scholarly journals Neuron type classification in rat brain based on integrative convolutional and tree-based recurrent neural networks

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Tielin Zhang ◽  
Yi Zeng ◽  
Yue Zhang ◽  
Xinhe Zhang ◽  
Mengting Shi ◽  
...  
Keyword(s):  
Neural Network ◽  
Rat Brain ◽  
Support Vector ◽  
Neuron Type ◽  
Data Formats ◽  
Proposed Model ◽  
Network Modules ◽  
Network Tree ◽  
Diverse Data ◽  
Type Classification

AbstractThe study of cellular complexity in the nervous system based on anatomy has shown more practical and objective advantages in morphology than other perspectives on molecular, physiological, and evolutionary aspects. However, morphology-based neuron type classification in the whole rat brain is challenging, given the significant number of neuron types, limited reconstructed neuron samples, and diverse data formats. Here, we report that different types of deep neural network modules may well process different kinds of features and that the integration of these submodules will show power on the representation and classification of neuron types. For SWC-format data, which are compressed but unstructured, we construct a tree-based recurrent neural network (Tree-RNN) module. For 2D or 3D slice-format data, which are structured but with large volumes of pixels, we construct a convolutional neural network (CNN) module. We also generate a virtually simulated dataset with two classes, reconstruct a CASIA rat-neuron dataset with 2.6 million neurons without labels, and select the NeuroMorpho-rat dataset with 35,000 neurons containing hierarchical labels. In the twelve-class classification task, the proposed model achieves state-of-the-art performance compared with other models, e.g., the CNN, RNN, and support vector machine based on hand-designed features.

Structured Pyramidal Neural Networks

2018 ◽  
Vol 28 (05) ◽  
pp. 1750021 ◽  
Author(s):  
Alessandra M. Soares ◽  
Bruno J. T. Fernandes ◽  
Carmelo J. A. Bastos-Filho
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Processing Time ◽  
Receptive Fields ◽  
Memory Capacity ◽  
The Other ◽  
Support Vector ◽  
Proposed Model ◽  
Novel Method ◽  
Original Approach

The Pyramidal Neural Networks (PNN) are an example of a successful recently proposed model inspired by the human visual system and deep learning theory. PNNs are applied to computer vision and based on the concept of receptive fields. This paper proposes a variation of PNN, named here as Structured Pyramidal Neural Network (SPNN). SPNN has self-adaptive variable receptive fields, while the original PNNs rely on the same size for the fields of all neurons, which limits the model since it is not possible to put more computing resources in a particular region of the image. Another limitation of the original approach is the need to define values for a reasonable number of parameters, which can turn difficult the application of PNNs in contexts in which the user does not have experience. On the other hand, SPNN has a fewer number of parameters. Its structure is determined using a novel method with Delaunay Triangulation and k-means clustering. SPNN achieved better results than PNNs and similar performance when compared to Convolutional Neural Network (CNN) and Support Vector Machine (SVM), but using lower memory capacity and processing time.

scholarly journals An Efficient and Fast Model Reduced Kernel KNN for Human Activity Recognition

2021 ◽  
Vol 2021 ◽  
pp. 1-9
Author(s):  
Zongying Liu ◽  
Shaoxi Li ◽  
Jiangling Hao ◽  
Jingfeng Hu ◽  
Mingyang Pan
Keyword(s):  
Neural Network ◽  
Activity Recognition ◽  
Human Activity ◽  
Processing Time ◽  
Kernel Method ◽  
Nearest Neighbors ◽  
Human Activity Recognition ◽  
Support Vector ◽  
K Nearest Neighbors ◽  
Proposed Model

With accumulation of data and development of artificial intelligence, human activity recognition attracts lots of attention from researchers. Many classic machine learning algorithms, such as artificial neural network, feed forward neural network, K-nearest neighbors, and support vector machine, achieve good performance for detecting human activity. However, these algorithms have their own limitations and their prediction accuracy still has space to improve. In this study, we focus on K-nearest neighbors (KNN) and solve its limitations. Firstly, kernel method is employed in model KNN, which transforms the input features to be the high-dimensional features. The proposed model KNN with kernel (K-KNN) improves the accuracy of classification. Secondly, a novel reduced kernel method is proposed and used in model K-KNN, which is named as Reduced Kernel KNN (RK-KNN). It reduces the processing time and enhances the classification performance. Moreover, this study proposes an approach of defining number of K neighbors, which reduces the parameter dependency problem. Based on the experimental works, the proposed RK-KNN obtains the best performance in benchmarks and human activity datasets compared with other models. It has super classification ability in human activity recognition. The accuracy of human activity data is 91.60% for HAPT and 92.67% for Smartphone, respectively. Averagely, compared with the conventional KNN, the proposed model RK-KNN increases the accuracy by 1.82% and decreases standard deviation by 0.27. The small gap of processing time between KNN and RK-KNN in all datasets is only 1.26 seconds.

