This study conducted research from the consumer perspective of commercial recommender system online, to explore the potential correlation between the accuracy of a product recommended by the system and the probability of purchase by the costumers online. From the aspect of a consumer motivation, the research obtained the data from a real E-commerce website, and use theory of UTAUT2 to deduce and rebuild a consumer motivation model. By refilling and testing the model with a big set of data, we made regression of the transparent model, and finally revealed how to improve a recommender system in correlation with its recommending accuracy. This study in aimed to cover the research gap of finding the exact degree of accuracy for an online recommender system at the maximum of increasing the probability of shopping online, rather from the aspect of only algorithm but from the perspective of consumer motivation, to build a practical model. In terms of that, the research planed to use the UTAUT2 model as the basic theory, and moderate it into a extensional model accordingly.
In real experiment of using a recommender system, it also requires the consumer to conduct input and there follows the output accordingly. See the Recommender System Operation in Fig. 3. . In terms of the sequence of inputs and outputs, we can divide the motivation into Accurate Motive and Ambiguous Motive. In other words, the Accurate Motive means the key words of inputs by the consumers, which exactly express the intention of a purchasing. For example, on the traveling E-commerce website, the consumer with the intention to go to the Maldives, a beautiful tropical island, input “Maldives”, which uncloses the exact purchasing purpose. And in this occasion, the setting of parameters could be displayed as the different results of “Maldives” (absolute accuracy), “ Hainan Island”(a similar tropical island, relatively high accuracy), “Guilin” (a tropical inland city, relatively low accuracy). As for the Ambiguous Motive, the input of the consumer can only reveal a rough picture of the desirable products, and the key words can be derivative from noun to adjectives, one to several. Taking the same traveling E-commerce website as well, when a consumer type in the keywords of “ island, relaxation, tropic”, the picture can be drawn into a tropic island with beautiful sunshine and suitable for a relaxing vacation. In accordance of different degrees of accuracy, the recommendations are respectively “Maldives” (a beautiful tropical island, relatively high accuracy), “Hong Kong” (a tropical island, suitable for shopping instead of relaxation, relatively medium accuracy), “Beijing” (a big temperate inland city, suitable for sightseeing rather than relaxation, relatively low accuracy).

Just described above, accuracy is tested as the parameters, and thus to build the transparent consumer model. We attempt to obtain the data from the real E-commerce website. With the huge set of data, we are able to refill the revised UTAUT2 model and therefore to the conclusion of the hypothesis we can make.

References

1. Liang T P, Yang Y F, Chen D N, et al. A semantic-expansion approach to personalized knowledge recommendation[J]. Decision Support Systems, 2008, 45(3): 401-412.

2. Bodapati A V. Recommendation systems with purchase data[J]. Journal of Marketing Research, 2008, 45(1): 77-93.

3. Rust R T, Thompson D V. How does marketing strategy change in a service-based world[J]. The service-dominant logic of marketing: Dialog, debate, and directions, 2006: 381-392.

4. Herlocker J L, Konstan J A, Terveen L G, et al. Evaluating collaborative filtering recommender systems[J]. ACM Transactions on Information Systems (TOIS), 2004, 22(1): 5-53.

5. Fleder D, Hosanagar K. Blockbuster culture's next rise or fall: The impact of recommender systems on sales diversity[J]. Management science, 2009, 55(5): 697-712.

6. Green P E, Krieger A M, Wind Y. Thirty years of conjoint analysis: Reflections and prospects[J]. Interfaces, 2001, 31(3_supplement): S56-S73.

7. Peterson R A, Merino M C. Consumer information search behavior and the Internet[J]. Psychology & Marketing, 2003, 20(2): 99-121.

8. McNee S M, Riedl J, Konstan J A. Being accurate is not enough: how accuracy metrics have hurt recommender systems[C]//CHI'06 extended abstracts on Human factors in computing systems. ACM, 2006: 1097-1101.

9. Ricci F, Rokach L, Shapira B. Introduction to recommender systems handbook[M]//Recommender systems handbook. Springer US, 2011: 1-35.

10. Ge M, Delgado-Battenfeld C, Jannach D. Beyond accuracy: evaluating recommender systems by coverage and serendipity[C]//Proceedings of the fourth ACM conference on Recommender systems. ACM, 2010: 257-260.

11. Atkins D, Briss P A, Eccles M, et al. Systems for grading the quality of evidence and the strength of recommendations II: pilot study of a new system[J]. BMC Health Services Research, 2005, 5(1): 25.

12. Rubenstein-Montano B, Liebowitz J, Buchwalter J, et al. A systems thinking framework for knowledge management[J]. Decision support systems, 2001, 31(1): 5-16.

13. Daly A K, Brockmoller J, Broly F, et al. Nomenclature for human CYP2D6 alleles[J]. Pharmacogenetics and Genomics, 1996, 6(3): 193-201.

