In this project, I worked directly with a customer in Japan to develop a recommendation algorithm that suggests similar liquors based on taste coefficients. The algorithm ultimately outperformed a well-known sake recommendation website (https://sakenowa.com/). This work also resulted in two publications in prestigious venues.
Please find more detail of our algorithms in our 2 publish papers below:
On the left, the search bar represents for the taste coefficients we want to find similar liquors. Right, the 3 remaining figures are the 3 most similar items that share the cloest distribution to the query input item.
References
2022
Solving distribution problems in content-based recommendation system with gaussian mixture model
@article{van2022solving,title={Solving distribution problems in content-based recommendation system with gaussian mixture model},author={Nguyen, Dat and Van Toan, Pham and Thanh, Ta Minh},journal={Applied Intelligence},volume={52},number={2},pages={1602--1614},year={2022},publisher={Springer},}
2020
Robust content-based recommendation distribution system with gaussian mixture model
Dat Nguyen, Van Toan Pham , and Ta Minh Thanh
In International Conference on Computational Collective Intelligence , Jun 2020
@inproceedings{van2020robust,title={Robust content-based recommendation distribution system with gaussian mixture model},author={Nguyen, Dat and Pham, Van Toan and Thanh, Ta Minh},booktitle={International Conference on Computational Collective Intelligence},pages={199--211},year={2020},organization={Springer},}
