Page 25 - CT_AI_Class-8

Refinement. The model can be improved by:

 Adding more data
 Removing incorrect data

 Handling missing values

 Selecting better features
 Changing model settings

 Training the model again

This stage ensures that the AI model becomes reliable, accurate and ready for real-world use.

Model Deployment

After the model has been trained, tested and refined, the next stage in the AI Project Life Cycle is
Model Deployment. In this stage, the AI model is implemented in a real-world environment where
it can start making predictions and assisting users.

Once deployed, the AI model can:
 Take input data (screen time, sleep hours, gadget type, etc.)

 Process and analyse the data

 Make predictions (High or Low screen time)

 Provide suggestions or alerts
 Help users make better decisions

For example:

In the Gadget Screen Time data, the trained model can be deployed in a school monitoring system
or student wellness application. Teachers or parents can enter student information such as:
 Gadget type  Screen time  Sleep hours

 Outdoor activity  Academic score

The system will then predict whether the student has High or Low screen time. Based on this
prediction:
 Teachers can guide students to reduce screen time

 Parents can monitor gadget usage at home

 Schools can promote outdoor activities

 Students can improve their study habits
Some real-life examples of deployment are as follows:

 Digital wellbeing apps  Screen time monitoring systems in schools

 Parental control apps  Health and wellness apps
 Study performance tracking systems  Smart classroom monitoring systems

AI Project Lifecycle 23