Generating keywords by associative context with input words
Systems and methods are provided for accessing a plurality of inventory item listings accessible over a network, wherein each of the plurality of inventory item listings includes one or more text strings, and creating inventory word vectors for at least some of the text strings of the plurality of inventory item listings. The systems and methods further provide for receiving a user input including an input word, wherein the user input is input by a user to access a primary media content, creating an input word vector for the input word, calculating cosine similarities between the input word vector and the inventory word vectors, and analyzing the calculated cosine similarities to determine one or more keywords from the one or more text strings, wherein the one or more keywords are from the text strings having inventory word vectors with the highest cosine similarities to the input word vector.
1. A computer-implemented method comprising:
receiving, from a client device, a search query corresponding to an item listed in an online marketplace;
determining, by one or more processors, one or more keywords based at least in part on a plurality of cosine similarities calculated between a text string included in the search query and a plurality of text strings corresponding to a plurality of listings available via the online marketplace;
identifying, by the one or more processors, a search result corresponding to one or more listings from the plurality of listings available via the online marketplace based at least in part on the one or more keywords; and
outputting, to the client device, the search result corresponding to the one or more listings in response to the search query.
2. The method of claim 1 , wherein determining the one or more keywords comprises:
calculating a plurality of cosine similarities between the text string included in the search query and a plurality of text strings corresponding to the plurality of listings to determine the one or more keywords.
3. The method of claim 2 , wherein determining the one or more keywords comprises:
ranking the plurality of cosine similarities to identify a subset of the plurality of cosine similarities, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.
4. The method of claim 2 , wherein determining the one or more keywords comprises:
identifying a subset of the plurality of cosine similarities that satisfy a threshold, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.
5. The method of claim 1 , further comprising:
selecting a secondary media content for display based at least in part on the one or more keywords, wherein outputting the search result comprises outputting the search result comprising the secondary media content.
6. The method of claim 5 , wherein the secondary media content comprises an advertisement.
7. The method of claim 1 , further comprising:
creating a first word vector from at least one listing of the plurality of listings;
creating a second word vector from the text string included in the search query; and
calculating one or more cosine similarities based at least in part on the first word vector and the second word vector.
8. The method of claim 1 , further comprising:
receiving an update to a listing of the plurality of listings; and
creating an updated word vector associated with the listing based at least in part on receiving the update, wherein identifying the search result corresponding to the one or more listings from the plurality of listings is based at least in part on the updated word vector.
9. A system, comprising:
at least one processor; and
a memory device storing instructions that, when executed by the at least one processor, cause the system to perform operations comprising:
receiving, from a client device, a search query corresponding to an item listed in an online marketplace;
determining one or more keywords based at least in part on a plurality of cosine similarities calculated between a text string included in the search query and a plurality of text strings corresponding to a plurality of listings available via the online marketplace;
identifying a search result corresponding to one or more listings from the plurality of listings available via the online marketplace based at least in part on the one or more keywords; and
outputting, to the client device, the search result corresponding to the one or more listings in response to the search query.
10. The system of claim 9 , wherein the instructions to determine the one or more keywords are further executable to perform operations comprising:
calculating a plurality of cosine similarities between the text string included in the search query and a plurality of text strings corresponding to the plurality of listings to determine the one or more keywords.
11. The system of claim 10 , wherein the instructions to determine the one or more keywords are further executable to perform operations comprising:
ranking the plurality of cosine similarities to identify a subset of the plurality of cosine similarities, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.
12. The system of claim 10 , wherein the instructions to determine the one or more keywords are further executable to perform operations comprising:
identifying a subset of the plurality of cosine similarities that satisfy a threshold, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.
13. The system of claim 9 , wherein the instructions are further executable to perform operations comprising:
selecting a secondary media content for display based at least in part on the one or more keywords, wherein outputting the search result comprises outputting the search result comprising the secondary media content.
14. The system of claim 13 , wherein the secondary media content comprises an advertisement.
15. The system of claim 9 , wherein the instructions are further executable to perform operations comprising:
creating a first word vector from at least one listing of the plurality of listings;
creating a second word vector from the text string included in the search query; and
calculating one or more cosine similarities based at least in part on the first word vector and the second word vector.
16. The system of claim 9 , wherein the instructions are further executable to perform operations comprising:
receiving an update to a listing of the plurality of listings; and
creating an updated word vector associated with the listing based at least in part on receiving the update, wherein identifying the search result corresponding to the one or more listings from the plurality of listings is based at least in part on the updated word vector.
17. A non-transitory computer-readable medium comprising instructions that, when executed by one or more processors, cause a machine to perform operations comprising:
receiving, from a client device, a search query corresponding to an item listed in an online marketplace;
determining one or more keywords based at least in part on a plurality of cosine similarities calculated between a text string included in the search query and a plurality of text strings corresponding to a plurality of listings available via the online marketplace;
identifying a search result corresponding to one or more listings from the plurality of listings available via the online marketplace based at least in part on the one or more keywords; and
outputting, to the client device, the search result corresponding to the one or more listings in response to the search query.
18. The non-transitory computer-readable medium of claim 17 , wherein the instructions to determine the one or more keywords are further executable to cause the machine to perform operations comprising:
calculating a plurality of cosine similarities between the text string included in the search query and a plurality of text strings corresponding to the plurality of listings to determine the one or more keywords.
19. The non-transitory computer-readable medium of claim 18 , wherein the instructions to determine the one or more keywords are further executable to cause the machine to perform operations comprising:
ranking the plurality of cosine similarities to identify a subset of the plurality of cosine similarities, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.
20. The non-transitory computer-readable medium of claim 18 , wherein the instructions are further executable to cause the machine to perform operations comprising:
identifying a subset of the plurality of cosine similarities that satisfy a threshold, wherein the one or more keywords correspond to the subset of the plurality of cosine similarities.