Text Analysis

On this page the graph is constructed and analyzed. Using the links_to attribute, a directional graph is created. This results in a graph containing 1.809 nodes and 11.151 edges. This reduction from 2111 philosophers is due to removal of isolated nodes, i.e., philosophers that neither refer to others or are referred to. Because of the directional nature, edges will be both ingoing and outgoing with respect to the nodes. First a scatter plot is shown, where each point represents the count of out-degree vs in-degree. This illustrates the correlation between the two sizes.

Most points are found on the first half of the x-axis, meaning that generally philosophers are referred to more than they refer to others. This is not surprising, since the most popular philosophers are referred to by many others, while they themselves may only refer to a few. The diluted occurrence of philosophers towards the right indicates a non-normal distribution. The following plot shows the density of the in-degree and out-degree. Notice the logarithmic scale on the x-axis.

Several things can be concluded from this graph. First by comparing the medians, it is observed that the in-degree is higher, reflecting the same as the scatter plot: More philosophers refer to a given philosopher than the other way around. A typical sign of heavy-tailed distributions is when the median aren’t equal to the means, which can clearly be seen here.

Before scrolling much further, we invite the reader to make a prediction. Who do you think will be the philosopher that is referred to the most? Which philosopher has the most outgoing mentions? Ask yourself the same with respect to all the 20 subfields.

Top 10 philosophers
In-degree		Out-degree
Aristotle	163	Friedrich Nietzsche	61
Immanuel Kant	161	Martin Heidegger	54
Plato	135	Jacques Derrida	52
Bertrand Russell	109	Boethius	51
Karl Marx	102	Georg Wilhelm Friedrich Hegel	51
John Stuart Mill	94	Charles Fourier	50
Ludwig Wittgenstein	82	Claude Henri de Rouvroy	49
David Hume	78	Eduard Bernstein	49
John Locke	77	Michel Foucault	49
Albert Einstein	64	Slavoj Žižek	49

Top 5 philosophers by degree across subfields
Language		Ethics		Socio-politics		Humanism
Ludwig Wittgenstein	108	Immanuel Kant	169	Karl Marx	122	Albert Einstein	64
Willard Van Orman Quine	58	Arthur Schopenhauer	60	Michel Foucault	106	Richard Dawkins	62
Richard Rorty	51	Charles Darwin	58	John Stuart Mill	105	Daniel Dennett	38
Saul Kripke	34	Cicero	44	Jacques Derrida	103	Carl Sagan	34
John McDowell	33	Martha Nussbaum	44	John Locke	83	Christopher Hitchens	33
Religion		Science		Logic		Utilitarianism
Benedictus de Spinoza	81	Rudolf Carnap	47	Bertrand Russell	129	Jeremy Bentham	66
Saadia Gaon	35	Paul Feyerabend	41	Gottlob Frege	72	Peter Singer	46
Solomon ibn Gabirol	34	Stephen Jay Gould	26	Hilary Putnam	72	Henry Sidgwick	26
Moshe ben Maimon	32	Paul Häberlin	24	Gottfried Wilhelm Leibniz	60	William Paley	19
Samuel ibn Tibbon	32	Thomas Kuhn	24	Kurt Gödel	60	Derek Parfit	17
Politics		Phenomenology		Mind		Feminism
David Hume	123	Edmund Husserl	93	Maurice Merleau-Ponty	50	Judith Butler	56
Friedrich Nietzsche	121	Adolf Reinach	13	Gilbert Ryle	32	Simone de Beauvoir	48
Martin Heidegger	116	William A. Earle	10	Donald Davidson	26	Luce Irigaray	37
Georg Wilhelm Friedrich Hegel	112	Hubert Dreyfus	9	David Chalmers	19	Hélène Cixous	33
Jean-Jacques Rousseau	96	Maurice Natanson	5	Jean Piaget	18	Bracha L. Ettinger	29
Epistemology		Metaphysics		Existentialism		Environment
William James	58	Aristotle	195	Jean-Paul Sartre	76	Aldo Leopold	5
Roger Bacon	48	Plato	149	Albert Camus	40	Anthony Weston	5
George Berkeley	35	William of Ockham	74	Søren Kierkegaard	30	Kathleen Dean Moore	2
Democritus	28	Alasdair MacIntyre	72	Emil Cioran	29	Michael P. Nelson	2
Paul Ziff	28	Boethius	72	Karl Jaspers	29	Ricardo Rozzi	2
Aesthetics		Critical theory		Political theory		Rationalism
Samuel Taylor Coleridge	33	Jürgen Habermas	56	Russell Kirk	21	Arnold Geulincx	4
Johann Gottlieb Fichte	29	Herbert Marcuse	43	Frank Meyer	14	Peter Ustinov	3
Walt Whitman	22	Jacques Lacan	35	John N. Gray	11	Yusuf Khass Hajib	3
Richard Shusterman	20	Erich Fromm	31	Seyla Benhabib	9	Anne Conway, Viscountess Conway	2
Friedrich Schleiermacher	17	Max Horkheimer	30	Andrew Feenberg	7	David Sosa	2

Enough with the tables. Let’s see some actual graphs. The following graphs were generated using the NetworkX library in Python. The nodes are sized by their degree.

To find communties we use the Louvain algorithm. This algorithm is a greedy optimization method that tries to maximize the modularity of the graph. The modularity is a measure of how well the graph is partitioned into communities. The algorithm is iterative and works by first assigning each node to its own community. Then it iterates over all nodes and tries to move it to the community that gives the largest increase in modularity.

The communties have the following number of nodes

0:	1:	2:	3:	4:	5:	6:	7:	8:	9:	10:
421	289	148	93	47	51	247	225	187	76	6

The algorithm finds 11 communities with varying sizes. We have community 10 as the smallest. It only includes 6 nodes, compared to the largest community 0 which includes 421 nodes. The modularity of the graph is 0.55 and indicates that the number of edges within communties are larger than they would be by chance.

Here we can see the communities visualized.

< Previous Page

Next Page >