Qualicision in association with artificial neural networks and techniques of machine learning
Neural networks – a method of machine learning ...
… to process complex structural data as well as to realise classification, description and recognition.
Deep neural networks – an extension of neural networks ...
… to extract prior data characteristics by interlacing of neural layers.
Convolutional neural networks – a characteristic of deep neural networks ...
… which is particularly suitable for processing and classifying 2D and 3D input data, e.g. images and 3D models.
Recurrent neural network – a characteristic of deep neural networks ...
… which is particularly suitable for processing and classifying sequential input data, e.g. language and time series.
Machine learning – a discipline of information technology ...
… which is often used as a synonym to the expression artifical intelligence to describe the prozess of automated data processing from the fields of big data, planning and intelligent assistence systems.
Supervised vs. unsupervised learning – are two methods of machine learning ...
… that on the one hand training data with fixed results is available to process unknown samples (supervised). On the other hand the determination of unknown structures in data without availability of training data with known characteristic (Unsuper-vised).
Shutter-Stock