About US

InSight Talk (37) | Geo-Spotting | Dr.Dmytro Karamshuk, Skyscanner, UK

Edit
x

Invitation

 MYRA-Insight37-Dr.Karamshuk-1w

 MYRA-Insight37-Dr.Karamshuk-2

InSight Special Lecture (37)

"Geo-Spotting: Mining Online Location-based Services for Optimal Retail Store Placement"

by

Dr. Dmytro Karamshuk

Senior Data Scientist
Skyscanner, UK

 

Date: Wednesday, December 27, 2017 | 03:30 PM

Venue: MYRA School of Business

Talk Summary: 

The problem of identifying the optimal location for a new retail store has been the focus of past research, especially due to its importance in the success of businesses. Traditional approaches to the problem have factored in demographics, revenue and aggregated human flow statistics from nearby or remote areas. However, the acquisition of relevant data is usually expensive. With the growth of location-based social networks, fine-grained data describing user mobility and popularity of places have recently become attainable.
In this talk, Dr.Karamshuk would focus on the predictive power of various machine learning features in making decisions regarding the location of retail stores. Based on his research on the popularity of retail stores in the city through the use of a dataset collected from Foursquare in New York, he is going to lay out how data mining can help make business decisions. The features they mined in their study included two general signals: geographic, where features are formulated according to the types and density of nearby places, and user mobility, which includes transitions between venues or the incoming flow of mobile users from distant areas. Their research suggests that the best performing features are common across the three different commercial chains considered in the study—Starbucks, Dunkin’ Donuts and McDonalds, although variations may exist too, as explained by heterogeneities in the way retail facilities attract users. We also show that performance improves significantly when combining multiple features in supervised learning algorithms, suggesting that the retail success of a business may depend on multiple factors.

About the Speaker:

Dima (Dr. Dmytro Karamshuk) is a Data scientist with considerable research and industry experience including startup experience working on understanding user behavior in social media and mobile networks to make the world slightly better place. Before joining Skyscanner, he spent more than three years as a Post Doc at King’s College London.

Dr. Karamshuk has worked on studying watching patterns of BBC iPlayer users (project in collaboration with BBC R&D), mining geo-tagged social networks (Foursquare and Twitter) for geographic retail analysis (collaboration with the University of Cambridge) and modeling user preferences in Twitter and Pinterest. Published in the top-tier academic conferences (KDD, WWW, INFOCOM, etc.) and journals (Communications Magazine, J-SAC, Research Policy, etc.) and featured in press (BBC, New Scientist, etc.). He is a frequently invited speaker at industrial labs (Google, Bell Labs, Yahoo Labs, Microsoft Research Asia, International Atomic Energy Agency, etc.) and a co-founder and former CEO of stanfy.com.

To know more about Dima (Dr. Dmytro Karamshuk) and the kind of interesting work that he has been doing read this interview: https://www.datascienceweekly.org/data-scientist-interviews/social-media-data-machine-learning-optimal-retail-location-dmytro-karamshuk-interview

You are cordially invited

Dr. Shalini Urs

Chairperson, MYRA School of Business

 

 

Report

   

 

Photos

DSC_8443
DSC_8492
DSC_8410
DSC_8448
DSC_8466
DSC_8416
DSC_8452
DSC_8510
1/8 
start stop bwd fwd

Back