lnu.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
The strength of social strength: an evaluation study of algorithmic versus user-defined ranking
Telenor Group, Norway.
Ericsson Research.
Luleå University of Technology.
2014 (English)In: SAC '14 Proceedings of the 29th Annual ACM Symposium on Applied Computing, ACM Press, 2014, p. 658-659Conference paper, Published paper (Refereed)
Abstract [en]

A family relation is generally considered to be stronger than a relation between coworkers in our society, but the strength of the relation is intrinsic and have been cumbersome to measure. The fact that we increasingly communicate electronically, such as through email, mobile phone calls or social media, has made it possible to automatically measure and analyze the relation between persons. This paper presents an evaluation study of social strength, where the social strength is defined as a metric that represents the tie strength of the relation between persons, calculated based on the frequency, duration, context and media type of the electronic communication between the persons.

The study found that the Utility Function performs better because it emphasize the communication frequency between persons. There is however a significant difference in results between the algorithms and the user-defined ranking. This indicates the inability of the algorithms to capture intrinsic knowledge (such as the importance of family bonds and non-electronic interaction). This would mean that the participants' ranking was colored by their interaction in real-life. It is however evident from the study that the functions provide more accurate results when they utilize multiple sources of communication history over only a single source.

Finally, capturing sufficient communication data from multiple data sources is very hard, as access to such data is restricted because of concerns regarding for instance business and privacy. A conclusion is that the algorithms requires a larger data set, preferably being captured continuously over a period longer than 2 weeks, to achieve a better accuracy that is closer to the ground truth. However, the study shows the feasibility of capturing social strength automatically and we believe that the results is an important step towards systems that reason about the relation between persons in order to make communications services more pervasive.

Place, publisher, year, edition, pages
ACM Press, 2014. p. 658-659
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Research subject
Computer and Information Sciences Computer Science, Media Technology
Identifiers
URN: urn:nbn:se:lnu:diva-41707DOI: 10.1145/2554850.2555158ISBN: 978-1-4503-2469-4 (print)OAI: oai:DiVA.org:lnu-41707DiVA, id: diva2:800451
Conference
29th Annual ACM Symposium on Applied Computing
Available from: 2015-04-04 Created: 2015-04-04 Last updated: 2015-04-22Bibliographically approved

Open Access in DiVA

No full text in DiVA

Other links

Publisher's full text

Authority records BETA

Rana, Juwel

Search in DiVA

By author/editor
Rana, Juwel
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar

doi
isbn
urn-nbn

Altmetric score

doi
isbn
urn-nbn
Total: 41 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf