Page 169 - Ai Book - 10

Answer the following questions.
1. In how many categories can we divide learning based approaches?

__________________________________________________________________________________________
2. Write any two advantages of Neural networks.
__________________________________________________________________________________________

K Keyey TTermserms

u AI Project Cycle
AI involves teaching machines to think and act as humans. The AI project cycle has stages from problem
scoping to evaluation.

u Problem Scoping
Define the problem’s scope using the 4W’s framework: Who, What, Where, and Why. Identify stakeholders,
nature, and complexity of the problem.

u Data Acquisition
Collect necessary data for the AI project. Data features such as facial characteristics, employee ID, etc., are
essential.

u Data Exploration
Understand data quality and characteristics. Identify trends and relationships in the collected data.
u Modelling
Analyse and train the AI model using appropriate machine learning algorithms. Two main modelling
approaches: rule based and learning based.

u Evaluation Stage
Final stage to test the model’s efficiency. Parameters include accuracy, precision, and complexity.

u Supervised Learning
Labeled dataset used for training, which includes classification and regression.
u Unsupervised Learning
Uses unlabeled datasets and includes clustering and dimensionality reduction.

u Reinforcement Learning
Models make a sequence of decisions. It learns to achieve goals in uncertain environments..

u Rule based approach
The Rule based approach is a simple and easy approach because it is directly based on rules.
u Learning Based Approach
Learning based approach or Adaptive Intelligence Approach refers to the model where relationship or
patterns in the data are not defined by the developer

43
43