1. Meaning of Project Cycle
The AI Project Cycle is a step-by-step process used to solve a problem with AI. It keeps the work organised and helps teams test whether the solution is useful.
2. Problem Scoping
Problem scoping defines what problem will be solved, who faces it, why it matters, and what success will look like. A clear problem statement prevents the team from collecting random data.
Example: Instead of saying "make a school AI", a clearer problem is "predict which library books may need more copies based on borrowing records."
3. Data Acquisition
Data acquisition means collecting relevant data from suitable sources. Data may come from surveys, sensors, records, observations, or public datasets.
Only collect data that is needed. Respect privacy and avoid personal details unless there is a clear and safe reason.
4. Data Exploration
Data exploration means studying the collected data to understand patterns, missing values, unusual entries, and relationships. Tables, charts, and summaries help in this step.
5. Modelling
Modelling means choosing a method that can learn patterns from data. A model may classify, predict, recommend, or group information depending on the problem.
A model should be trained with suitable data and tested with data it has not already seen.
6. Evaluation
Evaluation checks how well the model performs. Accuracy is useful, but it is not the only measure. Fairness, safety, consistency, and usefulness also matter.
7. Responsible Deployment
Deployment means using the solution in a real situation. The team should monitor results, collect feedback, fix problems, and explain limitations honestly.
8. Revision Questions
1. What is problem scoping?
It is the step of clearly defining the problem, users, need, and success measure.
2. Why is data exploration important?
It helps understand patterns, errors, missing values, and useful relationships.
3. What is evaluation?
Evaluation checks how well an AI model performs and whether it is useful and safe.