Machine learning is a science in which machines are becoming smarter and helping humans to make the best decisions based on previous data recommended practices. This technique is not new but is occupying fresh momentum. Machine Learning Algorithm learns from the previous records and analyses the data. Without any human interrupt, it will generate its own recommendation. A machine will add that recommendation as experience in its database and use it for further processing. In short, the machine learns from its own experience and gives you better and better output. Machine learning is an iterative process as the more data added to machines learn from fresh feeds of data and then independently adapt new features to handle new data without constant human intervention. Machine learning was earlier used to predict what’s happing with the business but now the machine learning algorithm will suggest what action needs be taken by moving our business forward. This PowerPoint presentation presents the results of a literature survey of machine learning applications nurturing the growth of various business domains. More specifically, it gives a brief introduction of Machine Learning, four major types of Machine Learning, enhancement in various business domains by the use of various machine learning algorithms.

1. Machine Learning Applications
Nurturing Growth of Various Business
Domains
Presented by:- Shrutika Suresh Oswal

2. Outline
Introduction
Types of machine learning
Machine Learning Applications in Various Business Domains
Conclusion
References

3. Introduction
 Machine learning is a science in which machines are becoming smarter and helping humans for
taking best decisions based on previous data recommended practices.
 It is concerned with computer programs that automatically improve their performance through
experience.
 Machine Learning Algorithms automatically adapts and customize themselves for individual users
 The concept of learning in ML system is
Learning = Improving with Experience at some Task
Improve over task T
With respect to performance measure, P
Based on experience, E

4. Types of Machine Learning
Supervised Learning
Unsupervised Learning
Semi-Supervised Learning
Reinforcement Learning

5. Supervised Learning
 Supervised learning is where you have input variables (x) and an output variable (Y) and you
use an algorithm to learn the mapping function from the input to the output.
Y = f(X)
 The goal is to approximate the mapping function so well that when you have new input data
(x) that you can predict the output variables (Y) for that data.
 Supervised learning problems can be further grouped into two problems.
 Classification: In this output variable is a category.
 Regression: In this output variable is a real value, such as “dollars” or “weight”.
Example: When you have a ton of labelled pictures of dogs and cats and you want to
automatically label new pictures of dogs and cats

6. Unsupervised Learning
 It is where you only have input data (X) and no corresponding output variables.
 The goal is to model the underlying structure or distribution in the data in order to learn more
about the data.
 Unsupervised learning problems grouped into two problems.
 Clustering: It is where you want to discover the inherent groupings in the data, such as
grouping customers by purchasing behaviour.
 Association: It is where you want to discover rules that describe large portions of your data,
such as people that buy X also tend to buy Y.
Example : When you want to see how the pictures structurally relate to each other by colour or
scene or whatever

7. Semi-Supervised Learning
 Here, you have a large amount of input data (X) and only some of the data is labelled (Y) .
 These problems sit in between both supervised and unsupervised learning.
 Many real world machine learning problems fall into this area.
 You can use unsupervised learning techniques to discover and learn the structure in the input
variables along with supervised learning techniques to make best guess predictions for the
unlabelled data.
Example: Where you have a ton of pictures and only some are labelled and you want to use the
unlabelled and labelled to help you in turn label new pictures in future

8. Reinforcement Learning
 Reinforcement learning algorithms try to find the best ways to earn the greatest reward.
Rewards can be winning a game, earning more money or beating other opponents. They
present state-of-art results on very human task, for instance, this paper from the University
of Toronto shows how a computer can beat human in old-school Atari video game.

9. Machine Learning Applications in Various Business Domains
 Machine learning applications in healthcare
 Machine Learning Applications in Finance
 Virtual Personal Assistants
 Predictions While Computing
 Video Surveillance
 Social Media Services
 Email Spam and Malware Filtering
 Online Customer Support
 Machine Learning Application in Retail

10. Machine learning applications in healthcare
 Drug Discovery :
Processing of thousands of compounds through series of tests becomes feasible, efficient and
time saving by the use of ML Algorithms.
Examples: Pfizer, Roche, Sanofi, GlaxoSmithKline, Johnson & Johnson and Benevolent
 Diagnostic in Medical Imaging :
Doctor’s can easily monitor activities inside patient’s body by inserting scope into patient’s
body which used to creates structures of diseases and & possible complexity that can occur in
future.
 Personalized Treatment :
Machine Learning algorithm, is used for giving personalized medicine and more effective
treatment based on individual’s health data, symptoms and genetic information.
 Clinical Trial Research :
It enable the doctors to take direct clinical trial research which will result in a predictive
analysis of critical diseases.

11. Google’s solution on diabetic retinopathy
 Diabetic Retinopathy is a condition which causes blindness but if you can detect it earlier
you can completely cure it otherwise this causes blindness it’s the fastest growing cause of
blindness in the world today you need advanced ophthalmologist to detect these conditions
but using machine learning we can detect it pretty accurately so that a regular doctor can
detect these conditions.

12. Machine Learning Applications in Finance
 Fraud Detection:
Machine learning algorithm scans the data and outputs the fraud-score i.e. ratio of an
unusual transaction. If that fraud-score is above a particular threshold then your account will be
triggered automatically.
Examples:
Citibank
PayPal
 Finance by Targeting Account Holder:
Machine Learning Algorithm identify patterns of customer interest that are used by various
financial companies and banks for offers deals which are very much close to induvial customer
needs.

