Sampling Online Social Networks with Tailored Mining Strategies

With the vast amount of daily generated data that expands every day, Online Social Network (OSN) is considered a key source of information for many Big Data applications. Despite that, companies behind OSNs resort to putting more constraints on their APIs gateways, decreasing the number of information researchers can gather while increasing the time of data mining procedures. This paper proposes a new platform to run sampling strategies with maximum scalability, to decrease the budget and time required for mining a representative sampling set. By comparing the accuracy of several strategies, our experiments demonstrate the relevance of the proposed platform in supporting OSN mining.