Others

An InnovativeConcept for Finding Friends in Social Networks based on their Lifestyles

Description
Existing social network services provide list of friends to users based on their request given. But it will not fulfill the user’s preferences in real life. Due to overloaded memory of the server memory size increases and lacking its efficiency. By
Categories
Published
of 3
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
  IJIRST  –  International Journal for Innovative Research in Science & Technology| Volume 1 | Issue 10 | March 2015   ISSN (online): 2349-6010 All rights reserved by www.ijirst.org    25 An Innovative Concept for Finding Friends in Social Networks based on their Lifestyles  Mrs.M.Kanchana S.P.Adaikkammai  Department of  Information Technology Department of  Information Technology    Karpaga Vinayaga College of Engineering and Technology Karpaga Vinayaga College of Engineering and Technology   V.Amirtham    Department of  Information Technology  Karpaga Vinayaga College of Engineering and Technology Abstract   Existing social network services provide list of friends to users based on their request given. But it will not fulfill the u ser’s  preferences in real life. Due to overloaded memory of the server memory size increases and lacking its efficiency. By implementing the Latent Dirichlet Allocation Algorithm we extracting their lifestyles and sensing the similarity of lifestyles  between users by using embedded sensors in the smartphones. Based on friend matching graphs we return a list of people with highest similarity of lifestyles. Feedback mechanism is integrated in this friendbook to get the results of users in choosing friends. We have implemented friendbook on the Android-based smartphones and evaluated its performance on both small scale experiments and large scale simulations. Finally, we reduce the memory size of the server and improving its performance. Keywords: social network services, Latent Dirichlet Allocation Algorithm, Android-based smartphones.  _______________________________________________________________________________________________________ I.   I NTRODUCTION   In olden days, people made friends in office (or) in working area and with their neighbours only because they get to know them  by the Geogr  aphical distances between each other. We can’t recommend a good friend user by using the existing social network services because fake people (or) duplication can be there and some people lifestyles will not match to the users. In that services, server will contain all the information about the users and get overloaded, while retrieving some information from the server it takes time. The main drawback is, it consuming much amount of time and lacking its efficiency. II.   R  ELATED WORK    In Facebook-statistics[1] people get started to using the service provided by social networks such as Facebook, Twitter, Google+ and they get a new way to making known and unknown friends in their friend list. J.Biagioni and J.Eriksson[2]GPS Traces collected from smartphones that are installed on transit vehicles to determined routes served, locate stops and infer schedules.K.Farrahi and D.Gatica-perez [3]proposed that they use the two wearable sensors but not smartphones to discover the daily routines J.Kwon [4] told that recommending friends based on current location of users has become increasingly popular in LMSN(location based mobile social networks).But they did not explain about what the physical and social contextZ.Wang, C.E.Taylor, [5]told that mobile social services utilize profile matching based on the preserving encryption to find friends with similar social attributes. But it is difficult to secure and information leakage problem also obtained. Personality plays an important role in advertisement and marketing environment but to detecting the personality from multiple social networks is difficult. M.Fritz and B.Schiel[6] proposed that they tried to discover daily location data they could not discover the daily routines of people who are staying at the same location. A.D.Sarma and A.R.Molla [7] Used iterative matrix vector multiplication method.X.Yu and A.Pan[8] uses GPS Data to understand the transportation mode of uses. L.Bian and H.Holtzman [9] presented matchmaker, a collaborative filtering friend recommendation system based on personality matching. III.   