Enhanced Meta – Interest Based Product Recommendation System by Using Myer – Briggs Method

Resumen

In today's ecommerce and business, recommender systems are quite important. Users are
recommended books, movies, videos, electronic gadgets, and a variety of other products
using recommender systems. Recommender systems assist users in receiving customized
recommendations. Assisting consumers in making the best decisions possible in their online
transactions, and boosting sales and reimagining people' web surfing experiences, retaining
customers, and improving their purchasing experience. In two ways, the existing system is
personality-aware: it uses the user's personality features to forecast his or her themes of
interest and to match the user's preferences. The system that has been used was compared to
more modern methods of recommendation, such as a deeplearning-based recommendation
system and a session-based recommendation system. The findings of the experiments reveal
that the existing strategy has the potential to improve the precision and recall of the data. The
existing system works well in cold start problems. The proposed system uses enhanced metainterest
process. The system integrates MyerBriggs personality trait with Big five traits. The
enhanced meta-interest process is compared with deep learning based recommendation
system, Session based recommendation system and Meta interest system. The time and speed
of the proposed system is evaluated by using parameters like precision, recall and F-Measure.
Precision and recall should be maintained stable and the enhanced system should alleviate
redundancy and cold start problem