Methods, systems, articles of manufacture, and apparatus to map client specifications with standardized characteristics
Methods, systems, articles of manufacture, and apparatus are disclosed to map client specifications to standardized characteristics. An example apparatus includes a cluster identifier to cluster client databases into client clusters based on a threshold quantity of overlapping universal product codes (UPCs) between respective ones of the client databases, a characteristic analyzer to identify custom characteristics from the respective ones of the client clusters, ones of the custom characteristics having dissimilar nomenclature, and a graph builder to cluster the ones of the custom characteristics based on a similarity metric, and normalize the ones of the custom characteristics as a proxy characteristic, the proxy characteristic having a common nomenclature to represent the ones of the custom characteristics, the characteristic analyzer to enable improved product marketing analysis by replacing dissimilar nomenclature with the proxy characteristic.
1. An apparatus to map client specifications with standardized characteristics, the apparatus comprising:
a cluster identifier to cluster client databases into client clusters based on a threshold quantity of overlapping universal product codes (UPCs) between respective ones of the client databases;
a characteristic analyzer to identify custom characteristics from the respective ones of the client clusters, ones of the custom characteristics having dissimilar nomenclature; and
a graph builder to:
cluster the ones of the custom characteristics based on a similarity metric; and
normalize the ones of the custom characteristics as a proxy characteristic, the proxy characteristic having a common nomenclature to represent the ones of the custom characteristics,
the characteristic analyzer to enable improved product marketing analysis by replacing dissimilar nomenclature with the proxy characteristic.
2. The apparatus as defined in claim 1 , wherein the graph builder is to:
generate first nodes associated with UPCs having first ones of the custom characteristics; and
generate second nodes associated with UPCs having second ones of the custom characteristics.
3. The apparatus as defined in claim 2 , wherein the first ones of the custom characteristics have a first nomenclature, and the second ones of the custom characteristics have a second nomenclature different than the first nomenclature.
4. The apparatus as defined in claim 2 , wherein the graph builder is to generate a micro-similarity score between pairs of (a) the first nodes and (b) the second nodes.
5. The apparatus as defined in claim 4 , wherein the graph builder is to generate the micro-similarity scores based on a threshold overlap of UPCs between the pairs of the first and second nodes.
6. The apparatus as defined in claim 4 , wherein the graph builder is to apply Maximum Weighted Bipartite Graph Matching (MWBGM) to the first nodes and the second nodes, respective first ones of the second nodes paired with respective ones of the first nodes based on a relative maximum micro-similarity score, and respective second ones of the second nodes unpaired with the respective ones of the first nodes based on not satisfying the relative maximum micro-similarity score.
7. The apparatus as defined in claim 6 , wherein the graph builder is to merge the second ones of the second nodes with respective ones of the first ones of the second nodes.
8. A non-transitory computer readable medium including instructions, which when executed, cause at least one processor to, at least:
cluster client databases into client clusters based on a threshold quantity of overlapping universal product codes (UPCs) between respective ones of the client databases;
identify custom characteristics from the respective ones of the client clusters, ones of the custom characteristics having dissimilar nomenclature;
cluster the ones of the custom characteristics based on a similarity metric;
normalize the ones of the custom characteristics as a proxy characteristic, the proxy characteristic having a common nomenclature to represent the ones of the custom characteristics; and
enable improved product marketing analysis by replacing dissimilar nomenclature with the proxy characteristic.
9. The non-transitory computer readable medium as defined in claim 8 , wherein the instructions, when executed, cause the at least one processor to:
generate first nodes associated with UPCs having first ones of the custom characteristics; and
generate second nodes associated with UPCs having second ones of the custom characteristics.
10. The non-transitory computer readable medium as defined in claim 9 , wherein the instructions, when executed, cause the at least one processor to identify first ones of the custom characteristics have a first nomenclature, and identify second ones of the custom characteristics have a second nomenclature different than the first nomenclature.
11. The non-transitory computer readable medium as defined in claim 9 , wherein the instructions, when executed, cause the at least one processor to generate a micro-similarity score between pairs of (a) the first nodes and (b) the second nodes.
12. The non-transitory computer readable medium as defined in claim 11 , wherein the instructions, when executed, cause the at least one processor to generate the micro-similarity scores based on a threshold overlap of UPCs between the pairs of the first and second nodes.
