A Novel deep ensemble classifier for recommending fashion products
Buradagunta Suvarna, Sivadi Balakrishna · 2022
The main aim of the recommender system is to recommend items that are similar to the given query image. Extracting similar items from a large collection of data is a challenging task. Systems for online shopping are exploring ways to recommend products according to the user's interests. Different statistical methods and similarity metrics were used in earlier days to get similar items, which is leading the recommendation of the products with less accuracy and precision. A Novel Deep ensemble classifier model is proposed for classifying the given product. The model is tested on Fashion product data set and the results are attractive. With this, it is possible to recommend the items with 88.32% accuracy. The suggested strategy outperforms current approaches in the form of accuracy, precision, recall, F1-score, and kappa statistics.