Journal Home > Volume 5 , issue 1

Intelligent Financial Advisors (IFAs) in online financial applications (apps) have brought new life to personal investment by providing appropriate and high-quality portfolios for users. In real-world scenarios, identifying potential clients is a crucial issue for IFAs, i.e., identifying users who are willing to purchase the portfolios. Thus, extracting useful information from various characteristics of users and further predicting their purchase inclination are urgent. However, two critical problems encountered in real practice make this prediction task challenging, i.e., sample selection bias and data sparsity. In this study, we formalize a potential conversion relationship, i.e., user activated user client and decompose this relationship into three related tasks. Then, we propose a Multitask Feature Extraction Model (MFEM), which can leverage useful information contained in these related tasks and learn them jointly, thereby solving the two problems simultaneously. In addition, we design a two-stage feature selection algorithm to select highly relevant user features efficiently and accurately from an incredibly huge number of user feature fields. Finally, we conduct extensive experiments on a real-world dataset provided by a famous fintech bank. Experimental results clearly demonstrate the effectiveness of MFEM.


menu
Abstract
Full text
Outline
About this article

Toward Intelligent Financial Advisors for Identifying Potential Clients: A Multitask Perspective

Show Author's information Qixiang ShaoRunlong YuHongke ZhaoChunli Liu( )Mengyi ZhangHongmei SongQi Liu
Department of Anhui Province Key Laboratory of Big Data Analysis and Application (BDAA), School of Computer Science and Technology, University of Science and Technology of China, Hefei 230027, China
Department of College of Management and Economics, Tianjin University, Tianjin 300072, China
School of Management, Hefei University of Technology, Hefei 230009, China
Department of Information Technology, China Merchants Bank, Shenzhen 518000, China

Abstract

Intelligent Financial Advisors (IFAs) in online financial applications (apps) have brought new life to personal investment by providing appropriate and high-quality portfolios for users. In real-world scenarios, identifying potential clients is a crucial issue for IFAs, i.e., identifying users who are willing to purchase the portfolios. Thus, extracting useful information from various characteristics of users and further predicting their purchase inclination are urgent. However, two critical problems encountered in real practice make this prediction task challenging, i.e., sample selection bias and data sparsity. In this study, we formalize a potential conversion relationship, i.e., user activated user client and decompose this relationship into three related tasks. Then, we propose a Multitask Feature Extraction Model (MFEM), which can leverage useful information contained in these related tasks and learn them jointly, thereby solving the two problems simultaneously. In addition, we design a two-stage feature selection algorithm to select highly relevant user features efficiently and accurately from an incredibly huge number of user feature fields. Finally, we conduct extensive experiments on a real-world dataset provided by a famous fintech bank. Experimental results clearly demonstrate the effectiveness of MFEM.

Keywords:

