Against the backdrop of continuously deepening the protection and utilization of industrial heritage and accelerating the integration of digital cultural tourism, the revitalization and regeneration of industrial heritage tourism resources face problems such as inadequate resource identification, inaccurate mechanism configuration, and poor cultural dissemination links. This article focuses on the innovation of the activation and regeneration mechanism of industrial heritage tourism resources and the optimization of cultural value dissemination paths. It constructs a resource element classification model, an activation and regeneration mechanism innovation model, and a cultural value dissemination path optimization model, and introduces computer methods such as multi-source data processing, feature fusion, and dynamic feedback updates into the analysis process. This paper evaluates the resource type identification, mechanism applicability and information transmission performance of the industrial site sample area. The results showed that the classification accuracy of the resource element classification model reached 91.83%, with an F1 value of 90.75%; The comprehensive activation benefit of the innovative model for activating and regenerating mechanisms is 0.88, and the accuracy of mechanism matching reaches 93.27%; The dissemination reach rate and node matching accuracy of the cultural value dissemination path optimization model reached 91.38% and 92.57%, respectively, and the secondary visit rate increased to 41.83%. The results show that computer-aided analysis methods can help improve the classification accuracy of industrial heritage tourism resources, enhance the matching degree between activation paths and resource attributes, and improve the organizational efficiency and diffusion stability of cultural value dissemination.
Povzetek: Članek obravnava aktivacijo in regeneracijo virov industrijske dediščine ter optimizacijo poti širjenja kulturnih vrednosti. Z uporabo večvišinskega podatkovnega povezovanja, združevanja značilk in dinamičnega povratnega posodabljanja so oblikovani modeli za razvrščanje virov, mehanizemsko ujemanje in širjenje vsebin. Rezultati kažejo izboljšano natančnost prepoznavanja, učinkovitost aktivacije ter stabilnost digitalnega komuniciranja.
