DM1 - Introduction

The process of discovering valid, useful, unexpected, and understandable patterns and models from large datasets.

Knowledge Discovery in Databases; a multi-step process including Data Cleaning, Data Integration, Data Selection, Data Transformation, Data Mining, Pattern Evaluation, and Knowledge Presentation.

The challenge of having tremendous amounts of data but lacking the knowledge or insights from it, leading to a "drowning in data, starving for knowledge" scenario.

Deductive query processing, expert systems, or small machine learning/statistical programs that don't involve discovering non-obvious patterns from large datasets.

An unsupervised learning task to group objects into sub-groups (clusters) based on similarity, aiming to maximize intra-class similarity and minimize inter-class similarity.

A supervised learning task to find models or rules from labeled training data that describe and distinguish classes, and predict class membership for new objects.

A type of machine learning where the training data includes known class labels, used to build models for prediction (e.g., classification).