Predictive Power of Online and Offline Behavior Sequences: Evidence from a Micro-finance Context

Microfinance based institutions have emerged as a potential solution to the financial exclusion problem in developing economies around the world. A key challenge facing such micro-lending firms is assessing the credit risk of borrowers, owing to the lack of formal financial histories and collaterals. A number of micro-lending companies have, therefore, started leveraging social media and digital communication data from applicants to assess their ability and willingness to repay loans. In our study, we demonstrate a novel approach of leveraging online and offline behavior sequences, as captured from the borrowers’ browsing logs and mobility traces to accurately predict the borrowers’ creditworthiness. Our preliminary results show that using such sequence data, we can provide micro-lending firms with a cheap and reliable strategy for assessing credit risk of borrowers at the time of loan creation. We contend that such bigdata based strategies are critical to the sustainability of micro-lending institutions

[1]  Catarina Sismeiro,et al.  A Model of Web Site Browsing Behavior Estimated on Clickstream Data , 2003 .

[2]  Alex Graves,et al.  Supervised Sequence Labelling with Recurrent Neural Networks , 2012, Studies in Computational Intelligence.

[3]  P. Chatterjee,et al.  Modeling the Clickstream: Implications for Web-Based Advertising Efforts , 2003 .

[4]  Yehuda Koren,et al.  Collaborative filtering with temporal dynamics , 2009, KDD.

[5]  Wendy W. Moe,et al.  The Influence of Goal‐Directed and Experiential Activities on Online Flow Experiences , 2003 .

[6]  Jure Leskovec,et al.  Finding progression stages in time-evolving event sequences , 2014, WWW.

[7]  LiKang,et al.  Prediction of Human Activity by Discovering Temporal Sequence Patterns , 2014 .

[8]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[9]  Brian D. Davison,et al.  Predicting Sequences of User Actions , 1998 .

[10]  Yoshua Bengio,et al.  Learning long-term dependencies with gradient descent is difficult , 1994, IEEE Trans. Neural Networks.

[11]  Razvan Pascanu,et al.  On the difficulty of training recurrent neural networks , 2012, ICML.

[12]  Cecilia Mascolo,et al.  Exploiting place features in link prediction on location-based social networks , 2011, KDD.

[13]  Christos Faloutsos,et al.  Fast mining and forecasting of complex time-stamped events , 2012, KDD.

[14]  Jure Leskovec,et al.  Friendship and mobility: user movement in location-based social networks , 2011, KDD.

[15]  Leora F. Klapper,et al.  The Global Findex Database 2014: Measuring Financial Inclusion Around the World , 2015 .

[16]  Hinrich Schütze,et al.  Book Reviews: Foundations of Statistical Natural Language Processing , 1999, CL.

[17]  Anna Monreale,et al.  WhereNext: a location predictor on trajectory pattern mining , 2009, KDD.

[18]  Cecilia Mascolo,et al.  Mining User Mobility Features for Next Place Prediction in Location-Based Services , 2012, 2012 IEEE 12th International Conference on Data Mining.

[19]  Yun Fu,et al.  Prediction of Human Activity by Discovering Temporal Sequence Patterns , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[20]  Cecilia Mascolo,et al.  An Empirical Study of Geographic User Activity Patterns in Foursquare , 2011, ICWSM.

[21]  S. Hartley.  Kiva.org: Crowd-Sourced Microfinance and Cooperation in Group Lending , 2010 .

[22]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[23]  Zachary Chase Lipton.  A Critical Review of Recurrent Neural Networks for Sequence Learning , 2015, ArXiv.

[24]  Jürgen Schmidhuber,et al.  Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks , 2006, ICML.

[25]  Sean R. Eddy,et al.  Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids , 1998 .

[26]  David J. Hand,et al.  A survey of the issues in consumer credit modelling research , 2005, J. Oper. Res. Soc..

[27]  Kannan Srinivasan,et al.  Modeling Online Browsing and Path Analysis Using Clickstream Data , 2004 .

[28]  Carsten Griwodz,et al.  Mobile video streaming using location-based network prediction and transparent handover , 2011, NOSSDAV.

[29]  Cecilia Mascolo,et al.  NextPlace: A Spatio-temporal Prediction Framework for Pervasive Systems , 2011, Pervasive.

[30]  Daniel Björkegren,et al.  Behavior Revealed in Mobile Phone Usage Predicts Loan Repayment , 2017, The World Bank Economic Review.

[31]  Tuan Phan,et al.  Credit-worthiness Prediction in Microfinance using Mobile Data: A Spatio-network Approach , 2016, ICIS.

[32]  Alex Pentland,et al.  Modeling and Prediction of Human Behavior , 1999, Neural Computation.

[33]  Niels Pinkwart,et al.  Predicting MOOC Dropout over Weeks Using Machine Learning Methods , 2014, EMNLP 2014.