Digital Information Research Centre Seminar

This event has taken place

View all upcoming events at Kingston University.

Time: 2.00pm - 3.00pm
Venue: John Galsworthy Building 3011, Penrhyn Road campus, Penrhyn Road, Kingston upon Thames, Surrey KT1 2EE
Price: free
Speaker(s): Martin Tunnicliffe - Senior Lecturer

A sample of fifty English texts were used to evaluate three models: (i) the long known empirical "Heap's Law", (ii) a model based on Bernoulli word-selection with Zipfian word frequency, and (iii) Bernoulli selection with Zipf-Mandelbrot frequency. The accuracies of these three models optimally fitted to parsed data were compared, as were their relative abilities to predict vocabulary growth.

For further information about this event:

Contact: Bakhtiyar Ahmed
Email: Bakhtiyar.Ahmed@kingston.ac.uk