Systems and methods for data narration

A method for data narration, comprising:

• obtaining, by a reinforcement learning model, a plurality of facts describing a dataset;
• generating, by a graph component, a graph including a plurality of nodes corresponding to the plurality of facts, respectively, and a plurality of edges connecting the plurality of nodes;
• computing, by an ordering component, an ordering of the plurality of facts based on the graph;
• and
• generating, by a language generator, a description of the dataset based on the ordering of the plurality of facts.

An apparatus for data narration, comprising:

• a processor;
• a memory including instructions executable by the processor;
• a reinforcement learning model configured to extract a plurality of facts from a dataset;
• a graph component configured to generate a graph including a plurality of nodes corresponding to the plurality of facts, respectively;
• and
• an ordering component configured to compute an ordering of the plurality of facts based on the graph.