Description:

This tutorial aims at introducing important concepts of Business Intelligence and Mathematical modelling to the targeted audience and opportunity for them implement their own BI systems in python. Business Intelligence may be defined as a set of mathematical models and analysis methodologies that systematically exploit the available data to retrieve information and knowledge which may be useful in supporting complex decision-making processes [1]. Implemented carefully and precisely, a business intelligence environment offers decision makers information and knowledge derived from processing data and through the application of complex mathematical models and algorithms.

As a part of this tutorial, I will cover various options available in Python for ETL, mining data, data preparation, data exploration, regression (linear as well as time series) and applying classification algorithms like SVM (Support Vector Machines), clustering algorithms on raw data in order to arrive at meaningful conclusions using various python libraries like NumPy, matplotlib, pandas, SciPy, scikit-learn and many more.

Not just introduction to possibilities for implementing full fledged BI system in python, this session is meant to implement simple BI system on your own using python libraries. Also, as a part of this session, audience will be given brief introduction to the fields of data science and machine learning which includes implementing machine learning algorithms on various problems staged on online competition platforms like HackerRank and Kaggle; just to help them to learn different models used for Business Intelligent systems. [See content and slides]

Towards the end, brief introduction to big data and pySpark has been added, so that audience will have slightest of hints on how to work with gigantic data!

What will attendees get out from the workshop?

1. How to make sense of the data.
2. How to replace common and/or commercial analytics tools with Python.
3. Getting business analytics done with Python for free of cost by creating models and great looking dashboards.
4. Most Importantly, sufficient introduction to fields of Machine Learning, Data Science, Big Data to get started!
5. Implementing your first Business Intelligence system in Python to make sense of transactional data generated by customers on retail stores.

References

[1] Vercellis, C., Business Intelligence: Data Mining and Optimization for Decision Making, Wiley (2011)

Prerequisites:

The attendees should have basic Python programming experience. Basic understanding of Probability and statistics and linear algebra concepts would be great! Though prior knowledge of mathematical modelling techniques is not required. Acquaintance with NumPy, SciPy, scikit-learn and matplotlib package is definitely a plus.

Content URLs:

Rough plan for the proposed workshop

Slides and Content for the workshop (Without codes, which will be added shortly)

Speaker Info:

I did my bachelors in System Science from IIT Jodhpur where I was introduced to python, since then I have been working on python language. My research areas lies in Business Intelligence and Financial engineering. During bachelors degree, my bachelor thesis focused on Business Intelligence where I worked on different decision support systems. I am currently working as a Software Development Engineer in a Bangalore based Big data start-up, working primarily on feature engineering and automated machine learning.

Speaker Links:

Github