Secure Image Sharing on Shared Web Sites Using Adaptive Privacy Policy Prediction System Miss Akanksha R.Watane Prof.P.B.Sambhare Abstract: - Many social media sites like Facebook, flicker are performing the vast amount of image uploading and storage mechanism. As the amount of uploading private images is growing on increasing that leads to the privacy concern. This need the tools to help users control access to their shared images. There are different privacy preserving mechanism applied by many researchers for secure social networking are studied in the literature. But all none of the methods seems perfect and performing secure social networking. So to overcome the drawbacks and make a secure social networking an Adaptive Privacy Policy Prediction (A3P) system is proposed. The system is develop for both user s interest and their security. The security is provided by uploading images by encrypting them. Sharing gets secure by earlier category wise classification of both users and images. Our experimental result proves that our encryption technique and classification method used for providing security offers significant improvements over current approaches available to provide privacy for users private images. Keywords: Social media, Content sharing sites, Privacy Preservation, Meta data, Adaptive Privacy Policy Prediction (A3P). I. INTRODUCTION IMAGES are now one of the key enablers of users connectivity. Most content sharing websites allow users to enter their privacy preferences. Recent studies have shown that users struggle to set up and maintain such privacy settings [1], Today, for every single piece of content shared on sites like Facebook every wall post, photo, status update, and video the up loader must decide which of his friends, group members, and other Facebook users should be able to access the content. Creating privacy controls for social networks that are both expressive and usable is a major challenge. Lack of user understanding of privacy settings can lead to unwanted disclosure of private information and, in some cases, to material harm. We propose a new paradigm which allows users to easily choose suites of privacy settings which have been specified by friends or trusted experts, only modifying them if they wish. Modern social network and services have become an increasingly important part of how users spend their time in the online world. The social network is a proper vehicle for people to share their interests, thoughts, pictures, etc. with their friends or the public. While sharing information about the self is intrinsically rewarding [2], the risk of privacy violation increases due to disclosing personal information [3,]. Recent cases, such as Canada's Privacy Commissioner challenge to Facebook's privacy policies and settings, have shown a growing interest on the part of the public with respect to how social network and services treat data entrusted to them. Some of the privacy violation incidents could be mitigated or avoided if people used more privacy setting options [4]. In this project propose an Adaptive Privacy Policy Prediction(A3P) system which aims to provide users a hassle free privacy settings experience by automatically generating personalized policies. The proposed A3Psystem is comprised of two main building blocks A3P-Social and A3P-Core. TheA3P-core focuses on analyzing each individual user s and their own images and metadata, while the A3P-Social offers a community perspective of privacy setting recommendations for a user s potential privacy improvement. II. LITERATURE SURVEY In 2009 Jonathan Anderson proposed a paradigm called Privacy Suites [5] which allows users to easily choose suites" of privacy settings they are using privacy programming. The disadvantage of a rich programming language is less understandability for other users. Given a sufficiently high-level language and good coding practice, motivated users should be able to verify a Privacy Suite. Main goal is transparency, which is essential for convincing influential users that it is safe to use. In 2012 Peter F. Klemperer developed a tag based access control of data [6] shared in the social media sites. A system that creates access-control policies from photo management tags. Every photo is incorporated with an access grid for mapping the photo with the participant s friends. The participants can select a suitable preference and access the information. Photo tags can be categorized as organizational or communicative based on the user needs. There are several important limitations to our study design. First, our results are limited by the participants we recruited and the photos they provided. A second set of limitations concerns our use of machine generated access-control rules. The algorithm has no access to the context and meaning of tags @IJRTER-2016, All Rights Reserved 347
