Edited by Aeneas Momoko
The results of CSRankings 2024 are out!The ranking of computer science institutions around the world has been greatly updated. In the AI sector, Tsinghua University, Peking University, and Shanghai Jiaotong ranked TOP 3, and CMU and Zhejiang University tied for 4th. The Chinese universities in the AI TOP 10 also include Renmin University, Nanjing University, Fudan University, Harbin Institute of Technology, etc.
The 2024 CSRankings Global Computer Science Rankings have just been released!
In the global CS rankings, Chinese universities/institutions occupy a number of seats, among which CMU ranks first, Tsinghua University ranks second, Shanghai Jiaotong ranks third, and Peking University ranks fourth.
In terms of the world's top 10 rankings in the field of AI, China's performance is very impressive, with a total of 8 universities/institutions in the top 10 (including joint rankings).
One of the most noteworthy points in the 2024 ranking of world universities/institutions in the field of AI is that the Chinese Academy of Science has fallen out of the top 10 in the world (ranked 8th in 2023).
CSRankings is a global ranking of computer science power in colleges and universities organized by Emery Berger, professor in the School of Computer and Information Science at Massachusetts State University Amherst, which is based entirely on research metrics and is relatively transparent.
The rankings include computer science majors at universities around the world, and are based on the number of papers published by universities and research institutions at top academic conferences in the field of computer science.
By geography, this ranking can be found in multiple countries separately, and can also be divided into North America, South America, Africa, Asia, Australia, Europe, and the world.
CSRanking's segment rankings are divided into 4 categories (27 subdivisions), which are AI, systems, theory, and interdisciplinary domains.
在AI板块，主要有5个细分领域：1.人工智能（Artificial intelligence）；2.计算机视觉（Computer vision）；3.机器学习（Machine learning）；4.自然语言处理（Natural language processing）；5.网页信息检索（The Web & information retrieval）。
2024 CS University Global Rankings
According to the latest CS Rankings, the top 10 computer science majors in the world in 24 years are as follows-
Carnegie Mellon University ranked first, Tsinghua University ranked second, Shanghai Jiao Tong University ranked third, Peking University and the University of Illinois at Urbana-Champaign tied for fourth, Zhejiang University ranked sixth, ETH Zurich and Georgia Institute of Technology tied for seventh, Korea Advanced Institute of Science and Technology, Stanford University, University of California, San Diego, University of Michigan, and University of Washington tied for ninth, and MIT, National University of Singapore, UC Berkeley, and University of Maryland, College Park tied for 14th.
Selecting the "AI" option, it can be seen that in the 2024 AI global ranking, many domestic universities will enter the top 20.
Tsinghua University ranked first, Peking University ranked 2nd, Shanghai Jiaotong University ranked 3rd, Carnegie Mellon University and Zhejiang University tied for 4th, Korea Advanced Institute of Science and Technology ranked 6th, Chinese Renmin University ranked 7th, Nanjing University ranked 8th, National University of Singapore ranked 9th, Fudan University and Harbin Institute of Technology tied for 10th.
Compared with the 2023 TOP 10 ranking of the world's AI institutions, the basic ranking remains unchanged.
Most notably, the Chinese Academy of Sciences (CAS) has fallen out of the top 10 in the world in the 2024 rankings (it ranked 8th in the world last year). In addition, UIUC has also been squeezed out of the top 10 in the world, and the National University of Singapore and Fudan University have been successfully promoted.
Specifically, Tsinghua University has published papers in the fields of machine Xi and NLP, far exceeding the number of papers in the fields of AI, computer vision, and web information retrieval.
Among them, the faculty members who have published more than 10 papers are: Huang Minlie (16), Sun Maosong (15), Zhu Jun (15), Li Yong (11), Liu Zhiyuan (11), Zhu Wenwu (11), and Liu Yang (10).
Peking University has published more than 40 papers in the fields Xi of AI, computer vision, and machine science.
Among them, the faculty members with more than 10 papers are Zhao Dongyan (12), Wang Yisen (11), and Shi Baixin (10).
Shanghai Jiao Tong University has the largest number of Xi papers in AI and machine learning.
The faculty members who have published more than 10 papers in the summit are: Yan Junchi (29), Zhao Hai (12), Wang Xinbing (11), and Yang Xiaokang (11).
It is worth mentioning that the number of top papers in the field of machine Xi at CMU has exceeded 100. In addition to the AI section, there are nearly 50 papers in the field of robotics.
Graham Neubig(16)、J. Zico Kolter (14)、Deva Ramanan(11)论文数量超过了10篇。
Zhejiang University, which is tied for fourth, has published more than 50 papers in AI and computer vision.
The faculty members who have published more than 10 papers are Zhao Zhou (20), Chen Huajun (14), and Yang Yi (14).
