How Google’s PageRank predicts Nobel Prize winners

pagerankings-graph

Ranking scientists by their citations–the number of times they are mentioned in other scientists’ papers– is a miserable business. Everybody can point to ways in which this system is flawed:

  • not all citations are equal. The importance of the citing paper is a significant factor
  • scientists in different fields of study use citations in different ways. An average paper in the life sciences is cited about six times, three times in physics, and about once in mathematics.
  • ground-breaking papers may be cited less often because a field is necessarily smaller in its early days.
  • important papers often stop being cited when they are incorporated into textbooks

The pattern of citations between papers forms a complex network, not unlike the one the internet forms. Might that be a clue that point us towards a better way of assessing the merits of the papers that it consists of?

Sergei Maslov from Brookhaven National Laboratory in New York state and Sidney Redner at Boston University have asked themselves just that question and suggest that Google’s PageRank algorithm might throw some light on the matter.

In essence, PageRank counts the number of citations a paper receives (or the number of links that point to a webpage). The more a paper receives, the higher it is ranked. But a citing is also weighted according to the ranking of the citing paper. So citations from important papers make another paper more important.

Maslov and Redner have applied the algorithm to 353,268 articles published by the American Physical Society since 1893 in journals such as Physical Review Letters . And the results are a breath of fresh air.

The top 10 papers by Google Pageranking are:

  1. Unitary Symmetry & Leptonic Decays by Cabibbo
  2. Theory of Superconductivity by Bardeen, Cooper & Schrieffer
  3. Self-Consistent Equations . . . by Kohn & Sham
  4. Inhomogeneous Electron Gas by Hohenberg & Kohn
  5. A Model of Leptons by Weinberg
  6. Crystal Statistics . . . by Onsager
  7. Theory of the Fermi Interaction by Feynman & Gell-Mann
  8. Absence of Diffusion in . . . by Anderson
  9. The Theory of Complex Spectra by Slater
  10. Scaling Theory of Localization by Abrahams, Anderson, et al.

That’s an impressive list, not least because most of these authors are Nobel Prize winners. (Curiously the author of the top paper, Nicola Cabibbo, is not. That ought to be of interest to the Nobel committee who awarded Makoto Kobayashi and Toshihide Maskawa the 2008 Nobel Prize for physics for work that was heavily based on Cabibbo’s ideas.)

All of which suggests an idea. Mining the later entries in this list might be an good way of predicting future prize winners. So get your bets in before the bookies get wind of it.

Redner and Maslov conclude: “Google’s PageRank algorithm and its modifications hold great promise for quantifying the impact of scientific publications.”

Can’t argue with that.

Ref: arxiv.org/abs/0901.2640: Promise and Pitfalls of Extending Google’s PageRank Algorithm to Citation Networks

32 Responses to “How Google’s PageRank predicts Nobel Prize winners”

  1. [...] (2009). How Google’s PageRank predicts Nobel Prize winners arxivblog.com, the physics arXiv blog [...]

  2. cosine says:

    But the real comparison question is: if one takes to top ten cited papers, how many of them are Nobel prizes.

  3. [...] analyze citation graphs in academic publishing. I’ll refrain from summarizing the paper, as a related post on the arXiv physics blog has already done a great job on that. But the upshot is that [...]

  4. anon says:

    And did it only use citations from prior to the Nobels being awarded, i.e. which came first the top 10 PageRank or the Nobel?

  5. andri says:

    pointless, because pagerank itself was motivated by researches related to the method of determining the degree of importance of papers.

  6. Reow says:

    I agree with anon and andri – scientists will be attracted to areas of research that have awarded Nobel prizes, as these are seen as topics of significance and fertile regions for further discoveries. Consequently, more researchers will cite papers by Nobel authors – so there is cause and effect, it’s just that Maslov and Redner have mistaken one for the other.

  7. [...] from the list: number 1 ranked Nicola Cabibbo. Perhaps this year’s his year. (Hat tip: the physics arXiv blog) Bookmark and Share These icons link to social bookmarking sites where readers can share and [...]

  8. [...] the physics arXiv blog » Blog Archive » How Google’s PageRank predicts Nobel Prize winners (tags: google web2.0) [...]

  9. no says:

    Unlike Reow, I believe scientists who focus on areas of research that have been awarded Nobel prizes, may ruin their careers. Those areas were perhaps topics of significance and fertile regions for further discoveries, but that was a decade or decades ago and probably have lost its prime significance.

  10. kyoro says:

    The big question is How is possible that Cabibbo is not a nobel prize!

  11. Alessio says:

    I think this issue is very important, because this measures are used to decide the career of researchers and the funding of institutions.
    Receiving a nobel is for sure a measure of quality for the research produced, but it’s a bit too sparse to indicate if a ranking method is effective.
    I would like to see this page ranking method used as a predictor of future production. For example a cumulative page rank for an author computed at a certain point in time (with all the citations until that time) could be correlated with the future page rank, or the future number of citations, or the future number of papers produced. And then I would like to see this compared with the other quality measures, such as citation index or impact factor.

  12. [...] KenntuckyFC writes: The top 10 papers by Google Pageranking are: [...]

  13. [...] the American Physical Society since 1893 (abstract). The results make interesting reading because almost all of the top ten papers resulted in (or were linked to) Nobel Prizes for their authors. Which means that studying the up-and-coming entries on the list ought to be a good way of [...]

