For the purpose of investigating the determinants of Taiwanese online shoppers’ intention to purchase recommended products, the quantitative approach was chosen to develop the survey. The online survey was distributed via social media platforms and messaging applications, and a total of 276 complete and valid responses were collected.
The findings indicate that consumers’ trust in recommender systems and performance expectancy of recommender systems are important factors influencing consumers’ purchase intention and that recommendation accuracy, novelty, and diversity indirectly affect purchase intention by shaping consumers’ perceptions about the performance expectancy of recommender systems. Thus, this study makes several suggestions for practices to facilitate the design and development of effective recommender systems by service providers:
- Building consumer trust
- Ensure that the proposed recommendations prioritize the interests of the consumer over those of the company. Since recommender system narrows the options for consumers, it may, to some extent, nudge consumers in the desired direction of the service provider.
- Handle product reviews with caution and weed out fake reviews and ratings in order to provide products with credibility and authenticity. Given that product ratings are important input for recommender systems to generate recommendations, if the service providers are not aware of the problem of fake reviews and address it, the products suggested by the recommender systems are likely to have fraudulent reviews, resulting in a loss of consumers’ trust.
- Improving the usefulness of recommender systems
- Enhance the ability of recommender systems to immediately identify consumers’ current interests and needs in order to support consumers in finding the ideal products quickly and help them increase the efficiency of product search and decision making.
- Optimize the machine learning models for more desirable recommendations, and strike a balance between recommendation accuracy, novelty, and diversity, rather than solely focusing on one of them.
This study compares Generation X and Generation Y consumers in Taiwan, as the online shopping penetration rates for both generations are higher than for any other age group and have the potential to continue growing. It is worth noting that consumers belonging to different generational cohorts have different perceptions toward recommended products, resulting in different standards for the usefulness of recommender systems. As a result, service providers should distinguish the generational cohort to which consumers belong before proposing recommendations, so as to present recommended products according to the characteristics of each generational cohort and effectively enhance consumers’ purchase intention. To give consumers peace of mind with a secure shopping experience, service providers should also enable consumers to have control over personal data and recommendations, clearly disclose their privacy policies, and keep consumers informed about the way in which recommender systems collect and use data.
With this blog post we want to motivate students to further analyze new features and functionalities of digital customer experience touchpoints such as websites, digital platforms, apps, chat bots and more. First of all, this is absolutely relevant, but in many areas still uncharted area for researchers. Secondly, students will benefit for understanding customer acceptance of new digital touchpoints and gain deep insight which will be a real booster for their further research or business career.
- Bagherifard, K., Rahmani, M., Nilashi, M., & Rafe, V. (2017). Performance improvement for recommender systems using ontology. Telematics and Informatics, 34(8), pp. 1772-1792.
- Cheng, L., Wang, H. (2014). A fuzzy recommender system based on the integration of subjective preferences and objective information. Applied Soft Computing, 18, pp. 290-301.
- Isinkaye, F., Folajimi, Y., & Ojokoh, B. (2015). Recommendation systems: Principles, methods and evaluation. Egyptian Informatics Journal, 16(3), pp. 261-273.
- Jung, H.H., Kraft, P. (2017). Vorwort, in, Jung, H.H., Kraft, P.: Digital vernetzt. Transformation der Wertschöpfung. Szenarien, Optionen und Erfolgsmodelle für Geschäftsmodelle, Produkte und Services, pp. 5-11. Munich, Hanser.
- Lops, P., Gemmis, M. D., & Semeraro, G. (2011). Content-based Recommender Systems: State of the Art and Trends. In Ricci F., Rokach L., Shapira B., Kantor P. (Eds.), Recommender Systems Handbook (pp. 73-105). Boston, MA: Springer.
- Lu, J., Wu, D., Mao, M., Wang, W., & Zhang, G. (2015). Recommender system application developments: A survey. Decision Support Systems, 74, pp. 12-32.
- Malhotra, N. K., Kim, S. S., & Agarwal, J. (2004). Internet Users’ Information Privacy Concerns (IUIPC): The Construct, the Scale, and a Causal Model. Information Systems Research, 15(4), pp. 336-355.
- Sivapalan, S., Sadeghian, A., Rahnama, H., & Madni, A. M. (2014). Recommender systems in e-commerce. 2014 World Automation Congress (WAC), pp. 179-184.
- Venkatesh, V., Morris, M. G., Davis, G. B., & Davis, F. D. (2003). User Acceptance of Information Technology: Toward a Unified View. MIS Quarterly, 27(3), pp. 425-478.