US 12,277,129 B2
Data recommender using lineage to propagate value indicators
Ted Dunning, Santa Clara, CA (US); Suparna Bhattacharya, Bangalore (IN); Glyn Bowden, Bristol (GB); Lin A. Nease, San Jose, CA (US); Janice M. Zdankus, San Jose, CA (US); and Sonu Sudhakaran, Bangalore (IN)
Assigned to Hewlett Packard Enterprise Development LP, Spring, TX (US)
Filed by HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP, Spring, TX (US)
Filed on Jul. 12, 2023, as Appl. No. 18/351,309.
Application 18/351,309 is a continuation of application No. 17/843,757, filed on Jun. 17, 2022, granted, now 11,907,241.
Prior Publication US 2023/0409586 A1, Dec. 21, 2023
Int. Cl. G06F 16/248 (2019.01); G06F 16/2455 (2019.01); G06F 16/25 (2019.01); G06F 16/28 (2019.01)
CPC G06F 16/248 (2019.01) [G06F 16/24556 (2019.01); G06F 16/254 (2019.01); G06F 16/288 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A system comprising:
a graphical user interface (GUI);
a data pipeline database;
a one or more processing resources; and
a non-transitory computer-readable medium, coupled to the one or more processing resources, having stored therein instructions that when executed by the one or more processing resources cause the system to:
generate lineal associations between an analytical model and ancestor data-related products used to create the analytical model based on extracting and characterizing metadata from data processing pipelines that create the analytical model from the ancestor data-related products;
assign a value indicator to the analytical model, the value indicator being a quantifiable measurement of the analytical model's value derived from human behavioral actions related to usage of the analytical model;
propagate the value indicator through the generated lineal associations, wherein propagating the value indicator through the generated lineal associations comprises attributing, to each ancestor data-related product, the value indicator for the analytical model;
encapsulate the generated lineal associations and attributed value indicators as one or more exportable data-related packages stored to the data pipeline database;
based on the propagated value indicators, rank the ancestor data-related products in order of attributed value;
receive, via the GUI, parameters related to a prospective analytical model;
reference the exportable data-related packages to predict a subset of ancestor data-related products by correlating the parameters with the attributed indicator;
based on the ranking, recommend, via the GUI, one or more of the ancestor data-related products from the subset of ancestor data-related products; and
create the prospective analytical model using the recommended one or more of the ancestor data-related products.