Abstract:
Online recommendation performance improvement challenges include recommendation efficiency and data sparsity. Most related work improves recommendation accuracy rather than efficiency. We propose a Deep Pairwise Hashing (DPH) to map users and items to binary vectors in Hamming space, where Hamming distance can efficiently calculate a user’s preference for an item, improving online recommendation. To overcome data sparsity and cold-start issues, user-item interactive information and item content information are unified. Specifically, we first pre-train robust item representation from item content data by a Denoising Auto-encoder instead of other deterministic deep learning frameworks; then we finetune the framework by adding a pairwise loss objective with discrete constraints; and finally, DPH aims to minimize a pairwise ranking loss that is consistent with the ultimate goal of recommendation. Finally, we optimize the discrete constraint model using alternating optimization. DPH improves data sparsity and item cold-start recommendation in three datasets.
Note: Please discuss with our team before submitting this abstract to the college. This Abstract or Synopsis varies based on student project requirements.
Did you like this final year project?
To download this project Code with thesis report and project training... Click Here
