kyegomez/tiktokx

Tiktok is an advanced multimedia recommender system that fuses the generative modality-aware collaborative self-augmentation and contrastive cross-modality dependency encoding to achieve superior performance compared to existing state-of-the-art multi-model recommenders.

TikTokX: Multi-Modal Recommentation Algorithm

Tiktok is an advanced multimedia recommender system that fuses the generative modality-aware collaborative self-augmentation and contrastive cross-modality dependency encoding to achieve superior performance compared to existing state-of-the-art multi-model recommenders.

Installation

pip install tiktokx

---

Usage

To start training and inference:

python example.py --dataset {DATASET}

Supported datasets include Amazon-Baby, Amazon-Sports, Tiktok, and Allrecipes.

---

Datasets

Dataset specifications are tabulated below:

Dataset	Modality	Embed Dim	User	Item	Interactions	Sparsity
Amazon	V T	4096 1024	35598	18357	256308	99.961%
Tiktok	V A T	128 128 768	9319	6710	59541	99.904%
Allrecipes	V T	2048 20	19805	10067	58922	99.970%

Datasets can be accessed from Google Drive. Note: The official website for the Tiktok dataset is no longer available. However, we've processed and made available various versions of the Tiktok dataset. Kindly cite our work if you utilize our preprocessed Tiktok dataset.