Knowledge representation for conceptual, motivational, and affective processes in natural language communication

• Main Authors: HO, Seng Beng, WANG, Zhaoxia, QUEK, Boon-Kiat, CAMBRIA, Erik
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2023
• Subjects: affective processes; knowledge representation; motivational processes; natural language communication; natural language understanding; Databases and Information Systems; Numerical Analysis and Scientific Computing
• Online Access: https://ink.library.smu.edu.sg/sis_research/8925; https://ink.library.smu.edu.sg/context/sis_research/article/9928/viewcontent/KnowledgeRep_CMAP_av.pdf
• Institution: Singapore Management University

Natural language communication is an intricate and complex process. The speaker usually begins with an intention and motivation of what is to be communicated, and what outcomes are expected from the communication, while taking into consideration the listener’s mental model to concoct an appropriate sentence. Likewise, the listener has to interpret the speaker’s message, and respond accordingly, also with the speaker’s mental model in mind. Doing this successfully entails the appropriate representation of the conceptual, motivational, and affective processes that underlie language generation and understanding. Whereas big-data approaches in language processing (such as chatbots and machine translation) have performed well, achieving natural language based communication in human-robot collaboration is non-trivial, and requires a deeper representation of the conceptual, motivational, and affective processes involved in conveying precise instructions to robots. This paper capitalizes on the UGALRS (Unified General Autonomous and Language Reasoning System) framework and the CD+ (Conceptual Dependency Plus) representational scheme to demonstrate how social communication through language can be supported by a knowledge representational scheme that handles conceptual, motivational, and affective processes in a deep and generalizable way. Through an illustrative set of concepts, motivations, and emotions, we show how these aspects are integrated into a general framework for knowledge representation and processing that could serve the purpose of natural language communication for an intelligent system.