Selecting diverse product titles to display on a website

1. A system comprising:

a non-transitory memory storing instructions; and

a processor configured to execute the instructions to cause the system to:

receive, via a communication with a device, a request for product information, the request including a product criteria for a product of interest and a display constraint that constrains a manner in which the device displays content;

identify, using a clustering engine, products that satisfy at least the product criteria;

cluster, using the clustering engine, the products identified based on a similarity score into a plurality of clusters;

sort the products from the plurality of clusters into buckets, a quantity of the buckets selected based on the display constraint and the products sorted such that each of the buckets includes at least one product from each cluster;

select one of the buckets to present in response to the request received; and

transmit, via the communication with the device, a response to the request, the response causing the selected bucket of products to appear on a display of the device as the content and according to the display constraint.

2. The system of claim 1 , executing the instructions further causes the system to:

generate a list of descriptive terms found in titles of the products identified;

generate, a frequency vector for the list of descriptive terms; and

determine the similarity score for the products identified based on the frequency vector.

3. The system of claim 2 , wherein the list of descriptive terms are ordered based on the frequency vector.

4. The system of claim 1 , wherein sorting the products into buckets includes using a random selection technique.

5. The system of claim 4 , wherein each of the products identified are assigned to one of the buckets.

6. The system of claim 1 , wherein the similarity score is a cosine similarity score.

7. The system of claim 5 , wherein the quantity of buckets is further based on the number of products identified.

8. A method comprising:

receiving, via a communication with a device, a request for product information, the request including a product criteria for a product of interest and a display constraint that constrains a manner in which the device displays content based on physical characteristics of the device;

identifying, using a clustering engine, products that satisfy at least the product criteria;

clustering, using the clustering engine, the products identified based on a similarity score into a plurality of clusters;

sorting the products from the plurality of clusters into buckets, a quantity of the buckets selected based on the display constraint and the products sorted such that each of the buckets includes at least one product from each cluster;

selecting one of the buckets to present in response to the request received; and

transmitting, via the communication with the device, a response to the request, the response causing the selected bucket of products to appear on a display of the device as the content and according to the display constraint.

9. The method of claim 8 , further comprising:

generating a list of descriptive terms found in titles of the products identified;

generating, a frequency vector for the list of descriptive terms; and

determining the similarity score for the products identified based on the frequency vector.

10. The method of claim 9 , wherein the list of descriptive terms are ordered based on the frequency vector.

11. The method of claim 8 ,

wherein sorting the products into buckets includes using a random selection technique.

12. The method of claim 11 , wherein each of the products identified are assigned to one of the buckets.

13. The method of claim 8 , wherein the similarity score is a cosine similarity score.

14. The method of claim 12 , wherein the quantity of buckets is further based on the number of products identified.

15. A non-transitory machine-readable medium having instructions stored thereon, the instructions executable to cause performance of operations comprising:

receiving, via a communication with a device, a request for product information, the request including a product criteria for a product of interest and a display constraint that constrains a manner in which the device displays content based on a user interface displayed by the device;

identifying, using a clustering engine, products that satisfy at least the product criteria;

clustering, using the clustering engine, the products identified based on a similarity score into a plurality of clusters;

sorting the products from the plurality of clusters into buckets, a quantity of the buckets selected based on the display constraint and the products sorted such that each of the buckets includes at least one product from each cluster;

randomly selecting one of the buckets to present in response to the request received; and

transmitting, via the communication with the device, a response to the request, the response causing the selected bucket of products to appear on a display of the device as the content and according to the display constraint.

16. The non-transitory machine-readable medium of claim 15 , further comprising:

generating a list of descriptive terms found in titles of the products identified;

generating, a frequency vector for the list of descriptive terms; and

determining the similarity score for the products based on the frequency vector.

17. The non-transitory machine-readable medium of claim 16 , wherein the list of descriptive terms are ordered based on the frequency vector.

18. The non-transitory machine-readable medium of claim 15 ,

wherein sorting the products into buckets includes using a random selection technique.

19. The non-transitory machine-readable medium of claim 18 , wherein each of the products identified are assigned to one of the product buckets.

20. The non-transitory machine-readable medium of claim 19 , wherein the quantity of buckets is further based on the number of products identified.