CS&E wins prestigious 10-Year Highest-Impact award at ICDM

Research from the Department of Computer Science & Engineering (CS&E) was recognized with the prestigious 10-Year Highest-Impact award at the 2020 International Conference on Data Mining (ICDM).

The paper, titled “SLIM: Sparse Linear Methods for Top-N Recommender Systems” was published by alumna Xia Ning (Ph.D. 2012) and her advisor, Professor George Karypis in 2011. This honor is given annually to the paper that has had the most substantial impact on the data mining community over the last decade.

“Xia’s Ph.D. research on recommender systems focused on developing machine learning models for solving the top-n recommendation problem,” shared Professor Karypis. “To solve this problem in a computationally scalable way, she developed an entirely novel class of methods that rely on learning from the data a sparse item-to-item similarity matrix. These methods build upon the traditional item-item based neighborhood schemes, but employ statistical learning techniques to estimate the similarity matrix directly from the data.”

“Her results, initially presented at ICDM in 2011, showed that this method, called SLIM, outperforms all previously published methods, and does so in a computationally efficient manner,” said Karypis, “Even today, the performance achieved by SLIM is extremely competitive, as it was shown in the best paper awardee “Are We Really Making Much Progress? A Worrying Analysis of Recent Neural Recommendation Approaches” at the 2019 ACM Recommender Systems Conference.”

Xia Ning is now an associate professor at Ohio State University in both Biomedical Informatics and Computer Science and Engineering. Her work focuses on data mining, machine learning and big data analytics with applications for emerging critical problems in drug discovery, medical informatics, health informatics and e-commerce.

This is the second time that CS&E has received this renowned recognition. In 2010, the inaugural 10-Year Highest Impact award was granted to the paper “Frequent Subgraph Discovery”, written by Michihiro Kuramochi (Ph.D. 2005) and Professor Karypis.