and no insight into the policy the participant intended when tagging for access control. As a result, some rules appeared strange or arbitrary to the participants, potentially driving them toward explicit policybased tags like private and public. In 2013 Kambiz Ghazinour designed a recommender system known as YourPrivacyProtector [7] that understands the social net behavior of their privacy settings and recommending reasonable privacy options. It uses user s personal profile, User s interests and User s privacy settings on photo albums as parameters and with the help of these parameters the system constructs the personal profile of the user. It automatically learned for a given profile of users and assign the privacy options. It allows users to see their current privacy settings on their social network profile, namely Facebook, and monitors and detects the possible privacy risks. Based on the risks it adopts the necessary privacy settings. In 2015 Anna Cinzia Squicciarini developed an Adaptive Privacy Policy Prediction (A3P) [8] system, IT is automatically generating personalized policies. The A3P system handles user uploaded images based on the person s personal characteristics and images content and metadata. In this system consists of two components: A3P Core and A3P Social. When a user uploads an image, the image will be first sent to the A3P-core. The A3P-core classifies the image and determines whether there is a need to invoke the A3P-social. Disadvantage of this system is inaccurate privacy policy generation in case of the absence of meta data information about the images. Also manual creation of meta data log data information leads to inaccurate classification and also violation privacy. III. SYSTEM ARCHITECTURE Fig. 3.1 Architecture diagram of working system Above is the architecture diagram of project. Basic input to this system are Users and Images and out generated from our system are predicted group of users and analysis of images. As we have used A3p privacy prediction system which mainly consist of two main Building blocks1) A3p Core and 2) A3p Social The first part AP3-core has input as category wised classified images and users. This this will goes to the second part as A3P-social for the purpose of social networking functionality and particular user selection for image sharing activity. And in the output functionality after making our friends, uploading images and sharing to respective groups it again comes to A3P core for prediction of particular users and images. All this activities are perform by keeping use of backend as Database and the space provide to all individual user for uploading our images @IJRTER-2016, All Rights Reserved 348
IV. PRPPOSED METHODOLOGY Fig.4.1 Workflow of propose system In this system propose a hierarchical classification of friends much initially at the time adding friends to our friend list. In System perform this by selection the category first and then sending the friend request to particular user. This will classify that user as our friend of that particular category after accepting the friend request. This classification is again repeated at the time of uploading images. User has to select the particular group first in which that image is belonging too, then he is able to browse and upload particular images. Now the uploaded image is securely stored to our space as we have apply encryption technique to it. Now when the time comes to share images, if anyone can come to share any image uploaded on out portal. It will get share to only that respective category friends which we have selected earlier at the time of uploading it. This is our basic idea to perform secure image sharing in our develop social network. After that we have perform analysis of each image that can come to our portal and shared by any of our friend or by self. For this we have use the concept of Natural Language Processing (NLP). Along with this we have perform prediction of individual user according to his likes and comments made on images comes to his portal. V. ALGORITHAM Data Encryption Standard (DES) The Data Encryption Standard (DES) is a symmetric-key algorithm for the encryption of electronic dataand published by the National Institute of Standards and Technology (NIST)[9].DES is an implementation of a Feistel Cipher. It uses 16 round Feistel structure. The block size is 64-bit. Though, key length is 64-bit, DES has an effective key length of 56 bits, since 8 of the 64 bits of the key are not used by the encryption algorithm (function as check bits only). Fig.5.1 of DES algorithm @IJRTER-2016, All Rights Reserved 349
