Systems and methods for vendor intelligence

Patent metadata

Field	Value
Publication number	US-12014308-B1
Application number	US-202117556266-A
Country	US
Kind code	B1
Filing date	Dec 20, 2021
Priority date	Mar 1, 2018
Publication date	Jun 18, 2024
Grant date	Jun 18, 2024

Abstract

A computing system includes a network interface configured to communicate data over a network, a vendor database configured to retrievable store vendor data related to a vendor of interest, and a processing circuit with a processor and memory. The memory is structured to store instructions that are executable by the processor and cause the processing circuit to identify a vendor of interest; conduct a search of internal documents and external resources to obtain the vendor data related to the vendor of interest, store the vendor data in the vendor database, and generate a vendor performance score based on the vendor data.

First claim

What is claimed is:

1. A method, comprising: extracting first entity-related data from a first document; determining an entity based on the first entity-related data; determining at least one entity performance indicator for the entity based on the first entity-related data; identifying a first author of the first document: generating an entity performance score based on the at least one entity performance indicator; generating a semantic summary regarding the entity based on the at least one entity performance indicator, comprising translating, via machine learning, at least one quantitative aspect of the at least one entity performance indicator into a natural language summary included in the semantic summary; generating and transmitting a user interface configured to display an entity scorecard to a computing device associated with the first author of the first document, the entity scorecard comprising the entity performance score, the semantic summary, and a relationships widget that includes a list of engagement names of persons who previously engaged with the entity; detecting a second document comprising second entity-related data created by a second author, the second entity-related data relating to the entity; extracting second entity-related data from the second document; updating the entity performance score based on the second entity-related data; and updating the entity scorecard based on the updated entity performance score, wherein updating the entity scorecard includes updating the list of engagement names to include a name of the second author.

2. The method of claim 1 , wherein the operation of extracting the first entity-related data from the first document include: generating search criteria; searching the first document with the search criteria; and downloading the results of searching the first document.

3. The method of claim 1 , wherein the operation of determining the entity based on the first entity-related data includes: maintaining an entity database comprising a list of entity names, and monitoring internally generated documents for a mention of the entity from the list of entity names.

4. The method of claim 3 , wherein the operation of determining the entity further comprises executing a natural language processing filter structured to identify at least one of emails and internally generated documents, the at least one of emails and internally generated documents being indicative of the entity.

5. The method of claim 3 , wherein the operation of determining the entity further comprises executing a natural language processing filter structured to identify at least one of emails and internally generated documents, the at least one of emails and internally generated documents being indicative of vetting or contracting of a second entity that is not within the list of entity names.

6. The method of claim 1 , wherein the operation of determining the at least one entity performance indicator for the entity includes executing sentiment analysis on the first document to determine an emotional reaction of the first author of the first document.

7. The method of claim 6 , further comprising: determining a status of the first author of the first document; and assigning a reliability weighting correction to the sentiment analysis based on the status of the first author; wherein the operation of generating the entity performance score includes weighting the entity performance score based on the reliability weighting correction.

8. The method of claim 1 , wherein the entity scorecard further comprises at least one of a related entity widget, a quotes and facts widget, and a graphical visualization of the entity performance score.

9. The method of claim 1 , wherein generating the semantic summary includes providing one of a warning about the entity, a praise of the entity, and advice on managing the entity.

10. The method of claim 1 , wherein the second author from the list of engagement names is hyperlinked such that clicking on the second author causes the computing device associated with the first author to display contact information for the second author.

11. An apparatus, comprising: a network interface; a processing circuit comprising at least one processor and memory, the memory structured to store instructions that are executable by the at least one processor and cause the processing circuit to: extract first vendor-related data from a first electronically-generated document; determine a vendor of interest based on the first vendor-related data; determine at least one vendor performance indicator for the vendor of interest based on the first vendor-related data; identify a first author of the first electronically-generated document; generate a vendor performance score based on the at least one vendor performance indicator; generate a semantic summary regarding the vendor of interest based on the at least one vendor performance indicator, where generating the semantic summary includes translating, via machine learning, at least one quantitative aspect of the at least one vendor performance indicator into a natural language summary; generate and transmit a user interface configured to display a vendor scorecard the first author of the first electronically-generated document, the vendor scorecard comprising the vendor performance score, the semantic summary, and a relationships widget that includes a list of engagement names of persons who previously engaged with the vendor; detect a second electronically-generated document comprising second vendor-related data created by a second author, the second vendor-related data relating to the vendor; extract the second vendor-related data from the second electronically-generated document; update the vendor performance score based on the second vendor-related data; and update the vendor scorecard based on the updated vendor performance score, wherein updating the vendor scorecard includes updating the list of engagement names to include a name of the second author.

12. The apparatus of claim 11 , wherein the instructions of extract the first vendor-related data from the first electronically-generated document further cause the at least one processor to: generate search criteria; search the first electronically-generated document with the search criteria; and download the results of searching the first electronically-generated document.

13. The apparatus of claim 11 , wherein the instructions of determine the vendor of interest based on the first vendor-related data further cause the at least one processor to: maintain a vendor database comprising a list of vendor names, and monitor internally generated documents for a mention of the vendor of interest from the list of vendor names.

14. The apparatus of claim 13 , wherein the instructions of determine the vendor of interest further causes the at least one processor to execute a natural language processing filter structured to identify at least one of emails and internally generated documents, the at least one of emails and internally generated documents being indicative of a second vendor of interest.

15. The apparatus of claim 13 , wherein the instructions of determine the vendor of interest further causes the at least one processor to execute a natural language processing filter structured to identify at least one of emails and internally generated documents, the at least one of emails and internally generated documents being indicative of vetting or contracting of a second vendor of interest that is not within the list of vendor names.

16. One or more non-transitory computer readable medium storing instructions t

Assignees

• Wells Fargo Bank Na

Inventors

• Anderson Ryan
• Kroll Bryan
• Rao Abhijit
• Singh Priyamvada
• Vakili Masoud

Classifications

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06F16/904

  Browsing; Visualisation therefor (for navigating the web G06F16/954; browsing optimisation for the web G06F16/957) · CPC title

• G06F16/93

  Document management systems · CPC title

• G06Q10/06393 Primary

  Score-carding, benchmarking or key performance indicator [KPI] analysis · CPC title