Grant: Strategic Projects - Group
University of Ottawa
$87,500.00
Analysis and Generation of Text with Personality and Emotions
EF 2010-11
Résumé
Many computer systems nowadays require verbal or textual interaction with the users. Examples of such systems are interactive training simulations (serious games), service robots, and automatic help desks. Traditionally, the sentences used for these interactions are written manually. The use of software that could automatically generate these sentences would enable the systems to adapt the expressed messages to the context of the conversation, and express past and future actions that may form these interactions. Such software implements a Natural Language Generation (NLG) process. For example, DISTIL Interactive creates business interview games which can range from simple sales calls to business audits or complex negotiations. Creating the dialogue for these interviews is a current bottleneck. An NLG system that can create realistic dialogue for the intelligent conversational agents can alleviate this bottle-neck and reduce the high human-labour cost. With such a technological solution, a writer can sketch out the skeleton of the dialogue and then the NLG component can create many different responses based on a defined set of rules. In addition, the use of NLG will permit to enrich the dialogue by automatically modifying the responses according to the character's personality, mood, and tone.
We already have a first proof-of-concept NLG system. In this project we propose to extend it in order to allow the creation of text with more individual personalities and emotions. We propose a new line of research in text analysis and text generations. The results of text analysis by automatic classification (machine learning) and information extraction will be used directly in the text generation phase. We are investigating ways of discovering emotions and tone patterns from text and integrating them in the NLG process that assists the dialogue content generation. We also propose to provide a tool that aids the human experts to build dialogue content templates to be used by the previously mentioned systems.
Programme
Strategic Projects - Group
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
380975-2009
Name nom
Inkpen, Diana
Department département
Information Technology and Engineering, School of
Organization id
28
Institution établissement
University of Ottawa
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2010
Competition year année de concours
2009
Award amount
87500
Program id
STPGP
Program name en
Strategic Projects - Group
Program name fr
Projets stratégiques - De groupe
Group en
Research partnerships programs
Group fr
Programmes de partenariat
Committee code
1041
Committee name en
Strategic Projects Selection Panel - Communications and Information
Committee name fr
Comité de sélection des projets stratégiques - Communication et information
Area of application code
802
Area of application group en
Information and communication services
Area of application group fr
Services d'information et de télécommunications
Area of application en
Information systems and technology
Area of application fr
Systèmes d'information et technologie de l'information
Research subject code
2801
Research subject group en
Artificial intelligence
Research subject group fr
Intelligence artificielle
Research subject en
Natural language and speech understanding
Research subject fr
Langage naturel et reconnaissance de la parole
Partie
2
Nb partie
2
Application title
Analysis and Generation of Text with Personality and Emotions
Keyword
NSERC
Application summary
Many computer systems nowadays require verbal or textual interaction with the users. Examples of such systems are interactive training simulations (serious games), service robots, and automatic help desks. Traditionally, the sentences used for these interactions are written manually. The use of software that could automatically generate these sentences would enable the systems to adapt the expressed messages to the context of the conversation, and express past and future actions that may form these interactions. Such software implements a Natural Language Generation (NLG) process. For example, DISTIL Interactive creates business interview games which can range from simple sales calls to business audits or complex negotiations. Creating the dialogue for these interviews is a current bottleneck. An NLG system that can create realistic dialogue for the intelligent conversational agents can alleviate this bottle-neck and reduce the high human-labour cost. With such a technological solution, a writer can sketch out the skeleton of the dialogue and then the NLG component can create many different responses based on a defined set of rules. In addition, the use of NLG will permit to enrich the dialogue by automatically modifying the responses according to the character's personality, mood, and tone.
We already have a first proof-of-concept NLG system. In this project we propose to extend it in order to allow the creation of text with more individual personalities and emotions. We propose a new line of research in text analysis and text generations. The results of text analysis by automatic classification (machine learning) and information extraction will be used directly in the text generation phase. We are investigating ways of discovering emotions and tone patterns from text and integrating them in the NLG process that assists the dialogue content generation. We also propose to provide a tool that aids the human experts to build dialogue content templates to be used by the previously mentioned systems.