Adaptive Transmission of Sensitive Information in Online Social Networks

Nowadays, there is an increase in sensitive data like rumors and private content in popular social media platforms like Twitter. So, it has become a severe issue. To limit the increase of sensitive information there is one approach like restricting the transmission among social network users. Transmission limiting methods, on the other hand, curb the transmission of insensitive data as well which results in poor experiences. To address the challenge of this kind by examining how we can reduce sensitive data dissemination at the same time as maintaining the insensitive data transmission in this research, and create this as a restricted reduction problem to maintain insensitive data transmission as the restraint. Now it is time to investigate this type of problem in the company of all user’s transmission capabilities in a fully known network which is known in advance and with partial users' transmission abilities in a semi known network which is unknown in advance. Now we can cooperatively develop a way out using a bandit framework in a fully known network and semi known network by representing sensitive transmission size as a reward of bandit. Furthermore, In the construction of algorithms it is impossible to define the data dissemination size because of the semi-known network’s uncertain diffusion capabilities. To address the problem, It will be good if we use the bandit framework that helps us to study unfamiliar transmission capabilities through a process of transmission and adjustably perform different measures to restrict the transmission on the basis of transmission capabilities. Extensive tests on natural and artificial datasets show the methods that we use have efficiently constrained sensitive data transmission while achieving forty percent reduction in non-sensitive information transmission loss when compared to four baseline techniques.