The ISI Web of Knowledge recently published the 2009 Journal Citation Reports. Following similar studies I performed in 2007, 2008, and 2009, here is my analysis of the current status and trends for the impact factor in computer science journals.
Let me start with an overview of the method I followed. I performed my analysis on the 426 journals appearing under the ISI's computer science subject categories: artificial intelligence; cybernetics; hardware & architecture; information systems; interdisciplinary applications; software engineering; theory & methods. Note that journals may appear in many categories. In particular, there are many overlaps between the above categories and "electrical and electronic engineering" and "operations research and management science".
Don't take this report too seriously. The impact factor is notoriously prone to misuse; let's not make the situation worse. Also note that in computer science there are tens of prestigious conferences whose impact factor has been calculated to be higher than that of most journals.
Rank | Name | ISSN | IF |
---|---|---|---|
1 | ACM COMPUT SURV | 0360-0300 | 7.667 |
2 | HUM-COMPUT INTERACT | 0737-0024 | 6.190 |
3 | COMPUT INTELL | 0824-7935 | 5.378 |
4 | IEEE T EVOLUT COMPUT | 1089-778X | 4.589 |
5 | VLDB J | 1066-8888 | 4.517 |
6 | MIS QUART | 0276-7783 | 4.485 |
7 | IEEE T PATTERN ANAL | 0162-8828 | 4.378 |
8 | J AM MED INFORM ASSN | 1067-5027 | 3.974 |
9 | J CHEM INF MODEL | 1549-9596 | 3.882 |
10 | J COMPUT AID MOL DES | 0920-654X | 3.835 |
This year three (up from just one in 2008) journals in the above list are published by a for-profit publisher: the Journal of Computer-Aided Molecular Design (Springer), Computational Intelligence (Wiley), and Human-Computer Interaction (Taylor & Francis). The rest are published by the IEEE (two) and other professional organizations. Apparently, some commercial publishers are chasing quality (or just the impact factor) more effectively. In contrast, the IEEE seems to be loosing the game for the top places, having gone from five to just two journals in the top-ten list.
ACM J EMERG TECH COM
ACM T APPL PERCEPT
ACM T ARCHIT CODE OP
ACM T AUTON ADAP SYS
ACM T COMPUT-HUM INT
ACM T EMBED COMPUT S
ACM T INFORM SYST SE
ACM T INTERNET TECHN
ACM T SENSOR NETWORK
ACM T WEB
AD HOC NETW
AD HOC SENS WIREL NE
ADV ELECTR COMPUT EN
AUTOMAT SOFTW ENG
BUS INFORM SYST ENG+
CURR COMPUT-AID DRUG
ENTERP INFORM SYST
FUZZY OPTIM DECIS MA
GENET PROGRAM EVOL M
IEEE SYST J
INF TECHNOL CONTROL
INT J AD HOC UBIQ CO
INT J AP MAT COM-POL
INT J COMPUT COMMUN
INT J COMPUT INT SYS
INT J FUZZY SYST
INT J INF SECUR
INT J UNCONV COMPUT
J ALGORITHMS
J OPT COMMUN NETW
KSII T INTERNET INF
LOG METH COMPUT SCI
MOB INF SYST
MODEL IDENT CONTROL
PROBL INFORM TRANSM+
SCI PROGRAMMING-NETH
SOFTW SYST MODEL
TRAIT SIGNAL
Thomson, the publisher of ISI Journal Citation Reports, seems to have changed its policy regarding the addition of new journals in its lists. During the past three years it was adding each year 18.7 journals to the list of computer science journals. This year it doubled that number to 38. ACM builds on the past year's success of adding three journals to the list with another 10 journals. Also, continuing the tradition of the past three years, the list includes one more security oriented publication.
COMPUT SPEECH LANG
IEE P-SOFTW
J ALGORITHM
J VISUAL-JAPAN
J ZHEJIANG UNIV-SC A
WIRTSCHAFTSINF
The journals dropped from the JCR include three non-English publications, signifying a shift toward publications written in English. This is a new trend; no such publications were dropped in the past three years.
Name | ISSN | Δ IF |
---|---|---|
HUM-COMPUT INTERACT | 0737-0024 | 3.285 |
INT J NEURAL SYST | 0129-0657 | 2.087 |
COMPUT INTELL | 0824-7935 | 2.068 |
INTEGR COMPUT-AID E | 1069-2509 | 1.425 |
J STAT SOFTW | 1548-7660 | 1.287 |
COMPUT-AIDED CIV INF | 1093-9687 | 1.242 |
INFORM RETRIEVAL | 1386-4564 | 1.145 |
COMPUT METHOD BIOMEC | 1025-5842 | 0.882 |
IEEE INTELL SYST | 1541-1672 | 0.866 |
USER MODEL USER-ADAP | 0924-1868 | 0.862 |
The journals with the highest increase in their impact factor seem to be mainly application-oriented. It would be interesting to track this shift together with economic cycles. Due to changes in demand for computer science work in industry, I would expect practice to do well during economic downturns, and theory to get a boost in opulent times. Interestingly, the Journal of Statistical Software is published only electronically.
Name | ISSN | Δ IF |
---|---|---|
VLDB J | 1066-8888 | -2.283 |
ACM COMPUT SURV | 0360-0300 | -2.253 |
ACM T SOFTW ENG METH | 1049-331X | -1.929 |
INT J COMPUT VISION | 0920-5691 | -1.85 |
IEEE T PATTERN ANAL | 0162-8828 | -1.582 |
ACM T COMPUT LOG | 1529-3785 | -1.554 |
IEEE T INFORM THEORY | 0018-9448 | -1.436 |
ACM T MULTIM COMPUT | 1551-6857 | -1.295 |
IBM J RES DEV | 0018-8646 | -1.203 |
COGN SYST RES | 1389-0417 | -1.01 |
This last table is mostly a result of year-to-year corrections. Five of the above journals appeared last year as in the list of those with the highest increase in their impact factor (ACM COMPUT SURV, VLDB J, IEEE T PATTERN ANAL, INT J COMPUT VISION, and IEEE T INFORM THEORY). Their increase occurred probably through the citation of a single highly-popular article; once the article's popularity wanes their impact factor suffered a drop.
Comments Toot! TweetLast modified: Sunday, June 20, 2010 0:13 am
Unless otherwise expressly stated, all original material on this page created by Diomidis Spinellis is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.