Target-Dependent Churn Classification in Microblogs

We consider the problem of classifying micro-posts as churny or non-churny with respect to a given brand. Using Twitter data about three brands, we find that standard machine learning techniques clearly outperform keyword based approaches. However, the three machine learning techniques we employed (linear classification, support vector machines, and logistic regression) do not perform as well on churn classification as on other text classification problems. We investigate demographic, content, and context churn indicators in microblogs and examine factors that make this problem more challenging. Experimental results show an average F1 performance of 75% for target-dependent churn classification in microblogs.

[1]  Bart Baesens,et al.  Social network analysis for customer churn prediction , 2014, Appl. Soft Comput..

[2]  Tiejun Zhao,et al.  Target-dependent Twitter Sentiment Classification , 2011, ACL.

[3]  Brendan T. O'Connor,et al.  Improved Part-of-Speech Tagging for Online Conversational Text with Word Clusters , 2013, NAACL.

[4]  Vasudeva Varma,et al.  Timespent based models for predicting user retention , 2013, WWW.

[5]  Marcel Karnstedt,et al.  Churn in Social Networks , 2010, Handbook of Social Network Technologies.

[6]  J. Fleiss.  Measuring nominal scale agreement among many raters. , 1971 .

[7]  Jie Gao,et al.  Modeling Attrition in Organizations from Email Communication , 2013, 2013 International Conference on Social Computing.

[8]  Matthew Rowe,et al.  The effect of user features on churn in social networks , 2011, WebSci '11.

[9]  Borko Furht,et al.  Handbook of Social Network Technologies and Applications , 2010, Handbook of Social Network Technologies.

[10]  David Lo,et al.  Collective Churn Prediction in Social Network , 2012, 2012 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining.

[11]  M. Tahar Kechadi,et al.  Customer churn prediction in telecommunications , 2012, Expert Syst. Appl..

[12]  Beth Levin,et al.  English Verb Classes and Alternations: A Preliminary Investigation , 1993 .

[13]  Jaideep Srivastava,et al.  Churn Prediction in MMORPGs: A Social Influence Based Approach , 2009, 2009 International Conference on Computational Science and Engineering.