DelphAI
English
English
  • Overview
    • Welcome!
    • About the component
    • How to install
    • About the developer
    • What's next
  • Documentation
    • AISelector
    • EasyAI
      • Regression
      • Classification
      • Recommendation for item
      • Recommendation for user
    • Models
      • Regression
        • KNN
        • Linear regression
        • Ridge regression
      • Classification
        • Decision Tree
        • KNN
        • Naive Bayes
      • Recommendation
      • Clustering
        • DBSCAN
        • K-Means
        • Mean Shift
    • Dataset
      • Regression
      • Classification
      • Recommendation
      • Clustering
    • FEEDBACK
Powered by GitBook
On this page
  1. Overview

Welcome!

NextAbout the component

Last updated 4 months ago

Welcome to the DelphAI component information center: an Artificial Intelligence component for Delphi, made entirely in Delphi!

Here we have an explanation of the and how to it, as well as a little about the .

If you don't have knowledge of artificial intelligence models, EasyAI is a great module for you.

If you want custom models, use to easily test multiple and choose the best one for you.

To see an example of each functionality of this component, open the 'DelphAIExamples' project from the github repository.

component
install
developer

Regression

Predicts a value based on input data.

Classification

Categorizes input data into predefined classes.

Recommendation for item

Suggests relevant items based on the similarities between items.

Recommendation for user

Suggests relevant items based on user preferences.

AISelector
models

AISelector

Allows testing multiple models with different parameters in a single function.

Regression

Predicts a value based on input data.

Classification

Categorizes input data into predefined classes.

Recommendation

Suggests relevant items based on the similarities between items or based on user preferences.

Clustering

Organizes data into groups with similar characteristics, without predefined labels.