TY  - JOUR
TI  - Generating Correlations among Different Modalities by Using Parallel Processing For Cross-Media Retrieval
AU  - R. Rajasekhar
AU  -   Rokkam Srikanth Reddy
JO  - International Journal of Scientific Research in Computer Science, Engineering and Information Technology
PB  - Technoscience Academy
DA  - 2017/08/31
PY  - 2017
DO  - https://doi.org/10.32628/IJSRCSEIT
UR  - https://ijsrcseit.com/CSEIT1724214
VL  - 2
IS  - 4
SP  - 869	
EP  - 874
AB  - Hashing strategies are very useful as they may be used for lots responsibilities including in search engines like Google and yahoo for cross media retrieval. Here hashing technique has been proposed to seize similarities among cross - media records such as textual, visible. As words of individual formations can also have proportional. Hence to clear up such troubles semantic hashing technique is used in this paper along with surf++ set of rules. The semantic level hashing is the gives best result on vector classification and word embedded. This paper propose a high-powered task formation for parallel object-oriented programming and represent the results from a source-to-source compiler and runtime system. With the insertion of one keyword, the serial code doesn't contain restructuring and asynchronous function management is executed on behalf of the programmer; the parallel code required realizing task parallelism looks significantly like the sequential counterpart. An instinctive result is provided to manage task dependencies as kindly as homogenize different task concepts into one model.