DEVELOPMENT OF A MODEL THAT LINKS HYPERTEXT/HYPERMEDIA USING MACHINE LEARNING

The active growth of digital technologies has demonstrated the importance of the association between the text and multimedia to make it more accessible and interesting to users. The traditional hypermedia systems rest on the manual organization that is time consuming and incoherent. This paper is referred to as a Development of a Model that Overlays Hypertext and Hypermedia with Machine Learning and proposes a bi-directional model that will make the text and multimedia linkage process automated. It was deployed in the Google Colab on Python programming language and presented through Gradio. The machine learning algorithms were off-the-shelf: BLIP for image-to-text captioning, and Stable Diffusion for text-to-image generation. Furthermore,hypermedia connectivity was createdthrough artificial videogeneration from text input. Testingshowed the system could produceappropriate captions touploaded photos, realisticimages from text prompts, and relatedvideo links. This washelpful in highlightingthe usabilityof the systemwith hypertexts andhypermedia. These limitations includeGPU usage andsporadic ineffective data; however, the projectdemonstrated that machinelearning could createlarge-scale and interactivehypermedia. It finally creates theautomated connectionof information, making it moreaccessible, more facilitated ineducational use, and opening upnew avenuesfor multimedia interaction. This enablesfuture enhancements, which mayinclude text-video modeling, datasetaugmentation, andapplication to thesystem onscaled platforms.