ملخص الدراسة:

We introduce a framework for data clustering and visualization called SOMvisua. The current thesis highlights a graph representation of data input for self-organizing map (SOM) and growing hierarchically self-organizing map (GHSOM). Sentences from an article will be represented as graph model instead of vector space model. The SOM and GHSOM clustering algorithm form knowledge about this article. We showed how to use the text similarity features natively and correctly in SOM and GHSOM clustering algorithms, developed a method to alleviate the hub problem and created a visualizing solution for the reviewed class of text similarity algorithms. SOMvisua provides a visual animation for eight famous graph algorithms execution with animation speed control. SOMvisua presents six types of visualization. For visualization for similarity Lists we used well-known methods that takes a similarity list as input and according to the used similarity measure an adjustable number of most similar sentences are arranged in visual form.

توثيق المرجعي (APA)