Professor Jalote to Conduct ACM Survey on PhD Production in India in Computer Science

Monday, July 16, 2012

PhD production in India in computer science has been an issue of concern over the last many years. In most discussions and reports, back-of-the-envelope calculations are used to estimate this data – often stated as less than 50.  The purpose of this study is to collect reasonably reliable data on PhD production in CS in India, and then use it to identify useful trends, do some analysis, etc. This exercise is heavily influenced and motivated from the Taulbee report in US.  Professor Pankaj Jalote, Director IIIT Delhi, will lead this effort on behalf of ACM India.

The report of the survey for academic year 2011-12 has now been accepted and released by ACM. A copy of the report can be found here.  

Methodology to be Used

List of Institutes to be included in the study (LOI):

  • All IITs, IISc
  • The following NITs: Trichy, Surathkal, Warangal, MNIT Allahabad, Nagpur, Bhopal 
  • IIITs with a PhD program: Allahabad, Bangalore, Hyderabad, Delhi
  • ISI Kolkata
  • Jadhavpur University
  • JNU
  • University of Pune
  • Amrita University
  • BHU IT
  • BITS Pilani
  • IITB-Monash Academy

A study like this must identify a list of Institutions (LoI) from which it will collect data. As India has an extremely large University setup with widely variable quality it is not feasible to identify LoI on the basis of “PhD granting Institutions” vs. not, as the Taulbee report does. We must find some other acceptable method to identify the list of Institutions to include in this study, not only to have a consistent set but also to ensure that the report focuses primarily on PhDs from places which are viewed as having the ability to produce decent quality.

The following approach is being used for identify the LoI: Some institutions, by virtue of their standing and reputation, will be included in LoI. Besides these Institutions, if an Institution has 5 or more faculty members with PhD in Computer Science, and is considered as worthy of inclusion by the review board, it will be included in LoI.

The Review Board consists of HODs of CSE of original five IITs, IISc, and TIFR; heads of Microsoft Research, IBM research, TCS Labs, Infosys Labs;  and some ACM India Council members.

This time, initially some Institutions were identified as being in LoI. This list was sent to Review Board members, who then suggested various other names. From these, an initial LoI was identified. There were some other institutions whose names were suggested by some but it was not clear whether they should be included or not. For these, an online poll was conducted of members of the review board. Those institutions who were voted by a majority were then included in the LoI (see right).

Process of Information Collection

Institutes in LoI will be sent a form to fill. They will be followed up. If they do not provide information within a period of time, then the information publicly available from their website will be used, and a filled form will be sent to the Institute, and they will be given one week for verification.

Any comments about this process – please send mail to Prof. Jalote.

List of Respectable Research Conferences in India

List of conferences (in no order): 

  • COMAD,
  • ISEC,
  • HiPC,
  • VLSI Design,
  • VDAT 
  • Indian Conference on Vision, Graphics and Image Processing
  • ACM Compute
  • Indian Conference on Logic and its Applications
  • ICISS (International Conference on Information Systems Security)
  • Indocrypt

A somewhat related exercise was done to identify the list of research conferences in CS in India that can be considered of decent quality.  For this exercise also, a criteria was first established, then a mail was sent to various senior researchers to suggest conferences - suggestions were then vetted by a set of senior academicians. 

The criteria for inclusion in this list was:  A conference will be included in the list of "good CS research conferences in India" if it has been running for a few years, has a tradition of having a strong program committee and proper reviews, and which is listed in DBLP, ACM DL, IEEE DL, or Springer DL. For the list, see the box to the right. If you want to suggest including a conference in this list, please send a mail to Prof. Jalote giving information about the conference (e.g. how long it has been running, composition of program committee, accept rate, etc). For including, views of senior academicians/researchers in that area will be taken.