Supervised Learning- using mathematical models pre-programmed that demonstrate the required outputs.

Training Data- for a supervised algorithm images with and without the product would be shared to demonstrate essentially- what good looks like and what bad looks like- then the system will predict from this data- which then leads into semi supervised, which develop models from this data.

Unsupervised Learning- has only input data, with no output product shared into the software- this type of system would be used to group data and discover linking factors and categories this as future learning.

Many forms of Machine learning can be designed using our Perceptor Software we are able to design algorithms to work to our customers challenges and take the time to design the most effective solution.

TPR True Positive Rate

TNR True Negative Rate

FNR False Negative Rate

TOC Total Operating characteristic

The data entered into the software can be from a direct process or from inputs which clarify the parameters of the system. Our dedicated software team will ensure they understand your full process to design a bespoke system to fulfil all requirements.