53
International Journal of Innovative and Emerging
Research in Engineering
e-ISSN: 2394 - 3343 p-ISSN: 2394 - 5494
Result On Social Networking Application Evaluator: Detecting
Malicious Social Networking Applications
Mr. Sumit Notawale,Prof. Juned A. Khan
Department of Computer Science and Engineering G H Raisoni College of Engineering & Management Amravati
Abstract:
With large number of users installed, third-party apps was a major reason for the popularity and habitual of using the Social networking application. Recently, hackers have started misuse of this third-party apps platform and deploying malicious applications, spreading Malicious application and extracting the user’s personal profile information data from the application because Malicious apps can provide a good business for hackers. This was the major issue, as we find that numbers of applications in our dataset was malicious. So that, Our purpose were creating a Social networking application, through which we understand it was malicious or not? Our goal are to develop a Social Networking Application Evaluator Detecting Malicious Social Networking Application which will help in detecting malicious applications on Social networking websites.
Keywords: Third-party apps, Social networking application, Evaluator Detecting
I. INTRODUCTION
Online social networks (OSN) enable and encourage third party applications (apps) to enhance the user experience on these platforms. Such enhancements include interesting or entertaining ways of communicating among online friends, and diverse activities such as playing games or listening to songs.
Recently, hackers had started taking advantage of the popularity of this third-party apps platform and deploying malicious applications. Malicious apps could provide a lucrative business for hackers, given the popularity of OSNs(Online Social Networking’s), with the leading the way with 900M active users[13]. There was many ways that hackers can benefit from a malicious app: (a) the app could reach large numbers of users and their friends to spread spam, (b) the app could obtain users’ personal information such as email address, home town, and gender, and (c) the app is “re-produce" by making other malicious apps popular. To make matters worse, the deployment of malicious apps are simplified by ready-to-use toolkits starting at $25[14]. In other words, there was motive and opportunity, and as a result, there were many malicious apps spreading on every day[12].
Despite the worrisome trends, today, a user has very limited information at the time of installing an app on social networking apps. In other words, the problem was: given an app’s identity number, could we detect if the app are malicious? Currently, there can no commercial service, publicly-available information, or research-based tool to advise a user about the risks of an app, malicious apps were widespread and they easily spread, as an infected user jeopardizes the safety of all its friends.
We developed Social Networking Application Evaluator for detecting Malicious Social Networking Application, a suite of efficient classification techniques for identifying whether an app was malicious or not. To build Social Networking Application Evaluator Detecting Malicious Social Networking Application, For example we use data from MyPageKeeper, a security app in Social networks that monitors the Social networks profiles of 2.2 million users[15]. We analyze 111K apps that made 91 million posts over nine months. This was arguably the first comprehensive study focusing on malicious Social networks apps that focuses on quantifying, profiling, and understanding malicious apps, and synthesizes this information into an effective detection approach [8,15].
II. Literature Survey Detecting and characterizing social spam campaigns
54 Towards online spam filtering in social networks. Gao et al. [2] and Efficient and Scalable Socware Detection in Online Social Networks. Rahman et al. [3] develop efficient techniques for online spam filtering on Online Social Networking sites such as Facebook. In this paper, we describe our work to provide online spam filtering for social networks. We use text shingling and URL comparison to incrementally reconstruct spam messages into campaigns, which are then identified by a trained classifier.
Detecting suspicious url’s in twitter stream
S. Lee and J. Kim. [4] And Design and Evaluation of a Real-Time URL Spam Filtering Service. K. Thomas, C. Grier, J. Ma, V. Paxson, and D. Song. [5] Presents the mechanisms for detection of spam URLs on Twitter. In contrast to all of these efforts, rather than classifying individual URLs or posts as spam, we focus on identifying malicious applications that are the main source of spam on Facebook. Die free or live hard? Empirical evaluation and new design for fighting evolving twitter spammers. Yang et al. [6] and Detecting spammers on Twitter. Benevenuto et al. [7] developed techniques to identify accounts of spammers on Twitter.
