NeurIPS Conference Analytics
The Microsoft Academic Graph makes it possible to gain analytic insights about any of the entities within it: publications, authors, institutions, topics, journals, and conferences. In this series, we present analytic insights about current conferences, which we hope will help you prepare for attending each event. All of the insights within are derived from the Microsoft Academic Graph and visualized in Microsoft Power BI. You can generate your own insights by accessing the Microsoft Academic Graph through the Academic Knowledge API or through Azure Storage (please contact us for the latter option). If you would like to learn how we generated the insights below, please see the repository with source code.
In this post, we present historical trend analysis about the conference NeurIPS – Conference on Neural Information Processing Systems, taking place in Montréal, Canada from December 2 – 8, 2018. We derive insights from 1996 to the latest available year.
Click on each image for current trends and data hosted by Microsoft Academic Graph.
NeurIPS paper output
The chart below shows the evolution of the number of conference papers for each conference year.
*Poster and workshop papers are also considered as conference papers in this chart.
In the following chart, the black bars represent the average number of papers a conference paper references for each conference year. The data shows that recent publications tend to reference more papers. The green bars show the average number of citations received by a conference paper for each conference year. Note that the citations are raw counts and not normalized by the age of publication. This is because the “correct” way to normalize the citation counts turns out to be a nontrivial problem and may well be application-dependent. Please treat the data presented as an invitation to conduct research on this topic!
*Average Citations: Average number of citations a NeurIPS paper received for a given conference year.
*Average References: Average number of references a NeurIPS paper references for a given conference year.
NeurIPS changed its submission instructions in 2009 to allow an additional page for references. The increase in average references in 2009 is likely due to the rule change.
Memory of references
How old are papers cited by NeurIPS papers?
Follow a given year’s column to see the age of papers cited in conference papers published that year. For example, in 2016, NeurIPS papers collectively cited 2,033 papers published in 2015; 1,392 papers published in 2014; and so on.
*If some years appear to cite publications from the future, it is likely due to two scenarios. First, they cited papers that are published in journals later. Second, they cited books. When a new edition of the book appeared, it replaced the previous one in the Microsoft Academic Graph, and the citation appears to be from the future. In this representation, we remove all instances of papers citing papers more than one year in the future, to generate a cleaner view.
What papers do NeurIPS papers cite? Are they usually from the same venue?
The charts below show the outgoing reference distribution by venue. The bar chart below shows the top 10 venues cited by NeurIPS papers. NeurIPS, ICML, Journal of Machine Learning Research, and CVPR emerge as the top four. The pie chart below shows that the top 10 most frequently cited venues makes up almost 40 percent of the total outgoing references from NeurIPS papers.
What papers cite NeurIPS papers? Are they usually from the same venue?
The charts below show incoming citation distribution by venue. The pie chart below shows that the top 10 venues that cite NeurIPS papers the most make up 24 percent of the total incoming citations. The bar chart below shows the top 10 venues that cite NeurIPS papers the most. Again, NeurIPS is at the top, followed by CVPR and ICML.
Which are the top institutions based on accepted papers by NeurIPS?
The bubble chart visualizes the top institutions at NeurIPS by citation count from their accepted NeurIPS papers. The color saturation of the bubble is proportional to the total number of publications from that institution that were accepted by NeurIPS.
Get the most current data and also explore the top institutions at the conference in more detail by clicking the chart. Once on the underlying Microsoft Power BI report, click on a column to rank the top institutions by publication or citation count.
Who are the top authors based on their NeurIPS papers?
The next three charts show author rankings according to different criteria.
The bubble chart displays NeurIPS authors ranked by citation count, with bubble color saturation being relative to publication count.
Get the most current data and also explore the top authors at the conference in more detail by clicking the chart. Once on the underlying Microsoft Power BI report, you can also explore the top conference authors in more detail. Click on a column to rank the top authors by Microsoft Academic rank, publication, or citation count.
The bubble chart below visualizes author rank, which is calculated by Microsoft Academic by using a formula that is less susceptible to citation counts than similar measures. The X axis shows author rank. The higher an author’s rank, the closer they are to the right side. The Y axis normalizes the rank by publication count and enables us to identify impactful authors who might not have had a very large number of publications. The closer an author is to the top, the higher their normalized rank. Of course, the area of the chart that represents the highest rank is the top right corner.
Most-cited scholars by NeurIPS papers
Who are the most-cited scholars of all time by NeurIPS papers?
The chart below ranks the most-cited scholars by using unique publications cited by NeurIPS papers and number of citations received from NeurIPS. Scholars do not have to have published in NeurIPS to appear on this chart.
Who are the rising stars among the top-cited scholars in NeurIPS? The 100 percent stacked bar chart below shows the NeurIPS citations distribution by the top 20 scholars, year by year.
We hope you have enjoyed the analytic insights into this conference made possible by the Microsoft Academic Graph! Please visit our Microsoft Academic Graph page to learn how you can use our knowledge graph to generate your own custom analytics about an institution, a topic, an author, a publication venue, or any combination of these.
As always, we would like to hear from you either through the feedback link at the bottom right of the website, or on Twitter. You can also find our project home page with this blog on the Microsoft Research site at aka.ms/msracad.