An Academic Social Network Friend Recommendation Algorithm Based on Decision Tree

With the rapid development of smart city services, social network plays a greater role in extensive areas with smart technology. How to use the smart technology to recommend friends from many users accurately and availably has been a research focus in the field of social recommendation. A friend recommendation algorithm based on Decision Tree with the background of the SCHOLAT, a large online academic social network site. The proposed algorithm translates the problem of friend recommendation into a prediction of binary-class. Firstly, the feature selection method, which the Relief and K-means algorithm are mainly applied, is used to remove the irrelevant and redundant features in the data preprocessing. After obtaining the effective features, we use a learning model to train the selected data. The Decision Tree method is introduced as a base classifier to predict the class of candidate user(recommendation or not recommendation). Secondly, combined with the AdaBoost enhanced algorithm, the results of Decision Tree classifier are weighted to adjusted for a better accurate result, which ultimately forms recommendation lists for the target users. Finally, the experimental results on the SCHOLAT for friend recommendation show that the effectiveness and practicability of the proposed algorithm.

[1]  Robert E. Schapire,et al.  The strength of weak learnability , 1990, Mach. Learn..

[2]  Sam Kwong,et al.  A noise-detection based AdaBoost algorithm for mislabeled data , 2012, Pattern Recognit..

[3]  Alejandro Bellogín,et al.  Using graph partitioning techniques for neighbour selection in user-based collaborative filtering , 2012, RecSys.

[4]  Walter D. Fisher.  On Grouping for Maximum Homogeneity , 1958 .

[5]  David A. Landgrebe,et al.  A survey of decision tree classifier methodology , 1991, IEEE Trans. Syst. Man Cybern..

[6]  Martha Larson,et al.  Exploiting user similarity based on rated-item pools for improved user-based collaborative filtering , 2009, RecSys '09.

[7]  D. Cox.  Note on Grouping , 1957 .

[8]  John Riedl,et al.  Item-based collaborative filtering recommendation algorithms , 2001, WWW '01.

[9]  Junjie Yao,et al.  Challenging the Long Tail Recommendation , 2012, Proc. VLDB Endow..

[10]  Igor Kononenko,et al.  Estimating Attributes: Analysis and Extensions of RELIEF , 1994, ECML.

[11]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[12]  Md Mohiuddin Khan,et al.  Friend recommendation framework for social networking sites using user's online behavior , 2015, 2015 18th International Conference on Computer and Information Technology (ICCIT).

[13]  George Karypis,et al.  Item-based top-N recommendation algorithms , 2004, TOIS.

[14]  F. O. Isinkaye,et al.  Recommendation systems: Principles, methods and evaluation , 2015 .

[15]  Benyamin Kusumoputro,et al.  EEG based patient emotion monitoring using relative wavelet energy feature and Back Propagation Neural Network , 2015, 2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[16]  Loren G. Terveen,et al.  People-to-People-to-Geographical-Places: The P3 Framework for Location-Based Community Systems , 2004, Computer Supported Cooperative Work (CSCW).

[17]  Larry A. Rendell,et al.  The Feature Selection Problem: Traditional Methods and a New Algorithm , 1992, AAAI.