Systems and methods for selecting and presenting products based on past purchases

1 . A computer-implemented system for generating and presenting product search results based on a user query, the system comprising:

a memory storing instructions; and

at least one processor configured to execute the instructions to perform operations comprising:

receiving, via a graphical user interface, a product search query by a user;

retrieving, from at least one data structure:

at least one set of historical purchase data associated with the user, and

at least one set of experimental data comprising at least one hierarchical list of product categories, wherein the experimental data is generated based on receiving the product search query;

normalizing the product search query, using a natural language processing technique comprising text tokenization, stemming, or lemmatization, to generate a normalized user query;

extracting, using at least one machine-learning algorithm and based on an analysis of the normalized user query:

a plurality of attributes associated with the product search query, and

at least one pattern associated with the product search query;

generating, using at least one machine-learning algorithm, a list of queried product categories based on the normalized user query, wherein the at least one machine-learning algorithm is configured to observe relationships between the plurality of attributes and the at least one pattern and to generate the list of queried product categories based on observed associations;

generating, using the at least one set of historical purchase data and the at least one set of experimental data, a hierarchical list of historical product categories associated with the user;

generating a list of product search results associated with the product search query, based on the list of queried product categories and the hierarchical list of historical product categories;

ranking, using at least one machine-learning algorithm, the list of product search results based on a relevancy metric to the product search query;

displaying the ranked list of product search results to the user via user interface elements, the user interface elements enabling the user to add one or more presented products to a virtual shopping cart, or to request a single detail page based on a particular presented product, the single detail page including recommendations for similar products based on an analysis of customers who bought the particular presented product and at least one other product; and

displaying the list of queried product categories or the hierarchical list of historical product categories alongside the ranked list of product search results.

2 . The system of claim 1 , wherein the at least one set of experimental data comprises a hierarchical list of products and product categories based on aggregated data from a plurality of users.

3 . The system of claim 1 , wherein the plurality of attributes associated with the product search query comprises at least a product name.

4 . The system of claim 1 , wherein the plurality of attributes associated with the product search query further comprises a price and a quantity.

5 . The processor of claim 1 , further comprising:

determining a history of prior access by the system for the historical purchase data associated with the user, and, upon determining that there is no history of prior access, storing the historical purchase data associated with the user in a cache memory.

6 . The system of claim 1 , wherein the list of queried product categories comprises at least one product category.

7 . The system of claim 1 , wherein the hierarchical list of historical product categories associated with the user is further ranked by a past purchase frequency of items within each category.

8 . The system of claim 1 , wherein the relevancy metric between the product search results and the user's product search query is at least based on past purchase volume.

9 . The system of claim 1 , wherein the relevancy metric between the product search results and the product search query is at least based on state of recency.

10 . The system of claim 1 , wherein an order of displaying the product search results to the user is based on the ranking of each product search result.

11 . A computer-implemented method for generating and presenting product search results based on a user query, comprising:

receiving, via a graphical user interface, a product search query by a user;

retrieving, from at least one data structure:

at least one set of historical purchase data associated with the user, and

at least one set of experimental data comprising at least one hierarchical list of product categories, wherein the experimental data is generated based on receiving the product search query;

normalizing the product search query, using a natural language processing technique comprising text tokenization, stemming, or lemmatization, to generate a normalized user query;

extracting, using at least one machine-learning algorithm and based on an analysis of the normalized user query:

a plurality of attributes associated with the product search query, and

at least one pattern associated with the product search query;

determining, using at least one machine-learning algorithm, a list of queried product categories based on the normalized user query, wherein the at least one machine-learning algorithm is configured to observe relationships between the plurality of attributes and the at least one pattern and to generate the list of queried product categories based on observed associations;

generating, using the at least one set of historical purchase data and the at least one set of experimental data, a hierarchical list of historical product categories associated with the user;

generating a list of product search results associated with the product search query, based on the list of queried product categories and the hierarchical list of historical product categories;

ranking, using at least one machine-learning algorithm, the list of product search results based on a relevancy metric to the product search query;

displaying the ranked list of product search results to the user via user interface elements, the user interface elements enabling the user to add one or more presented products to a virtual shopping cart, or to request a single detail page based on a particular presented product, the single detail page including recommendations for similar products based on an analysis of customers who bought the particular presented product and at least one other product; and

displaying the list of queried product categories or the hierarchical list of historical product categories alongside the ranked list of product search results.

12 . The method of claim 11 , wherein the at least one set of experimental data comprises a hierarchical list of products and product categories based on aggregated data from a plurality of users.

13 . The method of claim 11 , wherein the plurality of attributes associated with the product search query comprises at least a product name.

14 . The method of claim 11 , wherein the plurality of attributes associated with the product search query further comprises a price and a quantity.

15 . The method of claim 11 , further comprising:

determining a history of prior access by the system for the historical purchase data associated with the user, and, upon determining that there is no history of prior access, storing the historical purchase data associated with the user in a cache memory.

16 . The method of claim 11 , wherein the list of queried product categories comprises at least one product category.

17 . The method of claim 11 , wherein the hierarchical list of historical product categories associated with the user is further ranked by a past purchase frequency of items within each category.

18 . The method of claim 11 , wherein the relevancy metric between the product search results and the product search query is at least based on past purchase volume.

19 . The method of claim 11 , wherein the relevancy metric between the product search results and the product search query is at least based on state of recency.

20 . A computer-implemented system for generating and presenting product search results based on a user query, the system comprising:

a memory storing instructions; and

at least one processor configured to execute the instructions to perform operations comprising:

receiving, via a graphical user interface, a product search query by a user, the product search query comprising a text string;

retrieving, from at least one data structure:

at least one set of historical purchase data associated with the user, and

at least one set of experimental data comprising at least one hierarchical list of product categories collected from an aggregated set of users, wherein the experimental data is generated based on receiving the product search query;

determining a history of prior access by the system for the historical purchase data associated with the user, and, upon determining that there is no history of prior access, storing the historical purchase data associated with the user in a cache memory;

normalizing the product search query, using a natural language processing technique comprising text tokenization, stemming, or lemmatization, to generate a normalized user query;

extracting, using at least one machine-learning algorithm and based on an analysis of the normalized user query:

a plurality of attributes associated with the product search query comprising at least a product name, a price, and a quantity, and

at least one pattern associated with the product search query;

generating, using at least one machine-learning algorithm, a list of queried product categories based on the normalized user query, wherein the at least one machine-learning algorithm is configured to observe relationships between the plurality of attributes and the at least one pattern and to generate the list of queried product categories based on observed associations;

generating, using the at least one set of historical purchase data and the at least one set of experimental data, a hierarchical list of historical product categories associated with the user;

ranking the hierarchical list of historical product categories based on a past purchase frequency of items within each category;

generating a list of product search results associated with the product search query, based on the list of queried product categories and the hierarchical list of historical product categories;

ranking, using at least one machine-learning algorithm, the list of product search results based on a relevancy metric to the product search query, wherein the relevancy metric is at least based on past purchase frequency or state of recency;

displaying the ranked list of product search results to the user via user interface elements, the user interface elements enabling the user to add one or more presented products to a virtual shopping cart, or to request a single detail page based on a particular presented product, the single detail page including recommendations for similar products based on an analysis of customers who bought the particular presented product and at least one other product; and

displaying the list of queried product categories or the hierarchical list of historical product categories alongside the ranked list of product search results.