Home   LARC   News

Citation Count

One way that administrators measure the impact of researchers' publications is to count the number of places that they have been cited. Here is a much more thorough analysis of my citation count than any administrator is likely to do. They are probably happy looking at my h-index [Hirsch2005]. But the problem is, whose h-index do they use? Google Scholar appears to have the most thorough list of citations to my publications here. It says that my h-index is 23 (the last time I looked).

0. By Publication

I have approximately 100 cited publications. The average is about 23 cites per publication. The top 50 publications have an average of about 42 cites each. About half of my cited publications have 10 or more cites. Google Scholar calls the number of papers with at least 10 cites my i10-index, which is currently 52. About a quarter of my papers have 25 or more cites, about an eighth of them have 50 or more cites, and about a sixteenth of them have 100 or more cites. Fig 0.1 gives the number of cites for my 50 most-cited publications. The bars are colored by decade of publication, and the year of publication is given below each one.


Figure 0.1: Number of cites per publication.

1. By Citation Date

Fig. 1.1 shows citation count by year of citation. The bars are colored by decade of citation this time, and the year of citation is given below each one. The number inside each bar is the number of cites received in that year.


Figure 1.1: Number of cites received each year.

Fig. 1.2 shows cumulative citation count by year of citation. It looks like there are three distinct periods, 1986-89, 1990-2005, and 2006-present. The first period is easy to explain. In 1986-1989 I was a young Assistant Professor getting my research program up and running. My first ever publication was my PhD thesis in 1984, followed by my first conference paper in 1985 and my first journal paper in 1986. It took a few years for the first citations to accumulate. The second period was 1990-2005 which I would call my "mature research" phase in which my research in theoretical computer science peaked and started to drop as I helped pioneer the previously unknown research field area of game development. I started publishing game development papers with Tim Roden in 2003. The citations started picking up in 2006. Since then my citation count has been increasing linearly at a much higher rate.


Figure 1.2: Cumulative number of cites per year of citation, 1986-present.

2. By Publication Date

Figure 2.1 shows my total citation count by decade of publication. It makes sense that the older a publication is, the more citations it will receive, assuming that the research stays relevant.


Figure 2.1: Citation count by decade.

However, this may be because I was more productive when I was young. Figure 2.2 shows the number of cited publications per decade. It looks like my most productive period for turning out citable publications was the 1990s.


Figure 2.2: Number of cited publications by decade of publication.

Dividing the first number by the second number gives us the average number of cites per paper in each decade. Figure 2.3 below shows the average number of citations per publication for cited papers published in each decade.


Figure 2.3: Average citation count per publication by decade of publication.

3. By Subject

My top 25 most cited publications can be classified into 5 subjects, Artificial Neural Networks (ANN), Game Development (GD), Parallel Computing (PC), Design and Analysis of Algorithms (DAA), and Other (see Figure 3.1).


Figure 3.1: Citation counts for top 25 papers

Which subject has the most papers in my top 25? The pie chart in Figure 3.2 shows the distribution of papers by subject.


Figure 3.2: Distribution of top 25 papers by subject

This is good information, but perhaps my work in one subject is more citable than my work in others. Which subject has the most citations in my top 25? Figures 3.3 and 3.4 show the distribution of citations by subject, which gives us a different picture. My DAA and Other papers have fewer citations than the others.


Figure 3.3: Distribution of citations for top 25 papers by subject

Created December 1, 2011 Written in HTML 4.01 and CSS 3 using vi and Notepad++. Last updated January 5, 2017.

Valid HTML 4.01 Strict Valid CSS!