CAREA: Cotraining Attribute and Relation Embeddings for Cross-Lingual Entity Alignment in Knowledge Graphs

Figure 2

An example of entity attributes representation: (a) attribute triples in each KG are merged into a set of key-value pairs, (b) attribute keys and values are used to represent the frequency feature and type feature, respectively, (c) the frequency and type information are integrated into a sparse matrix, and (d) each attribute pair forms a feature vector.