Page 17 - CT_AI_Class-8

 AI-based: AI-based attendance systems use facial recognition technology to automatically
mark attendance when students enter the classroom. The system learns faces and
maintains accurate records.

 Security System
 Time-based (Automation): Security cameras record continuously but do not analyse the
footage. Security personnel must manually check recordings.

 AI-based: AI-based security systems detect unusual activities, recognise faces and send alerts
automatically. The system can learn patterns and improve detection accuracy over time.

 Smart Fan System
 Time-based (Automation): Fans run at fixed speed settings selected manually.

 AI-based: AI-based fans detect room temperature and occupancy. The fan automatically
adjusts speed and turns OFF when no one is present.

 Irrigation System:
 Time-based (Automation): In this irrigation system, a timer waters the plants every day at

7 AM, regardless of weather conditions or soil moisture levels.
 AI-based: An AI-based irrigation system waters plants based on soil moisture, weather
conditions and predictive analytics. The system analyses data and decides the best time

and amount of water required for plants.

Data Collection and Preparation

After defining the problem that requires AI, Data collection and preparation are the next important
phase of an AI project lifecycle. Data collection is a structured and systematic process of gathering
information from various sources such as observations, measurements, surveys and existing
records. This stage is an essential part of any AI project because the performance of an AI model
largely depends on the quality and quantity of data used. Accurate and relevant data helps in
building reliable AI systems, while poor-quality data can lead to incorrect predictions.
During this phase, data requirements are carefully reviewed based on the problem being solved.

The type of data, amount of data and sources of data are identified. If the available data is
insufficient, additional data is collected to meet project objectives.
There are mainly three types of data sources:

 Primary data source: Primary data is collected directly from the original source for a specific
purpose. This type of data is fresh, raw and highly accurate because it has not been processed
or modified by anyone else. Primary data is usually collected when specific information is
required that is not available from existing sources.
Common methods of primary data collection include:

 Surveys and interviews: Surveys and interviews are widely used to collect opinions,
feedback and preferences directly from individuals. Surveys may be conducted online,
offline or through phone calls.

AI Project Lifecycle 15