Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This paper introduces RAPI, a framework for inferring user characteristics from recommendation lists across four analytical scenarios with varying information availability. RAPI uses a surrogate recommendation model trained on content embeddings from BERT to simulate the original recommendation model and generate extended recommendation lists via a sample augmentation module. An adaptive weight classification model then infers user characteristics from these lists, achieving inference accuracies of 0.764 and 0.6477 on two datasets.
Extracting user profiles from recommendation lists is now more accurate thanks to RAPI, a new framework that leverages BERT embeddings and sample augmentation to boost inference accuracy by dynamically weighting user characteristics.
Recommender systems (RS) play a core role in various domains, including business analytics, helping users and companies make appropriate decisions. To optimize service quality, related technologies focus on constructing user profiles by analyzing users'historical behavior information. This paper considers four analytical scenarios to evaluate user profiling capabilities under different information conditions. A generic user attribute analysis framework named RAPI is proposed, which infers users'personal characteristics by exploiting easily accessible recommendation lists. Specifically, a surrogate recommendation model is established to simulate the original model, leveraging content embedding from a pre-trained BERT model to obtain item embeddings. A sample augmentation module generates extended recommendation lists by considering similarity between model outputs and item embeddings. Finally, an adaptive weight classification model assigns dynamic weights to facilitate user characteristic inference. Experiments on four collections show that RAPI achieves inference accuracy of 0.764 and 0.6477, respectively.