A product human-computer interaction adaptation system and method based on multi-source fusion

Abstract

This application provides a product human-computer interaction adaptation system and method based on multi-source fusion. It constructs an arbitrator based on contextual semantic understanding. The arbitrator receives context interpretation results from at least two independent analysis paths in parallel. The first analysis path extracts environmental and operational state features based on real-time sensor data streams, while the second analysis path infers implicit intent based on user interaction behavior sequences. The arbitrator performs cross-consistency verification on the context interpretation results from different analysis paths, calculates semantic consistency scores, and evaluates the data confidence level of the original data used by each path. Based on the semantic consistency scores and the corresponding data confidence levels of each path, it dynamically calculates the multi-source fusion credibility index for the current adaptation decision and matches preset adaptation strategy parameters according to the multi-source fusion credibility index. Using the scheme of this application, multi-source context fusion adaptation decisions can be achieved, thereby improving the accuracy of human-computer interaction commands and the robustness of the system.

Images