KDD

We are very excited to welcome you to the twenty-sixth meeting of KDD - arguably the most unique KDD experience in history. The world changed drastically due to the spread of COVID-19 disease at pandemic proportions. As of this writing, all gatherings are ruled out, social distancing and masking are now the mandated norms. Yet, our community has been drawn closer than ever, especially in response to the ongoing public health emergency to detect and mitigate the spreading disease. As to the signature meeting event itself, the organizing community and KDD steering committee took on the challenge and worked together so that KDD 2020 would be a brand new experience of a virtual conference with innovations that enable the community to socialize, make acquaintance and strike conversations among the presenters and attendees. Not surprisingly, among the program highlights are work by the KDD community on combating COVID-19 showcased through a number of sessions consisting of themed workshops, health day, a special call for papers and a research panel on data science for preventative measures, contact tracing, treatment recommendation, and so on. Sprinkled throughout the program, there are many papers that address Data Science for best social distancing, for epidemiological modeling of the spread and for optimized operation of healthcare facilities. The program includes keynotes by Alessandor Vespignani who leads national effort on modeling the spatial spread of epidemics, and Emery Brown, a statistician, a neuroscientist and an anesthesiologist whose work has been recognized by election to all three branches of the National Academies of Sciences. Among our compelling keynote speakers are Yolanda Gil, researcher and president of the Association for the Advancement of Artificial Intelligence (AAAI); and Manuela Veloso, head of AI Research at JP Morgan and former head of Machine Learning at CMU and past president of AAAI. Likewise, our applied data science track features a compelling montage of twenty invited speakers to provoke thoughtful dialogue and to keep you engaged in the program.

The pandemic crisis of the last five months has also found most of us locked in our homes and all movement cancelled. Amidst this stillness, we have been confronted with -- individually and as a society -- many moments of reflection on the issues that matter most. Deepest among these has been a realization that some amongst our fellow citizens face a different daily reality, of a deep racial divide that denies basic tenets of equal protection with heart-wrenching examples of their brutal treatment at the hands of the very law enforcement employed to protect them. As the realization of a society starkly divided in its basic guarantees of life and liberty to its black American citizens dawns, we can go beyond rejecting racism to bring meaningful change. We believe data science can serve as a power tool for transparency and support for systemic change. It can expose inequality and injustice while providing data-driven solutions for bringing lasting change. SIG-KDD as an organization brings together a community of researchers who can focus precisely on these important problems. KDD, as its main event, is the primary means by which KDD reaches out to this community of researchers, creates the context by curating a technical program that brings important technical and sociological challenges to the attention of a broader audience of researchers and practitioners.

This year's KDD continues the trend of record high number of submissions, the highest we have seen ever in our applied data science and research tracks. A total of 2035 valid submissions were made, which is the highest number in KDD history (over 13% more than the second highest one): 1279 in the research track and 756 in applied data science track. After a careful review process, 338 papers (217 in research track and 121 in applied data science track) were accepted for publication in the proceedings. The selectivity and attention to detail by the program committees is reflected in the program where sessions range from deeply theoretical topics in statistical analysis, machine learning to practical applications in health, finance etc. As the sessions from refereed papers emerged, the chairs have carefully curated the overall program through keynotes and invited speakers. The first two days of the conference include tutorials by experts in their field and workshops and theme days, followed by three rigorous yet enjoyable days of research papers, applied data science papers and over 21 applied invited speakers sharing views from the trenches of industrial applications.

Diversity and inclusion is a major theme in this year's KDD. For the first time we organized a dedicated full day event for diversity and inclusion. The day includes exciting invited talks by researchers in underrepresented groups or on diversity and inclusion research, as well as mentoring sessions for students raising awareness of diversity and inclusion and coaching career development. The final program composition itself reflects the diversity of talent in the KDD community: three out of four keynote speakers are in underrepresented groups, ten out of twenty-one ADS track invited speakers are female, close to fifty-percent of our organizing committee are female. We also host a women research panel, social impact program, all reflecting tremendous volunteer activities in support of the diversity and inclusion theme this year.