WikiTalk: A Spoken Wikipedia-based Open-Domain Knowledge Access System

WikiTalk is an open-domain knowledge access system that talks about topics using Wikipedia articles as its knowledge source. Based on Constructive Dialogue Modelling theory, WikiTalk exploits the concepts of Topic and NewInfo to manage topic-tracking and topic-shifting. As the currently talked-about Topic can be any Wikipedia topic, the system is truly open-domain. NewInfos, the pieces of new information to be conveyed to the partner, are associated with hyperlinks extracted from the Wikipedia texts. Using these hyperlinks the system can change topics smoothly according to the user’s changing interests. As well as user-initiated topics, the system can suggest new topics using for example the daily "Did you know?" items in Wikipedia. WikiTalk can be employed in different environments. It has been demonstrated on Windows, with an open-source robotics simulator, and with the Aldebaran Nao humanoid robot.

[1]  Joseph Weizenbaum,et al.  ELIZA—a computer program for the study of natural language communication between man and machine , 1966, CACM.

[2]  Alexander Franz,et al.  Searching the Web by Voice , 2002, COLING.

[3]  Paolo Rosso,et al.  Mining Knowledge fromWikipedia for the Question Answering task , 2006, LREC.

[4]  Dimitra Anastasiou,et al.  Evaluation of WikiTalk - User Studies of Human-Robot Interaction , 2013, HCI.

[5]  Kristiina Jokinen,et al.  Emergent verbal behaviour in human-robot interaction , 2011, 2011 2nd International Conference on Cognitive Infocommunications (CogInfoCom).

[6]  Emer Gilmartin,et al.  Multimodal conversational interaction with a humanoid robot , 2012, 2012 IEEE 3rd International Conference on Cognitive Infocommunications (CogInfoCom).

[7]  Kristiina Jokinen,et al.  Constructive Dialogue Modelling - Speech Interaction and Rational Agents , 2009, Wiley series in agent technology.

[8]  Kristiina Jokinen,et al.  Constructive Interaction for Talking about Interesting Topics , 2012, LREC.

[9]  Kathleen F. McCoy,et al.  Focus of attention: Constraining what can be said next , 1991 .

[10]  Olivier Galibert,et al.  Integrating spoken dialog and question answering: the ritel project , 2006, INTERSPEECH.

[11]  Kristiina Jokinen,et al.  Multimodal Open-Domain Conversations with the Nao Robot , 2012, Natural Interaction with Robots, Knowbots and Smartphones, Putting Spoken Dialog Systems into Practice.

[12]  Hideki Tanaka,et al.  Context Management with Topics for Spoken Dialogue Systems , 1998, COLING-ACL.

[13]  Raffaella Bernardi,et al.  Exploring Topic Continuation Follow-up Questions using Machine Learning , 2009, HLT-NAACL.