Product recommendation using deep learning in computer vision
Recently, recommendation models have gained popularity due to their effectiveness in improving customer satisfaction and deriving sales. However, current product recommendation models have a drawback: they lack personalized and targeted advertisements for individual users. Consequently, the recommen...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English English |
| Published: |
Institute of Electrical and Electronics Engineers Inc.
2023
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/40341/1/Product%20recommendation%20using%20deep%20learning%20in%20computer%20vision.pdf http://umpir.ump.edu.my/id/eprint/40341/2/Product%20recommendation%20using%20deep%20learning%20in%20computer%20vision_ABS.pdf http://umpir.ump.edu.my/id/eprint/40341/ https://doi.org/10.1109/ICSECS58457.2023.10256332 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Recently, recommendation models have gained popularity due to their effectiveness in improving customer satisfaction and deriving sales. However, current product recommendation models have a drawback: they lack personalized and targeted advertisements for individual users. Consequently, the recommendations provided are random and not tailored to users' preferences. This limitation negatively impacts the system's ability to deliver relevant and personalized advertisements, leading to reduced user engagement and potentially lower conversion rates. Moreover, the absence of personalized advertisements can result in user dissatisfaction as they may receive recommendations that are irrelevant or not aligned with their interests and needs. To address these challenges, this study proposed a targeted product recommendation model using Deep Learning (DL) techniques in computer vision. The study utilizes the dataset of human images obtained from the Kaggle website, which includes details such as gender, class, and age. Findings of the study demonstrated a high level of accuracy in product recommendations, indicating the potential for significant improvements in addressing the issues. In conclusion, the proposed method achieves good accuracy in predicting the gender and age, and provides appropriate product recommendations based on these features. |
|---|