13. The non-transitory computer readable medium as defined in claim 11 , wherein the instructions, when executed, cause the at least one processor to apply Maximum Weighted Bipartite Graph Matching (MWBGM) to the first nodes and the second nodes, respective first ones of the second nodes paired with respective ones of the first nodes based on a relative maximum micro-similarity score, and respective second ones of the second nodes unpaired with the respective ones of the first nodes based on not satisfying the relative maximum micro-similarity score.
14. The non-transitory computer readable medium as defined in claim 13 , wherein the instructions, when executed, cause the at least one processor to merge the second ones of the second nodes with respective ones of the first ones of the second nodes.
15. A method to map client specifications with standardized characteristics, the method comprising:
clustering, by executing an instruction with at least one processor, client databases into client clusters based on a threshold quantity of overlapping universal product codes (UPCs) between respective ones of the client databases;
identifying, by executing an instruction with the at least one processor, custom characteristics from the respective ones of the client clusters, ones of the custom characteristics having dissimilar nomenclature;
clustering, by executing an instruction with the at least one processor, the ones of the custom characteristics based on a similarity metric;
normalizing, by executing an instruction with the at least one processor, the ones of the custom characteristics as a proxy characteristic, the proxy characteristic having a common nomenclature to represent the ones of the custom characteristics; and
enabling, by executing an instruction with the at least one processor, improved product marketing analysis by replacing dissimilar nomenclature with the proxy characteristic.
16. The method as defined in claim 15 , further including:
generating first nodes associated with UPCs having first ones of the custom characteristics; and
generating second nodes associated with UPCs having second ones of the custom characteristics.
17. The method as defined in claim 16 , wherein the first ones of the custom characteristics have a first nomenclature, and the second ones of the custom characteristics have a second nomenclature different than the first nomenclature.
18. The method as defined in claim 16 , further including generating a micro-similarity score between pairs of (a) the first nodes and (b) the second nodes.
19. The method as defined in claim 18 , further including generating the micro-similarity scores based on a threshold overlap of UPCs between the pairs of the first and second nodes.
20. The method as defined in claim 18 , further including applying Maximum Weighted Bipartite Graph Matching (MWBGM) to the first nodes and the second nodes, respective first ones of the second nodes paired with respective ones of the first nodes based on a relative maximum micro-similarity score, and respective second ones of the second nodes unpaired with the respective ones of the first nodes based on not satisfying the relative maximum micro-similarity score.
21. An apparatus comprising:
at least one memory;
instructions in the apparatus; and
processor circuitry to execute the instructions to:
cluster client databases into client clusters based on a threshold quantity of overlapping universal product codes (UPCs) between respective ones of the client databases;
identify custom characteristics from the respective ones of the client clusters, ones of the custom characteristics having dissimilar nomenclature;
cluster the ones of the custom characteristics based on a similarity metric;
normalize the ones of the custom characteristics as a proxy characteristic, the proxy characteristic having a common nomenclature to represent the ones of the custom characteristics; and
enable improved product marketing analysis by replacing dissimilar nomenclature with the proxy characteristic.
22. The apparatus as defined in claim 21 , wherein the processor circuitry is to:
generate first nodes associated with UPCs having first ones of the custom characteristics; and
generate second nodes associated with UPCs having second ones of the custom characteristics.
23. The apparatus as defined in claim 22 , wherein the processor circuitry is to identify first ones of the custom characteristics have a first nomenclature, and identify second ones of the custom characteristics have a second nomenclature different than the first nomenclature.
24. The apparatus as defined in claim 22 , wherein the processor circuitry is to generate a micro-similarity score between pairs of (a) the first nodes and (b) the second nodes.
25. The apparatus as defined in claim 24 , wherein the processor circuitry is to generate the micro-similarity scores based on a threshold overlap of UPCs between the pairs of the first and second nodes.
26. The apparatus as defined in claim 24 , wherein the processor circuitry is to apply Maximum Weighted Bipartite Graph Matching (MWBGM) to the first nodes and the second nodes, respective first ones of the second nodes paired with respective ones of the first nodes based on a relative maximum micro-similarity score, and respective second ones of the second nodes unpaired with the respective ones of the first nodes based on not satisfying the relative maximum micro-similarity score.
27. The apparatus as defined in claim 26 , wherein the processor circuitry is to merge the second ones of the second nodes with respective ones of the first ones of the second nodes.