Document analysis/ retrieval system plays crucial role to strengthen any information retrieval system. There are various processing stages associated with a document analysis system, such as feature extraction stage, semantic representation stage, dimensionality reduction stage and similarity measure stage. Researchers are contributing well in every stage to improve the performance of the document analysis system. This short paper considers word affinity graph/ matrix for further improvement so that semantic representation can be given more precisely. This is accomplished by incorporating weight component in the word affinity matrix to provide significance for degree of distribution. Theoretical study on both word affinity matrix and weighted word affinity matrix shows the significance offering by them on widely distributed document terms.

Document analysis/ retrieval system plays crucial role to strengthen any information retrieval system. There are various processing stages associated with a document analysis system, such as feature extraction stage, semantic representation stage, dimensionality reduction stage and similarity measure stage. Researchers are contributing well in ever

[1] Song Mao, Azriel Rosenfeld, Tapas Kanungo, �Document structure analysis algorithms: a literature survey�, DRR 2003, 2003, p.p. 197-207
[2] Carsten Gorg, Zhicheng Liu, Jaeyeon Kihm, Jaegul Choo, Haesun Park, Member, and John Stasko, �Combining Computational Analyses and Interactive Visualization for Document Exploration and Sensemaking in Jigsaw�, IEEE Transactions on Visualization and Computer Graphics, Vol. 19, No. 10, 2013, p.p. 1646 � 1663
[3] Jinxi Xu Amherst, W. Bruce Croft, �Query expansion using local and global document analysis�, Proceedings of the 19th annual international ACM SIGIR conference on Research and development in information retrieval, 1996, p.p. 4-11
[4] G. Salton, M. McGill, Eds. �Introduction to Modern Information Retrieval�, New York: McGraw-Hill, 1983.
[5] S. Deerwester and S. Dumais, �Indexing by latent semantic analysis,� J. Amer. Soc. Inf. Sci., vol. 41, no. 6, 1990, pp. 391�407.
[6] Haijun Zhang, John K. L. Ho, Q. M. Jonathan Wu, Senior Member, IEEE, and Yunming Ye, �Multidimensional Latent Semantic Analysis Using Term Spatial Information�, IEEE Transactions on Cybernetics, Vol. 43, No. 6, 2013, p.p. 1625- 1640
[7] W. B. Frakes and R. Baeza-Yates, �Information Retrieval: Data Structures and Algorithms�, Prentice-Hall, Englewood Cliffs, NJ, 1992.
[8] Antoniol, G. ; Canfora, G. ; Casazza, G. ; De Lucia, A; �Information retrieval models for recovering traceability links between code and documentation�, Proceedings of International Conference on Software Maintenance, 2000, p.p. 40-49
[9] Yu-Gang Jiang ; Yang, J. ; Chong-Wah Ngo ; Hauptmann, A.G.; �Representations of Keypoint-Based Semantic Concept Detection: A Comprehensive Study�, IEEE Transactions on Multimedia, Vol. 12, No. 1, Jan. 2010, p.p. 42 � 53.
[10] Eaddy, M. ; Antoniol, G. ; Gueheneuc, Y.-G., �CERBERUS: Tracing Requirements to Source Code Using Information Retrieval, Dynamic Analysis, and Program Analysis�, 16th IEEE International Conference on Program Comprehension (ICPC 2008), 10-13 June 2008, p.p. 53 - 62
[11] G. Antoniol, G. Canfora, G. Casazza, A. De Lucia, E. Merlo, "Recovering Traceability Links between Code and
Documentation," IEEE Transactions on Software Engineering, Vol .28, No. 10, 2002, p.p.970�983
[12] D. Poshyvanyk, Y.-G. Gu�h�neuc, A. Marcus, G. Antoniol,
V. Rajlich, "Feature Location Using Probabilistic Ranking of Methods Based on Execution Scenarios and Information Retrieval," IEEE Transactions on Software Engineering, Vol. 33, No. 6, 2007, p.p.420�432.
[13] Akiko Aizawa, �An information-theoretic perspective of tf�idf measures�, Information Processing and Management, Vol. 39, 2003, p.p. 45�65
[14] Wray Buntine and Aleks Jakulin, �Applying discrete PCA in data analysis�, Proceedings of the 20th conference on Uncertainty in artificial intelligence, 2004, p.p. 59-66