SIGMOD reproducibility award for 2018

Prof Sayan Ranu’s (Computer Science) work titled Debunking the Myths of Influence Maximization: An In-depth Benchmarking Study has won the SIGMOD reproducibility award for 2018. Co-author Sainyam Galhotra is also a recent CSE B.Tech graduate from IIT Delhi.

The award recognizes the best papers in terms of reproducibility of the results claimed in the paper. The three most reproducible papers are picked every year. The award will be formally presented in the SIGMOD conference in Houston from June 10-15, 2018.

Giving details of the research Prof. Ranu said that Influence maximization (IM) on social networks was one of the most active areas of research in computer science.

“While various IM techniques proposed over the last decade have definitely enriched the field, unfortunately, experimental reports on existing techniques fall short in validity and integrity since many comparisons are not based on a common platform or merely discussed in theory. In this paper, we perform an in-depth benchmarking study of several prominent techniques for IM, and surface the open challenges even after a decade of research.”

In addition, he said that the research had unearthed several misconceptions and incorrect claims made in the literature. “Among them, we establish that CELF++, which claims to be 35%-50% faster than CELF, is in reality not faster than CELF. This is a particularly important result since Celf++ has garnered more than 350 citations till date based on this incorrect claim.”