Interview with Meituan’s Data Science Expert Yan Peng on KDD Cup Success and Machine Learning Insights
In this interview, Meituan senior data‑science expert Yan Peng—known as “Eureka,” a five‑star Kaggle competitor and 2017 KDD Cup champion—shares his career from Tsinghua to Meituan, the team strategies that won the traffic‑flow prediction challenge, and his advice to focus on strong mathematics, stay current with research, and study resources such as “The Elements of Statistical Learning” and NTU’s introductory machine‑learning course.
The 2017 KDD Cup, the most influential data mining competition, announced its results, and the Convolution team from 70 countries and 3,582 teams won both tasks. One of the champions is Yan Peng, a senior technical expert at Meituan-Dianping.
Yan Peng, known online as "Eureka", ranks 5th on Kaggle worldwide and holds the highest ranking among Chinese participants, with a record of 16 gold, 13 silver, and 3 bronze medals.
Yan Peng graduated with a master's degree from Tsinghua University in 2002, focusing on pattern recognition. He worked on computer vision at a startup (2002‑2005), founded his own company (2005‑2008), handled advertising at NetEase (2008‑2016), and joined Meituan in 2016, first working on hotel‑travel ranking and later on the financial services platform, focusing on machine learning.
He has participated in the KDD Cup three times. In 2015, as team captain, his multinational team (named InterContinental Ensemble) won first place. In 2016, a time‑zone error caused a late submission and missed the award. In 2017, he joined a new team with members from Microsoft and Beihang University, and they won the championship.
The 2017 competition featured a challenging traffic‑flow prediction task with limited data, making results unstable and partly dependent on luck. The team’s workflow involved each member exploring solutions independently based on their technical strengths, followed by thorough discussion and integration of ideas.
Yan emphasizes that success in such competitions requires strong mathematics, extensive practical experience with data mining and machine learning, and a keen sense for data patterns. He advises algorithm practitioners to keep up with the latest research from academia and industry and to apply those findings to real work.
He also recommends two learning resources: the textbook "The Elements of Statistical Learning" , which is challenging but valuable, and Professor Lin Hsin‑Tian’s machine learning course from NTU ( link ) for beginners.
Signed-in readers can open the original source through BestHub's protected redirect.
This article has been distilled and summarized from source material, then republished for learning and reference. If you believe it infringes your rights, please contactand we will review it promptly.
Meituan Technology Team
Over 10,000 engineers powering China’s leading lifestyle services e‑commerce platform. Supporting hundreds of millions of consumers, millions of merchants across 2,000+ industries. This is the public channel for the tech teams behind Meituan, Dianping, Meituan Waimai, Meituan Select, and related services.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.