Machine Learning on Azure

Both R and Python can connect to a wide variety of data sources. Under this module, learn how to establish a gateway between different databases with R and Python. Also, learn how to connect to external sources of data.

One of the major modules of Data Science is Machine Learning. Learn about the various modules that make up Machine Learning using the two most popular tools R and Python. Get introduced to the broad overview of ML and the various quality metrics with the help of R and Python.

In the real world, oftentimes, the datasets cannot be used as such and some amount of preprocessing activity needs to be done. Imbalance in the output classes is one of the common problems where sometimes the proportion may be as lopsided as 95% to 5% or even higher. Learn about the various methods and algorithms to address this problem of imbalanced data sets.

One of the integral parts of learning Data Science and working on Analytics projects is the sound understanding of Statistical tools. In this module learn about the need to know statistical measures and their application in Data Science. Also, learn how to visualize the data in a concise form to derive various meaningful insights.

The essence of analytics is to be able to get the story from the data. And for the data to be able to truly be useful one needs to munge the raw data to make it legible. Using tools like R and Python, learn how to manipulate data from the raw form to make it ready for subsequent ML algorithms. The topic is all the more important in the current context given that a lot of data is moving to the unstructured format.

Both R and Python while being classified under the object-oriented programming languages category, still require some traditional approach to programming whereby the user-defined function needs to be spelled out and the use of conventional program snippets is of utmost importance. In this module, learn how to create simple to complex user-defined functions and hone your programming skills in the context of machine learning.

As a data scientist one will be engaged in a multitude of data mining techniques in both supervised and unsupervised learning. One of the major variants of the same is reinforcement learning that enables machines to learn through rewards. Under this module, get introduced to all the supervised techniques of prediction and classification; learn about the major unsupervised learning methods and the application of reinforcement learning in Data Mining.

Under supervised learning one of the most popular methods of predicting numeric data is linear regression and for classifying categorical data is logistic regression. These two methods will be covered in detail under this module. And the participant will be introduced to multiple examples using R and Python.

Under the classification modeling, decision trees method has a special place even though they are not truly classification modeling techniques but rule-based algorithms. The popularity of Decision trees is in its simplicity, high accuracy, and most important the ability to explain behind- the- scenes working of the algorithm. Under this module, the participant will be introduced to working with Decision trees using R and Python.