Backgrounder on Artificial Intelligence and Machine Learning
There are several theories for processing vast multi-dimensional data:
- Machine Learning ML
- Artificial Intelligence
- AI Machine Vision MV
- Image Processing IP
- Wavelet Signal Analysis WSA
Machine Learning ML: is a method of data analysis that automates analytical model-building. It is a branch of artificial intelligence based on the construct that systems can learn from data, identify patterns and make decisions with minimal human intervention Machine Learning ML: Adaptive learning from examples. For example, we do not know how a dynamical system works and want to determine the specifics of this dynamical system. ML learns from samples and provides insights and inferences. Systems can learn from data, identify patterns and make decisions with minimal human intervention.
Artificial Intelligence AI: Mimics some behaviour of the human being. For example, an element or component of AI is Natural Language Processing that writes analysis as if it was written by a human.
Machine Vision MV: When AI algorithms are applied, visualization replaces the limited perspective of the human eye. For example, when microscopic circles are viewed in living tissue – the MV interprets these circles to highlight a tumor as there are no circles in tissues. This analysis occurs in a fraction of the time it takes a human being to perform such analysis.
Image Processing IP: Used to differentiate specific information from an image. IP removes all other aspects from the image to focus on a specific detail or image within the full picture. For example, to highlight areas of disease in a forest.
Note: There is new technology used by NASA – Lidar – that CDN integrates into its data analysis.
Wavelet Signal Analysis WSA: When big data signals are received, they are comprised of multiple signals. Using WSA algorithms, AI decomposes the original signals into many underlying signals to uncover trends, patterns and specific granular details. For example, periodic events and patterns over time, such as temperature or moisture data. This data is valuable to a range of industries such as manufacturing, agriculture, forestry, fisheries, etc.
The examples below pertain to scientific data – Hurricane Simon data received from NASA satellites. The volumetric four-dimensional data is downloaded from NASA and processed by CDN to form computational structures suitable for analysis. The images – that depict the data being analyzed – are generated by software rendering applications.
The above image depicts the network of droplets similarities in Hurricane Simon at a specific point in time.
In the above image, the green nodes indicate a central core to Hurricane Simon. The droplets form communities in terms of shape, volume, density and structure. This image is another view or version of the previous image – again at a point in time.