Computer Science Theses and Dissertations

Permanent URI for this collectionhttp://hdl.handle.net/1903/2756

Browse

Filters

20052

Settings

Subject: search.filters.subject.ontology

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Applications of Factorization Theorem and Ontologies for Activity ModelingRecognition and Anomaly Detection
    (2005-05-06) Akdemir, Umut; Chellappa, Rama; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    In this thesis two approaches for activity modeling and suspicious activity detection are examined. First is application of factorization theorem extension for deformable models in two dierent contexts. First is human activity detection from joint position information, and second is suspicious activity detection for tarmac security. It is shown that the first basis vector from factorization theorem is good enough to dierentiate activities for human data and to distinguish suspicious activities for tarmac security data. Second approach dierentiates individual components of those activities using semantic methodol- ogy. Although currently mainly used for improving search and information retrieval, we show that ontologies are applicable to video surveillance. We evaluate the domain ontologies from Challenge Project on Video Event Taxonomy sponsored by ARDA from the perspective of general ontology design principles. We also focused on the eect of the domain on the granularity of the ontology for suspicious activity detection.
  • Thumbnail Image
    Item
    Managing Uncertainty and Ontologies in Databases
    (2005-04-18) Hung, Edward; Subrahmanian, V.S.; Computer Science; Digital Repository at the University of Maryland; University of Maryland (College Park, Md.)
    Nowadays a vast amount of data is generated in Extensible Markup Language (XML). However, it is necessary for applications in some domains to store and manipulate uncertain information, e.g. when the sensor inputs are noisy, or we want to store data that is uncertain. Another big change we can see in applications and web data is the increasing use of ontologies to describe the semantics of data, i.e., the semantic relationships between the terms in the databases. As such information is usually absent from traditional databases, there is tremendous opportunity to ask new kinds of queries that could not be handled in the past. This provides new challenges on how to manipulate and maintain such new kinds of database systems. In this dissertation, we will see how we can (i) incorporate and manipulate uncertainty in databases, and (ii) efficiently compute aggregates and maintain views on ontology databases. First, I explain applications that require manipulating uncertain information in XML databases and maintaining web ontology databases written in Resource Description Framework (RDF). I then introduce the probabilistic semistructured PXML data model with two formal semantics. I describe a set of algebraic operations and its efficient implementation. Aggregations of PXML instances are studied with two semantics proposed: possible-worlds semantics and expectation semantics. Efficient algorithms with pruning are given and evaluated to show their feasibility. I introduce PIXML, an interval probability version of PXML, and develop a formal semantics for it. A query language and its operational semantics are given and proved to be sound and complete. Based on XML, RDF is a language used to describe web ontologies. RDQL, an RDF query language, is extended to support view definition and aggregations. Two sets of algorithms are given to maintain non-aggregate and aggregate views. Experimental results show that they are efficient compared with standard relational view maintenance algorithms.