Since DES is based on the Feistel Cipher, all that is required to specify DES is Round function Key schedule Any additional processing Initial and final permutation Initial and Final Permutation The initial and final permutations are straight Permutation boxes (P-boxes) that are inverses of each other. They have no cryptography significance in DES. Round Function The heart of this cipher is the DES function, f. The DES function applies a 48-bit key to the rightmost 32 bits to produce a 32-bit output. Fig 5.2 DES algorithm Expansion Permutation Box Since right input is 32-bit and round key is a 48-bit, we first need to expand right input to 48 bits. XOR (Whitener). After the expansion permutation, DES does XOR operation on the expanded right section and the round key. The round key is used only in this operation. Substitution Boxes. The S-boxes carry out the real mixing (confusion). DES uses 8 S-boxes, each with a 6-bit input and a 4-bit output. There are a total of eight S-box tables. The output of all eight s-boxes is then combined in to 32 bit section. Straight Permutation The 32 bit output of S-boxes is then subjected to the straight permutation Key Generation - The round-key generator creates sixteen 48-bit keys out of a 56-bit cipher key DES has proved to be a very well designed block cipher. There have been no significant cryptanalytic attacks on DES other than exhaustive key search. VI. Result Analysis Image no Image name Images Dimension Image Format Encrypted image time 1 aa 450*320 JPEG 0.010262 2 bb 525*350 JPEG 0.011389 3 cc 600*335 JPEG 0.013414 4 dd 640*427 JPEG 0.014425 5 ee 700*664 JPEG 0.020234 Table 6.1 encrypted image time @IJRTER-2016, All Rights Reserved 350
0.013262 0.0143890.0154140.016425 0.018234 0 Image format JPEG JPEG JPEG JPEG JPEG Images dimension 450*320525*350600*335640*427700*664 imag name aa bb cc dd ee img no. 1 2 3 4 5 Fig 6.1: graph of encrypted image VII. CONCLUSION & FUTURE SCOPE The security is provided by uploading images by encrypting them.sharing gets secure by earlier category wise classification of both users and images. Our experimental study prove that our encryption technique and classification method used for providing security offers significant improvements over current approaches to privacy. As the social networking is the vast area of users interest as seen from real world scenario and study performs in the literature. There is always space for improvement at different the facilities available / develop in social network. The develop system is not performing different activities needs to be available for social networking as like chatting and messaging facilities. Along with this, some GUI interfacing along with animation effects are necessary to attract user s interest to use our social networking application needs to develop. So the future scope of this application is mainly to develop all the functionality needs for social networking. Along with more eye catching interface and animation to help users in performing social networking. REFERENCES 1. A. Acquisti and R. Gross Imagined communities: Awareness, information sharing, and privacy on the facebook, in Proc.6th Int. Conf. Privacy Enhancing Technol. Workshop, 2006,pp. 36 58. 2. D. I. Tamir and J. P. Mitchell. (2012). Disclosing information about the self is intrinsically rewarding. PNAS 2012: 1202129109v1-201202129. 3. K. Ghazinour, M. Sokolova, and S. Matwin. "Detecting Health-Related Privacy Leaks in Social Networks Using Text Mining Tools." Advances in Artificial Intelligence. Springer Berlin Heidelberg, 2013. 25-39. 4. K. Lewis, J. Kaufman, and N. Christakis. (2008) The Taste for Privacy: An Analysis of College Student Privacy Settings in an Online Social Network. JCMC 14, 1. 5. J. Bonneau, J. Anderson, and L. Church, Privacy suites: Shared privacy for social networks, in Proc. Symp. Usable Privacy Security,2009. 6. Peter F. Klemperer, Yuan Liang, Michelle L. Mazurek, Tag, You Can See It! Using Tags for Access Control in Photo Sharing, Conference on Human Factors in Computing Systems, May 2012. 7. Kambiz Ghazinour, Stan Matwin and Marina Sokolova, Yourprivacyprotector: A Recommender System For Privacy Settings In Social Networks, International Journal of Security, Privacy and Trust Management ( IJSPTM) Vol 2, No 4, August 2013. 8. Anna Cinzia Squicciarini, Privacy Policy Inference of User-Uploaded Images on Content Sharing Sites, IEEE Transactions On Knowledge And Data Engineering, vol. 27, no. 1, January 2015. 9. Diffie, Whitfield; Hellman, Martin E. (June 1977). "Exhaustive Cryptanalysis of the NBS Data Encryption Standard" (PDF). Computer. 10 (6): 74 84. doi:10.1109/c-m.1977.217750. Archived from the original (PDF) on 2014-02-26. @IJRTER-2016, All Rights Reserved 351