{"id":149,"date":"2021-04-29T13:54:00","date_gmt":"2021-04-29T18:54:00","guid":{"rendered":"https:\/\/groups.cs.umass.edu\/zamani\/?p=149"},"modified":"2021-07-28T13:56:30","modified_gmt":"2021-07-28T18:56:30","slug":"sigir-2021-stats","status":"publish","type":"post","link":"https:\/\/groups.cs.umass.edu\/zamani\/2021\/04\/29\/sigir-2021-stats\/","title":{"rendered":"SIGIR 2021 Statistics"},"content":{"rendered":"\n

(This post was originally published in my former blog on April 29, 2021)<\/p>\n

It has been three years since I have started reporting some statistics on the SIGIR accepted papers. Here is the links to the\u00a0SIGIR 2018<\/a>\u00a0and the\u00a0SIGIR 2020<\/a>\u00a0report. I have intentionally skipped the SIGIR 2019 report as\u00a0Jimmy Lin<\/a>\u00a0released\u00a0his statistics<\/a>\u00a0before I get a chance to spend time on it and there was no point in redoing it.<\/p>\n

The following information has been automatically extracted from the\u00a0full list of accepted papers to ACM SIGIR 2021<\/a>. Please contact me if you find an error in the reported statistics.<\/p>\n

Unlike previous years, the SIGIR website has not yet released the author affiliations. Therefore, there is no institution-level statistics in this blog post.<\/p>\n

First, let\u2019s take a look at a word cloud I created from the titles of the accepted papers (all tracks, including full, perspective, short, and resource tracks):<\/p>\n

\"sigir<\/a><\/p>\n

\"SIGIR<\/p>\n

The high frequency of words such as \u201clearning\u201d, \u201cretrieval\u201d, \u201crecommendation\u201d, \u201csearch\u201d, \u201ctext\u201d, \u201cuser\u201d, and \u201cdeep\u201d is not surprising. However, I was personally shocked when I saw the high frequency of word \u201cgraph\u201d in SIGIR paper titles. It seems that graph neural nets are finding their ways into the information retrieval domain. We should have a better understanding of this observation once the proceedings is out.<\/p>\n

Now, let\u2019s take a look at some author statistics based on the number of papers they publish at SIGIR 2021. Similar to the blog post I wrote for SIGIR 2020, I will report the top 10 authors with respect to both publication count and normalized publication count.\u00a0Maarten de Rijke<\/a>\u00a0of University of Amsterdam and Ahold Delhaize has contributed to 9 papers and ranked #1 in the author list in terms of publication count. However, when the publication count is normalized by the number of authors on each paper,\u00a0Guido Zuccon<\/a>\u00a0of University of Queensland holds the first position.<\/p>\n

UMass Amherst researchers are also among the top contributors of SIGIR.\u00a0Hamed Zamani<\/a>\u00a0(Assistant Professor) is among the top 10 authors in terms of both publication count and normlaized publicatio count for all papers.\u00a0Daniel Cohen<\/a>\u00a0(PhD student) who has recently joined Brown University as a Postodtoral Researcher is the top author in terms of normalized publication count for full papers. He published a solo-author paper at SIGIR 2021 which is part of his PhD dissertation.<\/p>\n

TOP 10 AUTHORS – All papers (full, perspective, short, and resource tracks) – Based on Normalized Publication Count<\/h3>\n