Counterparty profile for an electronic agreement

A method for generating counterparty profiles for electronic agreements, the method comprising:

• based on an indication of a change to content of a pending clause of an electronic agreement associated with a first entity and a second entity, selecting, by processing circuitry and based on determining that metadata for a first historical document of a plurality of historical documents specifies the first entity and the second entity and metadata for a second historical document of the plurality of historical documents specifies the first entity and the second entity, the first historical document and the second historical document from the plurality of historical documents,
• wherein a historical documents database is configured to store, for each historical document of the plurality of historical documents, content of one or more clauses of the respective historical document and metadata specifying a plurality of entities for the respective historical document;
• based on selecting the first historical document and the second historical document, retrieving, by the processing circuitry and from a clause database configured to store vector embeddings for the plurality of historical documents, a first vector embedding for a first clause of the first historical document and a second vector embedding for a second clause of the second historical document, wherein a machine learning model is configured to generate the vector embeddings for the plurality of historical documents;
• generating, by the processing circuitry, with the machine learning model, and based on content of the pending clause with the change, a vector embedding for the pending clause;
• selecting, by the processing circuitry and based on the vector embedding for the pending clause with the change, the first vector embedding for the first clause of the first historical document, and the second vector embedding for the second clause of the second historical document, the first clause of the first historical document and the second clause of the second historical document;
• based on selecting the first clause of the first historical document and the second clause of the second historical document, determining, by the processing circuitry, that first content of the first clause of the first historical document indicates a first value for an attribute for the pending clause and determining, by the processing circuitry, that second content of the second clause of the second historical document indicates a second value for the attribute for the pending clause;
• generating, by the processing circuitry, a first expected execution time for the first value for the attribute based on a first execution time for the first historical document;
• generating, by the processing circuitry, a second expected execution time for the second value for the attribute based on a second execution time for the second historical document;
• generating, by the processing circuitry and based on the first expected execution time and the second expected execution time, a counterparty profile;
• and
• outputting, by the processing circuitry, an indication of the counterparty profile for the second entity.

Non-transitory computer-readable media encoded with instructions that, when executed by processing circuitry, cause the processing circuitry to:

• based on an indication of a change to content of a pending clause of an electronic agreement associated with a first entity and a second entity, select, based on a determination that metadata for a first historical document of a plurality of historical documents specifies the first entity and the second entity and metadata for a second historical document of the plurality of historical documents specifies the first entity and the second entity, the first historical document and the second historical document from the plurality of historical documents,
• wherein a historical documents database is configured to store, for each historical document of the plurality of historical documents, content of one or more clauses of the respective historical document and metadata specifying a plurality of entities for the respective historical document;
• based on a selection of the first historical document and the second historical document, retrieve, from a clause database configured to store vector embeddings for the plurality of historical documents, a first vector embedding for a first clause of the first historical document and a second vector embedding for a second clause of the second historical document, wherein a machine learning model is configured to generate the vector embeddings for the plurality of historical documents;
• generate, with the machine learning model and based on content of the pending clause with the change, a vector embedding for the pending clause;
• select, based on the vector embedding for the pending clause with the change, the first vector embedding for the first clause of the first historical document, and the second vector embedding for the second clause of the second historical document, the first clause of the first historical document and the second clause of the second historical document;
• based on the selection of the first clause of the first historical document and the second clause of the second historical document, determine that first content of the first clause of the first historical document indicates a first value for an attribute for the pending clause and determine that second content of the second clause of the second historical document indicates a second value for the attribute for the pending clause;
• generate a first expected execution time for the first value for the attribute based on a first execution time for the first historical document;
• generate a second expected execution time for the second value for the attribute based on a second execution time for the second historical document;
• generate, based on the first expected execution time and the second expected execution time, a counterparty profile;
• and
• output an indication of the counterparty profile for the second entity.