Intelligent Financial Advisor (IFA), potential client identification, MultiTask Learning (MTL), feature selection
Received: 20 October 2021 Accepted: 03 November 2021 Published: 27 December 2021 Issue date: March 2022
References(53)
[1]
H. K. Zhao, B. B. Jin, Q. Liu, Y. Ge, E. H. Chen, X. Zhang, and T. Xu, Voice of charity: Prospecting the donation recurrence & donor retention in crowdfunding, IEEE Trans. Knowl. Data Eng., vol. 32, no. 8, pp. 1652-1665, 2020.
[2]
H. K. Zhao, X. P. Liu, X. Zhang, Y. Y. Wei, and C. L. Liu, The effects of person-organization fit on lending behaviors: Empirical evidence from kiva, J. Manage. Sci. Eng., .
[3]
C. Y. Liu, Legal risks and the countermeasures of developing intelligent investment advisor in China, in Proc. 1st Int. Conf. Intelligent Human Systems Integration: Integrating People and Intelligent Systems, Dubai, United Arab Emirates, 2018, pp. 76-82.
[4]
B. Zadrozny, Learning and evaluating classifiers under sample selection bias, in Proc. 21st Int. Conf. Machine Learning, Banff, Canada, 2004, p. 114.
[5]
K. C. Lee, B. Orten, A. Dasdan, and W. T. Li, Estimating conversion rate in display advertising from past erformance data, in Proc. 18th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, Beijing, China, 2012, pp. 768-776.
[6]
G. M. Weiss, Mining with rarity: A unifying framework, ACM SIGKDD Explor. Newsl., vol. 6, no. 1, pp. 7-19, 2004.
[7]
N. V. Chawla, K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer, SMOTE: Synthetic minority over-sampling technique, J. Artif. Intell. Res., vol. 16, pp. 321-357, 2002.
[8]
R. Pan, Y. H. Zhou, B. Cao, N. N. Liu, R. Lukose, M. Scholz, and Q. Yang, One-class collaborative filtering, in Proc. of the 8th IEEE Int. Conf. Data Mining, Pisa, Italy, 2008, pp. 502-511.
[9]
S. Ruder, An overview of multi-task learning in deep neural networks, arXiv preprint arXiv: 1706.05098, 2017.
[10]
J. E. Fisch, M. Labouré, and J. A. Turner, The Emergence of the Robo-Advisor. Philadelphia, PA, USA: Wharton University of Pennsylvania, 2018.
[11]
B. J. Pine II, Mass Customization. Boston, MA, USA: Harvard business School Press, 1993.
[12]
D. Peppers and M. Rogers, The One to One Future: Building Relationships One Customer at a Time. New York, NY, USA: Bantam Press, 1997.
[13]
K. W. Cheung, J. T. Kwok, M. H. Law, and K. C. Tsui, Mining customer product ratings for personalized marketing, Decis. Support Syst., vol. 35, no. 2, pp. 231-243, 2003.
[14]
M. J. Pazzani and D. Billsus, Content-based recommendation systems, in The Adaptive Web: Methods and Strategies of Web Personalization, P. Brusilovsky, A. Kobsa, and W. Nejdl, eds. Berlin, Germany: Springer, 2007, pp. 325-341.
[15]
Q. Liu, E. H. Chen, H. Xiong, Y. Ge, Z. M. Li, and X. Wu, A cocktail approach for travel package recommendation, IEEE Trans. Knowl. Data Eng., vol. 26, no. 2, pp. 278-293, 2014.
[16]
Q. Liu, Y. Ge, Z. M. Li, E. H. Chen, and H. Xiong, Personalized travel package recommendation, in Proc. of IEEE 11th Int. Conf. Data Mining, Vancouver, Canada, 2011, pp. 407-416.
[17]
X. Y. Su and T. M. Khoshgoftaar, A survey of collaborative filtering techniques, Adv. Artif. Intell., vol. 2009, p. 421425, 2009.
[18]
F. D. Hodge, K. I. Mendoza, and R. K. Sinha, The effect of humanizing robo-advisors on investor judgments, Contemp. Account. Res., vol. 38, no. 1, pp. 770-792, 2021.
[19]
Y. Zhang and Q. Yang, A survey on multi-task learning, arXiv preprint arXiv: 1707.08114, 2017.
[20]
L. Di Persio and O. Honchar, Multitask machine learning for financial forecasting, Int. J. Circuits, Syst. Signal Process., vol. 12, pp. 444-451, 2018.
[21]
R. Sawhney, P. Mathur, A. Mangal, P. Khanna, R. R. Shah, and R. Zimmermann, Multimodal multi-task financial risk forecasting, in Proc. 28th ACM Int. Conf. Multimedia, Seattle, WA, USA, 2020, pp. 456-465.
