Top Banner
Multi-Layered Friendship Modeling for Location-based Mobile Social Networks Nan Li and Guanling Chen Department of Computer Science, University of Massachusetts Lowell July 14, 2009 Toronto, Canada MobiQuitous 2009
27

Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

May 11, 2015

Download

Documents

guanling

Presented at MobiQuitous, Toronto, Canada, July 2009.
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Multi-Layered Friendship Modeling for Location-based Mobile

Social NetworksNan Li and Guanling Chen

Department of Computer Science, University of Massachusetts Lowell

July 14, 2009Toronto, Canada

MobiQuitous 2009

Page 2: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Online Social Network Success• Popular (half billion ww users)• Sticky (26m per day)

Page 3: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

OSN Goes Mobile

• Already top Web destinations on smartphones

• Unique feature – location– GPS-enabled phones– Sharing current location– Attaching location to user-generated content

• Outlook– LSN >$3.3B revenue by 2013 (ABI)

• Dodgeball, Loopt, Brightkite, WhrrlGoogle Latitude, Foursquare

Page 4: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Brightkite

• Startup founded 2005, Denver CO– Angel funding $1M, 03/2008– Private beta, 04/2008– Opened to public, 10/2008

• User activity– Check in, status update, photo upload– All attached with current location– Updates through SMS, Email, Web, iPhone…

• Social graph with mutual connection– See your friends’ or local activity streams

Page 5: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Usage Snapshot

Page 6: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Contributions

• Data collection from Brightkite– 19k users; 1.5m updates

• Quantitative correlation model for friendship– User tags, social graph, location/activity

• Evaluation using 10m training data and 45d test data– Outperformed than Naïve Bayes classifier

or J48 decision tree algorithms

Page 7: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Data Collection

• Brightkite Web APIs• 12/9/08-1/9/09: 18,951 active users• Back traced to 3/21/08: 1,505,874

updates• Profile: age, gender, tags, friends list• Social graph: 41,014 nodes and

46,172 links• Testing data: next 45 days had 5,098

new links added

Page 8: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks
Page 9: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Tag Cloud

Page 10: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Basic Approach

• Coming up metrics that– Differentiate friends and non-friends– Tags, social graph, location, activities

• Combination of the metrics• Training and testing with traces

Page 11: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Using Metrics

Page 12: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Metric Combination

Page 13: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Social Graph

Page 14: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Social Graph Metric

Page 15: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Tag Graph

• 1000 most popular tags as the nodes• Complete graph• Link weight reflects likelihood of two

tags shared by friends

Page 16: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Tag Graph Metric

Page 17: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Location Graph

Page 18: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Location Graph Metric

Page 19: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Rank Value Result

Page 20: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Modeling Accuracy

• Take another 100,000 non-friend pairs– Not in training data

• Plus the newly added 5,098 friend pairs

• Sort the prediction values

Page 21: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

ROC Curve

Page 22: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Top Recommendations

Page 23: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Information Gain

Worldwide buzz: Planetary-scale views on an instant-messaging network. J. Leskovec and E. Horvitz, June 2007.

Page 24: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Discussions

• Model stability as Brightkite grows– Does not require frequent re-calculation

• On-demand recommendation– Heuristics to speed up metric calculation

• Possible improvement– Different metrics, or combination methods

• “Private” updates– Conjectured to be few, but no proof

Page 25: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Related Work

• Industrial solutions: Facebook, Twitter– Technical details unknown

• OSN structural analysis– Aggregated behavior not suitable for

individual recommendations

• Collective filtering– User-item vs. user-user

Page 26: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Conclusion

• Correlated attribute combination has good friendship recommendation power– Interests, social graph, location

• Location metric is important– Gender and age not so much

• Future work– System implementation– Real-user action-based evaluation

Page 27: Multi-Layer Friendship Modeling for Location-Based Mobile Social Networks

Acknowledgement

• Anonymous reviewers• Shepherd- Sharad Agarwal• Best Paper Award committee