I want to perform classification of alternating current (ac) wave-forms -to classify wave forms (such as type 1, type 2 etc.).
I request advice on best classification technique (SVM, Neural Networks etc.) for performing classification of ac, sinusoidal waveforms.
Hey there,
In the context of signal analysis, convolutional neural networks (ConvNets) have shown superior results regarding the SVM and other statistical classifiers (e.g., Naive Bayes). Therefore, I think you can use ConvNets in your task.
Below some relevant works of signal analysis using ConvNets:
Chapter Fuzzy Integral Optimization with Deep Q-Network for EEG-Base...
Article Human Activity Recognition Based on Wearable Sensor Data: A ...
Article Comparison of Feature Learning Methods for Human Activity Re...
If you have big data then use convolutional neural networks (CNNs) or LSTM else use SVMs. CNNs are expensive to train though.
- Badges
- Science topic