A Privacy Settings Prediction Model for Textual Posts on Social Networks

  • Lijun Chen
  • Ming XuEmail author
  • Xue Yang
  • Ning Zheng
  • Yiming Wu
  • Jian Xu
  • Tong Qiao
  • Hongbin Liu
Conference paper
Part of the Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering book series (LNICST, volume 252)


Privacy issues of social media are getting tricky due to the increasing volume of social media users sharing through online social networks (OSNs). Existing privacy policy mechanisms of OSNs may not protect personal privacy effectively since users are struggle to set up the privacy settings. In this paper, we propose a privacy policy prediction model to help users to specify privacy policies for their textual posts. We investigate the semantic of posts, social context, and keywords associated with users’ privacy preferences as possible indicators of decision making, and build a multi-class classifier based on their historical posts and decisions. During the cold-start periods, the proposed model integrates crowdsourcing and machine learning to recommend privacy policies for new users. Experimental results shows that the overall match rate for all the data with random forest classifier is over 70%, with more than 50% correct prediction rate for new users.


Social networks Privacy Policy recommendation 



This work is supported by the National Key R&D Plan of China under grant no. 2016YFB0800201, the Natural Science Foundation of China under grant no. 61070212 and 61572165, the State Key Program of Zhejiang Province Natural Science Foundation of China under grant no. LZ15F020003, the Key research and development plan project of Zhejiang Province under grant no. 2017C01065, the Key Lab of Information Network Security, Ministry of Public Security, under grant no C16603.


  1. 1.
    Abuelgasim, A., Kayem, A.: An approach to personalized privacy policy recommendations on online social networks. In: International Conference on Information Systems Security and Privacy, pp. 126–137 (2016)Google Scholar
  2. 2.
    Bilogrevic, I., Huguenin, K., Agir, B., Jadliwala, M., Hubaux, J.P.: Adaptive information-sharing for privacy-aware mobile social networks. In: ACM International Joint Conference on Pervasive and Ubiquitous Computing, pp. 657–666 (2013)Google Scholar
  3. 3.
    Chen, L.: Questionnaire about privacy preference on social networks (2016).
  4. 4.
    Fang, L., Lefevre, K.: Privacy wizards for social networking sites. In: International Conference on World Wide Web, pp. 351–360 (2010)Google Scholar
  5. 5.
    Goldberg, Y., Levy, O.: word2vec explained: deriving Mikolov et al’.s negative-sampling word-embedding method. Eprint Arxiv (2014)Google Scholar
  6. 6.
    Li, Q., Li, J., Wang, H., Ginjala, A.: Semantics-enhanced privacy recommendation for social networking sites. In: IEEE International Conference on Trust, Security and Privacy in Computing and Communications, pp. 226–233 (2011)Google Scholar
  7. 7.
    Lin, D., Wede, J., Sundareswaran, S.: Privacy policy inference of user-uploaded images on content sharing sites. IEEE Trans. Knowl. Data Eng. 27(1), 193–206 (2015)CrossRefGoogle Scholar
  8. 8.
    Madejski, M., Johnson, M., Bellovin, S.M.: A study of privacy settings errors in an online social network. In: IEEE International Conference on Pervasive Computing and Communications Workshops, pp. 340–345 (2012)Google Scholar
  9. 9.
    Naini, K.D., Altingovde, I.S., Kawase, R., Herder, E., Niederée, C.: Analyzing and predicting privacy settings in the social web. In: Ricci, F., Bontcheva, K., Conlan, O., Lawless, S. (eds.) UMAP 2015. LNCS, vol. 9146, pp. 104–117. Springer, Cham (2015). Scholar
  10. 10.
    Taboada, M., Brooke, J., Tofiloski, M., Voll, K., Stede, M.: Lexicon-based methods for sentiment analysis. Comput. Linguist. 37(2), 267–307 (2011)CrossRefGoogle Scholar
  11. 11.
    Toch, E.: Crowdsourcing privacy preferences in context-aware applications. Pers. Ubiquitous Comput. 18(1), 129–141 (2014)CrossRefGoogle Scholar
  12. 12.
    Watson, J., Besmer, A., Lipford, H.R.: +your circles: sharing behavior on Google+. In: Eighth Symposium on Usable Privacy and Security, p. 12 (2012)Google Scholar
  13. 13.
    Wu, H.C., Luk, R.W.P., Wong, K.F., Kwok, K.L.: Interpreting TF-IDF term weights as making relevance decisions. ACM Trans. Inf. Syst. 26(3), 55–59 (2008)CrossRefGoogle Scholar
  14. 14.
    Xie, J., Knijnenburg, B.P., Jin, H.: Location sharing privacy preference: analysis and personalized recommendation. In: International Conference on Intelligent User Interfaces, pp. 189–198 (2014)Google Scholar
  15. 15.
    Yeung, C.M.A., Kagal, L., Gibbins, N., Shadbolt, N.: Providing access control to online photo albums based on tags and linked data. In: AAAI-SSS: Social Semantic Web (2011)Google Scholar
  16. 16.
    Yuan, L., Theytaz, J., Ebrahimi, T.: Context-dependent privacy-aware photo sharing based on machine learning. In: IFIP International Conference on ICT Systems Security and Privacy Protection, pp. 93–107 (2017)CrossRefGoogle Scholar
  17. 17.
    Zerr, S., Siersdorfer, S., Hare, J., Demidova, E.: Privacy-aware image classification and search. In: International ACM SIGIR Conference on Research & Development on Information Retrieval, Portland, Oregon, pp. 35–44 (2012)Google Scholar
  18. 18.
    Zhao, Y., Ye, J., Henderson, T.: Privacy-aware location privacy preference recommendations. In: International Conference on Mobile and Ubiquitous Systems: Computing, NETWORKING and Services, pp. 120–129 (2014)Google Scholar

Copyright information

© ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2018

Authors and Affiliations

  • Lijun Chen
    • 1
  • Ming Xu
    • 1
    Email author
  • Xue Yang
    • 1
  • Ning Zheng
    • 1
  • Yiming Wu
    • 2
  • Jian Xu
    • 1
  • Tong Qiao
    • 2
  • Hongbin Liu
    • 1
  1. 1.Internet and Network Security Laboratory, School of Computer Science and TechnologyHangzhou Dianzi UniversityHangzhouChina
  2. 2.School of CyberspaceHangzhou Dianzi UniversityHangzhouChina

Personalised recommendations