An Approach to Similarity Search for Mathematical Expressions using MathML

Yokoi, Keisuke; Aizawa, Akiko

  • Towards a Digital Mathematics Library. Grand Bend, Ontario, Canada, July 8-9th, 2009, Publisher: Masaryk University Press(Brno), page 27-35

Abstract

top
The recent global computerization and digitization trend has helped to increase the numbers of documents with mathematical expressions on the Web. These mathematical expressions have their own unique structures, and therefore, it is not an easy task for traditional search systems targeting natural languages to deal with them. We propose a similarity search method for mathematical equations that is particularly adapted to the tree structures expressed by MathML based on this background. The similarity search system helps users acquire additional knowledge, discover concealed relationships to different fields, and compensate for some false recognition. Given an equation as a query, most of the conventional mathematical search systems return corresponding equations that exactly match the query. Contrarily, our proposed system makes it possible to return similar equations by measuring the similarity using tree-matching techniques and also by reforming the structure of Content-based MathML. In this paper, we examine our proposed techniques through preliminary experimentation using a prototype search system, and show this techniques’ effectiveness based on some conditions requested by the user.

How to cite

top

Yokoi, Keisuke, and Aizawa, Akiko. "An Approach to Similarity Search for Mathematical Expressions using MathML." Towards a Digital Mathematics Library. Grand Bend, Ontario, Canada, July 8-9th, 2009. Brno: Masaryk University Press, 2009. 27-35. <http://eudml.org/doc/221460>.

@inProceedings{Yokoi2009,
abstract = {The recent global computerization and digitization trend has helped to increase the numbers of documents with mathematical expressions on the Web. These mathematical expressions have their own unique structures, and therefore, it is not an easy task for traditional search systems targeting natural languages to deal with them. We propose a similarity search method for mathematical equations that is particularly adapted to the tree structures expressed by MathML based on this background. The similarity search system helps users acquire additional knowledge, discover concealed relationships to different fields, and compensate for some false recognition. Given an equation as a query, most of the conventional mathematical search systems return corresponding equations that exactly match the query. Contrarily, our proposed system makes it possible to return similar equations by measuring the similarity using tree-matching techniques and also by reforming the structure of Content-based MathML. In this paper, we examine our proposed techniques through preliminary experimentation using a prototype search system, and show this techniques’ effectiveness based on some conditions requested by the user.},
author = {Yokoi, Keisuke, Aizawa, Akiko},
booktitle = {Towards a Digital Mathematics Library. Grand Bend, Ontario, Canada, July 8-9th, 2009},
keywords = {MathML; MathFind},
location = {Brno},
pages = {27-35},
publisher = {Masaryk University Press},
title = {An Approach to Similarity Search for Mathematical Expressions using MathML},
url = {http://eudml.org/doc/221460},
year = {2009},
}

TY - CLSWK
AU - Yokoi, Keisuke
AU - Aizawa, Akiko
TI - An Approach to Similarity Search for Mathematical Expressions using MathML
T2 - Towards a Digital Mathematics Library. Grand Bend, Ontario, Canada, July 8-9th, 2009
PY - 2009
CY - Brno
PB - Masaryk University Press
SP - 27
EP - 35
AB - The recent global computerization and digitization trend has helped to increase the numbers of documents with mathematical expressions on the Web. These mathematical expressions have their own unique structures, and therefore, it is not an easy task for traditional search systems targeting natural languages to deal with them. We propose a similarity search method for mathematical equations that is particularly adapted to the tree structures expressed by MathML based on this background. The similarity search system helps users acquire additional knowledge, discover concealed relationships to different fields, and compensate for some false recognition. Given an equation as a query, most of the conventional mathematical search systems return corresponding equations that exactly match the query. Contrarily, our proposed system makes it possible to return similar equations by measuring the similarity using tree-matching techniques and also by reforming the structure of Content-based MathML. In this paper, we examine our proposed techniques through preliminary experimentation using a prototype search system, and show this techniques’ effectiveness based on some conditions requested by the user.
KW - MathML; MathFind
UR - http://eudml.org/doc/221460
ER -

References

top
  1. Munavalli, Rajesh, Miner, Robert, MathFind: A Math-Aware Search Engine, . SIGIR. pp. 735–735, 2006. (2006) 
  2. Youssef, Abdou, Information Search And Retrieval of Mathematical Contents: Issues And Methods, . the ISCA 14th Int’l Conf. on Intelligent and Adaptive Systems and Software Engineering (IASSE-2005), July 20–22, Toronto, Canada, 2005. (2005) 
  3. Adeel, Muhammad, Cheung, Hui Siu, Khiyal, Sikandar Hayat, Math GO! Prototype of A Content Based Mathematical Formula Search Engine, . Journal of Theoretical and Applied Information Technology, Vol4, No10, pp. 1002–1012, 2008. (2008) 
  4. Kohlhase, Michael, Sucan, Ioan A., A Search Engine for Mathematical Formulae, . Proceedings of Artificial Intelligence and Symbolic Computation, AISC’2006, Springer Verlag, pp. 241–253, 2006. (2006) Zbl1156.68306
  5. Asperi, Andrea, Guidi, Ferruccio, Coen, Claudio Sacerdoti, Tassi, Enrico, Zacchiroli, Stefano, A Content Based Mathematical Search Engine: Whelp Proceedings of TYPES 2004 conference: Types for Proofs and Programs, , LNCS 3839, Springer Berlin / Heidelberg, ISBN 3-540-31428-8, pp. 17–32, 2006. (2006) 
  6. Ichikawa, Hiroshi, Hashimoto, Taiichi, Tokunaga, Takenobu, Tanaka, Hozumi, New methods of retrieve sentences based on syntactic similarity, . IPSJ SIG Technical Reports, DBS-136, FI-79, pp. 39–46, 2005. (2005) 
  7. Hashimoto, Hideki, Hijikata, Yoshinori, Nishida, Shogo, A Survey of index formats for the search of MathML objects, . IPSJ SIG Technical Reports, DBS-142, FI-87, pp. 55–59, 2007. (2007) 
  8. Hijikata, Yoshinori, Hashimoto, Hideki, Nishida, Shogo, An Investigation of Index Formats for the Search of MathML Objects, . Proc. of Intelligent Web Interaction Workshop (IWI 2007), pp. 244–248, DOI 10.1109/WI-IATW.2007. 121, Silicon Valley, USA, November, 2007. (2007) 
  9. Otagiri, Kenichi, Murata, Tsuyoshi, Search of Mathematical Formulas using MathML, . The 22nd Annual Conference of the Japanese Society for Artificial Intelligence, 1F1-3, 2008. (2008) 
  10. Wolfram Research Inc., The Wolfram Functions Site, http://functions.wolfram.com. 
  11. World Wide Web Consortium, Mathematical Markup Language (MathML) Version 2.0 (Second Edition), . http://www.w3.org/TR/MathML2/. 

NotesEmbed ?

top

You must be logged in to post comments.

To embed these notes on your page include the following JavaScript code on your page where you want the notes to appear.

Only the controls for the widget will be shown in your chosen language. Notes will be shown in their authored language.

Tells the widget how many notes to show per page. You can cycle through additional notes using the next and previous controls.

    
                

Note: Best practice suggests putting the JavaScript code just before the closing </body> tag.