13. Virtual Personal Assistants
Google Assistant
• If you want to inquire about anything, it search for you.
• We can acquire information translated into many languages.
Siri:
Siri is software which helps us to do a list of tasks. For example, calling a
relati friend. You can also set your timer at a particular time. You can use this for
also sending or receiving messages.
Amazon Echo:
It is a speaker which is hands-free. With a connection to Alexa voice service,
you can control the Amazon-Echo by your voice. Your voice can be heard in the
whole room.

14. Predictions While Computing
Traffic Predictions
For building a map of current location &
velocities are saved into central database server
through GPS navigation services in our car. But there
is less number of cars that are well equipped with
GPS. Machine learning algorithm will estimate the
rate of congestion in particular area by analysing its
previous daily experiences.
Online Transportation Network
While booking a cab, the application directly
estimates the price of the ride. For minimizing detour
while providing services the companies like Uber use
machine learning algorithm to define price surge hours
by analysing rider’s demand.

15. Video Surveillance
machine learning algorithm used
at various surveillance services.
The machine learning algorithm
track the unusual behaviour of
people like standing in one place
for a long time, being motionless,
stumbling or any other action it
will automatically alert the
security guards &detect crime
before it occurs. When such
incident counted to be true, a
machine will add this experience
to a database, recall it and learn
from it.
Author Gaurav Goswami and et al. proposed a novel face verification algorithm. In this they have used
discrete wavelet transform and entropy computation technique. This algorithm yields the verification
accuracy of over 97% [1]

16. Social Media Services
People You May Know:
The Facebook continuously notices the
profiles that you are visiting, friend requests that
you are offering or accepting, your area of
interest using machine learning algorithm.
Facebook continuously learn from your collected
data & suggest you a list of Facebook users that
you can become a friend with
Face Recognition:
When any Facebook user uploads a
picture with friend Machine learning algorithm
observes the pose & projections in the picture
and finds out a unique pattern with your friend
list for identifying your friend.

17. Email Spam and Malware Filtering
More than 3, 25,000 malwares are found on each day. So far, the security of our
system there is various machine learning algorithm developed. One common
approach is that each, malware code is 90-98% similar to previous one. So, for
machine learning algorithm, it is easy to find out codes with 2-10% variation 7
protect our system against them.
Machine learning algorithm is
used in spam filtering will
continuously update the system

18. Online Customer Support (NLP)
Author Godson Michel and et al
have changed the concept of customer care
using automated chatbots that works on
Natural Language Processing and provide
the ease of use of various websites.
chatbot recalls the customer query & answer
to it by extracting information from the
website itself. Using machine learning
algorithm chatbot we can better understand
user’s query and serve them better answer.

19. Machine Learning Application in Retail
Product Recommendations: 42% of the retailers are providing personalized product recommendations.
Amazon Company uses machine learning algorithm of Artificial Neural Network for generating a
personalized recommendation.
Improved Customer Service
On an average, finding a new customer for the retailer’s company is 5 to 25% difficult
& expensive task than retaining old customers. The retailer company uses machine learning technique to
make a customer happy and give them satisfaction with its shopping.

20. Conclusion
 Machine learning is not a new concept but now a days, it receives that dramatic
computational power and availability of huge amount of data for processing
 Various business domains have been using machine learning algorithms to enhance the
existing systems. Machine Learning algorithms are helping the businesses to excel in term
of money as well as increased growth rate.
 As machine learning algorithms are self-learning algorithms they enhance the businesses-
oriented technologies by utilizing the past experiences and results.
 Google translator’s market, performance and efficiency is tremendously increased in one
year than past ten years – Sundar Pichai (Google CEO)

21. References
 Gaurav Goswami, Mayank Vatsa “Face Verification via Learned Representation on Feature-Rich
Video Frame”, IEEE Transactions on Information Forensics and Security ( Volume: 12, Issue: 7, July
2017 )
 Chen Qiu, Yanyan Zhang, Zhiyong Feng, Ping Zhang, Shuguang Cui “ Spatio-Temporal Wireless
Traffic Prediction with Recurrent Neural Network”, IEEE Wireless Communications Letters (
Volume: PP, Issue: 99 , 23 January 2018)
 Godson Michael D'silva ; Sanket Thakare ; Sharddha More ; Jeril Kuriakose “ Real world smart
chatbot for customer care using a software as a service (SaaS) architecture”, I-SMAC (IoT in Social,
Mobile, Analytics and Cloud) (I-SMAC), 2017 International Conference (INSPEC Accession
Number: 17224834 DOI: 10.1109/I-SMAC.2017.8058261, publisher:IEEE, 10-11 Feb. 2017.
 Chantal Fry, Sukanya Manna , “ Can We Group Similar Amazon Reviews: A Case Study with
Different Clustering Algorithms”, Semantic Computing (ICSC), 2016 IEEE Tenth International
Conference on INSPEC Accession Number: 15886708, DOI: 10.1109/ICSC.2016.71 , Publisher:
IEEE , Conference Location: Laguna Hills, CA, USA, 24 March 2016.