The hybrid clustering approach combining lexical and link-based similarities suffered for a long time from the different properties of the underlying networks. We propose a method based on noun phrase extraction using natural language processing to improve the measurement of the lexical component. Term shingles of different length are created form each of the extracted noun phrases. Hybrid networks are built based on weighted combination of the two types of similarities with seven different weights. We conclude that removing all single term shingles provides the best results at the level of computational feasibility, comparability with bibliographic coupling and also in a community detection application.
|Number of pages||6|
|Journal||CEUR Workshop Proceedings|
|Publication status||Published - Jan 1 2015|
|Event||1st Workshop on Mining Scientific Papers: Computational Linguistics and Bibliometrics, CLBib 2015 - co-located with 15th International Society of Scientometrics and Informetrics Conference, ISSI 2015 - Istanbul, Turkey|
Duration: Jun 29 2015 → …
ASJC Scopus subject areas
- Computer Science(all)