Incremental unsupervised name disambiguation in cleaned digital libraries.

Abstract

Name ambiguity in the context of bibliographic citation sisone of t h e hardest problems currently faced by the Digital Library ( DL) community. Here we deal with the problem of disambiguating new citations records insertedint o a cleaned DL, without the need t process the whole collection , which is usually necessary for un supervised methods. Although supervised solutions can deal with this situation , there is the costly burden of generating training data besides the fact that these methods cannot and le well the insertion of record s of new author not already existent in the repository. I n t h is article, we propose a new unsupervised method that identifies the correct author sof the new citation records to be inserted in a DL. The method is based on heuristics that are also used to identify whet her the new record s belong to authors already in t h e digital library or not , correctly identifying new authors in most cases. Our experiment al evaluation , using synthetic an d real data sets, shows gains of u p t o 19% when compared to a state- of- t h e- art method without the cost of having to disambiguate the whole DL at each new load ( as d on e by u n supervised methods) or the need for any train in g ( as d on e by supervised methods) .