Asian Science Citation Index is committed to provide an authoritative, trusted and significant information by the coverage of the most important and influential journals to meet the needs of the global scientific community.  
ASCI Database
308-Lasani Town,
Sargodha Road,
Faisalabad, Pakistan
Fax: +92-41-8815544
Contact Via Web
Suggest a Journal
Articles by Mohd. Hasan Selamat
Total Records ( 2 ) for Mohd. Hasan Selamat
  Aida Mustapha , Md. Nasir Sulaiman , Ramlan Mahmod and Mohd. Hasan Selamat
  Problem statement: Overgeneration-and-ranking architecture works well in written language where sentence is the basic unit. However, in spoken language where utterance is the basic unit, the disadvantage becomes critical as spoken language also render intentions, hence short strings may be of equivalent impact. Approach: In classification-and-ranking, response was deliberately chosen from dialogue corpus rather than wholly generated, such that it allows short ungrammatical utterances as long as they satisfy the intended meaning of input utterance. Because the architecture is intention-based, it adopted an open-domain knowledge representation, whereby response utterances were semantically represented using some ontology general enough for future reuse in another domain. Results: This study presented corpus-based analysis on cross-domain experimentation using different type of corpus to validate the consistency of the response classifier that delimits the searching space for ranking. The open-domain quality for classification-an-ranking architecture was tested on two mixed-initiative, transaction dialogue corpus in theater reservation and emergency planning. Results showed consistent distribution accuracies in both classification and ranking experiment, indicating that the approach is viable for cross-domain implementations. Conclusion: The ability of a response generation system to directly learn response utterances from the domain corpus suggested the possibility to build a dialogue system by feeding the learning module with a target corpus and the system learned the response behavior directly from the training corpus.
  Aida Mustapha , Md. Nasir Sulaiman , Ramlan Mahmod and Mohd. Hasan Selamat
  Problem statement: The first component in classification-and-ranking architecture is a Bayesian classifier that classifies user utterances into response classes based on their semantic and pragmatic interpretations. Bayesian networks are sufficient if data is limited to single user input utterance. However, if the classifier is able to collate features from a sequence of previous n-1 user utterances, the additional information may or may not improve the accuracy rate in response classification. Approach: This article investigates the use of dynamic Bayesian networks to include time-series information in the form of extended features from preceding utterances. The experiment was conducted on SCHISMA corpus, which is a mixed-initiative, transaction dialogue in theater reservation. Results: The results show that classification accuracy is improved, but rather insignificantly. The accuracy rate tends to deteriorate as time-span of dialogue is increased. Conclusion: Although every response utterance reflects form and behavior that are expected by the preceding utterance, influence of meaning and intentions diminishes throughout time as the conversation stretches to longer duration.
Copyright   |   Desclaimer   |    Privacy Policy   |   Browsers   |   Accessibility