Method and apparatus for synthesizing adaptive data visualizations

A computer-implemented method to dynamically update presentations based on context classification of voice inputs, comprising:

• displaying, using a presentation computing device, a first graphical user interface associated with a first context via a user interface;
• obtaining, using the presentation computing device, a first voice input;
• determining, using the presentation computing device, one or more first terms from the first voice input;
• determining, using the presentation computing device, that the first voice input is unrelated to the first context based on the one or more first terms;
• in response to determining that the first voice input is related to the first context:
• inputting, using the presentation computing device, the one or more first terms from the first voice input into a graph classifier to determine a graph type;
• and
• adding, using the presentation computing device, a graph to the first graphical user interface corresponding to the determined graph type;
• obtaining, using the presentation computing device, a second voice input;
• determining, using the one or more computing devices, one or more second terms from the second voice input;
• determining, using the presentation computing device, that the second voice input is unrelated to the first context based on the one or more second terms;
• and
• in response to determining that the second voice in put is unrelated to the first context:
• inputting, using the presentation computing device, the one or more second terms to a generative model to determine that a region of the displayed graph is statistically significant;
• and
• displaying, using the presentation computing device in response to determining the region that is statistically significant in the first graphical user interface, a close up view of the statistically significant portion of the graph as an overlay over the graph.

A computer program product, comprising:

• a non-transitory computer-readable medium having computer-readable program instructions embodied thereon that when executed by a processor dynamically update presentations based on context classification of voice inputs, the computer-readable program instructions comprising:
• computer-readable program instructions to display a first graphical user interface associated with a first context via a user interface;
• computer-readable program instructions to obtain a first voice input;
• computer-readable program instructions to determine one or more first terms from the first voice input;
• computer-readable program instructions to determine that the first voice input is related to the first context based on the one or more first terms;
• computer-readable program instructions to input, in response to determining that the first voice input is related to the first context, the one or more first terms from the first voice input into a graph classifier to determine a graph type;
• computer-readable program instructions to add a graph to the first graphical user interface corresponding to the determined graph type;
• computer-readable program instructions to obtain a second voice input;
• computer-readable program instructions to determine one or more second terms from the second voice input;
• computer-readable program instructions to determine that the second voice input is unrelated to the first context based on the one or more second terms;
• computer-readable program instructions to input, in response to determining that the second voice input is not related to the first context, the one or more second terms to a generative model to determine that a region of the displayed graph is statistically significant;
• and
• computer-readable program instructions to display, in response to determining the region that is statistically significant in the first graphical user interface, a close up view of the statistically significant portion of the graph as an overlay over the graph.