Maria Liakata

Inducing Domain Theories

Using Inductive Logic Programming to Learn Domain Specific Knowledge in the Form of Rules

• Broschiertes Buch

Produktdetails

• Verlag: VDM Verlag Dr. Müller
• Seitenzahl: 248
• Erscheinungstermin: Juni 2008
• Englisch
• Abmessung: 220mm x 150mm x 13mm
• Gewicht: 340g
• ISBN-13: 9783639014129
• ISBN-10: 363901412X
• Artikelnr.: 23899611

Autorenporträt

?Inducing Domain Theories? shows how world knowledge can belearnt from text through Inductive Logic Programming (ILP). What ismeant by a \"domain theory\" is a collection of facts andgeneralisations or rules which capture what commonly happens insome domain of interest. The domain of application was financialnews but the approach can be extended to the discovery of newknowledge from different domains. The learning paradigm employed,ILP, generalises over examples from the domain to obtain moregeneral patterns covering the majority of the input instances. ILPwas preferred over other machine learning techniques due to theexpressive power of the language specifications guiding the searchfor general patterns and the fact that it allows the inclusion ofbackground knowledge. The relational data mining algorithm WARMRgave the most satisfactory results as it was able to capturefrequent patterns of complex structure, often encoding causalrelations consisting of two or more verbs and informa

tion abouttheir respective arguments. Finite State Automata (FSA)minimisation techniques were employed to render the rules learntinto a more compact, human friendly format.