Detecting spam in a twitter network
Yardi et al. [10] In this paper analyzed behavioral patterns among spam accounts in Twitter. Instead of focusing on accounts created by spammers, our work enables detection of malicious apps that propagate spam and malware by luring normal users to install them. Is this app safe? A large scale study on application permissions and risk signals. Chia et al. [11] investigated the privacy intrusiveness of Facebook apps and concluded that currently available signals such as community ratings, popularity, and external ratings such as Web of Trust as well as signals from app developers are not reliable indicators.
Analyzing facebook privacy settings: user expectations vs. reality
Y. Liu, K. P. Gummadi, B. Krishnamurthy, and A. Mislove. [12] In this paper showed that privacy settings in Facebook rarely match users’ expectations. In this paper only represent how the application can protect the malicious message or malicious post. So from this paper we are unable to protect the malicious applications in this paper they are failed to protect the application from the malicious applications. Characterization Of Osn applications Gjoka et al. [8] study the user reach of popular Facebook applications. On the contrary, we quantify the prevalence of malicious apps, and develop tools to identify malicious apps that use several features beyond the required permission set. In this paper also showing the privacy security techniques to protect the user private data and also analyzing the whether the third party application is malicious or not. It is a technique which interprets between the benign application and the malicious application and showing the malicious applications contents and the characteristics of the malicious applications.
III. PROPOSED WORK
This section explains flow of proposed systems which will be uses.
Figure: DFD of proposed work
User post/user link
Make dictionary of
black listed words
Check with black
listed keywords
Determine malicious
If malicious add to black
list, else post
Start
55 When user’s inserting or posting the statement that time our system providing the check post, the check post will identifying that inserted post is statement or link. After identifying things the system forwarding control to next level. In this section it is also important to understand the type of posting, there is several kind of posting we can do on the wall of social networking sites. After identifying whether it is link or statement of post, here systems will applying the data mining techniques and web mining algorithm. System will be using data mining classifications techniques to extraction of words from the statement of the post, and web mining algorithm system using for web links to collect the information’s of the web link, and after collecting all information’s here system will checking out that particular web links or statements is malicious or not.
IV. IMPLEMENTATION
Figure: Flow Chart of proposed work
start
Stop Insert post
Is post link?
URL Contents
Black listed keywords
Malicious percentage
Extract information
If Percenta ge>67.33
Published
Admin
56 When user’s inserting or posting the statement that time our system providing the check post, the check post will identifying that inserted post is statement or link. After identifying the things the system forwarding control to next level and step wise system will be executed. Following steps involved.
Step 1: Inserting the post on the users time line or wall randomly. Step 2: Make dictionary(D) for word of post.
Step 3: Get blacklisted keywords from database(db).
Step 4: Compare dictionary(D) with keywords of database(db). Step 5: If matched increment B(count) by one.
Step 6: Determine malicious percentage(P)P ═ [∑ B(count) ∕ total no. of words in dictionary(D)]*100. Step 7: If P>67.33 consider as blacklisted post.
The system also having the provision to collect data in to back end like storing the blacklisted links into the data base, so that in future if any link is found as same as which is having the similar characteristics of the link, it will be identified immediately and also it will be very helpful to reduced the complexity and increasing the accuracy of the system to detect and evaluate the malicious post of the system.
V. RESULTS
In table showing analysis of graph and considered user U1, U2, U3, U4. Here showing the comparisons of malicious percentage of post and link of every user, these all aspect depends on the matching blacklisted keywords which is presents in the database.
Table: Comparison of link and keywords percentage
There is a lot differences between the malicious post and the malicious links, consider user U1 which showing the 50.25% for malicious post, and 46.66 for the malicious links which showing the both things are totally different from each other. Also by comparisons we can come to know the accuracy of the system.
If we are keeping more no. of blacklisted in the database table then our system giving more accuracy and reducing the system complexity and also we have the facility in the system that will automatically adding the blacklisted keywords directly into database tables, in the graph showing that how the system giving the comparatively results of the malicious links and post.
