The recommendation is the new and probably the only method to ensure any online application effectively caters to a global audience. An application that caters to a very wide group of people from different backgrounds so people can find the data they could be looking for. We see the use of recommendations almost everywhere in our digital lives. We can look at recommended apps, posts, dishes, songs, and this list extends till infinity. Personalizing an application for any specific user helps in achieving better conversion rates. This is achieved through recommending the user based on the data collected about the user. The more data is collected, the more precise the recommendation gets, thus resulting in greater conversion rates. Various machine learning algorithms could be used for creating a recommendation system based on the type of recommendation required. The recommendation could get better with the collection of more data and then training the algorithm on the particular dataset. More precise recommendations could lead to the reverse effect of the recommendation. It could be used to model the user behavior according to what the developer wants. This could lead to negative consumer patterns as user now does what the developer wants them to do. But this does not mean we forbid the idea of recommendation asa whole. A recommendation is a great tool for achieving more conversion rates. In this paper, the authors propose recommending using groups. This not only ensures that reverse consumer behavior is prevented from being achieved, but is also computationally cheaper.
Dergi Türü : Uluslararası
Benzer Makaleler | Yazar | # |
---|
Makale | Yazar | # |
---|