# US 2020/0404072 A1 · Generating attribute preference models based on disparate attribute spectrums

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

- Number: US 2020/0404072 A1
- Document ID: US20200404072A1
- Status: Published application
- Kind: A1
- Application number: 17/008,196
- Filed: 2020-08-31
- Published: 2020-12-24
- Earliest claimed: 2018-02-13
- Assignees: eBay Inc.
- Applicants: eBay Inc.
- Inventors: Steve YANKOVICH; Sergio Pinzon GONZALES, JR.
- Web page: https://patentgrep.com/patent/US20200404072A1

## CPC

- G06Q 50/04: Manufacturing
- G06F 16/9535: Search customisation based on user profiles and personalisation
- G06N 20/00: Machine learning
- G06Q 10/083: Shipping
- G06Q 30/0202: Market predictions or forecasting for commercial activities
- G06Q 30/0241: Advertisements
- G06Q 30/0256: {User search}
- G06Q 30/0603: {Catalogue creation or management}
- G06Q 30/0621: {by configuring or customising goods or services}
- G06Q 30/0627: {by specifying product or service characteristics, e.g. product dimensions}
- G06Q 30/0631: {Recommending goods or services}
- G06Q 30/0641: {utilising user interfaces specially adapted for shopping}
- G06Q 30/0643: {graphically representing goods, e.g. 3D product representation}
- G06V 20/20: in augmented reality scenes
- G06V 20/80: Recognising image objects characterised by unique random patterns
- G06V 40/20: Movements or behaviour, e.g. gesture recognition (recognition of facial expressions G06V40/16)
- H04L 67/306: User profiles
- H04L 67/535: {Tracking the activity of the user (network monitoring arrangements H04L43/00; recording of computer activity G06F11/34)}
- G06Q 10/40: Business processes related to social networking or social networking services
- Y02P 90/30: Computing systems specially adapted for manufacturing
- 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
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)
- Y02P: CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS

## Abstract

An attribute correlation system reduces network traffic and processing cycles associated with impromptu item selections by generating attribute preference models based on disparate attribute spectrums. The attribute correlation system deploys the attribute preference models to select individual items from various disparate “candidate item categories.” Generally described, the attribute preference models facilitate analyzing item sets across a wide variety of disparate “candidate” item categories to preemptively identify individual items for a user. In this way, the individual items may be identified and, ultimately, selected for the user even absent any indication that the user has searched for otherwise identified these items or even other items from within the disparate “candidate” item categories. The “candidate” item categories may be determined to be disparate from one another based on a relationship void existing such that predefined relationships are missing between these “candidate” item categories.

## Claims

### Claim 1

A computer-implemented method, comprising:
identifying at least one first attribute spectrum corresponding to at least one specimen of interest to a user, wherein individual first attribute spectrums define first values for one or more attributes in association with individual specimens of interest;
generating, based on the at least one first attribute spectrum, an attribute preference model in association with the user;
determining a candidate category from a plurality of item categories defined within item category data;
determining at least one second attribute spectrum corresponding to the candidate category, wherein individual second attribute spectrums define second values for the one or more attributes in association with individual items that are included within the candidate category;
analyzing the at least one second attribute spectrum based on the attribute preference model to designate a selected item from the candidate category; and
communicating attributes of the selected item to the user.

### Claim 2 (depends on 1)

The computer-implemented method of claim 1, wherein the analyzing the at least one second attribute spectrum includes deploying the attribute preference model to rank individual second attribute spectrums, of the at least one second attribute spectrum, in accordance with correlative strengths to the individual first attribute spectrums that define the first values for the one or more attributes in association with the individual specimens of interest.

### Claim 3 (depends on 1)

The computer-implemented method of claim 1, wherein the candidate item category is disparate from the at least one specimen of interest within the item category data.

### Claim 4 (depends on 1)

The computer-implemented method of claim 1, further comprising:
determining that the candidate category is disparate from the at least one specimen of interest based on a relationship void between the least one candidate category and the at least one specimen of interest within the item category data.

### Claim 5 (depends on 4)

The computer-implemented method of claim 4, wherein the relationship void corresponds to a hierarchical relationship lacking, within the item category data, between the candidate item category and the at least one specimen of interest.