scholarly journals Modelling the Vegetation Response to Climate Changes in the Yarlung Zangbo River Basin Using Random Forest

Water ◽  
2020 ◽  
Vol 12 (5) ◽  
pp. 1433 ◽  
Author(s):  
Kaige Chi ◽  
Bo Pang ◽  
Lizhuang Cui ◽  
Dingzhi Peng ◽  
Zhongfan Zhu ◽  
...  
Keyword(s):  
Neural Network ◽  
Support Vector Machine ◽  
River Basin ◽  
Partial Correlation ◽  
Climatic Factors ◽  
Vegetation Coverage ◽  
Support Vector ◽  
Partial Correlation Analysis ◽  
Proposed Model ◽  
Yarlung Zangbo River

Vegetation coverage variation may influence watershed water balance and water resource availability. Yarlung Zangbo River, the longest river on the Tibetan Plateau, has high spatial heterogeneity in vegetation coverage and is the main freshwater resource of local residents and downstream countries. In this study, we proposed a model based on random forest (RF) to predict the Normalized Difference Vegetation Index (NDVI) of the Yarlung Zangbo River Basin and explore its relationship with climatic factors. High-resolution datasets of NDVI and monthly meteorological observation data from 2000 to 2015 were used to calibrate and validate the proposed model. The proposed model was then compared with artificial neural network and support vector machine models, and principal component analysis and partial correlation analysis were also used for predictor selection of artificial neural network and support vector machine models for comparative study. The results show that RF had the highest model efficiency among the compared models. The Nash–Sutcliffe coefficients of the proposed model in the calibration period and verification period were all higher than 0.8 for the five subzones; this indicated that the proposed model can successfully simulate the relationship between the NDVI and climatic factors. By using built-in variable importance evaluation, RF chose appropriate predictor combinations without principle component analysis or partial correlation analysis. Our research is valuable because it can be integrated into water resource management and elucidates ecological processes in Yarlung Zangbo River Basin.

Deep Learning Based Sentiment Classification on User-Generated Big Data

2020 ◽  
Vol 13 (5) ◽  
pp. 1047-1056
Author(s):  
Akshi Kumar ◽  
Arunima Jaiswal
Keyword(s):  
Neural Network ◽  
Big Data ◽  
Deep Learning ◽  
Soft Computing ◽  
Sentiment Classification ◽  
Convolution Neural Network ◽  
Support Vector ◽  
Computing Technique ◽  
Proposed Model ◽  
Soft Computing Techniques

Background: Sentiment analysis of big data such as Twitter primarily aids the organizations with the potential of surveying public opinions or emotions for the products and events associated with them. Objective: In this paper, we propose the application of a deep learning architecture namely the Convolution Neural Network. The proposed model is implemented on benchmark Twitter corpus (SemEval 2016 and SemEval 2017) and empirically analyzed with other baseline supervised soft computing techniques. The pragmatics of the work includes modelling the behavior of trained Convolution Neural Network on wellknown Twitter datasets for sentiment classification. The performance efficacy of the proposed model has been compared and contrasted with the existing soft computing techniques like Naïve Bayesian, Support Vector Machines, k-Nearest Neighbor, Multilayer Perceptron and Decision Tree using precision, accuracy, recall, and F-measure as key performance indicators. Methods: Majority of the studies emphasize on the utilization of feature mining using lexical or syntactic feature extraction that are often unequivocally articulated through words, emoticons and exclamation marks. Subsequently, CNN, a deep learning based soft computing technique is used to improve the sentiment classifier’s performance. Results: The empirical analysis validates that the proposed implementation of the CNN model outperforms the baseline supervised learning algorithms with an accuracy of around 87% to 88%. Conclusion: Statistical analysis validates that the proposed CNN model outperforms the existing techniques and thus can enhance the performance of sentiment classification viability and coherency.

scholarly journals Data Driven Leakage Detection and Classification of a Boiler Tube

