A Novel Methodology to Filter Out Unwanted Messages from OSN User’s Wall Using Trust Value Calculation
Basic challenge in current Online Social Networks (OSNs) is to grant total control and access to its millions of customers (users) over the data and/or messages shared or highlighted on their personal accounts or private spaces. This control would allow users to have a check on its content and in turn would help in building a strong system facilitating OSN users to directly control the data/content. We need to update our training data on regular basis else it will misclassify any unwanted message which is not in our training data resulting into a negative impact on the accuracy of system. Hence to overcome this limitation we are proposing a new approach where-in an adjustable defined system that allows users to apply text filtering algorithms at preprocessing stage so as to categorize the message and trust value calculation. In this technique it will calculate the trust value for each message and give the trustworthiness of users. If that trust value is less than predefined threshold then it will block that user.
KeywordsOnline social networks Information filtering Short text classification
- 5.Sriram, B., Fuhry, D., Demir, E., Ferhatosmanoglu, H., Demirbas, M.: Short text classification in twitter to improve information filtering. In: Proceedings of the 33rd International ACM SIGIR Conference on Research and Development in Information Retrieval, pp. 841–842 (2010)Google Scholar
- 6.Beye, M., Jeckmans, A., Erkin, Z., Hartel, P., Lagendijk, R., Tang, Q.: Literature overview-privacy in online social networks. In: Distributed and Embedded System (DIES), pp. 1–19 (2010)Google Scholar
- 7.Hidalgo, J.M.G., Garcia, F.C., Sanz, E.P.: Named entity recognition for web content filtering. In: 10th International Conference on Applications of Natural Language to Information Systems, pp. 1–12 (2005)Google Scholar