Posting of online reviews play a dominant role in sharing the customer's opinion in social Medias. But the challenge is how to trust these reviews. Many researchers carried their work on sentimental analysis, predictions or forecasting but very few focused on fake reviews analysis. Fake reviews also change the mood of the people on their buying pattern. In the online shopping at a greater extent. In this paper, several conditions are applied on the reviews to identify fake reviews using support vector machines. Experimental results are validated using various accuracy measures and compared to state of the art methods to demonstrate the efficacy of the proposed method.

MANET nodes act as a host as well as a router which increases the significance of every node for their participation in communication. Loss of any node in the network results in failure of links connected to the node which brings the importance of increased lifespan of a node. Some nodes during frequent transaction at critical network scenario consume more energy and become ill with critical energy level. Special attention towards these nodes can improve the lifespan of the node. In this paper an ant colony-based pheromone deposition mechanism was proposed to extend the lifetime of ill nodes. Pheromone deposited for the neighbor in the pheromone table helps in identifying frequently communicated neighbor. The proposed algorithm identifies the ill node and requests its frequently communicated neighbor for a tie up. The neighbor shares the workload of the ill node with mutual agreement. This method also improves the performance of a network by limiting pheromone deposition practice for low weighted nodes with low energy and high density (packet in queue). The proposed method increases lifespan of ill nodes and thereby increases the lifetime of entire network. The proposed work was also compared with existing protocol and the results have proved that the proposed mechanism has increased lifetime and reduced energy consumption of the entire network.