Abstract: Data professionals (analysts, scientists, operators, …) utilize data to extract insights from it and subsequently to make decisions that impact day-to-day operations as well as long term strategy for organizations. The process of going from data to insights and using decisions typically involve (a) extracting data from varied structured and unstructured sources; (b) normalizing, cleaning, stitching such varied data sources to obtain “ground truth”; (c) extracting structure within data, interacting with it, visualizing it to obtain insights; (d) predicting, optimization, doing scenario analysis to make decisions, and (e) automating all of the above while allowing for human in the loop intervention.
In this hands-on workshop, we shall discuss all of these with the help of illustrative datasets in a low-code / no-code AI environment.
The session of 90 minutes will be divided into four 20 minutes sessions with an additional 10 minutes to allow for questions and answers as well as discussions. Each session discusses automation, with human in the loop, using low-code / no-code AI environment for the key four tasks involved in going from data to insights and using it for decisions.
Session 1. Extract.
Session 2. Data cleaning, stitching, normalizing.
Session 3. Interact, visualize and extract insights.
Session 4. Scenario analysis for decision making.
An important differentiator from “university classroom” learning of the above topics is the need for continuous attention required in such a data-driven process – data constantly changes, so do models, constraints and business rules. Therefore, the automation of such processes require human in the loop intervention, ideally minimal. This is precisely what this hands-on workshop focuses on.
A good, functional device that can connect to the internet via browser (preferably Chrome).
Ability to do basic “spreadsheets” operations.
And most importantly, a positive attitude and curiosity to learn.
Bio: Devavrat Shah is Andrew (1956) and Erna Viterbi Professor with the department of Electrical Engineering and Computer Science at Massachusetts Institute of Technology. He is the founding director of Statistics and Data Science at MIT. He is also a member of IDSS, LIDS, CSAIL and ORC at MIT. He co-founded Celect, Inc. (now part of Nike) in 2013 to help retailers decide what to put where by accurately predicting demand using omni-channel data. He is a co-founder and CTO of IkigaiLabs with the mission to build self-driving organizations by enabling data-driven operations with human-in-the-loop.
His research focuses on statistical inference and stochastic networks. His contributions span a variety of areas including resource allocation in communications networks, inference and learning on graphical models, algorithms for social data processing including ranking, recommendations and crowdsourcing and more recently causal inference. He has made foundational contributions to the development of “gossip” protocols and “message-passing” algorithms for statistical inference which have been the building blocks of modern distributed data processing systems.His work spans a range of areas across electrical engineering, computer science and operations research.
His work has received broad recognition, including prize paper awards in Machine Learning, Operations Research and Computer Science, and career prizes including 2010 Erlang prize from the INFORMS Applied Probability Society, awarded bi-annually to a young researcher who has made outstanding contributions to applied probability. He is a distinguished alumni of his alma mater IIT Bombay from where he graduated with the honor of President of India Gold Medal. His work has been covered in popular press including NY Times, Forbes, Wired and Reditt.