IntelligenceLab VCL

IntelligenceLab is a library component that provides artificial intelligence for data grouping, spam filters, computer vision, speech recognition, or classification related commands.

The library contains a series of classifiers, training elements, and a multi-purpose neuron component. Various applications can be built using this library: spam filters, OCR applications, computer vision, speech recognition, data classification, decision making.

IntelligenceLab is easily implemented using MMP optimized libraries, such as Intel MMX and Intel Performance Primitive (IPP). Also, it is compatible and integrates directly with SignalLab, VideoLab, AudioLab, VisionLab, InstrumentLab, and PlotLab.

This Library is a VCL - Firemonkey version designed for Delphi, C++ Builder, RAD Studio XE 4 to 10.1 Berlin.

The component contains Naive Bayesian, Nearest Neighbor, Neural Network, Self Organizing Map, and Radial Basis Function Network classifiers.

Moreover, the built-in converter pairs multiple data buffers together. The custom filters can be used on real, binary, and real matrix data.

Other components include Watch Dog Timers, a clock for other elements, a thread event (which executes code in a separate thread), a counter, and a frequency meter.

The training elements use various algorithms (Backprop and Rprop) to train neural networks and to prepare data for the classifiers.

The software component can be employed to perform some tasks, and it can easily be implemented into many applications. It is free for non-commercial use, teachers and students.