Journal of Intelligence and Information Systems (지능정보연구)

Korea Intelligent Information System Society (한국지능정보시스템학회)
권호 표지
DOI QR코드

DOI QR Code

Customer Behavior Prediction of Binary Classification Model Using Unstructured Information and Convolution Neural Network: The Case of Online Storefront

비정형 정보와 CNN 기법을 활용한 이진 분류 모델의 고객 행태 예측: 전자상거래 사례를 중심으로

  • Kim, Seungsoo (Dept. of Business Administration, Graduate School, Hanyang University) ;
  • Kim, Jongwoo (School of Business, Hanyang University)
  • 김승수 (한양대학교 일반대학원 경영학과) ;
  • 김종우 (한양대학교 경영대학 경영학부)
  • Received : 2018.05.21
  • Accepted : 2018.06.16
  • Published : 2018.06.30
Download PDF
⟨ Previous Next ⟩

Abstract

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

최근 딥러닝 기술이 주목을 받고 있다. 대중들의 관심을 받았던 국제 이미지 인식 기술 대회(ILSVR)와 알파고(AlphaGo)에서 사용된 딥러닝 기술이 바로 합성곱 신경망(CNN; Convolution Neural Network)이다. 합성곱 신경망은 입력 이미지를 작은 구역으로 나누어 부분적인 특징을 인식하고 이것을 결합하여 전체를 인식하는 특징을 가진다. 이러한 딥러닝 기술이 우리의 생활에 있어 많은 변화를 야기할 것이라는 기대를 주고 있지만 현재까지는 이미지 인식과 자연어 처리 등에 그 성과가 국한되어 있다. 비즈니스 문제에 대한 딥러닝 활용은 아직까지 초기 연구 단계로 향후 마케팅 응답 예측이나 허위 거래 식별, 부도 예측과 같은 전통적 비즈니스 문제들에 대해 보다 깊게 활용되고 그 성능이 입증된다면 딥러닝 기술의 활용 가치가 보다 더 주목받게 될 것으로 기대된다. 이러한 때 비교적 고객 식별이 용이하고 활용 가치가 높은 빅데이터를 보유하고 있는 전자상거래 기업의 사례를 바탕으로 하여 딥러닝 기술의 비즈니스 문제 해결 가능성을 진단해보는 것은 학술적으로 매우 의미 있는 시도라 할 수 있겠다. 이에 본 연구에서는 전자상거래 기업의 고객 행태 예측력을 높이기 위한 방안으로 합성곱 신경망을 활용한 '이종 정보 결합(Heterogeneous Information Integration)의 CNN 모델'을 제시한다. 이는 정형과 비정형 정보를 결합하여 다층 퍼셉트론 구조의 합성곱 신경망에서 학습시키는 모델로서 최적의 성능을 발휘하도록 '이종 정보 결합'과 '비정형 정보의 벡터 전환', 그리고 '다층 퍼셉트론 설계'로 하는 3개의 내부 아키텍처를 정의하고 각 아키텍처 단위로 구성되는 방식에 따른 성능을 평가하여 그 결과를 바탕으로 제안 모델을 확정하고 그 성능을 평가해보고자 한다. 고객 행태 예측을 위한 목표 변수는 전자상거래 기업에서 중요하게 관리하고 있는 재구매 고객, 이탈 고객, 고빈도 구매 고객, 고빈도 반품 고객, 고단가 구매 고객, 고할인 구매 고객 등 모두 6개의 이진 분류 문제로 정의한다. 제안한 모델의 유용성을 검증하기 위해서 국내 특정 전자상거래 기업의 실제 데이터를 활용하여 실험을 수행하였다. 실험 결과 정형과 비정형 정보를 결합하여 CNN을 활용한 제안 모델이 NBC(Naïve Bayes classification)과 SVM(Support vector machine), 그리고 ANN(Artificial neural network)에 비해서 예측 정확도와 F1 Measure가 높게 평가되었다. 또 NBC, SVM, ANN에서 정형 정보만을 사용할 때 보다 정형과 비정형 정보를 결합하여 입력 변수로 함께 활용한 경우에 예측 정확도가 향상되는 것으로 나타났다. 따라서 실험 결과로부터 비정형 정보의 활용이 고객 행태 예측의 정확도 향상에 기여한다는 점과 CNN 기법의 특징 추출 알고리즘이 VOC에 사용된 단어들의 분포와 위치 정보를 해석하여 문장의 의미를 파악하는데 효과적이라는 점을 실증적으로 확인하였다는데 그 의미가 있다고 할 수 있겠다. 이를 통해서 CNN 기법이 지금까지 소개된 이미지 인식이나 자연어 처리 분야 외에 비즈니스 문제 해결에도 활용 가치가 높다는 점을 확인하였다는데 이 연구의 의의가 있다 하겠다.

