Analysing Semantic Flow in Academic Writing

Many approaches have been proposed for providing feedback in academic writing, however, few of them are visually based. We describe a novel visualisation method for providing feedback to support formative essay assessment. The visualisation method makes use of text mining techniques to provide insight on the semantics of the topics in an essay. We propose that visualisation can be used to mitigate many of the problems associated with the subjectivity of formative essay assessment. The visualisation method involves a process of non-negative matrix factorisation (NMF), to uncover topics in an essay, followed by multidimensional scaling, to map the essay topics to a 2-dimensional representation. We evaluate our approach with a subset of the British Academic Written English corpus of 2761 assignments written by university students.

[1]  Carl D. Meyer,et al.  Matrix Analysis and Applied Linear Algebra , 2000 .

[2]  P. S. Gingrich,et al.  The writer's workbench: Computer aids for text analysis , 1982 .

[3]  H. Sebastian Seung,et al.  Algorithms for Non-negative Matrix Factorization , 2000, NIPS.

[4]  Lawrence M. Rudner.  Reducing Errors Due to the Use of Judges , 1992 .

[5]  Richard A. Harshman,et al.  Indexing by Latent Semantic Analysis , 1990, J. Am. Soc. Inf. Sci..

[6]  Chris North,et al.  Toward measuring visualization insight , 2006, IEEE Computer Graphics and Applications.

[7]  J. Leeuw.  Convergence of the majorization method for multidimensional scaling , 1988 .

[8]  Michael W. Berry,et al.  Algorithms and applications for approximate nonnegative matrix factorization , 2007, Comput. Stat. Data Anal..

[9]  Charles A. Perfetti,et al.  Using Intelligent Feedback to Improve Sourcing and Integration in Students' Essays , 2004, Int. J. Artif. Intell. Educ..

[10]  P. Groenen,et al.  Modern Multidimensional Scaling: Theory and Applications , 1999 .

[11]  Rhonda Liss,et al.  Effective Academic Writing , 2006 .

[12]  Patrick J. F. Groenen,et al.  Modern Multidimensional Scaling: Theory and Applications , 2003 .

[13]  Christopher A. Badurek,et al.  Review of Information visualization in data mining and knowledge discovery by Usama Fayyad, Georges G. Grinstein, and Andreas Wierse. Morgan Kaufmann 2002 , 2003 .

[14]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[15]  Rafael A. Calvo,et al.  Glosser: Enhanced Feedback for Student Writing Tasks , 2008, 2008 Eighth IEEE International Conference on Advanced Learning Technologies.

[16]  J. Leeuw.  Applications of Convex Analysis to Multidimensional Scaling , 2000 .

[17]  Download Book,et al.  Information Visualization in Data Mining and Knowledge Discovery , 2001 .

[18]  Peter W. Foltz,et al.  The Measurement of Textual Coherence with Latent Semantic Analysis. , 1998 .