DETECTION AND REMOVAL OF PUBLISHING FAKE POSTS & USERS IRRELEVANT COMMENTS ON SOCIAL MEDIA
- 1. DETECTION AND REMOVAL OF PUBLISHING FAKE POSTS & USERS IRRELAVENT COMMENTS IN SOCIAL MEDIA PREPARED BY: Prasanth J (622516104037) Santhanakumar P (622516104046) Sukumar R (622516104052) GUIDED BY: Mrs. R.Bhuvaneswari, M.E., (Ph.D.), Head of the Department, CSE Date : 22-09-2020 Review : Final
- 2. ABSTRACT Social media can be a double-edged sword for society, either as a convenient channel exchanging ideas or as an unexpected conduit circulating fake news through a large population. Social nuisances that place heavy financial burdens on society. Here we look at use of data mining followed by sentiment analysis on online social networks. Here we use Sentiment analysis to detect the fake news by using the block chain technique. The fake news are reported by the user and the algorithm will check the data and add it to the database. The irrelevant comments are automatically reported by the algorithm. Since, it has a set of unwanted word which should not be used publicly.
- 3. EXISTING SYSTEM However, most of the existing studies assume that each piece of information spreads independently regardless of the interactions between contagions. User spread the post through the social media in which some of them are fake or not needed. The irrelevant comments reported by the users are blocked by the admins after reviewing the comments.
- 4. PROPOSED SYSTEM The fake news are reported by the user and the algorithm checks the data and blocks the content. Sentiment analysis was used to determine a writer's/speaker's attitude with respect to either a topic or the overall contextual polarity of a text. The irrelevant comments are blocked automatically by the algorithm. Since, it has preprocessed datasets.
- 5. HARDWARE REQUIREMENTS Processor : Intel processor 3.0 GHz and above RAM : 4GB and above Hard disk : 50 to 500 GB Keyboard : Any Standard Keyboard Mouse : Any Standard Mouse Monitor : 15 inch color monitor and above
- 6. SOFTWARE REQUIREMENTS Front End : PHP Back End : MySQL (Ver 5.6) Operating System : Windows or Linux OS System type : 32-bit or 64-bit Operating System
- 7. SYSTEM ARCHITECTURE Data Acquisition Datasets Preprocessing Stop Removal Stemming words analysis Tokenization Hybrid Segmentation Global Context Local Context Pseudo Feedback POS tagger Named Entity Recognition Network Features Content Features Blog Features SVM classifiers Trained Datasets Rumors identification
- 8. LIST OF MODULES Registration Login Approved Update Status Check status Block Id Find location Logout
- 9. MODULES DESCRIPTION REGISTRATION This module describes details of the users. When a new user arrives the application user must enter the personal details such as user name, user age, package detail, no of days, and user contact details (address, telephone number). The user registers the details and gets consumer id and password. LOGIN This module describes the details of login. In this module the admin and consumer have an own unique login id and password. Login is used for protect the unauthorized access.
- 10. APPROVED The users register the detail then the admin has check the detail and approve the id. UPDATE STATUS The user updates the status of the social media then the admin the check the status the added to the status of social media. CHECK STATUS The user can update the status then main work of admin has checked the status. To check the status categories. The status has to a normal content to allow the update the status but the status has to unwanted categories so the status could not updated
- 11. BLOCK ID To the user can update the unwanted status initially admin send the warning message of user the user cant accepted that user do that recently the account has been blocked. FIND LOCATION To the user can update the unwanted status Then the account has to be blocked and find the user location. LOGOUT After complete the booking process the user can logout the process.
- 12. ALGORITHM Support Vector Machine(SVM) A Support Vector Machine (SVM) is a discriminative classifier formally defined by a separating hyperplane. In other words, given labeled training data (supervised learning), the algorithm outputs an optimal hyperplane. In this project we use SVM algorithm for classifying and fetch the data.
- 13. SCREENSHOT
- 14. User Registration Page
- 15. Admin Login Page
- 16. User Profile Page
- 17. User Post Page
- 18. Admin Page
- 19. CONCLUSION A crime pattern can be detected, nearly in real-time, when online social media is monitored. Crime can occur anywhere at any time. Previous statistics do not accurately identify the crime intensity of a specific location. More accurate results can be drawn from social media. Results from geographic data analysis conducted on various tweets provided a clear picture of the criminal trends in several different cities. The crime intensity day-wise positively correlated with crime statistics from cops, which ultimately prove the hypothesis. The Ferguson shooting case study clearly differentiates the city's safe and dangerous pattern
- 20. LITERATURE SURVEY S.NO TITLE AUTHORS ALGORITHM JOURNAL LIMITATION 1 IAD: Interaction- Aware Diffusion Framework in Social Networks{2018} C. Gentry, S. Halevi, and N. P. Smart routing algorithms arXiv:1709.01773 Forecast the average distance 2 Mining aspect- specific opinion using a holistic lifelong topic model{2016} E. Aktas, F. Afram, and K. Ghose Dummy- Location Selection (DLS) Proc. 25th Int. Conf. World Wide Web, pp. 167–176 Chance for lose of data
- 21. 3 TASC:topic- adaptive sentiment classification on dynamic tweets{2015} C. Gentry, S. Halevi, and N. P. Smart mixed-integer nonlinear program (MINLP) EEE Trans. Knowl. Data Eng., vol. 27, no. 6, pp. 1696– 1709 Limitations on recognition and its performance in behavioral verification 4 Modeling information diffusion over social networks for temporal dynamic prediction{2017} E. Aktas, F. Afram, and K. Ghose wireless service providers (WSPs IEEE Trans. Knowl. Data Eng, vol. PP, no. 99, pp. 1–1 Detect the time But don’t detect a location
- 22. REFERENCES Xi Zhang, Member, Yuan Su, Siyu Qu, Sihong Xi, Member, Binxing Fang, and Philip S. Yu, Fellow, “Interaction-Aware Diffusion Framework in Social Network,” IEEE Trans. Knowl. Data Eng. Vol. 14, no. 8, August 2018. D. Li, S. Zhang, X. Sun, H. Zhou, S. Li, and X. Li, and X. Li, “Modeling information diffusion over social networks for temporal dynamic prediction,” IEEE Trans. Knowl. Data Eng. Vol PP, no. 99, pp. 1-1,2017. R. Fan, J. Zhao, Y. Chen, and K. Xu, “Anger is more influential then joy: Sentiment correlation in weibo,” PloS one, vol. 9, no. 10, p. e110184, 2014. J. Zhang, B. Liu, J. Tang, T. Chen, and J. Li, “Social influence locality for modeling retweeting behaviors,” in proc. 23th Int. Joint Conf. Artificial intell., 2013.
- 23. THANK YOU