Searching a Social Networking Site – Shift from Centralized Database to Metadata

DOI : 10.17577/IJERTV3IS090525

Download Full-Text PDF Cite this Publication

Text Only Version

Searching a Social Networking Site – Shift from Centralized Database to Metadata

Aamir Junaid Ahmad Assistant Professor,

Department of Computer Science & Engineering, Maulana Azad College of Engineering & Technology, Patna, India

Sabina Priyadarshini Assistant Professor,

Department of Information Technology Birla Institute Of Technology, Mesra, India

Abstract Social Networking is becoming part of every individuals life be it students, research scholars, professionals from different domains or retired people living at home with their grandchildren. Everyone wants to share their information with their friends. The sociological theory says that social interaction creates similarity and vice versa, interaction creates similarity. We often want to look for a person belonging to a particular group or characteristics (similar to our profile) but land up with either a lot of information or no information at all. We can connect only with people who are known to us by name or the people who have been suggested by the system. Here we present a method using Semantic Web to represent and process social network information so that user-queries can be better answered. We have done a survey to know what are the information that members in Social Networking sites usually keep in their profile and how concerned are they about the privacy of their data.

Keywords FOAF, Resource Description Framework, Semantic Web, Social Networking, XML

INTRODUCTION

Social networking provides access to personal information sharing and online socialization. These sites allow members to create a profile with their basic information like contact details, education details, professional or work related details. The members can link to the profiles of their friends and invite others. They can browse the resulting social network in order to connect with common friends, find lost friends and to create new friendships based on shared interests.

Social Networking is an easy way for people to have a social connection with other people with similar interests. These sites serve communication purposes among special interest groups, but they do not have a proper searching option where we can look for people or groups having certain characteristics.

The following three searches in Facebook (02-May-2014)

couldnt give any result

  1. Assistant Professors in the Computer Science Department between 30 and 40 years of age

  2. Assistant Professors in Computer Science Departme nt who go to BIT Mesra

  3. Assistant Professors in the Computer Science department

Query 1. Assistant Professors in the Computer Science department between 30 and 40 years

The output is : Sorry we couldnt find any results for this

search

  1. Assistant Professors in Computer Science Department who go to BIT Mesra

    The output is : Sorry we couldnt find any results for this

    search

  2. Assistant Professors in the Computer Science department

    All of these results are not as expected by the user.

    Methodology

    The limitation in the present networking sites is the centralized storage of user profiles. The information is bound to the site in which the profile was created. It can only be accessed in the way the database owner has developed the site. This problem has been addressed with the use of Semantic Web technology. The Semantic Web can be used to improve the process of searching the relevant information from the vast store of continually updated information, as well as emergent data about individual. It is an effort to enhance current web so that computers can process the information, interpret and connect it, to help users to find required information.

    The objective of this research is to use Semantic Web technology to represent and process social network information (user profiles and friendship network) in such a way that it can be processed by machine. The RDF[1] and FOAF[2], Web Semantic technologies[3] can be used for such representation.

    The FOAF (friend-of-a-friend) initiative undertaken by the World Wide Web Consortium (W3C) has focused on developing ways to describe the properties of people such as date of birth, age, real name, nick name, contact details, professional details, educational details and their social relationships as expressed through their interests, people known, liking etc. Socially, the most important property in FOAF is foaf:knows which can be used in forming a FOAF- web.

    Some of the most common properties are the following:

    1. foaf: name – Full name of the person.

    2. foaf:firstname – Individuals first name.

    3. foaf: surname – Individuals family name.

    4. foaf: nick – provides a nickname for a particular person.

    5. foaf: homepage – specifies a link to the persons homepage, e.g. http://www.macet.net.in/aamir/

    6. foaf: phone – Individuals phone number.

    7. foaf: school – Link to the school homepage this individual is studying at.

    8. foaf: gender – provides gender information

    9. foaf: knows – specifies a person that this person knows.

    10. foaf: depiction – points to a URL of an image of the person in question.

Table 1

FOAF profiles are created and controlled by the individual user and shared in a distributed fashion. Much like the way web pages are linked to each other by anchors, these profiles link to the profiles of friends by using the knows and seeAlso properties, creating the FOAF-web.

FOAF IS AN RDF (RESOURCE DESCRIPTION FRAMEWORK) VOCABULARY. THE GOAL OF FOAF IS TO PROVIDE A MEANS FOR DESCRIBING PEOPLE AND THEIR PERSONAL INFORMATION SUCH AS NAME, E-MAIL ADDRESS AS WELL AS SOCIAL LINKAGE

TO THEIR FRIENDS BY USING XML AND RDF.

