Product Recommendation System (Targeted Recommendation using Deep Learning in Computer Vision)
Traditional product displays in shopping malls often fail to effectively engage customers due to their generic nature. This report presents a project on developing a product recommendation system that utilizes facial recognition technology to predict a user's age and gender. The system leverage...
Saved in:
| Main Author: | |
|---|---|
| Format: | Undergraduates Project Papers |
| Language: | English |
| Published: |
2023
|
| Subjects: | |
| Online Access: | http://umpir.ump.edu.my/id/eprint/40901/1/CB20174.pdf http://umpir.ump.edu.my/id/eprint/40901/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Traditional product displays in shopping malls often fail to effectively engage customers due to their generic nature. This report presents a project on developing a product recommendation system that utilizes facial recognition technology to predict a user's age and gender. The system leverages the Multi-task Cascaded Convolutional Networks (MTCNN) algorithm, along with a classifier and the OpenCV library, to accurately recognize faces and extract age and gender information. The primary objective of this project is to create a personalized recommendation system that suggests products tailored to an individual's age group and gender. By employing facial recognition techniques, the system is capable of identifying a user's face in real-time and making predictions regarding their age and gender. The project workflow involves several key steps. First, the MTCNN algorithm is utilized to detect and extract facial features from images or video streams. Once the face is successfully recognized, a classifier model is employed to predict the user's age and gender based on the extracted features. The OpenCV library provides the necessary tools for implementing these functionalities. The recommendations are then generated by mapping the predicted age and gender to specific age groups and gender categories. Each age group and gender category are associated with a set of products suitable for the corresponding demographic. These recommendations aim to enhance the user experience by providing relevant and personalized suggestions that align with their specific needs and preferences. To evaluate the effectiveness of the system, extensive testing and validation are conducted using various datasets. The performance metrics considered include accuracy, precision and Mean Absolute Error (MAE). The results of the project demonstrate the potential of utilizing facial recognition technology to develop accurate and efficient product recommendation systems. The system's ability to accurately predict a user's age and gender contributes to a more personalized and tailored user experience. The project's findings open up possibilities for further research and development in the field of recommendation systems, paving the way for improved user engagement and customer satisfaction in various industries. |
|---|