caa a22 4500 605469849 CHVBK 20210128100324.0 cr unu---uuuuu 210128e20150401xx s 000 0 eng 10.1007/s00500-014-1411-9 doi (NATIONALLICENCE)springer-10.1007/s00500-014-1411-9 Jo Taeho Department of Computer and Information Engineering, Inha University, Incheon, South Korea aut Normalized table-matching algorithm as approach to text categorization [Elektronische Daten] [Taeho Jo] This research is concerned with the improved version of table-based matching algorithm as the approach to text categorization tasks. It is intended to tackle the three problems in encoding texts into numerical vectors and the unstable performance by the fluctuations from text lengths in the previous version. In this research, we encode texts into tables rather than into numerical vectors, define the similarity measure between two tables which is always as a normalized value between zero and one, and apply it to the tasks of text categorization. As the benefits from this research, we expect better performance by solving the three problems resulting from encoding texts into numerical vectors, and more stable performance by improving the previous version. Therefore, we empirically validate the proposed approach through the four sets of experiments, with respect to both performance and stability. Springer-Verlag Berlin Heidelberg, 2014 Soft Computing Springer Berlin Heidelberg 19/4(2015-04-01), 839-849 1432-7643 19:4<839 2015 19 500 https://doi.org/10.1007/s00500-014-1411-9 text/html Onlinezugriff via DOI BK010053 XK010053 XK010000 Metadata rights reserved Springer special CC-BY-NC licence nationallicence 1 research-article jats NATIONALLICENCE NATIONALLICENCE NL-springer NATIONALLICENCE 856 40 https://doi.org/10.1007/s00500-014-1411-9 text/html Onlinezugriff via DOI NATIONALLICENCE 100 1- Jo Taeho Department of Computer and Information Engineering, Inha University, Incheon, South Korea aut NATIONALLICENCE 773 0- Soft Computing Springer Berlin Heidelberg 19/4(2015-04-01), 839-849 1432-7643 19:4<839 2015 19 500