Thirty years of the ATCM conference

The Asian Technology Conference in Mathematics is a friendly, genial conference which has been held almost every year since 1996. All the proceedings are freely available on that website, but there is no search method. If you want to find out if anybody has presented a paper about, say, teaching linear algebra with Python, you’re stuck.

To help, I created a database of all papers and authors since 1997. The 1996 proceedings only seem to exist as a PDF file for which each page is an image; to create a databse of authors, affiliations, papers, titles and abstracts would require more copying by hand that I want. Maybe some day…

This database has been put online using the excellent open-source self-hosted system Mathesar - named, you’ll be delighted to know, for Enrico Colantoni’s brilliant character in the utterly brilliant film Galaxy Quest.

My implementation is at https://mathesar.numbersandshapes.net, and if you want to play around you can log in with the username guest and password g.u.e.s.t.. currently off-line, but will be made available again soon (I hope).

The database has two tables: one consists of all papers (well, all presentations really), which includes papers, workshops, posters and panel sessions. There is one record for each paper. The other table lists all the authors: there is a record for every author/paper combination. So if a paper has four authors, there will be four records in this table.

There are many faults with this database:

• As much of the information has been scraped from the ATCM website using the Python library BeautifulSoup, there is text which hasn’t survived the scraping.
• A single author might exist in different forms, depending on whther initials are used, or various accented letters. For example, “Nguyễn Ngọc Trường Sơn” might appear as “Ngọc Trường Sơn Nguyễn”, or without any of the accents as “Truong Son Ngoc Nguyen” or as “Truong Son N Nguyen” or as “T S N Nguyen”, and so on. This partly depends on the spelling in the proceedings. Ensuring that all authors appear with only one spelling of their name is nearly impssible - at least for me. This in turn makes it impossible to determine precisely the number of different authors and speakers.
• Some titles and abstracts include mathematics typeset with LaTeX. This generally becomes garbled by the time it makes it into the database.
• A single author might appear several times with a different affiliation, if they’ve changed jobs, or maybe were on a sabbatical. Thus “Wolfgang A. Mozart, University of Vienna”, and “Wolfgang A. Mozart, University of Salzburg”, will appear as 2 different authors.
• Then again, two different people might share the same name. Are “Li Chen, Normal Universty of Beijing”, and “Li Chen, University of Shanghai” different people or not?

However, even with those caveats, we can obtain a pretty good idea of the conference statistics since 1997.

Some numbers

There have been 2027 papers published, by about 2013 different authors. (The exact number of authors will be a bit less than this, but I’m hoping not by too much.)

The country which has provided the largest number of different authors is China; the following table shows the top ten:

The most published author has appeared 37 times. (I’ve got a measly 19). Authors have come from 75 different countries. By region:

The “Independent states” includes Russia and previous members of the USSR.