Product Size Recommendation for Fashion E-commerce
Section: Crisp Talk Technical level: Intermediate
Recommending product sizes to customers is an important problem in the e-commerce domain. Though e-commerce is becoming increasingly popular, products such as apparel and shoes remain challenging to buy online and record high return rates. A key customer pain point that leads to excessive product returns is the size-fit problem. This talk (based on the linked WWW 2018 paper) describes some recent machine learning models that were built for product size recommendation.
I this talk, I would like to give a brief overview of the size recommendation problem in the domain of fashion e-commerce, discuss possible approaches and machine learning models to address this problem.
What is the size reecommendation problem
What are the challenges recommending size online
Possible approaches / Size recommendation ML models
Note that the speaker is not currently with Amazon - though the talk is based on recent work by the speaker publilshed in WWW 2018 while at Amazon.
Dr. Lavanya Tekumalla is a Machine Learning Scientist/independant consultant. She has worked in the industry for over 7 years in various roles at Amazon.com (Seattle, Bangalore), InMobi (Bangalore) and Myntra (Bangalore) that she thoroughly enjoyed. She has received a PhD from the Machine Learning Lab @ CSA, IISc, Bangalore and a masters in Computer Graphics from the University of Utah. Her research interests include Natural Language Understanding, Machine Learning - particularly Probabilistic Modeling and efficient techniques for approximate inference.
Make your own DL framework
We have all used all the high end frameworks that works really well. How about writing a small strip down version of one. In this session, I’ll walk you through how to write a small Deep Learning Framework in Pure python and numpy which has auto grad and optimizers and easy to create models. Also, the framework will be extendible so that you can easily play around with. more