AI Project Cycle
There is steps to be followed to make a cup of coffee. Software is also designed following some steps. There is also set of rules to be followed while making a AI projects. These are the steps to make AI projects:
 |
| AI Project Cycle |
1) Problem Scoping: This is first step in AI project cycle where problem is defined and it is identified that how AI technology can help to solve the problem. To understand the problem 4W’s technique is used.
i) Who: It identifies who will be affected by this AI solution. It defines the users and stakeholder.
ii) What: It finds out the specific problem which needs to be addressed by AI solution. It defines the outcome of the AI project.
iii) Where: It finds out where the AI project will work. It searches environment and specific domain to be used.
iv) Why: It finds out the reason to solve the problem. It looks for the impact of the solution on business and users.
2) Data Acquisition: In this stage raw data is gathered which needs to be analysed. A data could be anything like text, image, audio, video, email etc. The data is gathered from different sources like newspaper, internet, and journals, academic. Data gives valuable insight for developing the AI model.
3) Data Exploration: This is an important phase in AI project cycle because large amount of data is analysed to find the meaningful patters. There are many visualising tools like MS Excel to get the insight of data in the form of charts, graphs.
4) Modeling: In this phase AI system is designed using machine learning algorithm and the AI system is trained using any learning method. This is the crucial phase in AI project cycle because previous steps were conducted to design accurate AI system.
5) Evaluation: In this phase AI system’s performance is evaluated and ensured that it meets the predefined goal. If AI model doesn’t meet the objective of the project, it means some changes are required. When AI model gives accurate results and aligns with the user’s requirement, it means AI model is ready for deployment phase.
6) Deployment: In this AI system is integrated in working environment under the supervision of experts. It is also monitored that AI system is giving accurate result. Regular AI system’s performance is analysed. It may need some changes due to change in data.
i) Who: It identifies who will be affected by this AI solution. It defines the users and stakeholder.
ii) What: It finds out the specific problem which needs to be addressed by AI solution. It defines the outcome of the AI project.
iii) Where: It finds out where the AI project will work. It searches environment and specific domain to be used.
iv) Why: It finds out the reason to solve the problem. It looks for the impact of the solution on business and users.
2) Data Acquisition: In this stage raw data is gathered which needs to be analysed. A data could be anything like text, image, audio, video, email etc. The data is gathered from different sources like newspaper, internet, and journals, academic. Data gives valuable insight for developing the AI model.
3) Data Exploration: This is an important phase in AI project cycle because large amount of data is analysed to find the meaningful patters. There are many visualising tools like MS Excel to get the insight of data in the form of charts, graphs.
4) Modeling: In this phase AI system is designed using machine learning algorithm and the AI system is trained using any learning method. This is the crucial phase in AI project cycle because previous steps were conducted to design accurate AI system.
5) Evaluation: In this phase AI system’s performance is evaluated and ensured that it meets the predefined goal. If AI model doesn’t meet the objective of the project, it means some changes are required. When AI model gives accurate results and aligns with the user’s requirement, it means AI model is ready for deployment phase.
6) Deployment: In this AI system is integrated in working environment under the supervision of experts. It is also monitored that AI system is giving accurate result. Regular AI system’s performance is analysed. It may need some changes due to change in data.
AI Domains
Artificial Intelligence is divided into three domains Statistical Data, Computer Vision, and Natural Language Processing
 |
| AI Domains |
1) Statistical Data: Statistical Data is the important domain of AI which is used to collect, manage, and analysis the data set. Statistical Data AI system collect structured, unstructured, and unstructured data and uses machine learning algorithm to get the meaningful information in form of charts, graphs. The information gathered from statistical data analysis is used to take decision in real life.
Application of Statistical Data
i) Price Comparison Websites: Price Comparison sites work on statistical data analysis to compare the prices of same products of different website. It also tells trends, offers, and best prices.
ii) Weather Forecasting: Statistical data analysis also helps in weather forecasting. It examines the satellite images with AI system to predict the weather. It considers other factors responsible for weather change which is helpful for are farmers.
iii) Recommendation System: Statistical data analysis is also used in recommendation system. It examines user's history, interest, and behavior and recommends products, movies, and music. Amazon, YouTube, and Netflix are using statistical data analysis for better services.
i) Price Comparison Websites: Price Comparison sites work on statistical data analysis to compare the prices of same products of different website. It also tells trends, offers, and best prices.
ii) Weather Forecasting: Statistical data analysis also helps in weather forecasting. It examines the satellite images with AI system to predict the weather. It considers other factors responsible for weather change which is helpful for are farmers.
iii) Recommendation System: Statistical data analysis is also used in recommendation system. It examines user's history, interest, and behavior and recommends products, movies, and music. Amazon, YouTube, and Netflix are using statistical data analysis for better services.
2) Computer Vision: Computer Vision is the main domain of AI system. It collect information in image, video format. It analysis the data and takes decisions. It is used in face recognition, obstacle detection, and hand writing detection.
Application of Computer Vision
i) Face lock in Phone: Face recognition system is used to secure smart phone which uses computer vision. It stores all facial features which are always mapped to unlock the phone.
ii) Self-driving car: Self-driving cars are using computer vison to detect objects, road, and sign board. The AI system analysis the data and takes action like start, stop, and side parking.
iii) Surveillance System: Computer vision is also used in surveillance system like CCTV camera, drone. It analysis the objects, tracks motion, and continue monitor the places for security. It takes action in condition of any threat.
Application of Computer Vision
i) Face lock in Phone: Face recognition system is used to secure smart phone which uses computer vision. It stores all facial features which are always mapped to unlock the phone.
ii) Self-driving car: Self-driving cars are using computer vison to detect objects, road, and sign board. The AI system analysis the data and takes action like start, stop, and side parking.
iii) Surveillance System: Computer vision is also used in surveillance system like CCTV camera, drone. It analysis the objects, tracks motion, and continue monitor the places for security. It takes action in condition of any threat.
3) Natural Language Processing
NLP (Natural Language Processing) is the important domain of AI which is used to understand, process, and interact in human language. NLP enables machine to interact using text and speech in human language.
NLP (Natural Language Processing) is the important domain of AI which is used to understand, process, and interact in human language. NLP enables machine to interact using text and speech in human language.
Application of Natural Language Processing
i) Chatbots: Chatbots, ChatGPT or Deepseek interact with human in text or in speech to interact with human. Virtual assistant like Alexa, Siri also understand human language.
ii) Language Translator: Language translators like Google Translator converts sentences from one language to another language. Voice typing feature also types text spoken by human.
iii) Spam Filter: In emails NLP is used to filter spam emails. It analysis phrases and sender behavior to categorise the email. In Gmail, mails are automatically divided into three categories. Microsoft outlook uses copilot to read and answer the emails.
i) Chatbots: Chatbots, ChatGPT or Deepseek interact with human in text or in speech to interact with human. Virtual assistant like Alexa, Siri also understand human language.
ii) Language Translator: Language translators like Google Translator converts sentences from one language to another language. Voice typing feature also types text spoken by human.
iii) Spam Filter: In emails NLP is used to filter spam emails. It analysis phrases and sender behavior to categorise the email. In Gmail, mails are automatically divided into three categories. Microsoft outlook uses copilot to read and answer the emails.
No comments:
Post a Comment