With the rapid development of wireless communication technology, many users are accessing the internet from mobile appliances, such as notebooks, PDAs, and cellular phones. These devices are miscellaneously limited in computing resources, like CPU speed, memory, temporary storage, power supply, installed software, and communication bandwidth. The web accesses in these mobile devices thus encounter distorted user interface, broken images, slow responses, etc. To overcome this problem, we design and implement a semantics-based web content adaptation framework to provide automatically adapted contents to miscellaneous mobile devices. RDF Semantics of client device CC/PP configurations and web page tag structures are extracted to determine the proper parameters for the format and layout of web contents. Heuristic transcoding rules in the Jena Inference System are then applied to transform the web contents for each particular device. The functionality of this framework is illustrated by its capability of adjusting the layout of the main page of a shopping portal and of adjusting the parameters of images in the page.