2019 ◽  
Vol 9 (12) ◽  
pp. 2450 ◽  
Author(s):  
Muhammad Sohaib ◽  
Jong-Myon Kim
Keyword(s):  
Neural Network ◽  
Deep Neural Network ◽  
Wavelet Packet ◽  
Thermal Power ◽  
Leak Detection ◽  
Classification Model ◽  
Support Vector ◽  
Boiler Tube ◽  
Proposed Model ◽  
Ae Signals

Boiler heat exchange in thermal power plants involves tubes to transfer heat from the fuel to the water. Boiler tube leakage can cause outages and huge power generation loss. Therefore, early detection of leaks in boiler tubes is necessary to avoid such accidents. In this study, a boiler tube leak detection and classification mechanism was designed using wavelet packet transform (WPT) analysis of the acoustic emission (AE) signals acquired from the boiler tube and a fully connected deep neural network (FC-DNN). WPT analysis of the AE signals enabled the extraction of features associated with the different conditions of the boiler tube, that is, normal and leak conditions. The deep neural network (DNN) effectively explores the salient information from the wavelet packet features through a deep architecture instead of considering shallow networks, such as k-nearest neighbors (k-NN) and support vector machines (SVM). This enhances the classification performance of the leak identification and classification model developed. The proposed model yielded a 99.2 % average classification accuracy when tested with AE signals from the boiler tube. The experimental results prove the efficacy of the proposed model for boiler tube leak detection and classification.

Abnormal Event Detection in a Surveillance Scene Using Convolutional Neural Network

2021 ◽  
Vol 11 (4) ◽  
pp. 1-20
Author(s):  
Kinjal V. Joshi ◽  
Narendra M. Patel
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Convolutional Neural Network ◽  
Event Detection ◽  
Characteristic Curve ◽  
Research Area ◽  
Support Vector ◽  
Abnormal Event Detection ◽  
Proposed Model ◽  
Benchmark Datasets

Automatic abnormal event detection in a surveillance scene is very significant because of more consciousness about public safety. Because of usefulness and complexity, currently, it is an open research area. In this manuscript, the authors have proposed a novel convolutional neural network (CNN) model to detect an abnormal event in a surveillance scene. In this work, CNN is used in two ways. Firstly, it is used for both feature extraction and classification. In a second way, CNN is used for feature extraction, and support vector machine (SVM) is used for classification. Without any pre-processing, the proposed model gives better results compared to state-of-the-art methods. Experiments are carried out on four different publicly available benchmark datasets and one combined dataset, which contains all images of four datasets. The performance is measured by accuracy and area under the ROC (receiver operating characteristic) curve (AUC). The experimental results determine the efficacy of the proposed model.

City Fire Forecasts and Analysis Based on Nonlinear Auto-Regressive Time-Series Model

2012 ◽  
Vol 241-244 ◽  
pp. 1550-1555 ◽  
Author(s):  
Sheng Peng Liu ◽  
Ye Zhang
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Prediction Model ◽  
Bp Neural Network ◽  
Support Vector ◽  
Future Developments ◽  
Proposed Model ◽  
Network Method ◽  
Auto Regressive ◽  
The City

The forecasting to future developments of the city fire time series is a challenging task that has been addressed by many researchers due to the importance. In this paper, a Nonlinear Auto-Regressive (NAR) prediction model is applied to forecast the city fire data based on support vector regression. The performances of the NAR prediction model in city fire forecasting are compared with the BP neural network method. The experimental results show that the proposed model performs best.

scholarly journals Three-Class Mammogram Classification Based on Descriptive CNN Features

2017 ◽  
Vol 2017 ◽  
pp. 1-11 ◽  
Author(s):  
M. Mohsin Jadoon ◽  
Qianni Zhang ◽  
Ihsan Ul Haq ◽  
Sharjeel Butt ◽  
Adeel Jadoon
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Curvelet Transform ◽  
Data Matrix ◽  
Support Vector ◽  
Discrete Wavelet ◽  
Accuracy Rate ◽  
Data Set ◽  
Proposed Model ◽  
Mammogram Images

