Product recommendations based on characteristics from end user-generated text

1. A method for generating a related product interface portion in an electronic user interface based on end user-generated text, the method comprising:

applying, to end user-generated text associated with each product of a plurality of products, a machine learning model trained to output portions of the end user-generated text that include one or more respective characteristics, whereby a plurality of characteristics are identified for the plurality of products and whereby the end user-generated text comprises text generated by a plurality of users;

determining a plurality of characteristic groups, each group comprising two or more of the plurality of characteristics, wherein the characteristics within a group are similar to each other;

receiving, from a user through the electronic user interface, an input related to a product category; and

in response to receiving the user input, generating and presenting the related product interface portion that includes a respective product from one or more of the plurality of characteristic groups related to the product category.

2. The method of claim 1 , wherein each characteristic comprises one or more plaintext words from the end user-generated text.

3. The method of claim 1 , wherein determining the plurality of characteristic groups comprises:

generating respective embeddings for each characteristic;

calculating similarities of the embeddings to each other; and

assigning characteristics to groups according to the similarities of the embeddings.

4. The method of claim 1 , wherein determining the plurality of characteristic groups further comprises:

determining a representative word for each group;

determining a pairwise similarity of each characteristic to the representative word for each group; and

adding products to groups according to matches in the pairwise similarity.

5. The method of claim 1 , further comprising:

selecting, for each characteristic group, a representative product according to one or more of respective popularities of the products within the group or a confidence score associated with the determination of the characteristic from the end user-generated text associated with the product.

6. The method of claim 1 , wherein the plurality of products are within a same product category.

7. A system for generating a related product interface portion in an electronic user interface based on user text, the system comprising:

a non-transitory, computer-readable medium storing instructions; and

a processor configured to execute the instructions to:

apply, to end user-generated text associated with each product of a plurality of products, a machine learning model trained to output portions of the end user-generated text that include one or more respective characteristics, whereby a plurality of characteristics are identified for the plurality of products and whereby the end user-generated text comprises text generated by a plurality of users;

determine a plurality of characteristic groups, each group comprising two or more of the plurality of characteristics, wherein the characteristics within a group are similar to each other;

receive, from a user through the electronic user interface, an input related to a product category; and

in response to receiving the user input, generate and present the related product interface portion that includes a respective product from one or more of the plurality of characteristic groups related to the product category.

8. The system of claim 7 , wherein each characteristic comprises one or more plaintext words from the end user-generated text.

9. The system of claim 7 , wherein determining the plurality of characteristic groups comprises:

generating respective embeddings for each characteristic;

calculating similarities of the embeddings to each other; and

assigning characteristics to groups according to the similarities of the embeddings.

10. The system of claim 7 , wherein determining the plurality of characteristic groups further comprises:

determining a representative word for each group;

determining a pairwise similarity of each characteristic the representative words of the groups; and

adding products to groups according to matches in the pairwise similarity.

11. The system of claim 7 , wherein the processor is further configured to:

select, for each characteristic group, a representative product according to one or more of respective popularities of products within the group or a confidence score associated with the determination of the characteristic from the end user-generated text associated with the product.

12. The system of claim 7 , wherein the plurality of products are within a same product category.

13. A method for generating a related product interface portion in an electronic user interface based on end user-generated text, the method comprising:

receiving a first text associated with a first product;

applying, to the first text, a machine learning model trained to output portions of the first text that include one or more respective characteristics of the first product from the first text;

determining a first characteristic group and a second characteristic group, each group comprising at least one of the respective characteristics, wherein:

characteristics within the first group are similar to each other;

characteristics within the second group are similar to each other; and

the characteristics within the first group are different than the characteristics within the second group; and

presenting the related product interface portion that includes a respective product of each of the first group and of the second group.

14. The method of claim 13 , further comprising:

receiving a second text associated with a second product;

identifying one or more second characteristics from the second text;

comparing the second characteristics to a first representative of the first group and to a second representative of the second group; and

assign the second product to at least one of the first group or the second group based on the comparison.

15. The method of claim 14 , wherein comparing the second characteristics to the first representative and the second representative comprises determining a first similarity score for the second characteristics and the first representative and a second familiarity score for the second characteristics and the second representative.

16. The method of claim 13 , wherein determining the first characteristic group and the second characteristic group comprises:

generating respective embeddings for each characteristic;

calculating similarities of the embeddings to each other; and

assigning characteristics to the first group and the second group according to the similarities of the embeddings.

17. The method of claim 13 , wherein presenting the first product comprises:

determining a metric of each product within the first group;

ordering the products within the first group based on the metric; and

displaying at least one of the products within the first group in the order.

18. The method of claim 17 , wherein the metric is a confidence score associated with the determination of each characteristic from the end user-generated text associated with the product.

19. The method of claim 1 , wherein the machine learning model comprises a natural language processing module.

20. The system of claim 7 , wherein the machine learning model comprises a natural language processing module.