The world's top 10 universities in major AI subdivisions
The top 10 universities in the world are Shanghai Jiao Tong University, Zhejiang University, Nanjing University, Peking University, Chinese National University, Tsinghua University, Harbin Institute of Technology, National University of Singapore, Wuhan University and Korea Advanced Institute of Science and Technology.
In the AI category, the top 10 universities in the world are Zhejiang University, Peking University, Korea Advanced Institute of Science and Technology, Carnegie Mellon University, Mohammed Bin Zayed University of Artificial Intelligence, Shanghai Jiao Tong University, ETH Zurich, Stanford University, Chinese University of Hong Kong, Nanyang Technological University.
The top 10 universities in the world are Carnegie Mellon University, Stanford University, Korea Advanced Institute Xi of Science and Technology, MIT, University of California, San Diego, University of California, Berkeley, University of California, Los Angeles, Tsinghua University, Peking University, and Shanghai Jiao Tong University.
The top 10 universities in the world are Harbin Institute of Technology, Carnegie Mellon University, Fudan University, University of Washington, Stanford University, Tsinghua University, Peking University, University of Illinois at Urbana-Champaign, ETH Zurich, and University of California, Los Angeles.
Select "Web Information Retrieval" in the AI category, and the top ten universities in the world are Tsinghua University, Chinese University of People's University, University of Science and Technology of China, Zhejiang University, University of Amsterdam, Fudan University, University of Illinois at Urbana-Champaign, Peking University, University of Chinese Academy of Sciences, and Beijing University of Posts and Telecommunications.
China's AI TOP10: Tsinghua University dominates the list, and the Chinese Academy of Sciences falls to 9th
In the field of AI, the top 10 institutions in China are:
Tsinghua University ranked 1st, Peking University ranked 2nd, Shanghai Jiao Tong University ranked 3rd, Zhejiang University ranked 4th, Chinese Renmin University ranked 5th, Nanjing University ranked 6th, Fudan University ranked 7th, Harbin Institute of Technology ranked 8th, Chinese Academy of Sciences ranked 9th, University of Electronic Science and Technology of China ranked 10th.
Compared to last year, the top 10 institutions as a whole remained unchanged. Fudan University has improved (ranked 9th last year) and the Chinese Academy of Sciences has dropped significantly (6th last year).
Ranking basis: the number of papers published in the top conferences
Currently, the U.S. News and World Report, which began in 1983, is ranked among the most prestigious.
However, the U.S. News rankings are based entirely on reputation and rely on surveys of department heads and graduate school directors.
Take the 2023 US News world rankings out of the general spectrum, and make the ranking of 213 institutions wrong, and the study abroad circle is also noisy.
Including the previous ranking data falsification incident of Columbia University in US News, it also set off a bloody storm.
In addition, the index based on the number of citations is also suspected of irrigation. For example, some universities encourage faculty and staff to quote each other, and the "Citation Cartels" is a shining name.
Not only that, but not all citations are free, and they change quickly, like the citation statistics system in Google Scholar that doesn't do a very good job of author ambiguity.
In order to provide a meaningful and transparent ranking system for all, Emery Berger's Global Rankings of Computer Science are based solely on "research indicators".
Specifically, CSRankings is measured by the number of papers published by the vast majority of faculty members in the field of computer science.
The top conferences in the field of natural language processing include ACL, EMNLP, NAACL, the top conferences in the field of computer vision include CVPR, ECCV, ICCV, the papers in the machine Xi and data mining conference are from ICML, KDD, NIPS, and the top conference on artificial intelligence include AAAI and IJCAI.
It can be seen that this method is designed to motivate faculty and staff to publish papers at the summit meeting, and all CSrankings code and data are shared, while also preventing fraud.
What are the criteria for faculty and staff selection?
The criteria for inclusion in the database are that any full-time, tenure-track faculty member on a particular campus who can independently provide supervision to doctoral students in computer science can be included in the database.
As a result, this approach extends the coverage of the database to some faculty members in other faculties who have part-time contracts with computer science departments or similar faculties that can provide guidance to doctoral students at CS.
Please note that full-time means working at least 75% of the time throughout the school year.
How to calculate the author's signature and the number of papers?
A faculty member can receive a score of 1/N for a paper, where N is the number of authors, regardless of their affiliation or status (faculty, student, or other status). That number will never change.
In cases where all authors are/eventually become faculty members in the database, then a paper can only be counted as a maximum of 1.0 points.
If you don't divide the academic output of a paper by the number of authors, and simply count the number of papers, then the author can easily manipulate and exaggerate the output of a single paper by increasing the number of authors.
In order to avoid this, it is necessary to divide the academic output of the paper. This can motivate authors to treat the attribution of academic output appropriately and not to abuse multiple authors to exaggerate the output of a single paper.
Resources:
https://csrankings.org/#/index?all&us