  14. [...] Cómo el PageRank de Google puede predecir a los Premios Nobelarxivblog.com/?p=1123 por mezvan hace pocos segundos [...]

  15. [...] idea detrás de un reciente artículo en un Journal of Physics de utilizarlo para saber de antemano quienes serán los ganadores del Premio Nobel de Física es por lo menos intrigante. Recordemos que el algoritmo es el que se basa el rank establece el [...]

  16. [...] How Google’s PageRank predicts Nobel Prize winners "In essence, PageRank counts the number of citations a paper receives (or the number of links that point to a webpage). The more a paper receives, the higher it is ranked. But a citing is also weighted according to the ranking of the citing paper. So citations from important papers make another paper more important." (Tags: via:mento.info citation research pagerank science polkarobot) [...]

  17. HabitTrail says:

    Why do you credit PageRank and Google?

    The weighting of citation authorities using exactly the same “circular” mathematics as PageRank was pioneered by the Library Science community in the 1970s.

    All that Page and Brin did was apply that longstanding mathematical work on citation weighting to the link structure of the web. In no way did they invent the mathematical approach.

    Here, you’re actually applying the Library science to standard citations, so it’s bizarre to credit the Google founders.

    Please do your homework, and give credit where credit is due.

  18. HabitTrail says:

    My goodness, you wrote a peer reviewed article that made *publication* in the Journal of Neuroscience, and no one realized these algorithms originated in Library Science?

  19. [...] How Google’s PageRank predicts Nobel Prize winners, the physics arXiv blog [...]

  20. [...] 原文:How Google’s PageRank predicts Nobel Prize winners [...]

  21. [...] via the physics arXiv blog » Blog Archive » How Google’s PageRank predicts Nobel Prize winners. [...]

  22. [...] the physics arXiv blog » How Google’s PageRank predicts Nobel Prize winners [...]

  23. [...] Google’s PageRanking of scientific papers predicts Nobel Prize winners. [...]

  24. Joe nahhas says:

    E=mc²/2
    2009 is the end of Einstein’s space-jail of time and Fraud symbol E=mc²
    Joenahhas1958@yahoo.com
    Time is not a structure like space to allow space-to time-back to space jumping claimed by Physicists regardless of what physicists have to say about it because Physics is a business and not necessarily science or scientific and like every business it comes with fraud and fraud is Einstein’s space-time (x, y, z, it) continuum that led to fraud symbol E=mc² and yes I am saying that 109 years of Nobel prize winners physics and physicists are all wrong and space-time physics is based on scientific fraud. When “results” expected and “No” discovery, Physicists rigged Physics for grant money since the start of the industrial revolution. Physics today is at least 51 % fraud!
    r ——————>>Exp (ì w t) ———->> S=r Exp (ì wt) Nahhas’ Equation
    Orbit——–>> Orbit light sensing——>> Visual Orbit; Exp = Exponential
    Particle —->> light sensing of moving objects———— >> Wave
    Newton———>>light sensing———->> Quantum
    Quantum = Newton x Visual Effects
    Quantum – Newton = Relativistic = Optical Illusions
    E (Energy by definition) = mv²/2 = mc²/2; if v = c
    m = mass; v= speed; c= light speed; w= angular velocity; t= time
    S = r Exp (ì w t) = r [cos (wt) + ì sin (wt)] Visual effects
    P = visual velocity = change of visual location
    P = d S/d t = v Exp (ì w t) + ì w r Exp (ì w t)
    = (v + ì w r) Exp (ì w t) = v (1 + ì) Exp (ì w t) = visual speed; v = wr
    E (visual energy= what you see in lab) = m p²/2; replace v by p in E = mv²/2
    = m p²/2 = m v²/2 (1 + ì) ² Exp (2ì wt)
    = mv²/2 (2ì) [cosine (2wt) + ì sine (2wt)]
    =ì mv² [1 - 2 sine² (wt) + 2 ì sine (wt) cosine (wt)];v = speed; c = light speed
    wt = π/2
    E (visual) = ìmv² (1 – 2 + 0)
    E (visual) = -ì mc² ≡ mc² (absolute value;-ì = negative complex unit) If v = c
    w t = π/4
    E (visual) = imv² [1-1 +ỉ] =-mc²; v = c
    wt =-π/4+ỉln2/2; 2ỉ wt=-ỉπ/2 – ln2
    Exp (2i wt) = Exp [-ỉπ/2] Exp [ln(1/2)]=[-ỉ (1/2)]
    E (visual) = imv² (-ỉ/2) =1/2mc² v = c
    Conclusion: E = mc² is the visual Illusion of E = mc²/2 joenahhas1958@yahoo.com. All rights reserved.
    PS: In case of E=mc² claims to be rest energy claims then
    E=1/2m (m v + m’ r) ² = (1/2m) (m’ r) ²; v = 0
    E = (1/2m) (mc) ²; m’ r =mc
    E=mc²/2

  25. [...] Le PageRank Google prédit les prix Nobel [...]

  26. Simon Nazrul says:

    I think this issue is very important, because this measures are used to decide the career of researchers and the funding of institutions.

  27. [...] por la American Physical Society desde 1893 (resumen). Los resultados son interesantes porque casi todos los documentos en el top ten resultaron (o fueron ligados a) Premios Nobel. Lo cual significa que estudiando esta lista sería una buena forma de predecir a los futuros [...]