Extract features
Feature Extraction Agent
The feature extration agent can be used to mine information from data over a range of modalities: text, JSON, image and audio (video coming soon).
To use first add your input nodes and supply a prompt detailing what information you want extracted from your data. The agent will reply with a schema with features that it will extract from the data. You can then propose any changes or accept the schema and begin the extraction job.
To use external files (image, audio etc) select the Use node content
field. Each input node should have a URL to an uploaded file.
Input nodes can be supplied in three ways:
* A single node can be given which is used as a header for the input nodes.
* The input nodes can be directly provided.
* If the Consider the successors
field is set then the nodes are interpreted as feature nodes. Their successors are used as the input nodes and the data encoded in the subgraph is used as input. The features of each node are encoded as a JSON object which is used by the extractor.
The resulting view will show the extracted features alongside your input nodes.
Example (old): Using the On selected
mode to extract movie genres from descriptive plot overviews
Feature Describer
The Describer tells you the predecessor nodes that best “describe” your selection of nodes. By default, this description is by comparison to the other “background” nodes in the active view; open the settings by clicking the button in the bottom right for more advanced options.
Describing predecessors are presented in a table alongside their importance
: large values of importance,
whether positive or negative, indicate that a given predecessor is a good description of the selection.
Positive importance values indicate that the predecessor is likely to connect to the selection of nodes or
does so with a greater edge weight than the background nodes, negative values indicate the converse.
To see all of the describing predecessors and their importances in a view, click “Create View”.