Knowledge Bases (KBs) and textual documents contain rich and complementary information about real-world objects, as well as relations among them. While text documents describe entities in freeform, KBs organizes such information in a structured way. This makes these two information representation forms hard to compare and integrate, limiting the possibility to use them jointly to improve predictive and analytical tasks. In this article, we study this problem, and we propose KADE, a solution based on a regularized multi-task learning of KB and document embeddings. KADE can potentially incorporate any KB and document embedding learning method. Our experiments on multiple datasets and methods show that KADE effectively aligns document and entities embeddings, while maintaining the characteristics of the embedding models.
CITATION STYLE
Baumgartner, M., Zhang, W., Paudel, B., Dell’Aglio, D., Chen, H., & Bernstein, A. (2018). Aligning knowledge base and document embedding models using regularized multi-task learning. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11136 LNCS, pp. 21–37). Springer Verlag. https://doi.org/10.1007/978-3-030-00671-6_2
Mendeley helps you to discover research relevant for your work.