# US 2021/0065150 A1 · Systems and methods for assigning attribution weights to nodes

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2021/0065150 A1
- Document ID: US20210065150A1
- Status: Published application
- Kind: A1
- Application number: 17/009,777
- Filed: 2020-09-02
- Published: 2021-03-04
- Earliest claimed: 2019-09-03
- Assignees: Linen Software Inc.
- Applicants: Linen Software Inc.
- Inventors: JACOB MATHEW; YESHWANT NAYAK
- Web page: https://patentgrep.com/patent/US20210065150A1

## CPC

- G06Q 20/20: Point-of-sale [POS] network systems
- G06Q 20/405: {Establishing or using transaction specific rules}
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

Implementations described herein relate to methods, systems, and computer-readable media to assign an attribution weight to an agent node. In some implementations, a method includes receiving a plurality of transaction data records, determining a crediting profile associated with each transaction data record, creating a plurality of clusters of transaction data records, assigning each of the plurality of clusters of transaction data records to a respective processor of a plurality of processors, loading rules associated with a respective crediting profile into a memory associated with the corresponding processor, and for each cluster of transaction data records, determining a match of one or more rules from a set of rule, selecting a rule from the determined one or more rules, and assigning the attribution weight to the at least one agent node based on the selected rule.

## Claims

### Claim 1

A computer-implemented method comprising:
receiving a plurality of data records, wherein each data record comprises a plurality of attribute values associated with a transaction;
grouping the plurality of data records into two or more clusters, wherein each cluster is associated with a respective set of attribute values, and wherein no two clusters share a same set of attribute values;
identifying a respective set of rules for each cluster of the two or more clusters;
loading the data records and the respective set of rules for each cluster into a respective cache; and
analyzing, in parallel with a plurality of program threads that execute on a respective processor associated with the respective cache and without accessing an external memory, such that the cache provides data locality for the data records using the respective set of rules, wherein the analyzing includes:
determining a rule associated with each data record of the cluster based on a subset of attribute values in the data record that match a particular rule of the respective set of rules; and
determining one or more transaction-specific parameters based on the rule.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein grouping the plurality of data records into two or more clusters comprises determining that a profile associated with a first data record is a profile type that includes two or more embedded data records, and wherein the method further comprises:
creating the two or more embedded records; and
applying a subset of attribute values of the first data record to the two or more embedded data records.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein determining one or more transaction-specific parameters based on the rule comprises determining one or more attribution weights associated with a node.

### Claim 4

A computer-implemented method to assign an attribution weight to an agent node, wherein the method comprises:
receiving a plurality of transaction data records, wherein each transaction data record comprises a plurality of attribute values corresponding to a plurality of attributes associated with a transaction;
determining a crediting profile associated with each transaction data record;
wherein the crediting profile is determined based on one or more attribute values in the transaction data record;
creating a plurality of clusters of transaction data records, wherein each cluster of the plurality of transaction data records is associated with a corresponding crediting profile;
assigning each of the plurality of clusters of transaction data records to a respective processor of a plurality of processors;
loading rules associated with a respective crediting profile into a memory associated with the corresponding processor; and
for each cluster of transaction data records:
determining a match of one or more rules from a set of rules associated with a respective crediting category of a plurality of crediting categories of the respective crediting profile based on a match of attribute values of a subset of attributes in the transaction data record with attribute values of the subset of attributes in the set of rules, wherein the plurality of crediting categories are associated with a predetermined priority order and the sets of rules are evaluated in the priority order of the plurality of crediting categories;
selecting a rule from the determined one or more rules, for each transaction data record, wherein the selected rule identifies at least one agent node and specifies an attribution weight for at least one agent node; and
assigning the attribution weight to the at least one agent node based on the selected rule.

### Claim 5 (depends on 4)

The computer-implemented method of claim 4, wherein the method further comprises:
prior to determining the crediting profile, comparing attribute values of the transaction data records to a predetermined set of attribute values indicative of transaction synthesis;
based on a match of attribute values of a first transaction data record to the predetermined set of attribute values indicative of transaction synthesis:
creating two or more synthesized transaction data records, wherein each of two or more synthesized transaction data records includes a reference to the first transaction data record; and
applying one or more attribute values associated with the first transaction data record to the two or more synthesized transaction data records.

