Modern Information Retrieval
Probabilistic models of information retrieval based on measuring the divergence from randomness
ACM Transactions on Information Systems (TOIS)
TREC: Experiment and Evaluation in Information Retrieval (Digital Libraries and Electronic Publishing)
Hi-index | 0.00 |
In this article we present a model of human written text based on statistical mechanics consideration. The empirical derivation of the potential energy for the parts of the text and the calculation of the thermodynamic parameters of the system, show that the "specific heat" corresponds to the semantic classification of the words in the text, separating keywords, function words and common words. This can give advantages when the model is used in text searching mechanisms.