Text 2 Multilevel MindMap

Brain Halfs according to Tony Buzan's claims (There are doubts about it)

MindMapping is powerful pictorial technique for representing knowledge, concepts and ideas so it became a well-known technique used in note taking and is known to encourage learning and studying. With today's vast and rapid increase in the use of electronic document readers, smart phone, tablets, there is an eminent need to develop tools for visualizing and summarizing textual contents. MindMapping is not only a note-taking tool, but also a very powerful tool for text summarization and visualization. Converting a text paragraph to a MindMap would provide an easier way to visually access the knowledge and ideas in the text. However, many people do not use MindMaps because making MindMaps needs huge extra mental efforts and full concentration for a very long time specially if someone wants to make MindMaps for a long script or a book. Besides, only few people are creative enough to draw good MindMaps. Unfortunately, there is no reliable automated system that can generate MindMaps from Natural Language text. This work fills in this gap by developing the first evaluated automated system that takes a text input and generates a MindMap visualization out of it. The system also could visualize large text documents in multilevel MindMaps in which a high level MindMap node could be expanded into child MindMaps. The proposed approach involves understanding of the input text converting it into intermediate Detailed Meaning Representation (DMR). The DMR is then visualized with two proposed approaches; Single level or Multiple levels which is convenient for larger text. The generated MindMaps from both approaches were evaluated based on Human Subject experiments performed on Amazon Mechanical Turk with various parameter settings.


                                    System Architecture

Single Level Mind Map from Pure Text  


MultiLevel MindMap from Pure Text  


The dataset created for the evaluation could be downloaded through here If you used these dataset please cite this paper.


+  Mohamed Elhoseiny, Ahmed Elgammal, Radwa Elsahn, “MultilevelMindMap Generation”, US PatentPending, 2013

(Oral Presentation) Mohamed Elhoseiny,Ahmed Elgammal, "English2MindMap: Automated system for Mind Map generationfrom Text”, International Symposium on Multimedia (ISM), 2012          


+  Asmaa Hamdy, Mohamed H.Elhoseiny, Radwa Elsahn, Eslam Kamal, "Mind Map Automation (MMA)System". SWWS, Las Vegas, Nevada, USA, 2009

Citation : Please cite the "English2MindMap: Automated system for Mind Map generation from Text", and/or" Mind Map Automation (MMA) System", 2009


Code could be find here

Presentation in 3rd NYC Multimedia and Computer Vision Meeting, June 2013: http://gnymv.org/#pd
          + Poster https://sites.google.com/site/mhelhoseiny/hmmapfinalposter.pdf 
          + Flier  https://sites.google.com/site/mhelhoseiny/hmmapfiler.pdf


1) Supplementary Materials (Many inputs and outputs in a pdf): https://sites.google.com/site/mhelhoseiny/supp.pdf

2) Ontology in OWL format (1.owl, 1.pprj )
It was created by Protege stanfford Ontology Editor.
It could be downloaded at http://protege.stanford.edu/
3) The dataset created for the evaluation could be downloaded through here

5) System Outputs
The outputs are compressed in two-part compressed file


Mind Map Automation (MMA) System (2009), It was just a demo version (prototype) that did not support hierarchy

YouTube Video

Video (An old Version), please open the first item in download to show see many single and hierarchical examples

YouTube Video

This video is created thanks to previous Demo by Mai Samy , Mona Othman.

Mohammad Elhoseiny,
Apr 22, 2014, 5:01 PM
Mohammad Elhoseiny,
Apr 22, 2014, 5:01 PM