Big Data Analytics: Theory & Practice
from 05:00 PM to 06:00 PM
Big Data Analytics: Theory & Practice
March 07, 2017 Smurfit School and University College Dublin (UCD)
By Shaku Atre, Atre Group, Inc., Manhattan, New York
March 07, 2017
All along with Big Data or without Big Data enterprises have made forecasts, they have launched products, served their customers and run businesses successfully. They also had to finish fulfillment of products or services to keep customers happy and at the same time make profit to keep the businesses afloat. Then what is different now?
Predictions based on “Big Data”. If a prediction indicates that something will go wrong, businesses will do everything possible to prevent it. The main use of Big Data is to be able to look at data and predictions in much faster ways and change the business processes midstream.
Time to Results (TTR) is of paramount importance for businesses to succeed. If you think volume of big data is the only problem – you are mistaken. Bigger than Big Data the bigger problem is Speed – meaning the velocity with which the big data is arriving today, with which it is supposed to be worked on, and with which the insights are supposed to be provided to the decision makers. THE MAIN PROBLEM is a dual problem - Big Data and Big Speed.
Data goes mainly through four phases in a circle and the real challenge with the big speed of this big data is now to manage this “Disruption” of the circular process & having to be able to change the course of running the business instantly with understanding what the data is trying to tell us:
- Phase 1: Data is generated by transactions
- Phase 2: Data is received by various recipients
- Phase 3: Data is stored and processed
- Phase 4: Insights are created & acted upon
In order to master this double headed tsunami of big data and big speed our businesses have to have expertise in data science’s main building blocks: mathematics and statistical analysis, skills which many of today’s data analysts typically lack.
Besides the mathematics and statistical analysis other Technical skills needed are, among others:
- Ability to manage hardware with hundreds or thousands of “small’ CPUs with Master/Slave (very unfortunate nomenclature) architecture.
- Open Source Software in various forms and components of Hadoop, Spark, NoSQL, Cassandra, HBase and many names you can’t even pronounce
- Database architecture of terabytes and more of data and taming of the elephant
- Expertise with analytics programming languages and facilities such as very important languages R or Pig and a few more animals you will meet on this journey
And, soft skills having not much to do with Big Data, but are needed also with Big Data and Big Speed, are lacking in many organizations:
- Data story telling skills - Communications skills to explain the analytics results without the audience having the urge of “texting”
- Understanding of the ”ins and outs” of the business
- Ability to discern which analytics will answer the bottom-line questions
- Understanding not only transactions (as we have been doing all along) but also interactions (such as people buying products on the web) and observations (such as machines or sensors measuring and reporting about happenings or not-happenings).
Shaku Atre is an exceptional speaker, with the reputation of capturing the attention of audiences and maintaining their interest while guiding her listeners painlessly through sophisticated material. Ms. Atre is the president of Atre Group Inc. which is a leading consulting, training, and publishing company specializing in business intelligence (BI), Data Warehouses and Big Data. Before heading her present company, Ms. Atre was a Partner with Price Waterhouse Coopers. She also has fourteen years of experience in various fields with IBM. Ms. Atre is an acknowledged expert in the data warehousing and database field. She has extensive practical experience in database projects, has helped a number of clients in establishing successful data warehouses and client/server installations, and has taught at IBM’s prestigious Systems Research Institute.
She has lectured on the subject to professional organizations in the USA and Canada, as well as close to 40(yes – forty) countries around the world. Ms. Atre is frequently quoted in reputable publications such as Computerworld and Information Week. She has written an award-winning outstanding book on database management systems that has become a classic on the subject: “Database: Structured Techniques for Design, Performance and Management,” published by John Wiley and Sons, New York. The book has sold over 250,000 copies ( yes – a quarter million - not including its Spanish and Russian translations) and has been selected by several book clubs and leading universities including Harvard, Columbia, Cornell, MIT, New York University, Stanford, and U.C. Berkeley as well as by the Moscow University. Her book, “Information Center: Strategies and Case Studies,” published by Atre International Consultants Inc., has also been very well received by the industry. “Database Management Systems” is another successful book authored by Ms. Atre used by the , now Big Four accounting companies as their, so to say, ”Bible” of the database projects . Her fourth book, “Distributed Databases, Cooperative Processing & Networking,” was published by McGraw-Hill. She has also authored a very well received book: “Atre’s Roadmap for Data Warehouse/Data Mart Implementations,” published by Gartner Group, and is co-author of her latest BI book, “Business Intelligence Roadmap: The Complete Project Lifecycle for Decision-Support Applications,” published by Addison Wesley.
Shaku has Master’s Degree in Statistics Suma Cum Laude from the University of Poona, India. Her majors were Physics, Chemistry and Mathematics, Suma cum Laude, at her Bachelors of Science Degree at S. P. College in Poona, India. She has done Research in Applied Mathematics with thesis in Astronomy on Merit Scholarship at University of Heidelberg, Germany.