[22]
D. W. Zhou, L. C. Zheng, Y. D. Zhu, J. B. Li, and J. R. He, Domain adaptive multi-modality neural attention network for financial forecasting, in Proc. the Web Conf. 2020, Taipei, China, 2020, pp. 2230-2240.
[23]
W. T. Chu and Y. H. Liu, Thermal facial landmark detection by deep multi-task learning, in Proc. of the IEEE 21st Int. Workshop on Multimedia Signal Processing, Kuala Lumpur, Malaysia, 2019, pp. 1-6.
[24]
X. Ma, L. Q. Zhao, G. Huang, Z. Wang, Z. L. Hu, X. Q. Zhu, and K. Gai, Entire space multi-task model: An effective approach for estimating post-click conversion rate, in Proc. of the 41st Int. ACM SIGIR Conf. Research & Development in Information Retrieval, Ann Arbor, MI, USA, 2018, pp. 1137-1140.
[25]
H. Wen, J. Zhang, Y. Wang, F. Y. Lv, W. T. Bao, Q. Lin, and K. P. Yang, Entire space multi-task modeling via post-click behavior decomposition for conversion rate prediction, in Proc. of 43rd Int. ACM SIGIR Conf. Research and Development in Information Retrieval, Virtual Event, China, 2020, pp. 2377-2386.
[26]
R. L. Yu, Q. Liu, Y. Y. Ye, M. Y. Cheng, E. H. Chen, and J. H. Ma, Collaborative list-and-pairwise filtering from implicit feedback, IEEE Trans. Knowl. Data Eng., .
[27]
R. L. Yu, Y. Z. Zhang, Y. Y. Ye, L. Wu, C. Wang, Q. Liu, and E. H. Chen, Multiple pairwise ranking with implicit feedback, in Proc. 27th ACM Int. Conf. Information and Knowledge Management, Torino, Italy, 2018, pp. 1727-1730.
[28]
H. K. Zhao, X. P. Wu, C. Zhao, L. Zhang, H. P. Ma, and F. Cheng, CoEA: A cooperative-competitive evolutionary algorithm for bidirectional recommendations, IEEE Trans. Evol. Comput., .
[29]
I. Guyon and A. Elisseeff, An introduction to variable and feature selection, J. Mach. Learn. Res., vol. 3, pp. 1157-1182, 2003.
[30]
A. Bommert, X. D. Sun, B. Bischl, J. Rahnenführer, and M. Lang, Benchmark for filter methods for feature selection in high-dimensional classification data, Comput. Stat. Data Anal., vol. 143, p. 106839, 2020.
[31]
D. Dernoncourt, B. Hanczar, and J. D. Zucker, Analysis of feature selection stability on high dimension and small sample data, Comput. Stat. Data Anal., vol. 71, pp. 681-693, 2014.
[32]
Q. Q. Gu, Z. H. Li, and J. W. Han, Generalized fisher score for feature selection, in Proc.27th Conf. Uncertainty in Artificial Intelligence, Barcelona, Spain, 2011, pp. 266-273.
[33]
C. Lazar, J. Taminau, S. Meganck, D. Steenhoff, A. Coletta, C. Molter, V. de Schaetzen, R. Duque, H. Bersini, and A. Nowe, A survey on filter techniques for feature selection in gene expression microarray analysis, IEEE/ACM Trans. Comput. Biol. Bioinform., vol. 9, no. 4, pp. 1106-1119, 2012.
[34]
M. M. Kabir, M. M. Islam, and K. Murase, A new wrapper feature selection approach using neural network, Neurocomputing, vol. 73, no. 16-18, pp. 3273-3283, 2010.
[35]
S. Maldonado and R. Weber, A wrapper method for feature selection using support vector machines, Inf. Sci., vol. 179, no. 13, pp. 2208-2217, 2009.
[36]
J. L. Tang, S. Alelyani, and H. Liu, Feature selection for classification: A review, in Data Classification: Algorithms and Applications, C. C. Aggarwal, ed. Boca Raton, FL, USA: CRC Press, 2014, p. 37.
[37]
J. R. Quinlan, Induction of decision trees, Mach. Learn., vol. 1, no. 1, pp. 81-106, 1986.
[38]
J. R. Quinlan, C4.5: Programs for Machine Learning. San Francisco, CA, USA: Morgan Kaufmann, 1993.
[39]
N. Gui, D. N. Ge, and Z. Y. Hu, AFS: An attention-based mechanism for supervised feature selection, Proc. AAAI Conf. Artif. Intell., vol. 33, no. 1, pp. 3705-3713, 2019.
[40]
Y. F. Li, C. Y. Chen, and W. W. Wasserman, Deep feature selection: Theory and application to identify enhancers and promoters, J. Comput. Biol., vol. 23, no. 5, pp. 322-336, 2016.
[41]
