Embedding Learning with Events in Heterogeneous Information Networks

Abstract

In real-world applications, objects of multiple types are interconnected, forming <i>Heterogeneous Information Networks</i>. In such heterogeneous information networks, we make the key observation that many interactions happen due to some <i>event</i> and the objects in each event form a complete semantic unit. By taking advantage of such a property, we propose a generic framework called <b>H</b><i>yper</i><b>E</b><i>dge</i>-<b>B</b><i>ased</i><b>E</b><i>mbedding</i> (Hebe) to learn object embeddings with events in heterogeneous information networks, where a <i>hyperedge</i> encompasses the objects participating in one event. The Hebe framework models the proximity among objects in each event with two methods: (1) predicting a target object given other participating objects in the event, and (2) predicting if the event can be observed given all the participating objects. Since each hyperedge encapsulates more information of a given event, Hebe is robust to data sparseness and noise. In addition, Hebe is scalable when the data size spirals. Extensive experiments on large-scale real-world datasets show the efficacy and robustness of the proposed framework.