Detecting and Normalizing Formulas in Electronic Literature Resources

Abstract:

Formulas exist in various kinds of documents with different formats. Extracting and normalizing them into a unique form are the precondition of mathematical retrieval. In this paper, an extraction and conversion method of formulas in Word documents is built for mathematical expression retrieval. Firstly, the mathematical expressions in Word documents are detected through the processing of OLE objects. Then, the matching rules of formula format conversion are defined. Finally, the extracted mathematical expressions in OMML format are converted into LaTeX format follow the defined rules and stored in a txt file. Furthermore, the formulas exist in MathType format are stored in bitmap documents and converted into LaTeX documents through formula recognition and reconstruction module. Experiments show the effectiveness of the designed approach.

• Periodical: Applied Mechanics and Materials (Volume 740)
• Pages: 875-880
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.740.875
• Online since: March 2015
• Authors: Wei Xin Xu, Xue Dong Tian*, Xin Fu Li
• Keywords: Detecting, Electronic Literature Resources, Formula, Latex, Normalizing, Word

References

[1] LaTeX-A Document Preparation System on http: /www. latex-project. org.

[2] J. Nie, T. Y. Chen and H. G. Fu. Journal of Computer Applications Vol. 30(2) (2010), pp.312-315. In Chinese.

[3] Y. P. Qin, Y. W. Tang, S. X. Lun and X. K. Wang. Computer Science Vol. 40(5) (2013), p.251, 252, 278. In Chinese.

[4] L. A. Sobreviela: A Reduce-based OpenMath↔ MathML Translator. ACM SIGSAM Bulletin Vol 34(2) (2000), pp.31-32.

DOI: 10.1145/362001.362018

[5] C. M. So and S.M. Watt: On the Conversion between Content Mathml and Openmath. International Conference on Applications of Computer Algebra. Raleigh, North Carolina. July (2003).

[6] H. Stamerjohanns, D. Ginev, C. David, et al: MathML-aware Article Conversion from LaTeX. Towards a Digital Mathematics Library. Grand Bend, Ontario, Canada, July 8-9th, 2009, (2009), pp.109-120.

[7] A dynamic LaTeX Mathematics to Mathml Converter on http: /www. Maths. nottingham. ac. uk/ personal /drw/lm. html.

[8] S. B. Sun, C. J. Hua. Computer Application Vol. 17(5) (1997), pp.54-55. In Chinese.

[9] H. F. Wang, R. Ran, Y. Y. Dai and Y. Zhang. Computer Systems Application Vol. 11 (2008), pp.76-79. In Chinese.

[10] H. J. Wang. Journal of Lanzhou Institute of Technology Vol. 20(2) (2013), pp.28-30. In Chinese.

[11] B. Zhou. Journal of Kunming Metallurgy College Vol. 29(3) (2013), pp.45-49. In Chinese.

[12] Math in Office 2010 and OMML Specification on http: /blogs. msdn. com/b/murrays.

[13] Omml2mml on http: /www. dcarlisle. demon. co. uk/omml2mml.

[14] Word2MathML on http: /www. codeplex. com/Word2mathml.

[15] L. W. Cui: The Research on Extraction of Mathematics Expressions on Web. Lanzhou University, (2012). In Chinese.

[16] F. H. Li, X. Huang. Development and Application of the Computer Vol. 20(3) (2007), pp.27-29. In Chinese.

[17] L. H. Wu: Analysis and Reconstruction of Special Structure in Printed Mathematical Expressions. Hebei University, (2009). In Chinese.