Aristotle University Of Thessaloniki (AUTH)
Στατιστικές μέθοδοι πολυμεταβλητής ανάλυσης δεδομένων από βιολογικά κείμενα και οντολογίες
Abstract
dc:descriptionThe research involves text mining of biological texts using statistical methods of classification and clustering. The classification involves the use of Linear Discriminant Analysis, LDA, Support Vector Machines, SVA and Multinomial Logistic Regression, MLR, LDA was found to perform the best. Non Linear Canonical Correlation, Analysis, NLCCA was also used in order to describe the information of the words of the texts, their gene ontology and Medical Subject Headings with only one dataset, with reduced number of variables. The clustering was based on a stochastic algorithm, namely Markov clustering (MCL) and represented the results to the end user in a 2d or 3d environment.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2008
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Theodosiou, T.
- Θεοδοσίου, Θεοδόσιος
Subjects
dc:subject × 14- Στατιστική πολυμεταβλητή ανάλυση
- Ομαδοποίηση
- Κατηγοροποίηση
- Βιολογικά κείμενα
- Βιολογικές οντολογίες
- Statistical multivariate analysis
- Clustering
- Classification
- Biological texts
- Biological ontologies
- Φυσικές Επιστήμες
- Επιστήμη Ηλεκτρονικών Υπολογιστών και Πληροφορική
- Natural Sciences
- Computer and Information Sciences
Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/18932