Multi-motive emotion generation method based on SPFA algorithm

Abstract

The invention relates to a multi-motive emotion generation method based on an SPFA algorithm, and belongs to the technical field of computer. First, mixed emotions are produced by adopting a multi-motive emotion building method and taking the influence of interpersonal relationship as one of the motives, and the multiple motives restrain one another and reinforce one another. An improved SPFA algorithm is adopted in the calculation of the influence of interactive relationship on individual emotions, and the influence of the interactive relationship network on individuals under the condition of negative weight can be calculated. Then, motives produced by the interactive relationship are refined to maintaining, repairing and revenging, and emotions produced under the influence of interactive relationship are calculated. Finally, emotions produced by the interactive relationship and native emotions of individuals are mixed to produce a mixed emotion under multiple motives.

Description

technical field [0001] The invention relates to a multi-motive emotion generation method based on an SPFA algorithm, which belongs to the field of computers. Background technique [0002] As a social group, a human individual is a node in a social network. Individuals are influenced by interactive relationships to form unique views on things or other individuals; or to maintain their own social networks to cover up their original views on things. This network of interacting relationships forms the motivations that individuals need to generate emotions. However, the existing emotion models are all intended to generate a primitive emotion and ignore the social nature of emotion, which leads to the lack of high intelligence in the interaction process of emotional virtual humans. [0003] The generation of emotion is one of the problems to be solved in the field of affective computing. However, it has always been a technical problem to be solved by those skilled in the art t...

AI Technical Summary

Problems solved by technology

Guo Shangbo's "Research on Personalized Emotional Modeling Method" mainly used dimensional space and incorporated personality factors. Liu Zhen's "A Virtual Character Emotional Cognition Evaluation Model" constructed fuzzy reasoning, but still could not solve the problem of interaction. The problem of the influence of relationships on individual emotions; both are emotional reasoning processes based on the OCC model

Xiang Nan's "Research on Virtual Interactive Emotional Simulation Model with Control Mechanism" can solve the impact of interactive relationships on individual emotions, but the algorithm efficiency is not high, and it is not suitable for the calculation of unfriendly relationships. In addition, the impact on interpersonal relationships is superimposed. Yes, no mutual inhibition

In addition, "An Emotional Modeling Method Applicable to Shaping Non-player Controlled Game Characters" and "An Emotional Modeling Method Applicable to Modeling Non-player Controlled Game Characters" published by Huang Deheng constructed motivation to drive individuals to produce emotions, but The former model is simple and cannot adapt to the emotion generation under complex interactive conditions, and the latter does not give a calculation method for the impact of specific social motivations on individuals