Hello I am student working on NLP project. The top ranked words under a given topic number which is obtained from LDA (Latent Dirchlet allocation), I am trying to assign topic names to each topic number using wikipedia as a knowledge base.
Wikipedia category graph contains links between categories which are having some relationship but do not have an hierarchical structure. From this graph I removed the non hierarchical links to get a DAG (directed acyclic graph) as a consequence a given category can have one or more parent, after this I applied BFS like algorithm to get a taxonomy but this misses relevant hierarchical links.
Is there any factor which I consider to get more accurate and meaningful taxonomy.
Thank you in advance
Please answer of dr. Mehrnoush Shamsfard
Shahid Beheshti University, Tehrān · Faculty of Electrical & Computer Engineering
Follow
You can use TAG approach
take a look in this IEEE paper
http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=6987557&url=http%3A%2F%2Fieeexplore.ieee.org%2Fiel7%2F6973089%2F6985764%2F06987557.pdf%3Farnumber%3D6987557
There is some work on taxonomy extraction from Wikipedia (identifying taxonomic links in the Wikipedia catetory graph).
You might want to check our paper
García, Renato Domínguez, et al. "Automatic taxonomy extraction in different languages using wikipedia and minimal language-specific information." Computational Linguistics and Intelligent Text Processing. Springer Berlin Heidelberg, 2012. 42-53.
see http://link.springer.com/chapter/10.1007/978-3-642-28604-9_4
or ftp://ftp.kom.tu-darmstadt.de/papers/DSRS12-1.pdf