Key Concepts Explained

Welcome to ML Clever! To get the most out of the platform, it helps to understand some fundamental concepts from the world of Artificial Intelligence (AI), Machine Learning (ML), and specific terms we use within the platform.

This page provides clear definitions for these key terms. Familiarizing yourself with this vocabulary will make navigating ML Clever and interpreting your results much easier.

Core Concepts Defined

Artificial Intelligence (AI)

A broad field of computer science focused on creating systems that can perform tasks typically requiring human intelligence. This includes learning, problem-solving, decision-making, and understanding language. ML Clever utilizes AI techniques throughout the platform, especially in features like AutoML and AI Dashboard Generation.

Machine Learning (ML)

A subset of AI where systems learn from data without being explicitly programmed. Algorithms identify patterns in data (training) to build a Model capable of making predictions or decisions on new, unseen data (inference). ML Clever is designed to make ML accessible without needing to code these algorithms yourself.

Key Idea

ML focuses on learning patterns from examples (data) rather than following pre-written instructions for every possible scenario.

Supervised Learning

The most common type of ML used in ML Clever. In supervised learning, the algorithm learns from a labeled Dataset – meaning the historical data includes the correct "answers" or outcomes (the Target Variable). The goal is to learn a mapping from input Features to the known Target Variable.

Primary Tasks in ML Clever:

  • Classification: Predicting categories (e.g., 'Yes'/'No', 'Spam'/'Not Spam').
  • Regression: Predicting numerical values (e.g., Price, Temperature).

Classification

A supervised learning task where the goal is to predict a discrete category or class label for a given input. The Target Variable belongs to a finite set of possibilities.

Examples in ML Clever:

  • Predicting if a customer will churn (Yes/No).
  • Identifying the type of product based on features (Electronics/Clothing/Home Goods).
  • Classifying emails as Spam or Not Spam.
Learn about Classification Models

Regression

A supervised learning task where the goal is to predict a continuous numerical value for a given input. The Target Variable can take on any value within a range.

Examples in ML Clever:

  • Predicting the price of a house based on its features.
  • Forecasting future sales volume.
  • Estimating the temperature based on sensor readings.
Learn about Regression Models

Dataset

The starting point for any analysis in ML Clever. A Dataset is a structured collection of data (usually rows and columns, like a spreadsheet or database table) that you upload or connect to. It contains the observations and Features used for analysis, visualization, and Model Training.

Learn about Datasets

Feature

An individual, measurable characteristic or property represented as a column in your Dataset. Features (also called variables or attributes) are the inputs used by ML Models to make Predictions about the Target Variable. Examples: 'Age', 'Price', 'Location', 'Click Count'.

Target Variable

The specific Feature (column) in your Dataset that you want your ML Model to predict. Selecting the Target Variable defines the objective of your ML task (e.g., predicting 'Revenue', 'Churn Status', 'Risk Level'). This is crucial for Supervised Learning.

Learn about setting up Model Training

Preprocessing

The essential step of cleaning, transforming, and preparing your raw Dataset to make it suitable for ML Model Training. This often involves handling missing values, converting text data into numbers (encoding), scaling numerical Features, and sometimes creating new, more informative features (feature engineering). ML Clever's AutoML automates many preprocessing tasks.

Learn about Data Preprocessing

Model (Machine Learning Model)

The output of the Model Training process. A Model is essentially a mathematical representation of the patterns learned from your Dataset. It takes new input Features and generates a Prediction for the Target Variable. ML Clever allows you to train various types of models (e.g., Random Forest, XGBoost, Linear Regression).

Explore Models in ML Clever

Model Training

The process where an ML algorithm learns patterns from your prepared (preprocessed) Dataset. The algorithm adjusts its internal parameters to minimize errors in predicting the known Target Variable values in the training data. This results in a trained Model. ML Clever offers both manual configuration and automated (AutoML) training options.

Learn about Model Training

AutoML (Automated Machine Learning)

A powerful capability in ML Clever that automates many steps of the Model Training process. AutoML intelligently handles Preprocessing, tries different algorithms (Models), tunes their settings (hyperparameters), and evaluates them to find the best-performing model for your Dataset and Target Variable, requiring minimal user intervention.

Learn about AutoML Training

Prediction (or Inference)

The output generated by a trained Model when it receives new, unseen input data (Features). It's the model's estimation or classification for the Target Variable based on the patterns learned during Training. ML Clever provides tools to easily generate predictions using your trained models.

Learn about Making Predictions

Dashboard

A visual workspace in ML Clever used to display insights, monitor metrics, and explore data through various interactive visualizations called Dashboard Components. Dashboards help communicate findings effectively. ML Clever's AI Dashboard Generator can create comprehensive dashboards automatically from your Dataset or Model results.

Learn about Dashboards

Dashboard Component

An individual, configurable building block used to construct a Dashboard. Each component serves a specific purpose, such as displaying a bar chart, line chart, KPI metric, data table, map, or text area. You add and configure components to visualize your data and insights.

Learn about Dashboard Components

Pipeline / Workflow

An automated sequence of connected steps that perform a complete data processing or ML task. In ML Clever, you can build Workflows to chain together actions like data loading, Preprocessing, Model Training (including AutoML), and generating Predictions. This allows for reproducible and scalable operations.

Connecting the Concepts

While you can use features independently (like creating a Dashboard directly from a Dataset), a common Machine Learning workflow in ML Clever involves using these concepts together:

You start with a Dataset, identify Features and a Target Variable, then apply Preprocessing (often automated via AutoML). Next, you perform Model Training to create a Model. Finally, you use this model to make Predictions on new data and visualize the results or model performance using Dashboard Components within a Dashboard. This entire process can often be automated using a Pipeline/Workflow.

Next Steps

With these core concepts in mind, you're ready to explore specific areas of the ML Clever platform in more detail:

Managing Data

Learn how to import, connect, explore, and prepare your datasets for analysis.

Go to Data Docs

Building Models

Dive into training, evaluating, and understanding machine learning models.

Go to Model Docs

Creating Dashboards

Explore AI dashboard generation, component configuration, and visual design.

Go to Dashboard Docs

Full Documentation

Browse all documentation categories for comprehensive guides and tutorials.

Browse All Docs

Was this page helpful?

Need help?Contact Support
Questions?Contact Sales

Last updated: 5/3/2025

ML Clever Docs