Systems and methods for identifying and using micro-intents

Patent metadata

Field	Value
Publication number	US-12026728-B2
Application number	US-201916531659-A
Country	US
Kind code	B2
Filing date	Aug 5, 2019
Priority date	Aug 6, 2018
Publication date	Jul 2, 2024
Grant date	Jul 2, 2024

Abstract

Systems and methods including one or more processing modules and one or more non-transitory storage modules storing computing instructions configured to run on the one or more processing modules and perform acts of accessing first transaction data stored in a transaction database, the first transaction data describing first transactions for first items from first users; determining, using the first transaction data, first micro-intents associated with the first transaction data; grouping the first micro-intents into clusters; labeling each cluster of the first micro-intents with a respective label; receiving second transaction data of a user, the second transaction data describing second transactions for second items for the user; determining, using the second transaction data, second micro-intents present in the second transactions; receiving current transaction data from a user interface of an electronic device of the user; determining, using the current transaction data, that the user is expressing a current micro-intent, the current micro-intent having at least one associated label; and transmitting an instruction to display, on the user interface of the electronic device, a user interface element correlated with the at least one associated label. Other embodiments are disclosed herein.

First claim

What is claimed is:

1. A system comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computing instructions that, when executed on the one or more processors, cause the one or more processors to perform:
accessing, using a distributed network, first transaction data stored in a transaction database, the first transaction data describing first transactions for first items for first users, wherein the transaction database comprises aggregated historical transaction datasets from multiple sources from multiple users, and wherein the distributed network comprises a distributed memory architecture and a distributed processing architecture to reduce congestion from the first transaction data on the distributed network while allowing access to the aggregated historical transaction datasets from another location;
determining, using the distributed processing architecture and using the first transaction data, first micro-intents associated with the first transaction data using a transpose of a matrix comprising the first transaction data and a transpose of a respective mean vector for each item of the first items, wherein the first micro-intents associated with the first transaction data further comprises creating a respective correlation matrix from a respective diagonal matrix for each item of the first items and a respective covariance matrix for each item of the first items;
grouping, using the distributed processing architecture, the first micro-intents into clusters, wherein each of the clusters is created by using:
a first transaction data matrix;
the respective mean vector for each item of the first items;
the respective diagonal matrix for each item of the first items; and
a respective eigenvector of respective eigenvectors, wherein the respective eigenvector corresponds to a respective eigenvalue decomposition of the respective correlation matrix, and wherein the respective eigenvalue decomposition of the respective correlation matrix comprises percentages of transformed vectors;

labeling, using the distributed processing architecture, each cluster of the first micro-intents for each of the first transactions with a respective label and a respective label pattern, wherein the respective label pattern comprises a sticky preference when the respective label reoccurs within a history of one of the first users;
receiving, using the distributed memory architecture, second transaction data of a user, the second transaction data describing second transactions for second items for the user;
determining, using the distributed processing architecture and using the second transaction data, second micro-intents present in the second transactions;
receiving, using the distributed memory architecture, current transaction data from a user interface of an electronic device of the user, wherein the current transaction data comprises datasets of a number of items currently added to an electronic shopping cart of the user during a current browsing session of a website;
determining, using the distributed processing architecture and using the current transaction data, third micro-intents present in the current transaction data;
determining, using the distributed processing architecture, that the user is expressing a current micro-intent based on at least one of (i) the first micro-intents, (ii) the second micro-intents, and (iii) the third micro-intents; and
transmitting an instruction to display a user interface element on the electronic device of the user, wherein the user interface element is correlated with the current micro-intent of the user.

2. The system of claim 1 , wherein determining, using the distributed processing architecture and using the first transaction data, the first micro-intents associated with the first transaction data comprises:
creating the first transaction data matrix using the first transaction data, wherein rows of the first transaction data matrix correspond to transactions of the first transactions and columns of the first transaction data matrix correspond to items of the first items;
creating the respective mean vector for each item of the first items using the first transaction data matrix;
creating the respective covariance matrix using the first transaction data matrix and the respective mean vector for each item of the first items;
creating the respective diagonal matrix for the respective covariance matrix for each item of the first items, wherein diagonals of the respective diagonal matrix for each item of the first items are equal to that of the respective covariance matrix for each item of the first items;
creating the respective eigenvalue decomposition of each of the respective correlation matrixes, wherein columns of the respective eigenvalue decompositions of the respective correlation matrixes are the respective eigenvectors of the respective correlation matrixes, and the respective eigenvectors of the respective correlation matrixes represent the first micro-intents; and
localizing and scaling each transaction of the first transactions using the first transaction data matrix, the respective mean vector for each item of the first items, and the respective diagonal matrix for each item of the first items.

3. The system of claim 2 , wherein the respective covariance matrix for each item of the first items is created based on:

Σ
=

1
N

⁢

X
T

⁢
X

-

X
_

T

⁢

X
_

wherein N is a total number of transactions in X, X is the first transaction data matrix, X T is a transpose of the first transaction data matrix, X is the respective mean vector for each item of the first items, and X T is the transpose of the respective mean vector for each item of the first items.

4. The system of claim 2 , wherein each of the respective correlation matrixes is created based on:
R=D −1/2 ΣD −1/2
wherein D −1/2 is the respective diagonal matrix for each item of the first items and Σ is the respective covariance matrix.

5. The system of claim 2 , wherein each of the clusters is grouped based on:
{ x :( x− X ) D −1/2 Q .,j ∈( v 1 ,v 2 ]}

wherein x is the first transaction data matrix, X is the respective mean vector for each item of the first items, D −1/2 is the respective diagonal matrix for each item of the first items, Q .,j is the respective eigenvector corresponding to the respective eigenvalue decomposition of the respective correlation matrix, v 1 is a 25 th percentile of transformed vectors, and v 2 is a 50 th percentile of the transformed vectors; and
wherein grouping the first micro-intents into the clusters comprises comparing similar micro-intents that have differing values.

6. The system of claim 1 , wherein:
the computing instructions, when executed on the one or more processors, further cause the one or more processors to perform:
determining, using the distributed processing architecture and using the first micro-intents for the first transactions, a label pattern for the user,
wherein the label pattern comprises at least one of:
a new interes

Assignees

• Walmart Apollo Llc

Inventors

• Achan Kannan
• Mitra Abhimanya
• Kumar Sushant
• Korpeoglu Evren

Classifications

• G06F18/2321

  using statistics or function optimisation, e.g. modelling of probability density functions · CPC title

• G06F17/16

  Matrix or vector computation {, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization (matrix transposition G06F7/78)} · CPC title

• G06Q30/0201 Primary

  Market modelling; Market analysis; Collecting market data · CPC title