【Title】京东推荐系统及图片属性预测
 
【Speaker】Dr. Yan Jin, Research Engineering, JD.com Silicon Valley Office
 
【Host】Dr. Kaibo Wang
 
【Time】10:00-11:00am, January 11th (Thursday) , 2018
 
【Location】 Room N510, Shunde Building
 
【Abstract】In this talk, I will mainly introduce the framework of recommendation system at JD.com, and also one of my main focuses: fashion image attribute prediction. The recommendation plays an important role at JD.com in many different ways: it contributes huge amount of advertisement profits, strongly attract customers to JD.com platform, improve the customer shopping experience, merchants promotions, etc. The recommendation system at JD.com is also a very so sophisticated. It is very important to us to build a reliable and quickly responsible to provide meaningful merchants to our customers. On the other hand, the attributes of merchants are vital when doing content-based filtering. I will talk about how we make use of huge amount of images to do fashion attribute prediction.
 
【Short Bio】Yan Jin is now a research engineer at JD.com Silicon Valley office. He received his PhD degree in industrial engineering from University of Washington (UW), Seattle, WA, in 2017, and his bachelor degree in geophysics from University of Science and Technology of China (USTC), Hefei, China, in 2011. His PhD dissertation is “Diagnostic Monitoring of High-dimensional Networked Systems (with Applications in Manufacturing and Healthcare System)”. His interests are image recognition and attribute prediction with deep learning and computer vision, and personalized advertisement recommendation.
 
All interested are welcome!