### Claim 6 (depends on 1)

The computer-implemented method of claim 1, wherein the generating the attribute preference model comprises:
determining a plurality of attribute value distributions of the first values with respect to the at least one first attribute spectrum; and
assigning weights to individual attributes, of the one or more attributes, based on the plurality of attribute value distributions.

### Claim 7 (depends on 6)

The computer-implemented method of claim 6, wherein the weights are inversely related to deviations of the plurality of attribute value distributions.

### Claim 8 (depends on 1)

The computer-implemented method of claim 1, wherein the first values for the one or more attributes comprise at least one of:
craftmanship values that are indicative of a user-defined quality of design for the individual specimens of interest;
complexity values that are indicative of a level of intricacy for the individual specimens of interest;
scarcity values that are indicative of a limited availability for the individual specimens of interest; or
demand values that are indicative of a market demand for the individual specimens of interest.

### Claim 9 (depends on 1)

The computer-implemented method of claim 1, wherein the analyzing the at least one second attribute spectrum includes scoring a plurality of individual items within the candidate item category based on the attribute preference model.

### Claim 10 (depends on 1)

The computer-implemented method of claim 1, further comprising generating an item order to schedule a preemptive delivery of the selected item to a physical address that corresponds to the user.

### Claim 11

A system, comprising:
one or more processors; and
a memory in communication with the one or more processors, the memory having computer-readable instructions stored thereupon that, when executed by the one or more processors, cause the one or more processors to:
identify a plurality of first attribute spectrums corresponding to a plurality of specimens of interest to a user, wherein individual first attribute spectrums correspond to individual specimens of interest to the user, and wherein the individual first attribute spectrums define first values for one or more attributes in association with individual specimens of interest;
generate an attribute preference model that corresponds to the user based on the plurality of first attribute spectrums that correspond to the plurality of specimens of interest to the user;
determine a candidate category from a plurality of item categories defined within item category data;
determine a plurality of second attribute spectrums corresponding to the candidate category, wherein individual second attribute spectrums define second values for the one or more attributes in association with individual items that are included within the candidate category;
analyze the plurality of second attribute spectrums based on the attribute preference model to designate a selected item, from the candidate category, for the user; and
transmit, to the user, a communication that is associated with the selected item.

### Claim 12 (depends on 11)

The system of claim 11, wherein the computer-readable instructions further cause the one or more processors to analyze the item category data with respect to an interest profile that corresponds to the user to select the candidate category from the plurality of item categories.

### Claim 13 (depends on 12)

The system of claim 12, wherein the plurality of item categories includes a first item category and second item category that is disparate from the first item category.

### Claim 14 (depends on 11)

The system of claim 11, wherein the computer-readable instructions further cause the one or more processors to:
determine attribute value distributions of the first values across the plurality of first attribute spectrums; and
assign weights to the plurality of attributes within the attribute preference model based on the attribute value distributions.

### Claim 15 (depends on 14)

The system of claim 14, wherein the weights that are assigned to individual attributes are inversely related to the attribute value distributions for the individual attributes.

### Claim 16 (depends on 11)

The system of claim 11, wherein the computer-readable instructions further cause the one or more processors to score the individual items based on the attribute preference model.

### Claim 17

A computer-implemented method, comprising:
identifying first attribute spectrums corresponding to specimens of interest to a user, wherein individual first attribute spectrums define first values for attributes in association with individual specimens of interest;
generating, based on the first attribute spectrums, an attribute preference model in association with the user;
determining a candidate category from a plurality of item categories defined within item category data;
determining second attribute spectrums corresponding to the candidate category, wherein individual second attribute spectrums define second values for the attributes in association with individual items in the candidate category; and
analyzing the second attribute spectrums based on the attribute preference model to designate a selected item, from the candidate category, in association with the user.

### Claim 18 (depends on 17)

The computer-implemented method of claim 17, wherein the candidate category is disparate from the specimens of interest within the item category data.

### Claim 19 (depends on 18)

The computer-implemented method of claim 18, wherein the attribute preference model comprises weights that are inversely related to attribute value distributions associated with individual attributes of a plurality of attributes.

### Claim 20 (depends on 17)

The computer-implemented method of claim 17, wherein the candidate category is designated based at least in part on one or more relationship voids between the specimens of interest.
