Selecting diverse product titles to display on a website
A system and method for selecting diverse product titles to display on a website are disclosed. In some example embodiments, the methods and systems described herein identify available products to be displayed, cluster the identified products based on their similarity to one another, select one or more products from each of the clusters, and display information, such as a title, associated with the randomly selected products.
1. A system, comprising:
a non-transitory memory storing instructions; and
a processor configured to execute the instructions to cause the system to:
group a plurality of available products into a plurality of clusters;
determine a number of products, from the plurality of available products, to display on a client device;
generate a set of buckets to which to allocate the plurality of available products, wherein a particular number of buckets in the set of buckets is based on the number of products to display on the client device;
allocate individual ones of the plurality of available products, from the plurality of clusters, to the set of buckets, wherein, for a first cluster of the plurality of clusters, the allocating includes allocating at least one product from the first cluster to each of the set of buckets; and
select a particular bucket from the set of buckets.
2. The system of claim 1 , wherein the particular bucket is selected using a random selection technique.
3. The system of claim 1 , wherein grouping the plurality of available products into the plurality of clusters is performed based on similarity scores indicative of a similarity between the plurality of available products.
4. The system of claim 1 , wherein executing the instructions further causes the system to:
transmit, to the client device, information corresponding to one or more items allocated to the particular bucket.
5. The system of claim 1 , wherein grouping the plurality of available products into the plurality of clusters is performed based on lists of descriptive terms associated with plurality of available products.
6. The system of claim 5 , wherein the grouping the plurality of available products into the plurality of clusters includes:
generating frequency vectors using the lists of descriptive terms; and
determining similarity scores indicative of a similarity between the plurality of available products using the frequency vectors.
7. A method, comprising:
grouping, by a computer system, a plurality of available products into a plurality of clusters;
determining, by the computer system, a number of products, from the plurality of available products, to display on a client device;
generating, by the computer system, a set of buckets to which to allocate the plurality of available products, wherein a particular number of buckets in the set of buckets is based on the number of products to display on the client device;
allocating, by the computer system, individual ones of the plurality of available products, from the plurality of clusters, to the set of buckets, wherein, for a first cluster of the plurality of clusters, the allocating includes allocating at least one product from the first cluster to each of the set of buckets; and
selecting, by the computer system, a particular bucket from the set of buckets.
8. The method of claim 7 , wherein the particular bucket is selected using a random selection technique.
9. The method of claim 7 , wherein the grouping the plurality of available products into the plurality of clusters is performed based on similarity scores indicative of a similarity between the plurality of available products.
10. The method of claim 7 , further comprising:
transmitting, to the client device, information corresponding to one or more items allocated to the particular bucket.
11. The method of claim 7 , wherein the grouping the plurality of available products into the plurality of clusters is performed based on lists of descriptive terms associated with the plurality of available products.
12. The method of claim 11 , wherein the grouping the plurality of available products into the plurality of clusters includes:
generating frequency vectors using the lists of descriptive terms; and
determining similarity scores indicative of a similarity between the plurality of available products using the frequency vectors.
13. A non-transitory machine-readable medium having instructions stored thereon, the instructions executable to cause performance of operations comprising:
grouping a plurality of available products into a plurality of clusters;
determining a number of products, from the plurality of available products, to display on a client device;
generating a set of buckets to which to allocate the plurality of available products, wherein a particular number of buckets in the set of buckets is based on the number of products to display on the client device;
allocating individual ones of the plurality of available products, from the plurality of clusters, to the set of buckets, wherein, for a first cluster of the plurality of clusters, the allocating includes allocating at least one product from the first cluster to each of the set of buckets; and
selecting a particular bucket from the set of buckets.
14. The non-transitory machine-readable medium of claim 13 , wherein the particular bucket is selected using a random selection technique.
15. The non-transitory machine-readable medium of claim 13 , wherein the grouping the plurality of available products into the plurality of clusters is performed based on similarity scores indicative of a similarity between the plurality of available products.
16. The non-transitory machine-readable medium of claim 13 , wherein the grouping the plurality of available products into the plurality of clusters is performed based on lists of descriptive terms associated with the plurality of available products.
17. The non-transitory machine-readable medium of claim 13 , the operations further comprising:
transmitting, to the client device, information corresponding to one or more items allocated to the particular bucket.
18. The method of claim 7 , wherein the selecting the particular bucket is performed subsequent to a request, from the client device, for information associated with products that satisfy a search query.
19. The method of claim 18 , further comprising:
identifying, by the computer system, the plurality of available products, from a larger set of products, based on the search query specified in the request.
20. The method of claim 19 , wherein the determining the number of products to display on the client device is based on an available display area on a webpage used to send the request to the computer system.