AI Modeling:-
The ability to mathematically describe the relationship between
data and parameters forms the core of every AI Model.
Modeling is the process in which different models based on the
visualized data can be created and even checked for the
advantages and disadvantages.
2 Type of Modeling Approaches:-
1. Rule Based Approach [Model driven]
2. Learning Based Approach [Data Driven]
Rule Based Approach:-
● Rule Based Approach Refers to the AI modeling where the
relationship or pattern in data are defined by developer.
● This means the machine works on the rules and information given
by the developer and performs the task accordingly.
● The rule based AI is used when we have known or labeled dataset
Ex. You trained you model with 100 images of apples and
bananas. Now if you test it by showing an apple, it will figure
out and tell of it’s an apple or not. Here Labeled images of apple
and banana were fed,due to which the model could detect the
fruit.
Labeled Image:- Simply when model is told about what the image is.
Rule Based Approach Drawback:-
❖ Lot of manual Work
❖ Consume a lot of time
❖ Suitable only for less complex domain
❖ Limited ability and learning capacity
❖ Static and not scalable
Data Set:-
Dataset is a collection of related sets of information that is composed of
separate elements but can me manipulated by computer as a unit.
Training Data- A subset required to train the model
Testing Data- A subset required while testing the trained model
S No. Structured Data Unstructured Data
1 It comes from existing data model It does not come from pre existing
data model
2 It has well defined relationship among It has undefined relationship
its elements among its elements
3 It does not require extra preprocessing It requires more preprocessing
before being analysed or searched before analysed or searched
Data Set:-
Learning Based Approach :-
The learning-based Approach is based on a machine learning experience with
data fed [input huge data].
Machine Learning[ML]:-Machine learning is a subset of artificial intelligence
(AI) that provides machine the ability to learn automatically and improve from
experience based on input data without being programed for it.
In ML , there are no patterns, rules and relationships predefined by the
developer , rather machine learns with each new input and come with own
algorithm.
Machine Learning[ML] :-
3 types of machine learning:-
★ Supervised Learning #Task Driven
○ Regression
○ Classification [KNN ,Naive Bayes,SVM]
★ Unsupervised Learning #Data Driven
○ Clustering [ Continuous data-Algo: K-Means]
○ Dimensionality Reduction
★ Semi supervised or Reinforcement Learning #True AI
○ Clustering
○ Association
Supervised Learning :-
Supervised Learning where a computer algorithm is
trained on input data that has been labeled for a
particular output.
2 types of Supervised Learning:-
○ Regression
○ classification
Classification :-
Data is categorized under different labels according to some
parameters given in the input and then the labels are
predicted for the data.
It can classify the category of new unlabelled test inputs on
the basis of training data
The classification models use
non-contiguous i.e. discrete data
Ex. To predict which of them
is apple and banana.
Regression :-
It is used to predict continuous value.
It is based on a mathematical approach used to find the
relationship between two or more variables and predict the
outcome
Ex.1. To predict your next salary
Based on your previous salary,
Increment etc.
2. Weather prediction using past
data.
Unsupervised Learning :-
It works with unlabelled data and creates cluster of items
having similar features,characteristic or patterns.
It is used to identify relationship patterns and trends out of
the data which is fed into it
It helps the user in understanding
- What the data is about
- What are the major feature identified by the machine
Ex. boy observe someone performing trick with ball,so he
also learn the trick by himself
Clustering :-
Its an algorithm which can cluster the unknown data
according to the pattern or trends identified out of it.
The patterns observed can be known to the developer or it
can be unique.
Note: Classification ≈ Division, Clustering ≈ Grouping
Dimensionality Reduction :-
● We can visualize up to 3-Dimensions only.
● To reduce the dimensions and still be able to make sense of the data,
we use Dimensionality Reduction.
● The ball in our hand is 3-Dimensions. But if we click its picture, the
data transform to 2-D.
Reinforcement Learning:-
● Learning through feedback or trial and error method is called
Reinforcement Learning.
● The system works on Reward or Penalty policy. In this an agent
performs an action positive or negative, in the environment which is
taken as input from the system, then the system changes the state
in the environment and the agent is provided with a reward or
penalty.
Example:- A very good example
of these is Vending machines.
Reinforcement Learning:-
Example:- A very good example of these is Vending machines.
Suppose you put a coin (action) in a Juice Vending machine(environment),
now the system detects the amount of coin given (state) you get the
drink corresponding to the amount(reward) or if the coin is damaged or
there is any another problem, then you get nothing (penalty).
EVALUATION:-
Evaluation is the process of understanding the reliability of any AI model,
based on outputs by feeding the test data into the model and comparing it
with actual answers.
There can be different Evaluation techniques, depending on the type and
purpose of the model.
“Arise! Awake! and stop not until the goal
is reached.”
-Swami Vivekananda
Stay Safe. Stay aware. Stay Healthy. Stay Alert
Questions:-
1. During Data Acquisition feeding previous data into the machine is called…
a. Training Data b. Testing Data c. Predicting Data d. Evaluating Data
2. ________ Involves collecting data from various authentic sources such as reliable website,
observation surveys.
a. Data Acquisition b. Data Exploration c. Data Tasting d. Evaluating Data
[Link]