KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Settings

Author: search.filters.author.Netisopakul, PonrudeeSubject: search.filters.subject.Ontology

Search Results

Now showing 1 - 2 of 2
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Event extraction using ontology directed semantic grammar
    (2016-01-01)
    Kuptabut, Suthasinee
    ;
    Netisopakul, Ponrudee
    The task of extracting and constructing knowledge base from news is still a subject of ongoing research. The obtained knowledge base is useful for many applications, such as a question answering system. Football news always gains enormous interest from many football fan clubs. Hence, there are needs to extract certain information from this news in timely fashion. This paper proposes a new approach to extract football events from football news webpages and represents them as frames. One of the main contributions of the new approach is the process of dynamically generating semantic grammar from football domain ontology. A semantic parser uses this grammar to recognize events of interest and their details. This paper explains the overall architecture of our system called Ontology Directed Event Extraction System and explains detail implementation of our new proposed approach for extracting football events. The system is demonstrated by extracting events from 40 football news webpages, resulting in 699 frames which evaluated over 85% precision and recall.
  • Loading...
    Thumbnail Image
    Some of the metrics are blocked by your 
    consent settings
    Item type:Item,
    Ontology directed semantic annotation process
    (2010-10-11)
    Kuptabut, Suthasinee
    ;
    Netisopakul, Ponrudee
    This paper proposes a process for annotating semantic concepts to sentences excerpted from webpages. The annotation process is guided by a domain specific ontology and an entities knowledge base. The overall process has four steps: extracting textual contents from a webpage, selecting relevant sentences, extracting clauses and phrases from a sentence and assigning concepts to phrases. The process is demonstrated using weather news webpages.