Keywords

References

  1. Ahn, S., "Deep learning architectures and applications," Journal of Intelligence and Information Systems, 22(2), (2016), 127-142. https://doi.org/10.13088/jiis.2016.22.2.127
  2. Chu, H., S. Ahn, and S. Kim, "AlphaGo's artificial intelligence algorithm analysis", Software Policy & Research Institute, (2016).
  3. Coussement, K., D. Van den Poel, "Integrating the voice of customers through call center emails into a decision support system for churn prediction," Information & Management, 45(3), (2008), 164-174. https://doi.org/10.1016/j.im.2008.01.005
  4. Gridach, M., H. Haddad, and H. Mulki, "Churn identification in microblogs using convolutional neural networks with structured logical knowledge," Paper presented at the Proceedings of the 3rd Workshop on Noisy User-Generated Text, (2017), 21-30.
  5. Kim, K., B. Lee, and J. Kim, "Feasibility of Deep Learning Algorithms for Binary Classification Problems," Journal of Intelligence and Information Systems, 23(1), (2017), 95-108. https://doi.org/10.13088/jiis.2017.23.1.095
  6. Kim, S., J. Song, and K. Lee, "A Study of customer churn by analysing CRM customer data," Asia Marketing Journal, 7(1), (2005), 21-42.
  7. Kim, Y., "Convolutional neural networks for sentence classification," In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), (2014), 1746-1751
  8. Krizhevsky, A., I. Sutskever, and G. E. Hinton, "Imagenet classification with deep convolutional neural networks," Paper presented at the Advances in Neural Information Processing Systems, (2012), 1097-1105.
  9. Le, Q., and T. Mikolov, "Distributed representations of sentences and documents," Paper presented at the International Conference on Machine Learning, (2014), 1188-1196.
  10. LeCun, Y., Y. Bengio, and G. Hinton, "Deep learning," Nature, 521(7553), (2015), 436-444. https://doi.org/10.1038/nature14539
  11. Lee, J., J. Kim, "Integrated use of classification and association rule for real-time CRM: Application of predicting credit card customer churn," KMIS International Conference, (2007), 135-140.
  12. Mikolov, T., I. Sutskever, K. Chen, G. S. Corrado, and J. Dean, "Distributed representations of words and phrases and their compositionality," Paper presented at the Advances in Neural Information Processing Systems, (2013), 3111-3119.
  13. Schmidhuber, J. "Deep learning in neural networks: An overview," Neural Networks, 61, (2015), 85-117. https://doi.org/10.1016/j.neunet.2014.09.003
  14. Yigit, I. O., A. F. Ates, M. Guvercin, H. Ferhatosmanoglu, and B. Gedik, "Call center text mining approach," Paper presented at the Signal Processing and Communications Applications Conference (SIU), 2017 25th, (2017), 1-4.
  15. Yu, E., J. Kim, C. Lee, and N. Kim, "Using ontologies for semantic text mining," Journal of Information Systems, 21(3), (2012), 137-161. https://doi.org/10.5859/KAIS.2012.21.3.137
  16. Zhang, X., J. Zhao, and Y. LeCun, "Character-level convolutional networks for text classification," Paper presented at the Advances in Neural Information Processing Systems, (2015), 649-657.

Cited by

  1. 정확히 재가중되는 온라인 전체 에러율 최소화 기반의 객체 추적 vol.25, pp.4, 2018, https://doi.org/10.13088/jiis.2019.25.4.053
  2. 증권 금융 상품 거래 고객의 이탈 예측 및 원인 추론 vol.5, pp.2, 2018, https://doi.org/10.36498/kbigdt.2020.5.2.215