Joseph Kruskal

From Wikipedia, the free encyclopedia

Joseph Bernard Kruskal (b. 1929 in New York) is an American mathematician, statistician, and psychometrician. He was a student at the University of Chicago and at Princeton University, where he completed his Ph.D. in 1954, nominally under Albert W. Tucker and Roger Lyndon, but de facto under Paul Erdős with whom he had two very short conversations. Kruskal has worked on well-quasi-orderings and multidimensional scaling.

He is a Fellow of the American Statistical Association, former president of the Psychometric Society, and former president of the Classification Society of North America. He also initiated and was first president of the Fair Housing Council of South Orange and Maplewood in 1963, and actively supported civil rights in several other organizations.

In statistics, Kruskal's most influential work is his seminal contribution to the formulation of multidimensional scaling. In computer science, his best known work is the Kruskal's algorithm for computing the minimal spanning tree (MST) of a weighted graph. The algorithm first orders the edges by weight and then proceeds through the ordered list adding an edge to the partial MST provided that adding the new edge does not create a cycle. Minimal spanning trees have applications to the construction and pricing of communication networks.

Joseph Kruskal should not be confused with his two brothers Martin David Kruskal (born 1925; co-inventor of solitons and of surreal numbers) and William Kruskal (1919–2005; developed the Kruskal-Wallis one-way analysis of variance).