### Claim 6 (depends on 4)

The computer-implemented method of claim 4, wherein determining a match of one or more rules from a set of rules associated with a respective crediting category comprises:
comparing a number of transaction data records in the cluster of transaction data records with a number of the set of rules in the respective crediting category; and
based on the comparison, if the number of transaction data records in the cluster of transaction data records is greater than or equal to the number of the set of rules in the respective crediting category, evaluating the set of rules against the cluster of transaction data records, else evaluating the cluster of transaction data records against the set of rules.

### Claim 7 (depends on 4)

The computer-implemented method of claim 4, wherein selecting the rule from the determined one or more rules comprises selecting the rule with the highest rule priority from the one or more rules.

### Claim 8 (depends on 7)

The computer-implemented method of claim 7, wherein determining a match of one or more rules from a set of rules associated with a respective crediting category comprises determining a match of two or more rules based on a match of attribute values of the subset of attributes in the transaction data record with attribute values of the subset of attributes in the set of rules.

### Claim 9 (depends on 8)

The computer-implemented method of claim 8, further comprising storing the two or more rules and the attribute values associated with the transaction data record.

### Claim 10 (depends on 4)

The computer-implemented method of claim 4, wherein the at least one agent node is one of a plurality of agent nodes in an ordered hierarchy of agent nodes.

### Claim 11 (depends on 4)

The computer-implemented method of claim 4, wherein receiving a plurality of transaction data records comprises:
receiving a sequence of transaction data records; and
adding the sequence of transaction data records to a memory buffer.

### Claim 12 (depends on 4)

The computer-implemented method of claim 4, wherein the at least one agent node is associated with at least one territory node, and wherein the at least one agent node inherits one or more properties of the at least one territory node.

### Claim 13 (depends on 4)

The computer-implemented method of claim 4, wherein the selected rule identifies at least two agent nodes and specifies an attribution weight for each of at least two agent nodes, and wherein a sum of the attribution weights for the at least two agent nodes is greater than 100%.

### Claim 14 (depends on 4)

The computer-implemented method of claim 4, wherein determining a match of one or more rules from a set of rules associated with a respective crediting category comprises:
assigning each rule of the set of rules to a respective computational process of a plurality of parallel computational processes;
evaluating, utilizing each of the plurality of parallel computational processes, attribute values in the transaction data record against corresponding attribute values specified in each rule; and
determining a match of one or more rules based on the evaluation.

### Claim 15 (depends on 4)

The computer-implemented method of claim 4, further comprising loading a superset of attribute values included in the transaction data records that is associated with the rules associated with the respective crediting profile into the memory associated with the corresponding processor.

### Claim 16 (depends on 4)

The computer-implemented method of claim 4, further comprising identifying one or more rules that were not selected as applicable to any transaction data record, one or more rules that were selected with a frequency that met a threshold frequency, and one or more transaction data records for which a rule was not selected.

### Claim 17 (depends on 4)

The computer-implemented method of claim 4, further comprising:
determining that the identified at least one agent node is associated with an overlay node; and
tagging the transaction data record as an overlay eligible transaction data record.

### Claim 18

A non-transitory computer-readable medium comprising instructions that, responsive to execution by a processing device, causes the processing device to perform operations comprising:
receiving a plurality of transaction data records, wherein each transaction data record comprises a plurality of attribute values corresponding to a plurality of attributes associated with a transaction;
determining a crediting profile associated with each transaction data record;
wherein the crediting profile is determined based on one or more attribute values in the transaction data record;
creating a plurality of clusters of transaction data records, wherein each cluster of the plurality of transaction data records is associated with a corresponding crediting profile;
assigning each of the plurality of clusters of transaction data records to a respective processor of a plurality of processors;
loading rules associated with a respective crediting profile into a memory associated with the corresponding processor; and
for each cluster of transaction data records:
determining a match of one or more rules from a set of rules associated with a respective crediting category of a plurality of crediting categories of the respective crediting profile based on a match of attribute values of a subset of attributes in the transaction data record with attribute values of the subset of attributes in the set of rules, wherein the plurality of crediting categories are associated with a predetermined priority order and the sets of rules are evaluated in the priority order of the plurality of crediting categories;
selecting a rule from the determined one or more rules, for each transaction data record, wherein the selected rule identifies at least one agent node and specifies an attribution weight for at least one agent node; and
assigning the attribution weight to the at least one agent node based on the selected rule.

### Claim 19 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein the at least one agent node is associated with at least one territory node, and wherein the at least one agent node inherits one or more properties of the at least one territory node.

### Claim 20 (depends on 18)

The non-transitory computer-readable medium of claim 18, wherein selecting the rule from the determined one or more rules comprises selecting the rule with the highest rule priority from the one or more rules.
