Affiliations: [a] School of Software, Xinjiang University, Urumqi, China
| [b]
Xinjiang Xinnong Network Information Center, Meteorological Bureau of Xinjiang Uygur Autonomous Region, Urumqi, China
Correspondence:
[*]
Corresponding author. Xingyao Yang, School of Software, Xinjiang University, Urumqi 830046, China. E-mail: yangxy@xju.edu.cn.
Abstract: In existing sequential recommendation systems, user behavior data are directly used as training data for the model to complete the training process and address recommendation tasks. However, user-generated behavioral data inevitably contains noise, and the use of the Transformer’s recommendation model may lead to overfitting on such noisy data. To address this issue, we introduce a sequence recommendation algorithm model named FAT-Rec, which incorporates fusion filters and converters through joint training. By employing joint training methods, we establish both a transformer prediction layer and a CTR prediction layer. Toward the end of the model, we assign weights and sum up the losses from the Transformer and CTR prediction layers to derive the final loss function. Experimental results on two widely used datasets, MovieLens and Goodbooks, demonstrate a significant enhancement in the performance of the proposed FAT-Rec recommendation algorithm compared with seven comparative models. This validates the efficacy of the fusion filter and transformer within the context of sequence recommendation tasks under the joint training mechanism.
Keywords: Filter, self-attention mechanism, transformer, joint training, user sequence
