Internet Protocol Television (IPTV) ubiquitously provides multimedia video services through mobile device, and integrates contents over cloud systems. For allowing users to browse their favorite multimedia videos on cloud networks without the trouble of repeating and complex search, this study presents a personalized TV program-oriented video system automatically searching for the data of high similarity to the one that user is browsing, and reconstructing separated video contents in accordance with a seamless video reconstruction ranking (SVRR). By employing SVRR, the presented system can integrate video data for continuous playing. The study focuses on the system implementation, the design of SVRR, and the accuracy validations of the theories and the algorithm. According to the experiment results, the proposed system was tested and validated with success.