14. Adomavicius G, Tuzhilin A. Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions[J]. Knowledge and Data Engineering, IEEE Transactions on, 2005, 17(6): 734-749.

15. Singleton J A, Lloyd J C, Mootrey G T, et al. An overview of the vaccine adverse event reporting system (VAERS) as a surveillance system[J]. Vaccine, 1999, 17(22): 2908-2917.

16. Yavas U, Shemwell D J. Modified importance-performance analysis: an application to hospitals[J]. International Journal of Health Care Quality Assurance, 2001, 14(3): 104-110.

17. Yavas U, Romanova N. Assessing performance of multi-hospital organizations: a measurement approach[J]. International Journal of Health Care Quality Assurance, 2005, 18(3): 193-203.

18. Gokhale A, Balasubramanian K, Krishna A S, et al. Model driven middleware: A new paradigm for developing distributed real-time and embedded systems[J]. Science of Computer programming, 2008, 73(1): 39-58.

19. Parsons J, Ralph P, Gallagher K. Using viewing time to infer user preference in recommender systems[J]. 2004.

20. Resnick P. Impersonal sociotechnical capital, ICTs, and collective action among strangers[J]. Transforming enterprise: the economic and social implications of information technology, 2005: 399-412.

21. Dellarocas C. Building trust on-line: The design of reliable reputation reporting: Mechanisms for online trading communities[J]. 2001.

22. Liebowitz J. Will knowledge management work in the government?[J]. Electronic Government, An International Journal, 2004, 1(1): 1-7.

23. Chen Y F. Herd behavior in purchasing books online[J]. Computers in Human Behavior, 2008, 24(5): 1977-1992.

24. Komiak S Y X, Benbasat I. The effects of personalization and familiarity on trust and adoption of recommendation agents[J]. Mis Quarterly, 2006: 941-960.

25. Koufaris M. Applying the technology acceptance model and flow theory to online consumer behavior[J]. Information systems research, 2002, 13(2): 205-223.

26. Corritore C L, Kracher B, Wiedenbeck S. On-line trust: concepts, evolving themes, a model[J]. International Journal of Human-Computer Studies, 2003, 58(6): 737-758.

27. Chin W W, Todd P A. On the use, usefulness, and ease of use of structural equation modeling in MIS research: a note of caution[J]. MIS quarterly, 1995: 237-246.

28. Yang S. Understanding Undergraduate Students’ Adoption of Mobile Learning Model: A Perspective of the Extended UTAUT2[J]. Journal of Convergence Information Technology, 2013, 8(10).

29. Slade E, Williams M, Dwivdei Y. Extending UTAUT2 To Explore Consumer Adoption Of Mobile Payments[C]//Proceedings of the UK Academy for Information Systems Conference, March. 2013.

30. Alazzam M B, Samad A, Basari H, et al. Review of Studies with Utaut2 as Conceptual Framework[J]. 31.Yang S. Understanding Undergraduate Students’ Adoption of Mobile Learning Model: A Perspective of the Extended UTAUT2[J]. Journal of Convergence Information Technology, 2013, 8(10).

31. Slade E L, Williams M D, Dwivedi Y K. Devising a research model to examine adoption of mobile payments: An extension of UTAUT2[J]. The Marketing Review, 2014, 14(3): 310-335.

32. Shao X, Siponen M. Consumer acceptance and use of information technology: Adding consumption theory to UTAUT2[J]. 2011.

33. C. Hanson, J. West, B. Neiger, R. Thackeray, M. Barnes, and E. Mcintyre, “among Health Educators,” vol. 42, no. 4, pp. 197–204, 2011.

34. B. Holtz and S. Krein, “Understanding Nurse Perceptions of a Newly Implemented Electronic Medical Record System,” J. Technol. Hum. Serv., vol. 29, no. 4, pp. 247–262, Oct. 2011.

35. M. Kang, B. T. Liew, H. Lim, J. Jang, and S. Lee, “Emerging Issues in Smart Learning,” pp. 209–216, 2015.

36. M. Technologies, S. D. Impact, and M. I. Usage, “Mobile Technologies and Services Development Impact on Mobile Internet Usage in Latvia « Mobile Technologies and Services Development Impact on Mobile Internet Usage in Mobile Technologies and Services Development Impact on Mobile Internet Usage in Latvia,” 2013

37. A. Raman and Y. Don, “Preservice Teachers’ Acceptance of Learning Management Software: An Application of the UTAUT2 Model,” Int. Educ. Stud., vol. 6, no. 7, Jun. 2013.

38. S. Yang, “Understanding Undergraduate Students’ Adoption of Mobile Learning Model: A Perspective of the Extended UTAUT2,” J. Converg. Inf. Technol., vol. 8, no. 10, pp. 969–979, May 2013.