Video

Meet the team

Data Management, Exploration, and Mining group (DMX) at Microsoft Research
"We make it easy for information workers to find relevant transformations for their sophisticated data wrangling tasks."
Yeye He

Garage Team

Yeye He, Kris Ganjam, Surajit Chaudhuri, Vivek Narasayya, Kukjin Lee

Data Management, Exploration, and Mining group (DMX) at Microsoft Research

Redmond, WA

Backstory

We are motivated by the fact information workers everywhere have the need to transform, clean and standardize their data so they can better understand and analyze it. Users often work with data from common domains such as date and time, phone numbers, names, addresses and URLs. Although programmers over the years have written relevant data transformations for such domains, today they are locked up in code libraries and web services and are not easily discoverable and usable, particularly for users who are not sophisticated programmers themselves. Our approach is to use an example-driven paradigm for information workers to find transformations easily. To make this paradigm succeed for the world of complex, domain specific data transformations, we developed a backend service on Azure that can crawl, index and search over repositories of transformation functions and web services efficiently. Our current release exposes this functionality as an Excel Add-In. Like web search engines, we will continuously improve our service by leveraging user feedback.

Video https://youtu.be/cl7RXSHdsyo

More information from the Microsoft Research team https://aka.ms/transform-data-by-example

Download from the Office Store https://aka.ms/transform-data-by-example-download