Reconstructing Evolution Process of Documents in Spatio-Temporal Analysis

• Main Authors: Chang-Keon Ryu, Hyong-Jun Kim, Hwan-Gue Cho
• Format: Conference Proceeding
• Language: English
• Subjects: Bioinformatics; Biological cells; Content based retrieval; DNA; Genetic mutations; Genomics; Information retrieval; Plagiarism; Sequences; Text analysis
• DOI: 10.1109/ICCIT.2008.233

Assuming that document evolution is process transfer document into another document, it is very similar to gene evolution. Although document evolution can be document plagiarism, but it can be descriptions to same events such as reports, news, assignments. The process of document evolution, including insertion, deletion, transposition and transformation is similar to genetics process as by mutation, of one or more nucleotides to a chromosome. Previous systems for detecting documents used word-based detecting method that detects exact match word by quick-indexing in documents. Also current plagiarism detection systems decide plagiarism by using similarity value according to ratio of same words. In this paper, we propose an approach to reconstruct the document evolution in order to detect valuable concerned documents. For this, we propose the approach to detect document evolution similarity using DNA sequence alignment in two dimensions: spatial and temporal similarity. We can get document evolution similarity by combining these two similarities. The main difference of our study to the previous studies is that our method applies to genomic sequence matching and temporal event. Also, we can use our systems to content-based information retrieval.