2023-03-24T13:26:09Z
https://meral.edu.mm/oai
oai:meral.edu.mm:recid/3629
2021-12-13T03:38:49Z
1582963302567:1597824273898
user-ucsy
Classifying Textual Documents Using Two Dimensional Probabilistic Model
Than, Wai Me Me
Kham, Nang Saing Moon
This paper presents the probabilistic model named TwodimensionalProbabilistic Model (2DPM). In this model, terms are seen as disjoinevents, and terms and categories are realeated to each other. Since the documentsare represented as the union of terms, disjoint event, document and categories arealso rreleated. Terms are measured with their presence and expressiveness. Thepresentce and expressivencess of a term is defined as the peculiarity of that term. Adocument is defined as set of terms and it also has presence and expressivenessfor a category. So, the 2DPM model defines a direct relationship between theprobability of a document given a category of interest and a point on atwodimensionalspace. With the points, entire collections of documents are graphed ona Cartesian plane and documents are classifie directly on the two-dimensionalrepresentation. To experiment the system, Reuters-21578 newswire dataset is usedfor text classification.
2011-12-29
http://hdl.handle.net/20.500.12678/0000003629
https://meral.edu.mm/records/3629