	US 12,254,522 B2
	Contract recommendation platform
Lijuan Marissa Zhou, Dublin (IE); Qirun Chen, Dublin (IE); Tianhua Hu, Dalian (CN); Mark Dineen, Dublin (IE); and Stephen Redmond, Dublin (IE)
Assigned to Accenture Global Solutions Limited, Dublin (IE)
Filed by Accenture Global Solutions Limited, Dublin (IE)
Filed on Mar. 30, 2021, as Appl. No. 17/217,504.
Claims priority of application No. PCT/CN2020/091320 (WO), filed on May 20, 2020.
Prior Publication US 2021/0366065 A1, Nov. 25, 2021
Int. Cl. G06Q 50/18 (2012.01); G06F 16/338 (2019.01); G06F 40/166 (2020.01); G06Q 10/067 (2023.01)

CPC G06Q 50/188 (2013.01) [G06F 16/338 (2019.01); G06F 40/166 (2020.01); G06Q 10/067 (2013.01)]

20 Claims

1. A method, comprising:

receiving, by a device, historical contract data;

reading, by the device and from the historical contract data, numerical categorical data identifying explicit historical contract data;

processing, by the device, the numerical categorical data, with one or more artificial intelligence models, to convert the numerical categorical data into vector format;

creating, by the device, a taxonomy in vector format based on the numerical categorical data in vector format;

reading, by the device and from the historical contract data, nonnumerical noncategorical data identifying implicit historical contract data;

processing, by the device, the nonnumerical noncategorical data, with one or more artificial intelligence models, to convert the nonnumerical noncategorical data into vector format;

creating, by the device, a knowledge graph in vector format based on the nonnumerical noncategorical data in vector format;

receiving, by the device and from a client device, a request that includes data identifying a new contract to be created;

identifying, by the device and in the taxonomy, a first set of similar contracts that include data similar to the data identifying the new contract;

identifying, by the device and in the knowledge graph, a second set of similar contracts that include data similar to the data identifying the new contract;

determining, by the device, similarity scores of a first similarity between the data identifying the new contract and data identifying the first set of similar contracts and a second similarity between the data identifying the new contract and data identifying the second set of similar contracts;

receiving, from the client device, feedback data associated with a feedback on the similarity scores;

determining weights for the similarity scores based on the feedback data;

updating the similarity scores of the first similarity and the second similarity based on the weights to generate the final similarity scores;

generating, by the device and based on the final similarity scores, a draft of the new contract by modifying particular terms in one or more historical contract clauses from one or more contracts of the first set of similar contracts and the second set of similar contracts,

wherein generating the draft of the new contract comprises:

identifying and selecting the one or more historical contract clauses based the particular terms in on the one or more historical contract clauses corresponding to first features meeting a threshold level of match with respect to second features of the new contract and

generating the draft of the new contracts based on modifying the particular terms in the historical contract clauses to bring the particular terms in conformity with the data identifying the new contract and to generate modified clauses and including the modified clauses in the draft of the new contract; and

providing, by the device and to the client device, the draft of the new contract and the final similarity scores, the data identifying the first set of similar contracts, and the data identifying the second set of similar contracts.