The first challenge in using FOAF for displaying information is the availability of basic user information on the social networking site. The second important issue is users privacy in making their information public in the FOAF format. Since some information in a persons FOAF document may include the private information which the user doesnt want to make public, as the metadata extracted from the web will be given some sort of publicity. It is therefore important to know what are the information which the users usually keep in their profile and secondly how concerned are users about the privacy of the following information they submit on social networking sites[4]. To reach to the conclusion a survey was conducted in August 2014.

Survey Methodology: A questionnaire was prepared and distributed through

  • Email

  • Online Form

  • Distribution in Classroom and

  • Distribution in market

Total number of individuals who returned or gave response is 2120

The individuals range from 18 years old to retired persons.

Question: Please indicate what information you include on your social networking sites

From the above survey(Table 1) it is clear that a user usually keep those information in their profiles which can be easily converted to FOAF properties discussed above.

Sl.No.

User Information

Size

Percentage

1

Real name

1295

71%

2

Nick name

510

28%

3

Gender

1623

89%

4

ate of Birth

875

48%

5

Email address

875

48%

6

Mobile number

291

16%

7

Home town/city

1185

65%

8

Photos of yourself

1368

75%

9

Photos of others

1076

59%

10

Political views

218

12%

11

People You Know

1696

93%

12

Sexual orientation

565

31%

13

Relationship status

1276

70%

14

School/Education related

1568

86%

16

Your Profession/Work related

1495

82%

17

Your Company(Organisation)

1423

78%

Total/ respondents 1824

The second challenge is to know how concerned the users are about the privacy of their information. To check this another question was used in the survey.

Question) How active are you in safeguarding (protecting) the following information submitted by you on social networking sites?

Not at all

A

little

Somewhat

Highly

Friend List

Your contact details (email-id, mobile- no, Hometown, city etc)

Education Detail

Date of Birth

Lifestyle related (eg. photos, views etc)

Professional / work related information

Relationship status / Sexual orientation

Sample Size

Percentage

Not at all

967

53%

A little

383

21%

Somewhat

218

12%

Highly

256

14%

Total 1824

100%

The response submitted by 1824 users are as follows Privacy Concern for Friend List

Privacy Concern for Contact Details

Sample Size

Percentage

Not at all

863

47%

A little

487

27%

Somewhat

228

13%

Highly

246

13%

Total 1824

100%

Sample Size

Percentage

Not at all

894

49%

A little

496

27%

Somewhat

213

12%

Highly

221

12%

Total 1824

100%

Privacy Concern for Education Details

Privacy Concern for Professional / work related information

Sample Size

Percentage

Not at all

994

55%

A little

426

23%

Somewhat

183

10%

Highly

221

12%

Total 1824

100%

From the above chart we can see that most of the users are not very concerned about their privacy related to the people they know, their contact details, their education detail as well as professional/work related information. Therefore we can

publish these information in the form of metadata by extracting them from the Web without worrying about the privacy quotient. The social networking sites can even add this to their terms and conditions that the information will be published in the form of metadata which can be used in the searching process.

In addition to the privacy problem, another potential problem may be of copyrights. Some Social Networking Sites may prevent the utilization of their contents for revision or manipulation especially for commercial objectives. We have to consider this problem before we publish the data.

CONCLUSION AND FUTURE WORK

Once we are sure that the required data is available in the database the Social Networking sites can create FOAF profile that contains user information including email address, location, interests, a list of friends, etc., and place that file in a web-accessible location. Users can edit their profiles using web-based forms, hence, most users may be completely unaware of the technologys existence and their use of it.

Now these FOAF files can be used in different ways to process and find answer to user queries. Some research are going on like Indexing FOAF files as well as Flink system for the extraction, aggregation and visualization of online social networks which employs semantic technology for reasoning with personal information extracted from a number of electronic information sources including web pages, emails, publication archives and FOAF profiles.

Thus we can say that FOAF profiles can make ways for answering user queries in a better way.

Acknowledgment

The authors would like to thank all faculty and students of Maulana Azad College of Engineering & Technology, Patna who have shown interest and supported in this research.

References

  1. www.obitko.com/tutrials/ontologies-semantic-web/resource- description-

    framework.html

  2. John C. Paolillo and Elijah Wright. The Challenges of FOAF Characterization.

  3. Peter Mika, Social Network and the Semantic Web

  4. John P. Scott. Social Network Analysis: A Handbook. Sage Publications

Leave a Reply