Collection, evaluation and selection of scientific literature : machine learning, bibliometrics and the World Wide Web

[摘要] ENGLISH ABSTRACT:We present a system that uses statistical machine learning to identify and extractbibliography information from scientific literature. Techniques for finding and gatheringuseful information from the ever growing volume of knowledge on the World Wide Web(WWW), are investigated.We use hidden Markov models both for recognition of bibliography styles and extractionof bibliographic information with an accuracy of up to 97%. The accuracy with whichwe are able to extract this information allows us to present a case study in whichwe apply methods of citation analysis to information extracted from three areas ofmachine learning. We use this information to identify core sets of papers that havemade significant contributions to the fields of hidden Markov models, neural networksand recurrent neural networks.

[发布机构] Stellenbosch University