In the recent decades multi core processor systems are in great demand in order to achieve the high performance in Real time Embedded applications, A network-on chip (NoC) provides a packet switched fabric for multicore on chip communication and has proved efficient, many core interconnect solution compared to shared bus approach System on Chip(SoC).In order to implement various real time Multimedia networking applications on single chip that involves streaming of real time audio/video are more sensitive to delay. Efficient Mapping of these application tasks on various cores contribute to latency minimization as well as achieving increased throughput.In this paper, a given set of static application mapping techniques such as Random, NMAP a constructive heuristic search mapping approach, BB a systematic Search approach,nature inspired mapping approaches GFF works on metaheuristic search and PSO that works on transformative heuristic search approach are executed on different multimedia and networking embedded traffics under similar simulation environments.The Support Vector Machine(SVM) model is executed on the results generated by the simulator in order to predict the best mapping option for a given application considering either only minimum latency or Maximum Throughput,Minimum Latency,Power, Energy Considerations. This study may be help NoC designer to select appropriate mapping technique for particular application weather it is either compute sensitive or communication sensitive.