Adobe PDF (7.27 MB)
Title Details:
Categorization and Prediction
Authors: Verykios, Vasileios
Kagklis, Vasileios
Stavropoulos, Ilias
Reviewer: Kalles, Dimitrios
Description:
Abstract:
The main objective of this chapter is to introduce and deepen the concept of categorization which aims to induce a categorization model using a training set and a learning algorithm, through which values can be assigned to the category attribute in uncategorized records. There are various kinds of categorization models, such as rules, lists, decision trees, set of subsamples or example data, neural networks, clustering methods, etc. In this chapter we will deal with the induction of decision tree models, and we will look at the non-purity measures and decomposition techniques used to develop these trees. Familiarity with issues related to decision tree induction such as overfitting or underfitting a model to the data, calculating the generalization error of a model in different ways are some of the objectives of this chapter.
Linguistic Editors: Spanaka, Adamantia
Graphic Editors: Filoni, Valentina
Type: Chapter
Creation Date: 2015
Item Details:
License: http://creativecommons.org/licenses/by-nc-nd/3.0/gr
Handle http://hdl.handle.net/11419/2970
Bibliographic Reference: Verykios, V., Kagklis, V., & Stavropoulos, I. (2015). Categorization and Prediction [Chapter]. In Verykios, V., Kagklis, V., & Stavropoulos, E. 2015. Data science through the R language [Undergraduate textbook]. Kallipos, Open Academic Editions. https://hdl.handle.net/11419/2970
Language: Greek
Is Part of: Data science through the R language
Publication Origin: Kallipos, Open Academic Editions