Welcome back to our course - AI for everyone
I’m Miss.Vidhya Suriyamoorthy from National Vidyalaya Senior Secondary School.
So far we have seen the introduction of AI, ML, Neural Network and about preceptors. Right?
Now we are in the second week of our course. In this week we are going to see about,
AI project cycle framework
Example for AI project cycle
What machine learning can and cannot do?....and about
Data Science
Let’s get into it.
We are going to see about AI project cycle framework. Here AI project refers any application/system which we create using AI technology. And framework is nothing but, whenever we are doing any task we should follow some steps to achieve that task.Right?... In the same way whenever we are creating an AI project we should follow 5 stages/steps in general. This is actually called AI project cycle framework .
And…here these are the five stages.
Problem scoping
Data acquisition
Data exploration
Modelling and
Evaluation
Let’s see about each stage
First stage of AI project cycle is Problem scoping,
Problem scoping is the stage where we have to select the problem which we want to solve using AI technology.
Then second stage is Data acquisition,
Data acquisition is the stage where we have to collect data (ie., information/details) related to our problem.
Then the third stage is Data Exploration,
Actually the acquired data might be very large. The acquired data might be a big data. Therefore, Data Exploration is the stage where we have to filter out only the useful information out of that acquired data and represent it in the simplified format (i.e., either in the form of table/graph/charts/maps)
That simplified format will be very helpful to understand the data and to sense the trends and patterns out of it.
Then the fourth stage of AI project cycle is Modelling,
Modelling is the stage where we have to select and implement the AI algorithm or the machine learning algorithm or techniques which we are going to use to develop the project. Therefore, at the end of this stage we will come out with a complete model of the project.
Then the last and the fifth stage of AI project cycle is Evaluation,
It is a very important stage because before deploying the model in a real world we have to test our project as many ways as possible. After ensuring the accuracy and efficiency of the application it will be allowed to deploy in the real world. So it is a very important stage.
Let’s recap
There are 5 stages in AI project cycle. Problem scoping is the first stage where we have to select the problem. In Data acquisition stage we have to collect data related to it. Then in the Data Exploration stage we have to filter out the useful information and represent it in the simplified format….in any of the visual form. Then in the modeling stage we have to select algorithm and implement it to make the complete model. Then at the last stage evaluation we have to test our model and after that…after ensuring the accuracy we should deploy it in real world.
So that’s all about AI project cycle framework.
Thank you all.
