TC8 Ranking of IS Conferences

TC8 Conference List - Rationale, usage and evolution

1. Rationale

In contrast to many other scientific disciplines, computer science and information systems considers conference publications as important, and as illustrated in [1] one can show formally that a few computer science conferences are equally important in terms of impact as the top journals in the field. In the last few years, a debate has started in the field on publication venues and on the role of conferences and journals in the field. It has been claimed that the focus on publications through conferences is a sign that the fields of Computer Science / Information Systems is not mature, and it is indicated also CS/IS should shift to have journals as the main publication channel, and that "Computer science should refocus the conference system on its primary purpose of bringing researchers together" [2]. We think that this is to a large degree happening already (e.g. in most countries, it is the best journals that are the most prestigious publication outlets, in particular when research evaluations and selections for positions are performed) although the set of focused workshops and conferences is very important, especially for PhD candidates to get feedback from the research community quickly enough for it to be useful towards their PhD thesis. It is also useful for a discipline like Information Systems, which deals with designed (and not natural) artifacts in interaction with human reality to have robust mechanism to build arenas for bringing people together to bring the research ahead. In particular for new research directions it is not sufficient to only read each other works in scientific journals and discussion and interaction are needed.
[2] lists a number of reasons for having conferences
  1. To rate publications and researchers.
  2. To disseminate new research results and ideas.
  3. To network, gossip, and recruit.
  4. To discuss controversial issues in the community.
In most fields, items 2, 3 and partly 4 are most important, although also the first is relevant as we have seen.
On the other hand, an increasing number of conferences in the Information Systems area are emerging, focusing either on general topics or on more specialized issues. The challenge therefore is not only to have a conference to provide an interesting discussion environment, but also a conference that is considered a high quality forum for exchanging ideas and meeting other researchers.

2. TC8 conference list

There are a large and increasing number of events, and it is very varying what it gives back to the participant. The list presented here is meant to in particular guide new researchers on the best events for scientific discussion. Several general lists exist, and we have partly built upon these when developing the current list. On the other hand we have limited the listed venues according to
  1. Relevance to Information Systems (as covered by the TC8 Working Groups). With this we also try to contribute to delimit the IS-field and sub-fields;
  2. Only list international events (i.e. events with international organizations and participations, thus not including some national and regional events
In ranking, we have used the following classifications:
  • A,B,C, unrated, discouraged.
Only conferences currently ranked as A or B is listed on the web. On a high level, an A conference is regarded as one of a few main conferences for a defined research community, and a PhD-student would only be expected to have one or two publications on this level. A B conference is less prestigious and competitive, but is still well regarded and generally looked upon as a good place for scientific discourse in a specific field. A B-level conference is one that you would encourage your PhD-students to submit to if it is within their field to get
  • Detailed feedback from the reviews
  • A quality mark since they will be selecting less than 40 % of the papers
  • Get detailed feedback also at the event from researchers in the same field
We also keep an overview also other relevant conferences on C, unrated (new conferences), and discouraged. Discouraged conferences are bogus conferences without a scientific review process. Separate lists of such conferences can be found other places.
There are several indicators that can be taken into consideration in the evaluation of a conference. For community-building, one possible metric is the ratio attendants/papers. An A conference would typically gather 5 times or more people than papers in the main conference. A B level conference should gather at least double the number of people than papers. Another important point here is that the conference is relatively focused as for topics. A conference accepting papers across a large number of areas will seldom be central to a specific community. Here some main conferences have the necessary depth in each sub-area to still regard (part of) the conference to be on an A level. The conference should have international paper submission and attendants.
  • A top conference should have good possibilities for interaction (e.g., long discussions, special interactive sessions, doctoral consortium etc.). This should also have a positive impact for conferences and workshop that would be less prestigious as independent events:
  • A top conference (both A and B) should have a rigorous review process that ensures fair review and paper selection and detailed review comments to all authors.
  • Impact: A top level conference typically has a 20 year + history, and the papers from the conference are also cited quite a lot. A top level conference would thus be expected to have a H-index above 50 (i.e. on the same level as the second-best journals). Tools such as 'Journal Impact on Publish or Parrish (based on google scholar)’ can be used for calculating H-index.
  • A top level conference should have a low acceptance rate (<20). A B-conference can have a higher acceptance rate (Up to around 40 % , often markek by the publisher of the conference proceedings (e.g. Springer LNCS or LNBIP mandate a consistent acceptance level at this level over years).
And A conference do not have to qualify on all of the above criteria, but should score high on most of these.

3. Current state and future evolution

The published list only provides the current rating of A and B conferences and a link to further information on the conference. The list is maintained by John Krogstie on behalf of IFIP TC8. A major overhaul of the list will be done every second year, first time in 2016. If you miss an event on the list that should be on level A or B (and this event is international and within the scope of the interests of the TC8 Working groups), please provide information on the event including:
  • Name and acronym
  • Website for the conference series
  • Statistics for # attendants, # papers and acceptance rate for the last 5 years
  • Description of publication format and review process
  • Description of the event, and how it contribute to the 4 goals of conferences listed in section 1
  • H-index

4. Reference

[1] Pham, M. C., Klamma, R., and Jarke, M. (2011). Development of computer science disciplines: a social network analysis approach. Social Netw. Analys. Mining, 1(4):321–340.
[2] Fortnow L. (2011) CACM Viewpoint: Time for computer science to grow up, Communications of the ACM, Vol. 52 No. 8, Pages 33-35