, are used in the model. In conclusion, our model outperforms previous models using raw ECGs from a small data set and, so, we believe that our model can significantly contribute to mobile healthcare for stress management in daily life.

scholarly journals BraggNet: integrating Bragg peaks using neural networks

2019 ◽  
Vol 52 (4) ◽  
pp. 854-863 ◽  
Author(s):  
Brendan Sullivan ◽  
Rick Archibald ◽  
Jahaun Azadmanesh ◽  
Venu Gopal Vandavasi ◽  
Patricia S. Langan ◽  
...  
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Negative Control ◽  
Protein Crystal ◽  
Data Sets ◽  
Data Set ◽  
X Ray ◽  
X Ray Crystallography ◽  
Integrated Intensities ◽  
Neutron Crystallography

Neutron crystallography offers enormous potential to complement structures from X-ray crystallography by clarifying the positions of low-Z elements, namely hydrogen. Macromolecular neutron crystallography, however, remains limited, in part owing to the challenge of integrating peak shapes from pulsed-source experiments. To advance existing software, this article demonstrates the use of machine learning to refine peak locations, predict peak shapes and yield more accurate integrated intensities when applied to whole data sets from a protein crystal. The artificial neural network, based on the U-Net architecture commonly used for image segmentation, is trained using about 100 000 simulated training peaks derived from strong peaks. After 100 training epochs (a round of training over the whole data set broken into smaller batches), training converges and achieves a Dice coefficient of around 65%, in contrast to just 15% for negative control data sets. Integrating whole peak sets using the neural network yields improved intensity statistics compared with other integration methods, including k-nearest neighbours. These results demonstrate, for the first time, that neural networks can learn peak shapes and be used to integrate Bragg peaks. It is expected that integration using neural networks can be further developed to increase the quality of neutron, electron and X-ray crystallography data.

scholarly journals SYNTHESIS OF SELF-ORGANIZING MAP AND FEEDFORWARD NEURAL NETWORK FOR BETTER FORECASTING

2014 ◽  
pp. 68-75
Author(s):  
Oles Hodych ◽  
Yuriy Shcherbyna ◽  
Michael Zylan
Keyword(s):  
Neural Network ◽  
Feedforward Neural Network ◽  
Share Price ◽  
Data Sets ◽  
Self Organizing Map ◽  
Data Set ◽  
Price Fluctuation ◽  
Price Data ◽  
Comparison Results ◽  
Self Organizing

In this article the authors propose an approach to forecasting the direction of the share price fluctuation, which is based on utilization of the Feedforward Neural Network in conjunction with Self-Organizing Map. It is proposed to use the Self-Organizing Map for filtration of the share price data set, whereas the Feedforward Neural Network is used to forecast the direction of the share price fluctuation based on the filtered data set. The comparison results are presented for filtered and non-filtered share price data sets.

scholarly journals Core looseness fault identification model based on Mel spectrogram-CNN

2021 ◽  
Vol 2137 (1) ◽  
pp. 012060
Author(s):  
Ping He ◽  
Yong Li ◽  
Shoulong Chen ◽  
Hoghua Xu ◽  
Lei Zhu ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Convolution Neural Network ◽  
Data Sets ◽  
Data Set ◽  
Recognition Model ◽  
Model Based ◽  
Fault Recognition ◽  
Transformer Core ◽  
Identification Model

Abstract In order to realize transformer voiceprint recognition, a transformer voiceprint recognition model based on Mel spectrum convolution neural network is proposed. Firstly, the transformer core looseness fault is simulated by setting different preloads, and the sound signals under different preloads are collected; Secondly, the sound signal is converted into a spectrogram that can be trained by convolutional neural network, and then the dimension is reduced by Mel filter bank to draw Mel spectrogram, which can generate spectrogram data sets under different preloads in batch; Finally, the data set is introduced into convolutional neural network for training, and the transformer voiceprint fault recognition model is obtained. The results show that the training accuracy of the proposed Mel spectrum convolution neural network transformer identification model is 99.91%, which can well identify the core loosening faults.

scholarly journals A Comparison of Emotional Neural Network (ENN) and Artificial Neural Network (ANN) Approach for Rainfall-Runoff Modelling

2019 ◽  
Vol 5 (10) ◽  
pp. 2120-2130 ◽  
Author(s):  
Suraj Kumar ◽  
Thendiyath Roshni ◽  
Dar Himayoun
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Data Sets ◽  
Rainfall Runoff ◽  
Data Set ◽  
Discharge Data ◽  
Artificial Neural ◽  
Artificial Neural Network Ann ◽  
Emotional Neural Network ◽  
Selection Of

Reliable method of rainfall-runoff modeling is a prerequisite for proper management and mitigation of extreme events such as floods. The objective of this paper is to contrasts the hydrological execution of Emotional Neural Network (ENN) and Artificial Neural Network (ANN) for modelling rainfall-runoff in the Sone Command, Bihar as this area experiences flood due to heavy rainfall. ENN is a modified version of ANN as it includes neural parameters which enhance the network learning process. Selection of inputs is a crucial task for rainfall-runoff model. This paper utilizes cross correlation analysis for the selection of potential predictors. Three sets of input data: Set 1, Set 2 and Set 3 have been prepared using weather and discharge data of 2 raingauge stations and 1 discharge station located in the command for the period 1986-2014.  Principal Component Analysis (PCA) has then been performed on the selected data sets for selection of data sets showing principal tendencies.  The data sets obtained after PCA have then been used in the model development of ENN and ANN models. Performance indices were performed for the developed model for three data sets. The results obtained from Set 2 showed that ENN with R= 0.933, R2 = 0.870, Nash Sutcliffe = 0.8689, RMSE = 276.1359 and Relative Peak Error = 0.00879 outperforms ANN in simulating the discharge. Therefore, ENN model is suggested as a better model for rainfall-runoff discharge in the Sone command, Bihar.

COMPARATIVE ANALYSIS OF PREDICTION TECHNIQUES ON THE BASIS OF TELECOM CUSTOMER CHURN

2021 ◽  
Vol 16 (9) ◽  
Author(s):  
Yasser Khan
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Performance Metrics ◽  
Learning Algorithm ◽  
Data Sets ◽  
Data Set ◽  
Broadband Service ◽  
Customer Churn ◽  
Artificial Neural ◽  
Prediction Techniques

Telecommunication customer churn is considered as major cause for dropped revenue and customer baseline of voice, multimedia and broadband service provider. There is strong need on focusing to understand the contributory factors of churn. Now considering factors from data sets obtained from Pakistan major telecom operators are applied for modeling. On the basis of results obtained from the optimal techniques, comparative technical evaluation is carried out. This research study is comprised mainly of proposition of conceptual frame work for telecom customer churn that lead to creation of predictive model. This is trained tested and evaluated on given data set taken from Pakistan Telecom industry that has provided accurate & reliable outcomes. Out of four prevailing statistical and machine learning algorithm, artificial neural network is declared the most reliable model, followed by decision tree. The logistic regression is placed at last position by considering the performance metrics like accuracy, recall, precision and ROC curve. The results from research has revealed main parameters found responsible for customer churn were data rate, call failure rate, mean time to repair and monthly billing amount. On the basis of these parameter artificial neural network has achieved 79% more efficiency as compare to low performing statistical techniques.

Sign in / Sign up

Export Citation Format

Share Document