Page 174 - AI Computer 10

The primary role of a training dataset is to allow the model to learn the underlying patterns and relationships
between the features and the labels. The size of the training dataset can significantly impact the performance
of the model. Generally, a larger training dataset can allow the model to learn more robust patterns, but an
excessively large dataset can also lead to diminishing returns.

Testing Dataset
A Testing Dataset is a subset of data used to evaluate the performance and generalisation ability of a trained
machine learning model. It is distinct from the training dataset and is crucial for understanding how well the
model can perform on unseen data. The primary purpose of the testing dataset is to provide an unbiased
evaluation of the model’s performance after it has been trained. The testing dataset also plays an essential role
in ensuring that the model can be effectively deployed in real-world applications.
AI MODELLING

AI Modelling refers to the process of creating algorithms and models that allow machines to mimic human
cognition and perform tasks involving understanding, reasoning, and learning from data. This includes various
techniques from machine learning, deep learning, natural language processing, and other areas of Artificial
Intelligence.
The goal of AI Modelling is to develop systems that can perform tasks intelligently and adaptively, often exceeding
the capabilities of traditional programming techniques.

Types of AI Models

In general, AI Modelling techniques can be broadly classified into two approaches:
Supervised
Learning

Machine Unsupervised
Learning Learning

Reinforcement
Learning Based Learning

AI Models Artificial Neural
Rule Based Deep Networks

Learning
Convolution Neural
Networks

Rule Based Approach

The Rule based approach to AI modelling is a simple and easy approach because it is directly based on simple
“if – then” rules. This approach is used to build an AI system that works on the basis of a predefined hierarchy of
rules that govern how to transform user input into desired course of action or automated actions. The two major
components of rule-based AI models are “a set of rules” and “a set of facts”.
Examples of Rule-based AI Models

Example: To develop a rule-based chatbot for an e-commerce website that helps customers with common
inquiries about their orders and other issues.

40
40