Understanding Latent Factors Using a GWAP
Information Retrieval
2018-08-31 v1 Machine Learning
Machine Learning
Abstract
Recommender systems relying on latent factor models often appear as black boxes to their users. Semantic descriptions for the factors might help to mitigate this problem. Achieving this automatically is, however, a non-straightforward task due to the models' statistical nature. We present an output-agreement game that represents factors by means of sample items and motivates players to create such descriptions. A user study shows that the collected output actually reflects real-world characteristics of the factors.
Cite
@article{arxiv.1808.10260,
title = {Understanding Latent Factors Using a GWAP},
author = {Johannes Kunkel and Benedikt Loepp and Jürgen Ziegler},
journal= {arXiv preprint arXiv:1808.10260},
year = {2018}
}
Comments
Proceedings of the Late-Breaking Results track part of the Twelfth ACM Conference on Recommender Systems (RecSys '18), Vancouver, BC, Canada, October 2-7, 2018, 2 pages