Balancing the trade-off between privacy and profitability in Social Media using NMSANT

Social Media is being used as a key platform by advertisers to improve business by providing targeted and personalized advertising. There exist a trade-off between productivity in advertising and invasion of user´s privacy in the existing approaches. Due to these privacy concerns, there were many law suits filed against the Beacon[1] advertising model used by Facebook resulting in its discontinuation. The new approaches need to address targeted audience while preserving the privacy in order to build appropriate revenue model for their operations. In this paper we propose an innovative model that leverage the trade-off. The model effectively interacts with user´s linked data present in the web structured format, retrieve it and integrate data from marketing partners. Then it broadcasts the advertisement in social graph with flexible sentences For targeted advertising and privacy our model maintains interaction records among users in virtual containers for finding tie-strength[2] and also active friends ¹ by Association Rules Mining [ARM] algorithm. We applied and validated our approach using a real data set obtained from 506 active social media users.