|
|
Bitmap picture classification
Lukšová, Ivana ; Surynek, Pavel (advisor) ; Holan, Tomáš (referee)
In this work we propose a method of automatic content-based bitmap pictures classification into classes like buildings, landscapes, photographs etc. The method uses concepts of the machine learning, concretely the mechanism of a decision tree. We extract characteristics like the contrast, the distribution of colors, the presence of straight lines. On the basis of these characteristics the decision tree is built using the algorithm ID3. Creation of the program, which allows operations with the database of images based on the classification results - retrieval, sorting into directories etc. is also a part of the work. The method shows rate 75-85% correct within the classes.
|
|
|
Ontology Enrichment Based on Unstructured Text Data
Lukšová, Ivana ; Nečaský, Martin (advisor) ; Kozák, Jakub (referee)
Title: Ontology Enrichment Based on Unstructured Text Data Author: Ivana Lukšová Department: Department of Software Engineering Supervisor: Mgr. Martin Nečaský, Ph.D., Department of Software Engi- neering Abstract: Semantic annotation, attaching semantic information to text data, is a fundamental task in the knowledge extraction. Several ontology-based semantic annotation platforms have been proposed in recent years. However, the process of automated ontology engineering is still a challenging problem. In this paper, a new semi-automatic method for ontology enrichment based on unstructured text is presented to facilitate this process. NLP and ma- chined learning methods are employed to extract new ontological elements, such as concepts and relations, from text. Our method achieves F-measure up to 71% for concepts extraction and up to 68% for relations extraction. Keywords: ontology, machine learning, knowledge extraction 1
|
|
|
Bitmap picture classification
Lukšová, Ivana ; Surynek, Pavel (advisor) ; Holan, Tomáš (referee)
In this work we propose a method of automatic content-based bitmap pictures classification into classes like buildings, landscapes, photographs etc. The method uses concepts of the machine learning, concretely the mechanism of a decision tree. We extract characteristics like the contrast, the distribution of colors, the presence of straight lines. On the basis of these characteristics the decision tree is built using the algorithm ID3. Creation of the program, which allows operations with the database of images based on the classification results - retrieval, sorting into directories etc. is also a part of the work. The method shows rate 75-85% correct within the classes.
|
guest ::
