This paper provides a comprehensive review of statistical relational learning (SRL) methods for knowledge graphs. It discusses two main types of SRL models: latent feature models and graph feature models, both of which can scale to large datasets. Latent feature models, such as tensor factorization and multiway neural networks, capture the correlation between nodes/edges using latent variables. Graph feature models, on the other hand, directly capture the correlation using observable properties of the graph. The paper also explores methods to combine these two approaches to enhance modeling power while reducing computational cost. Additionally, it discusses how SRL models can be combined with text-based information extraction methods to automatically construct knowledge graphs from the web, using Google's Knowledge Vault project as an example. The paper covers various aspects of SRL, including link prediction, entity resolution, and link-based clustering, and provides a detailed analysis of different models and their applications.This paper provides a comprehensive review of statistical relational learning (SRL) methods for knowledge graphs. It discusses two main types of SRL models: latent feature models and graph feature models, both of which can scale to large datasets. Latent feature models, such as tensor factorization and multiway neural networks, capture the correlation between nodes/edges using latent variables. Graph feature models, on the other hand, directly capture the correlation using observable properties of the graph. The paper also explores methods to combine these two approaches to enhance modeling power while reducing computational cost. Additionally, it discusses how SRL models can be combined with text-based information extraction methods to automatically construct knowledge graphs from the web, using Google's Knowledge Vault project as an example. The paper covers various aspects of SRL, including link prediction, entity resolution, and link-based clustering, and provides a detailed analysis of different models and their applications.