2. What is Machine Learning
• how the computers can learn
• how the computer gains the knowledge,
• how the computer gains experience from the
past experiences.
3. Variety of Data
Nature of data Source of Data
What to perform
on data
Learning
Mechanism
Numeric Transactional Classification Supervised
Text Survey Prediction Unsupervised
Image Log Trend Semi Supervised
Hybrid Observed Classification
Supervised /
Semi Supervised
4. Models in Machine Learning
Model Learning Purpose Algorithms
Predictive Supervised Classification
• Nearest Neighbour
• Naive Bayes
• Decision Trees
• Classification rule learners
Predictive Supervised Prediction
• Liner regression
• Regression trees
• Model trees
Predictive Supervised
Classification /
Prediction
• Neural Networks
• Support Vector Machines
Descriptive Unsupervised Pattern discovery Association rules
Descriptive Unsupervised Clustering K-means clustering
5. Top 4 applications in
Machine Learning
• Google maps.
• Auto / suggestions of tagging of friends in
images posted in facebook
• Ads recommendations.
• Security applications – face recognition on
mobiles
6. Top 5 applications in Machine
Learning
• Google maps.
• Auto / suggestions of tagging of friends in
images posted in facebook
• Ads recommendations.
• Security applications – face recognition on
mobiles
• Auto pilots
7. Trends in Machine Learning
• Machine Learning is any where and
everywhere
• For example, Google’s TensorFlow, for instance,
is an “open source software library for Machine
Intelligence.” There are other API’s
• Also, API’s are available to the enterprises to
store the data, and analyze the data.
8. Trends in Machine Learning
contd...
• Integration of data science with business
operations
• Single algorithm applied to marketing, IT,
analytics, and a number of other areas.
• For example, - an open source platform for
Machine Learning
9. Trends in Machine Learning
contd...
• Chatbots for Daily Operations
• For example, IRCTC, SBI, etc.,
• Cyber security