In this paper, a novel classification technique for large data set of mammograms using a deep learning method is proposed. The proposed model targets a three-class classification study (normal, malignant, and benign cases). In our model we have presented two methods, namely, convolutional neural network-discrete wavelet (CNN-DW) and convolutional neural network-curvelet transform (CNN-CT). An augmented data set is generated by using mammogram patches. To enhance the contrast of mammogram images, the data set is filtered by contrast limited adaptive histogram equalization (CLAHE). In the CNN-DW method, enhanced mammogram images are decomposed as its four subbands by means of two-dimensional discrete wavelet transform (2D-DWT), while in the second method discrete curvelet transform (DCT) is used. In both methods, dense scale invariant feature (DSIFT) for all subbands is extracted. Input data matrix containing these subband features of all the mammogram patches is created that is processed as input to convolutional neural network (CNN). Softmax layer and support vector machine (SVM) layer are used to train CNN for classification. Proposed methods have been compared with existing methods in terms of accuracy rate, error rate, and various validation assessment measures. CNN-DW and CNN-CT have achieved accuracy rate of 81.83% and 83.74%, respectively. Simulation results clearly validate the significance and impact of our proposed model as compared to other well-known existing techniques.

scholarly journals Fault Diagnosis and Fault Frequency Determination of Permanent Magnet Synchronous Motor Based on Deep Learning

Sensors ◽  
2021 ◽  
Vol 21 (11) ◽  
pp. 3608
Author(s):  
Chiao-Sheng Wang ◽  
I-Hsi Kao ◽  
Jau-Woei Perng
Keyword(s):  
Neural Network ◽  
Feature Extraction ◽  
Deep Learning ◽  
Convolutional Neural Network ◽  
Permanent Magnet ◽  
Support Vector ◽  
One Dimensional ◽  
Proposed Model ◽  
Wide Range ◽  
Fault State

The early diagnosis of a motor is important. Many researchers have used deep learning to diagnose motor applications. This paper proposes a one-dimensional convolutional neural network for the diagnosis of permanent magnet synchronous motors. The one-dimensional convolutional neural network model is weakly supervised and consists of multiple convolutional feature-extraction modules. Through the analysis of the torque and current signals of the motors, the motors can be diagnosed under a wide range of speeds, variable loads, and eccentricity effects. The advantage of the proposed method is that the feature-extraction modules can extract multiscale features from complex conditions. The number of training parameters was reduced so as to solve the overfitting problem. Furthermore, the class feature map was proposed to automatically determine the frequency component that contributes to the classification using the weak learning method. The experimental results reveal that the proposed model can effectively diagnose three different motor states—healthy state, demagnetization fault state, and bearing fault state. In addition, the model can detect eccentric effects. By combining the current and torque features, the classification accuracy of the proposed model is up to 98.85%, which is higher than that of classical machine-learning methods such as the k-nearest neighbor and support vector machine.

scholarly journals Prediction of Energy-Efficient Production of Coalbed Methane Based on Chaotic Time Series and Bayes-Least Squares-Support Vector Machine

2020 ◽  
Vol 38 (4) ◽  
pp. 933-940
Author(s):  
Yan Wang ◽  
Zhongshui Man ◽  
Meihua Lu
Keyword(s):  
Neural Network ◽  
Time Series ◽  
Support Vector Machine ◽  
Phase Space ◽  
Energy Efficient ◽  
Coalbed Methane ◽  
Phase Space Reconstruction ◽  
Support Vector ◽  
Proposed Model ◽  
Bayesian Evidence

The productivity of coalbed methane (CBM) depends heavily on the heat environment, and directly reflects the quality of the well. Following the theories of phase space reconstruction and Bayesian evidence framework, this paper puts forward a Bayes-least squares-support vector machine (Bayes-LS-SVM) model for the prediction of energy-efficient productivity of CBM under Bayesian evidence network based on chaotic time series. The energy-efficient productivity stands for the gas and water production of CBM wells at a low energy consumption, despite the disturbance from the heat environment. The proposed model avoids the local optimum trap of backpropagation neural network (BPNN), and overcomes the main defects of the SVM: high time consumption of parameter determination, and proneness to overfitting. In our model, the model parameters are optimized through three-layer Bayesian evidence inference, and the input vector for prediction is selected adaptively. In this way, the model construction is not too empirical, and the constructed model is highly adaptive. Then, the theory on phase space reconstruction was applied to investigate the chaotic property of the time series on CBM production, and the Bayes-LS-SVM was adopted to predict the time series after phase space reconstruction, in comparison with neural network prediction methods like SVM and BPNN. Experimental results show that the proposed model boast quick computing, accurate fitting, flexible structure, and strong generalization ability.

Sign in / Sign up

Export Citation Format

Share Document