Sharif Digital Repository

Nowadays the importance of ontologies in various domains such as semantic web and NLP is obvious. Manually construction of ontologies is a time-consuming, costly and error prone. The non-manual building of ontologies approaches are divided to the expand approach and the merge approach, each has pros and cons. The expand approach is preferred upon the merge approach for languages with few lingual resources. The most important thing in the expand approach, which is building an ontology from another existing ontology, is making distinction between the different meanings of polysemous words. This activity is called word sense disambiguation(WSD). The existing approaches usually give attention to... 

Recognition of complex visual events has attracted much interest in recent years. Compared to somehow similar tasks like action recognition, event recognition is much more complex, primarily because of huge intra-class variation of events, variable video durations, lack of pre-imposed video structures, and severe preprocessing noises. To deal with these complexities and improve the state of the art approaches to the problem of video understanding, this thesis is focused on video event recognition based on frame level CNN descriptors. Using transfer learning, the image trained descriptors are applied to the video domain to make event recognition feasible in scenarios with limited...