B. Liu, C. X. Zhu, G. L. Li, W. N. Zhang, J. C. Lai, R. M. Tang, X. Q. He, Z. G. Li, and Y. Yu, AutoFIS: Automatic feature interaction selection in factorization models for click-through rate prediction, in Proc. 26th ACM SIGKDD Int. Conf. Knowledge Discovery & Data Mining, Virtual Event, CA, USA, 2020, pp. 2636-2645.
[42]
D. Roy, K. S. R. Murty, and C. K. Mohan, Feature selection using deep neural networks, in Proc. of the 2015 Int. Joint Conf. Neural Networks, Killarney, Ireland, 2015, pp. 1-6.
[43]
C. W. Chen, Y. H. Tsai, F. R. Chang, and W. C. Lin, Ensemble feature selection in medical datasets: Combining filter, wrapper, and embedded feature selection results, Expert Syst., vol. 37, no. 5, p. e12553, 2020.
[44]
H. Sun, J. Jin, R. Xu, and A. Cichocki, Feature selection combining filter and wrapper methods for motor-imagery based brain-Computer interfaces, Int. J. Neural Syst., vol. 31, no. 9, p. 2150040, 2021.
[45]
R. L. Yu, Y. Y. Ye, Q. Liu, Z. H. Wang, C. F. Yang, Y. C. Hu, and E. H. Chen, XCrossNet: Feature structure-oriented learning for click-through rate prediction, in Proc. of the 25th Pacific-Asia Conf. Advances in Knowledge Discovery and Data Mining, Cham, Germany, 2021, pp. 436-447.
[46]
S. K. Chao and G. Cheng, A generalization of regularized dual averaging and its dynamics, arXiv preprint arXiv: 1909.10072, 2019.
[47]
Q. Liu, G. F. Wang, H. K. Zhao, C. R. Liu, T. Xu, and E. H. Chen, Enhancing campaign design in crowdfunding: A product supply optimization perspective, in Proc. 26th Int. Joint Conf. Artificial Intelligence, Melbourne, Australia, 2017, pp. 695-702.
[48]
P. Covington, J. Adams, and E. Sargin, Deep neural networks for youtube recommendations, in Proc. 10th ACM Conf. on Recommender Systems, Boston, MA, USA, 2016, pp. 191-198.
[49]
H. X. Liu, K. Simonyan, and Y. M. Yang, Darts: Differentiable architecture search, arXiv preprint arXiv: 1806.09055, 2018.
[50]
H. B. McMahan, G. Holt, D. Sculley, M. Young, D. Ebner, J. Grady, L. Nie, T. Phillips, E. Davydov, D. Golovin, et al., Ad click prediction: A view from the trenches, in Proc. of the 19th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, Chicago, IL, USA, 2013, pp. 1222-1230.
[51]
T. Q. Chen, T. He, M. Benesty, V. Khotilovich, Y. Tang, H. Cho, Xgboost: Extreme Gradient Boosting., , 2020.
[52]
H. T. Cheng, L. Koc, J. Harmsen, T. Shaked, T. Chandra, H. Aradhye, G. Anderson, G. Corrado, W. Chai, M. Ispir, et al., Wide & deep learning for recommender systems, in Proc. 1st Workshop on Deep Learning for Recommender Systems, Boston, MA, USA, 2016, pp. 7-10.
[53]
R. X. Wang, B. Fu, G. Fu, and M. L. Wang, Deep & cross network for ad click predictions, in Proc. ADKDD’17, Halifax, Canada, 2017, p. 12.
Publication history
Copyright
Acknowledgements
Rights and permissions

Publication history

Received: 20 October 2021
Accepted: 03 November 2021
Published: 27 December 2021
Issue date: March 2022

Copyright

© The author(s) 2022

Acknowledgements

This research was partially supported by the National Key Research and Development Program of China (No. 2018YFC0832101), the National Natural Science Foundation of China (Nos. 71802068, 61922073, and U20A20229), and the financial supports of Tianjin University (No. 2020XSC-0019). Runlong Yu and Qi Liu thank the support of USTC-CMB Joint Laboratory of Artificial Intelligence. The personalized recommendation functions in IFAs could be turned off by apps’ users. Furthermore, the collection and usage of user privacy data have been authorized by users to comply with Personal Information Protection Law of the P.R.C.

Rights and permissions

The articles published in this open access journal are distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/).

Reprints and Permission requests may be sought directly from editorial office.

Return