Figure: Comparison of link and post maliciousness
0 10 20 30 40 50 60 70
U1 U2 U3 U4
Malicious post
Malicious Link's
Users Post Malicious(%)
Link Malicious(%)
U1 50.25 46.66
U2 42.85 65.33
U3 59.42 55.33
57 In above graph, blue line is showing that malicious post of the user and red line shows malicious links, for user U1 it contains the 50% of the malicious post while for same users showing the malicious links that is approximately 46.66%. So from this parameter of comparison we can understand that ratio of malicious percentage of post and links are depending on the malicious keywords.
VI. Conclusion
Applications present a convenient means for hackers to spread malicious content on Social networks. However, little is understood about the characteristics of malicious apps and how they operate. In this work, using a large corpus of malicious Social networks app observed over a nine month period, we showed that malicious apps differ significantly from benign apps with respect to several features. For example, malicious apps are much more likely to share names with other apps, and they typically request less permission than benign apps. Leveraging our observations, we developed Social Networking Application Evaluator: Detecting Malicious Social Networking Applications, an accurate classifier for detecting malicious Social networks applications. Most interestingly, we highlighted the emergence of large groups of tightly connected applications that promote each other.
References
[1] H. Gao, C. Wilson, Z. Li, Y. Chen, and B. Y. Zhao, “Detecting and characterizing social spam campaigns,” Internet Measurement Conference, Melbourne, Australia, Volume 4, pp 35-87,
November 2010.
[2] Y. Zhang, J. Hong, and L. Cranor, “CANTINA: A Content-Based Approach to Detecting Phishing Web Sites,” In Proceedings of the International World Wide Web Conference (WWW), Banff, Alberta, Canada, Volume4, pp 639-648, May 2007.
[3] F. Benevento, G. Mango, T. Rodrigues, and V. Almeida, “Detecting spammers on Twitter,” In Collaboration Electronic messaging Anti-Abuse and Spam Conf. (CEAS), Volume 2, pp 140-148, 2010. [4] S. Lee and J. Kim, “Warningbird: Detecting suspicious urls in twitter stream,” In Network and
Distributed System Security (NDSS), pp 42-48, 2012.
[5] K. Thomas, C. Grier, J. Ma, V. Paxson, and D. Song, “Design and Evaluation of a Real-Time URL Spam Filtering Service,” In Proceedings of the IEEE Symposium on Security and Privacy, pp 447-462, 2011 ISSN: 1081-6011.
[6] C. Yang, R. Harkreader, “Die free or live hard? Empirical evaluation and new design for fighting evolving twitter spammers,” Research in Attacks Intrusions and Defenses, pp 1-20, 2011.
[7] K. Rieck, T. Krueger, and A. Dewald, “CUJO: Efficient Detection and Prevention of Drive-by-Download Attacks,” In Proceedings of the Annual Computer Security Applications Conference (ACSAC), pp- 208-15, 2010.
[8] Priyanka Vijay Patankar, Prof. Smita Jangale, “Malicious Web Crawler Detection Using Intrusion Detection System,” International Journal of Modern Trends in Engineering and Research, Volume 3, pp.365-371, 2016.
[9] R E Fan, K W Chang, C J Hsieh, X R Wang, and C J Lin, “LIBLINEAR: A library for large linear classification,” Journal of Machine Learning Research, Volume 9, pp1871–1874, 2008.
[10] V. S. Dhaka, Sanjeev Kumar Singh, “Web Crawler: A Review,” International Journal of Computer Applications, Volume 63– No.2, pp 0975 – 8887 February 2013.
[11] W. Enck, M. Ongtang, and P. McDaniel, “On lightweight mobile phone application certification,” In Proc. of the 16th ACM conf. on Computer and Communications Security, Volume 09, pp 235–245, 2009. [12] Y. Liu, K. P. Gummadi, B. Krishnamurthy, and A. Mislove, “Analyzing facebook privacy settings: user
expectations vs. reality,” In Internet Meas urement Conference, pp 61-70, 2011.
[13] Facebook tops 900 million users. http://money.cnn.com/2012/04/23/technology/facebook-q1/index.htm. [14] Hackers selling $25 toolkit to create malicious Facebook apps. http://zd.net/g28HxI.
