Title

Artificial Intelligence – I: A preliminary framework for human-agent communication in electronic negotiations

Abstract/Description

Electronic negotiations are business negotiations conducted via electronic means using information and communications technologies (ICT). Two dominant types of electronic negotiation systems are automated negotiation systems for software agents and negotiation support systems (NSSs) for humans. However, the integration of two types for human-agent negotiations is an important task. In this paper, an extended communication model for human-agent business negotiations is presented. For this purpose, the underlying communication models of automated negotiations and NSSs are analyzed. The extended communication model is based on a common negotiation ontology which captures the negotiation agenda and paves the way for such hybrid communication, a natural language processing (NLP) component which process natural language negotiation content, and a translator to covert human message to agent message and vice versa. NLP component has been added as an alternative approach to manual annotation of message content. Our aim is to show that using an ontology-based negotiation approach as implemented in the NSS Negoisst and NLP techniques integrated into it can make human-agent communication better.

Location

Room C5

Session Theme

Artificial Intelligence – I

Session Type

Other

Session Chair

Dr. Sajjad Haider

Start Date

23-7-2011 2:15 PM

End Date

23-7-2011 2:35 PM

Share

COinS
 
Jul 23rd, 2:15 PM Jul 23rd, 2:35 PM

Artificial Intelligence – I: A preliminary framework for human-agent communication in electronic negotiations

Room C5

Electronic negotiations are business negotiations conducted via electronic means using information and communications technologies (ICT). Two dominant types of electronic negotiation systems are automated negotiation systems for software agents and negotiation support systems (NSSs) for humans. However, the integration of two types for human-agent negotiations is an important task. In this paper, an extended communication model for human-agent business negotiations is presented. For this purpose, the underlying communication models of automated negotiations and NSSs are analyzed. The extended communication model is based on a common negotiation ontology which captures the negotiation agenda and paves the way for such hybrid communication, a natural language processing (NLP) component which process natural language negotiation content, and a translator to covert human message to agent message and vice versa. NLP component has been added as an alternative approach to manual annotation of message content. Our aim is to show that using an ontology-based negotiation approach as implemented in the NSS Negoisst and NLP techniques integrated into it can make human-agent communication better.