DOCUMENT SIMILARITY WITH AUTHOR USING STYLOMETRY

Computer engineering

Similarity measures have been used for document classification and clustering. To calculate similarity measure in two documents with respect to the distinctive attribute the similarity measure takes the following three occurrences: a) The distinctive attribute appears in both documents. b) The distinctive attribute appears in only one document and c) The distinctive attribute appears in none of the documents. In the first occurrence, similarity increases as the dissimilarity between the two documents associated with a present feature decreases. In the first occurrence, similarity decreases when the number of presence-absence features increases. In the second occurrence, a fixed value is contributed to the similarity. In the third occurrence, an absent feature has no contribution to the similarity. Proposed system utilizes TFIDF and Stylometry features which will increase accuracy and Identification of Author. By utilization of NB, ID3 classifiers proper the calculation .This classifiers also employ to test performance.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.