Predicting positive and negative links in signed social networks via transfer learning.
之前和社交網絡相關的研究,大多數都非常關注積極正面的用戶鏈接關係;與這些研究不同,我們研究同時含有正面與負面鏈接關係的帶符號社交網絡。具體來講,我們特別關注如何在一個帶符號的社交網絡(該網絡又稱為“目標網絡“)中可信並且有效地去預測鏈接關係的符號為正或是為負,且該網絡中僅有一小部份的鏈接關係符號已知,作為訓練樣本。我們採取遷移學習的機器學習方法,借助於另外一個帶符號社交網絡(該網絡又稱為“源網絡“)中充足的鏈接關係符號信息,從而訓練得到一個有效的鏈接關係分類器;需要注意的是,該 “源網絡同“目標網絡,在鏈接關係樣本和鏈接關係符號的聯合分佈上,並不相同。 === 由於在帶符號社交網絡中沒有事先定...
Other Authors: | Ye, Jihang. |
---|---|
Format: | Others |
Language: | English Chinese |
Published: |
2012
|
Subjects: | |
Online Access: | http://library.cuhk.edu.hk/record=b5549173 http://repository.lib.cuhk.edu.hk/en/item/cuhk-328643 |
Similar Items
-
Deep Belief Network-Based Approaches for Link Prediction in Signed Social Networks
by: Feng Liu, et al.
Published: (2015-04-01) -
Online Social Networks as Formal Learning Environments: Learner Experiences and Activities
by: George Veletsianos, et al.
Published: (2012-01-01) -
Prediction and influence maximization in location-based social networks.
Published: (2012) -
Unsupervised Negative Link Prediction in Signed Social Networks
by: Pengfei Shen, et al.
Published: (2019-01-01) -
Homophily-Based Link Prediction in The Facebook Online Social Network: A Rough Sets Approach
by: Elkabani Islam, et al.
Published: (2015-12-01)