1. First presentation of artificial intelligence Page 1
Topic: A brief introduction to supervised learning, its types, advantages &
disadvantages which case we should use it and which case is not used.
Group members:
Ali zafar roll #: 19
M. Zaman roll # 34 (retainer)
Asif Ali
2. First presentation of artificial intelligence Page 2
What is the machine learning?
Machine learning is an application of artificial intelligence (AI) that provides systems the
ability to automatically learn and improve from experience without being explicitly
programmed. Machine learning focuses on the development of computer programs that can
access data and use it learn from them. The primary aim is to allow the computers learn
automatically without human intervention and adjust actions accordingly.
4. First presentation of artificial intelligence Page 4
• Supervised learning is the learning in which we teach or train the machine using
data which is well labeled. A supervised learning algorithm analyze training data
and produces an inferred function which can be used for mapping new examples.
• The correct classes of the training data are known.
Here the human experts acts as the teacher where we feed the computer with training
data containing the input/predictors and we show it the correct answers (output) and
from the data the computer should be able to learn the patterns.
we can predict the output values for new data based relationships which it learned
from the previous data sets.
6. First presentation of artificial intelligence Page 6
Supervised learning uses classification and regression techniques to develop predictive models.
Classification techniques
When the output variable is categorical i.e with 2 or more classes (yes/no , true/false”) we make use
of classification. Classification is the task of predicting a discrete class label.
Predict discrete responses—for example, whether an email is genuine or spam,
or whether a tumor is cancerous or benign. Classification models classify input
data into categories. Typical applications include medical imaging, speech
recognition, and credit scoring.
Use classification if your data can be tagged, categorized, or separated into
specific groups or classes. For example, applications for hand-writing recognition
7. First presentation of artificial intelligence Page 7
use classification to recognize letters and numbers. In image processing and
computer vision, unsupervised pattern recognition techniques are used for object
detection and image segmentation.
Common algorithms for performing classification include support vector machine
(SVM), boosted and bagged decision trees, k-nearest neighbor, Naïve
Bayes, discriminant analysis, logistic regression, and neural networks.
Regression techniques
Relations hip between two or more variables where a change in one variable is associated with a
change in other variable. Regression is the task of predicting a continuous quantity.
Predict continuous responses—for example, changes in temperature or
fluctuations in power demand. Typical applications include electricity load
forecasting and algorithmic trading.
Use regression techniques if you are working with a data range or if the nature of
your response is a real number, such as temperature or the time until failure for a
piece of equipment.
Common regression algorithms include linear model, nonlinear
model, regularization, stepwise regression, boosted and bagged decision
trees, neural networks, and adaptive neuro-fuzzy learning.
8. First presentation of artificial intelligence Page 8
Supervised learning algorithms
Nearest Neighbor
Naive Bayes
Decision Trees
Linear Regression
Support Vector Machines (SVM)
Neural Networks
Applications of supervised learning
9. First presentation of artificial intelligence Page 9
How Do You Decide Which Machine
Learning Algorithm to Use?
Choosing the right algorithm can seem overwhelming—there are dozens of
supervised and unsupervised machine learning algorithms, and each takes a
different approach to learning.
There is no best method or one size fits all. Finding the right algorithm is partly
just trial and error—even highly experienced data scientists can’t tell whether an
algorithm will work without trying it out. But algorithm selection also depends on
the size and type of data you’re working with, the insights you want to get from
the data, and how those insights will be used.
.
Here are some guidelines on choosing between supervised and unsupervised
machine learning:
Choose supervised learning if you need to train a model to make a
prediction--for example, the future value of a continuous variable, such
as temperature or a stock price, or a classification—for example, identify
makes of cars from webcam video footage.
Choose unsupervised learning if you need to explore your data and
want to train a model to find a good internal representation, such as
splitting data up into clusters.
10. First presentation of artificial intelligence Page 10
Advantages
1- You can get very specific about the definition of the classes, which means that you
can train the classifier in a way which has a perfect decision boundary to distinguish
different classes accurately.
2- You can specifically determine how many classes you want to have.
3- After training, you don’t necessarily need to keep the training examples in a memory.
You can keep the decision boundary as a mathematical formula and that would
be enough for classifying future inputs.
Disadvantages:
1- Your decision boundary might be overtrained. Which means that if your training
set is not including some examples that you want to have in a class, when you use those
examples after training, you might not get the correct class label.
2- When this an input which is not from any of the classes in reality, then it might get
a wrong class label after classification.
3- You have to select a lot of good examples from each class while you are training the
classifier. If you consider classification of big data that can be a real challenge.
4- Training needs a lot of computation time, so do the classification.
5- You might need to use a cloud and leave the training algorithm work over a night or
nights before obtaining a good decision boundary model.