Event Abstract

Analyses of Topics Maps derived from Neuroanatomical publications

  • 1 University of Southern California, Information Sciences Institute, USA
  • 2 Indiana University, School of Library and Information Science, USA

Using Latent Dirichlet Allocation (LDA) (Steyvers and Griffiths, 2007) we derived the topical composition of large corpora of full-text neuroanatomical articles that we extracted from PDF documents. Based on the derived topic mixture of each document, we analzed the relationships between documents in a given corpus. We applied Distrivuted Recursive Layout (DrL) (Martin et al., 2011), a graph layout algorithm, to visualize these inter-document relationships. Finally, we replicated this process focusing on different sections of an article (e.g., Abstracts, Methods and Materials) and looked for correspondences between the topics derived from the different sections.

Our poster presents the results of our analyses and describes the details of our methodology. We discus the challenges we faced and how we overcome them. Our ultimate goal is to streamline this procedure and incorporate it into our literature management application to enable other investigators to explore text collections of interest.

We conducted these analyses with articles from two journals: The Journal of Comparative Neurology and Brain Research.


This work was supported by the BioScholar (GM083871) and the NeuARt (MH079068) projects. This work was also supported in through the Biomedical Informatics Research Network (1 U24 RR025736-01, NCRR).


Steyvers, M., & Griffiths, T. (2007). Probabilistic topic models. Handbook of latent semantic analysis, 427(7), 424–440.

Martin, S., Brown, W. M., Klavans, R., & Boyack, K. W. (2011). OpenOrd: an open-source toolbox for large graph layout. In P. C. Wong, J. Park, M. C. Hao, C. Chen, K. Börner, D. L. Kao, & J. C. Roberts (Eds.), Conference on Visualization and Data Analysis. doi:10.1117/12.8714

Keywords: Topic maps, topic models, Journal of Comparative Neurology, brain research, Distributed Recursive Layout, Latent Dirichlet Allocation

Conference: Neuroinformatics 2013, Stockholm, Sweden, 27 Aug - 29 Aug, 2013.

Presentation Type: Poster

Topic: General neuroinformatics

Citation: Tallis M, Kong C, Borner K and Burns GA (2013). Analyses of Topics Maps derived from Neuroanatomical publications. Front. Neuroinform. Conference Abstract: Neuroinformatics 2013. doi: 10.3389/conf.fninf.2013.09.00056

Received: 29 Apr 2013; Published Online: 11 Jul 2013.

* Correspondence: Dr. Gully A Burns, University of Southern California, Information Sciences Institute, Marina del Rey, California, 90292, USA, gully@usc.edu

© 2007 - 2017 Frontiers Media S.A. All Rights Reserved