A Trio Neural Model for Dynamic Entity Relatedness Ranking
This addresses the need for dynamic entity relatedness in NLP and information retrieval applications, but appears incremental as it builds on prior static and unsupervised methods.
The paper tackles the problem of measuring dynamic entity relatedness, where entity relationships change over time, by proposing a neural network approach that uses collective attention as supervision. It demonstrates better results than competitive baselines on large-scale datasets.
Measuring entity relatedness is a fundamental task for many natural language processing and information retrieval applications. Prior work often studies entity relatedness in static settings and an unsupervised manner. However, entities in real-world are often involved in many different relationships, consequently entity-relations are very dynamic over time. In this work, we propose a neural networkbased approach for dynamic entity relatedness, leveraging the collective attention as supervision. Our model is capable of learning rich and different entity representations in a joint framework. Through extensive experiments on large-scale datasets, we demonstrate that our method achieves better results than competitive baselines.