Business

` Analytics
Machine Learning
 Definition of Machine Learning

a type of artificial intelligence

(AI) that allows software
applications to become more
accurate at predicting
outcomes without being
explicitly programmed to do
so. Machine learning 
algorithms use historical data
as input to predict new
output values.
 types of machine learning

supervised unsupervised semi-supervised reinforcement

learning learning learning learning
• data scientists • involves • Data scientists • to teach a
supply algorithms that may feed an machine to
algorithms train on algorithm complete a
with labeled unlabeled mostly labeled multi-step
training data data. data sets training data, process for
and define the looking for any but the model which there
variables they meaningful is free to are clearly
want the connection. explore the defined rules.
algorithm to data on its own
assess for and develop its
correlations. own
understanding

Logo
supervised machine learning How does supervised machine learning work

Binary
Dividing data into two categories.
classificatio
 Combining the n
predictions of multiple
machine learning
models to produce an
accurate prediction. Choosing between
Multi-class more than two types of
Ensembling classificatio answers
n

Regression
Predicting continuous values modeling

Logo
 unsupervised machine learning

 Splitting the dataset into

groups based on similarity.
Clustering

Reducing the
number of variables
in a data set Identifying
unusual data
Dimensionalit Anomaly points in a
y reduction detection data set.

Association
 Identifying sets of items in a data set mining
that frequently occur together.

Logo
 semi-supervised learning

Machine Teaching algorithms to translate

language based on less than a full
translatio dictionary of words.
n
Algorithms trained on
small data sets can
learn to apply data
labels to larger sets
automatically.

Labelling Fraud Identifying cases of fraud

when you only have a few
data detection positive examples

Logo
 reinforcement learning

Robots can learn to perform tasks

the physical world using this
Robotics technique.

Given finite resources and a defined

goal, reinforcement learning can help
enterprises plan out how to allocate
resources.

Reinforcement learning has

Resource been used to teach bots to
Video
managemen play a number of video
gameplay games.
t

Logo
 what's it used for?
Customer relationship management. Human resource information systems

Self-driving cars Virtual assistants.

Logo
advantages and disadvantages of machine learning

help enterprises it can be expensive,

understand their who command high
customers at a salaries, require
deeper level software
• Ex: Uber in finding infrastructure
drivers • contain errors can lead to
• Facebook finding friends inaccurate models
recommendation

Logo
Machine Learning Steps

Making Collecting
Predictions Data

Parameter Preparing
Tuning the Data

Evaluating Choosing a
the Model Model

Training
the Model

Logo
 How to Implement Machine Learning Steps in Python

start by importing any necessary modules

you will import the data

Logo
How to Implement Machine Learning Steps in Python….
Cont.
clean your data by removing
duplicate values, and
transforming columns into
numerical values to make them
easier to work with.

split your dataset into training and testing sets.

Logo
 How to Implement Machine Learning Steps in Python….
Cont.

Now, predict your testing dataset and find how accurate

your predictions are

1.0 is the highest level of accuracy you can get

Logo