P ROPOSED SYSTEM   People lifestyles (e.g., attitude, taste, economic level, etc…) are extracted by implementing the latent Dirichlet Allocation  Algorithm to create a document. We are using the embedded sensors such as GPS, Accelerometer, Microphone, Gyroscope and Camera in the smartphones. That sensor will measure the similarity of lifestyles between users. We are using the friend matching graph to return a list of friends with highest similarity of lifestyles in an order. In this friendbook there is a feedback mechanism to further improve the recommendation accuracy. Then we are compressing the document size with the help of Hadoop tool, to reduce the memory size of the server and increase the performance of the server. Fig.1.Explaining about the Architecture of the Proposed system that such that Collecting the Data from the user and creating a document, getting user query, comparing   An Innovative Concept for Finding Friends in Social Networks based on their Lifestyles   (IJIRST/ Volume 1 / Issue 10 / 007) All rights reserved by www.ijirst.org    26 lifestyles of the users, provide ranking based on their lifestyles with the help of friend matching graph and providing feedback control. Fig. 1: .An analogy between word documents and people’s daily lives.   IV.   M ODULES      User Module    Messages for User Module    User Behavior Module    User Topic Awareness User Module: A. There are three different user roles in friendbook: System Administrator (server), Team members, Individual Person Messages for User Module: B. The messages posted in the course discussion forum can be roughly divided into three categories: The first is the small conversation which is not relevant to course content (such as self-introduction, or to initiate a study group).The second is the questions about the course arrangements. (Example: what they Download and Upload).The third is Question &Answer about the course-related content, so it is the most important and informative. In this paper, we only focus on messages belonging to the third category User Behaviour Model: C. As we know, some people prefer to ask more questions while some people prefer to answer more questions. We propose the user model to represent user’s behavior and classify users into three classes:  Class 1 (prefer to ask questions),Class 2(Normal Class), Class 3 (prefer to answer questions).Obviously, recommending an Class 3 person to a Class 1 person is a more effective recommendation and will maximize the understanding for both person. User Topic Awareness: D. To recommend friends for a person, we should also consider the relevance of person’s document in addition to the behavior characteristics of individual person. Therefore, we proposed a topic model to measure the similarity of lifestyles between the  people. The data flow diagram shows that the process of user queries, data collection, feedback mechanism for the user.   An Innovative Concept for Finding Friends in Social Networks based on their Lifestyles   (IJIRST/ Volume 1 / Issue 10 / 007) All rights reserved by www.ijirst.org    27 Fig. 2: Data Flow Diagram. V.   C ONCLUSION   In this paper, we presented the An Innovative Concept for finding friends in Social Networks Based on Their Lifestyles. One challenge with existing social networking services is how to recommend a good friend to a user. Most of them rely on pre-existing user relationships to pick fri end candidates. Friend book is the first friend recommendation system exploiting a user’s life style information discovered from smart phone sensors. We integrate a linear feedback mechanism that exploits the user’s feedback to improve recommendation accuracy. Then we are compressing the document size with the help of Hadoop tool, to reduce the memory size of the server and increase the performance of the server. R  EFERENCES   [1]   Facebook statistics. http://www.digitalbuzzblog.com/ facebook-statistics-stats-facts-2011/. [2]   J.Biagioni, T. Gerlich, T. Merrifield, and J. Eriksson. EasyTracker: Automatic Transit Tracking, Mapping, and Arrival Time Prediction Using S martphones. Proc. of SenSys, pages 68-81, 2011. [3]   T. Huynh, M. Fritz, and B. Schiel. Discovery of Activity Patterns using Topic Models. Proc. of UbiComp, 2008. [4]   J. Kwon and S. Kim. Friend recommendation method using physical and social context. International Journal of Computer Science and Network Security, 10(11):116-120, 2010 [5]   Z. Wang, C. E. Taylor, Q. Cao, H. Qi, and Z. Wang. Demo: Friendbook: Privacy Preserving Friend Matching based on Shared Interests. Proc. of ACM SenSys, pages 397-398, 2011. [6]   K. Farrahi and D. Gatica-Perez. Discovering Routines from Largescale Human Locations using Probabilistic Topic Models. ACM Transactions on Intelligent Systems and Technology (TIST), 2(1), 2011. [7]   A. D.Sarma, A. R.Molla, G.Pandurangan, and E.Upfal. Fast distributed computation. Springer Berlin Heidelberg, pages 11-26, 2013. [8]   Y. Zheng, Y. Chen, Q. Li, X. Xie, and W.-Y. Ma. Understanding Transportation Modes Based on GPS Data for Web Applications. ACM Transactions on the Web (TWEB), 4(1):1-36, 2010. [9]   L. Bian and H. Holtzman. Online friend recommendation through personality matching and collabor  ative filtering. Proc. of UBI - COMM, pages 230-235, 2011.
Search
Tags
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks