Granted Patent B2

US 12,724,817 · App. 18/386,228 · Granted Sep 1, 2026

Inventors: Lewis C. Lee (Seattle, WA); Jeffrey Brendan Ryan (Liberty Lake, WA)

Assignee: Moat Metrics, Inc.

G06F16/358G06Q10/0637G06Q50/184

Abstract

Claims (90)

1 . A method comprising:

receiving intellectual property assets, individual ones of the intellectual property assets including first respective portions of text;

transforming the first respective portions of text into first vectors;

generating a custom vector space configured to maintain the first vectors instead of maintaining data representing the intellectual property assets, wherein the custom vector space represents a lower dimensionality than the data representing the intellectual property assets such that storage requires are decreased by generation and use of the custom vector space;

receiving a search prompt including second respective portions of text;

transforming the second respective portions of text into second vectors in the custom vector space;

generating a similarity score for a data set that includes first intellectual property (IP) assets based at least in part on comparing the first vectors to the second vectors;

sending a search request including the first data set that includes the first IP assets and the search prompt to a generative artificial intelligence (AI) model, the search request including an instruction to perform a search based on the prompt and to use the first vectors and the second vectors as input data;

receiving, from the generative AI model, a second data set that includes second IP assets;

displaying, to a user, a generated graphical user interface (GUI) configured to display the second data set that includes the second IP assets;

receiving, from the GUI, additional input data indicating a selection of a generative AI summary button;

sending a request to the generative AI model for a word summary of the second data set; and

receiving the word summary of the second data set, wherein the displayed GUI is updated based at least in part on the word summary of the second data set.

2 . The method of claim 1 , further comprising:

generating a first vector representation of the first data set based at least in part on the first IP assets;

generating a second vector representation of the second data set based at least in part on the second IP assets;

determining a ranking of the second IP assets based at least in part on comparing the first vector representation to the second vector representation;

sending the generative AI model a third data set that includes the ranking of the second IP assets as a feedback input; and

sending a second search request including a fourth data set that includes the first IP assets to the generative AI model.

3 . The method of claim 2 , further comprising:

receiving, from the generative AI model, a fifth data set that includes third IP assets;

generating a third vector representation of the fifth data set based at least in part on the third IP assets;

generating a fourth vector representation of the fourth data set based at least in part on the first IP assets;

determining a ranking of the third IP assets based at least in part on comparing the third vector representation to the fourth vector representation;

sending the generative AI model a sixth data set that includes the ranking of the third IP assets as a feedback input; and

sending a third search request including a seventh data set that includes the first IP assets to the generative AI model.

4 . The method of claim 2 , wherein generating the ranking of the second IP assets include generating a similarity score associated with the second IP assets based at least in part on comparing the first vector representation to the second vector representation.

5 . The method of claim 1 , wherein the first data set includes at least one non-IP based assets.

6 . The method of claim 5 , wherein the at least one non-IP based asset includes a 10-K filing associated with an entity.

7 . A method comprising:

receiving intellectual property assets, individual ones of the intellectual property assets including first respective portions of text;

transforming the first respective portions of text into first vectors;

generating a custom vector space configured to maintain the first vectors instead of maintaining data representing the intellectual property assets, wherein the custom vector space represents a lower dimensionality than the data representing the intellectual property assets such that storage requires are decreased by generation and use of the custom vector space;

receiving a search prompt including second respective portions of text;

transforming the second respective portions of text into second vectors in the custom vector space;

generating a similarity score for a data set that includes first intellectual property (IP) assets based at least in part on comparing the first vectors and the second vectors;

sending a market outlook search request including the first data set that includes the first IP assets and the search prompt to a generative artificial intelligence (AI) model, the search request including an instruction to perform a search based on the prompt and to use the first vectors and the second vectors as input data;

receiving the market outlook from the generative AI model;

displaying, to a user, a generated graphical user interface (GUI) configured to display the second data set that includes the second IP assets;

receiving, from the GUI, additional input data indicating a selection of a generative AI summary button;

sending a request to the generative AI model for a word summary of the second data set; and

receiving the word summary of the second data set, wherein the displayed GUI is updated based at least in part on the word summary of the second data set.

8 . The method of claim 7 , further comprising receiving additional input data including selection of a at least one of a first time period or a second time period via a slider mechanism.

9 . The method of claim 8 , further comprising displaying the second data set in response to receiving the additional input data via the slider mechanism.

10 . The method of claim 7 , wherein generating a ranking of the second IP assets include generating a similarity score associated with the second IP assets.

11 . The method of claim 10 , wherein the ranking of the second IP assets is based at least in part on at least one of:

geographical data;

breadth data;

expiration data;

diversity data;

revenue alignment data;

invalidity data;

filing velocity data;

spending data;

predictive analytics data;

precedence data;

litigation data;

market data; or

revenue alignment data.

12 . The method of claim 10 , further comprising:

receiving additional input data representing the input, the input data indicating a selection of a generative AI summary button;

sending a request to the generative AI model for a word summary of the second data set;

receiving the word summary of the second data set; and

causing the GUI to display at least the word summary.

13 . A system comprising:

one or more processors; and

one or more non-transitory computer-readable media comprising instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

receiving intellectual property assets, individual ones of the intellectual property assets including first respective portions of text;

transforming the first respective portions of text into first vectors in a custom vector space;

receiving a search prompt including second respective portions of text;

transforming the second respective portions of text into second vectors;

generating a custom vector space configured to maintain the first vectors instead of maintaining data representing the intellectual property assets, wherein the custom vector space represents a lower dimensionality than the data representing the intellectual property assets such that storage requires are decreased by generation and use of the custom vector space;

generating a similarity score for a data set that includes first intellectual property (IP) assets based at least in part on comparing the first vectors and the second vectors;

sending a search request including the first data set that includes the first IP assets and the search prompt to a generative artificial intelligence (AI) model, the search request including an instruction to perform a search based on the prompt and to use the first vectors and the second vectors as input data; and

receiving, from the generative AI model, a second data set that includes second IP assets; and

displaying to a user, a generated graphical user interface (GUI) configured to display the second data set that includes the second IP assets, wherein the displayed GUI is updated based at least in part on a selection of at least one IP asset displayed on a first portion of the GUI.

14 . The system of claim 13 , the operations further comprising:

generating a first vector representation of the first data set based at least in part on the first IP assets;

generating a second vector representation of the second data set based at least in part on the second IP assets;

determining a ranking of the second IP assets based at least in part on comparing the first vector representation to the second vector representation;

sending the generative AI model a third data set that includes the ranking of the second IP assets as a feedback input; and

sending a second search request including a fourth data set that includes the first IP assets to the generative AI model.

15 . The system of claim 14 , the operations further comprising:

receiving, from the generative AI model, a fifth data set that includes third IP assets;

generating a third vector representation of the fifth data set based at least in part on the third IP assets;

generating a fourth vector representation of the fourth data set based at least in part on the first IP assets;

determining a ranking of the third IP assets based at least in part on comparing the third vector representation to the fourth vector representation;

sending the generative AI model a sixth data set that includes the ranking of the third IP assets as a feedback input; and

sending a third search request including a seventh data set that includes the first IP assets to the generative AI model.

16 . The system of claim 14 , wherein generating the ranking of the second IP assets include generating a similarity score associated with the second IP assets based at least in part on comparing the first vector representation to the second vector representation.

Assignments (3)

Continuity (1)

References Cited (11)