Systems, apparatuses, and methods for providing a ranking based recommendation
One or more of the systems, apparatuses, or methods discussed herein can include a quality score for a plurality of item listings or collections of item listings. Data sparseness can be avoided, as the quality score is based on inherent properties of the listing. An item listing can be recommended to a user based on the quality score. In one or more embodiments, a method can include determining a plurality of quality scores including a quality score for each of a plurality of item listings or a plurality of collections of item listings, the quality scores determined independent of a user's attributes and independent of the user's contextual information, the contextual information corresponding to details of the user's access to a website, and recommending an item listing or collection of item listings to a user based on the quality scores and the contextual information.
1. A system comprising:
one or more hardware processors; and
a non-transitory machine-readable medium storing instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors of the system to perform operations comprising:
determining, using a machine-learning model, a quality score for each item listing of a plurality of item listings, the quality score being based on an item listing attribute including item freshness and a user attribute including a user engagement, via user interfaces of client devices, with one or more item listings included in a collection of item listings, wherein the item freshness indicates how recently a corresponding item listing was posted for sale;
identifying, based on the determined quality scores of the plurality of item listings, a plurality of candidate item listings from among the plurality of item listings;
determining a browser type of a plurality of browser types, the determined browser type used by a client device to access a website hosting the plurality of item listings;
generating contextual information for the client device that indicates one or more device parameters, the one or more device parameters indicating the determined browser type;
selecting one or more item listings from among the plurality of candidate item listings by filtering the plurality of candidate item listings, wherein the filtering includes:
determining whether user attributes are available;
responsive to determining that user attributes are not available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and then filtering based on the contextual information; and
responsive to determining that user attributes are available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and the user attributes, and then filtering based on the contextual information, wherein the contextual information indicates the determined browser type and the user attributes include a user interest relationship for the one or more item listings indicated by the determined browser type; and
causing display of the one or more selected item listings on a user interface displayed on the client device.
2. The system of claim 1 , wherein the quality score is further based on the item listing attribute further including at least one of an age of the corresponding item listing or an update time of the corresponding item listing, and
wherein the operations further comprise:
comparing the quality score of each item listing of the plurality of item listings to a threshold; and
filtering out each item listing having a quality score less than the threshold.
3. The system of claim 1 , wherein the filtering further includes at least one of:
ranking and removing of a particular item listing using a random forest classifier; or
comparing ranks of the plurality of item listings to a threshold value to determine that the particular item listing is to be removed.
4. The system of claim 1 , wherein the plurality of candidate item listings is identified before the client device accesses the website hosting the plurality of item listings.
5. The system of claim 1 , wherein the contextual information includes a device type of the client device.
6. The system of claim 1 , wherein the item freshness indicates how recently each item listing was put on sale or posted for sale.
7. The system of claim 1 , wherein the machine-learning model comprises a random forest classifier.
8. A computer-implemented method comprising:
determining, using a machine-learning model, a quality score for each item listing of a plurality of item listings, the quality score being based on an item listing attribute including item freshness and a user attribute including a user engagement, via user interfaces of client devices, with one or more item listings included in a collection of item listings, wherein the item freshness indicates how recently a corresponding item listing was posted for sale;
identifying, based on the determined quality scores of the plurality of item listings, a plurality of candidate item listings from among the plurality of item listings;
determining, by one or more processors, a browser type of a plurality of browser types, the determined browser type used by a client device to access a website hosting the plurality of item listings;
generating, by the one or more processors, contextual information for the client device that indicates one or more device parameters, the one or more device parameters indicating the determined browser type;
selecting one or more item listings from among the plurality of candidate item listings by filtering the plurality of candidate item listings, wherein the filtering includes:
determining whether user attributes are available;
responsive to determining that user attributes are not available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and then filtering based on the contextual information; and
responsive to determining that user attributes are available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and the user attributes, and then filtering based on the contextual information, wherein the contextual information indicates the determined browser type and the user attributes include a user interest relationship for the one or more item listings indicated by the determined browser type; and
causing display of the one or more selected item listings on a user interface displayed on the client device.
9. The computer-implemented method of claim 8 , wherein the filtering further includes at least one of:
ranking and removing of a particular item listing using a random forest classifier; or
comparing ranks of the plurality of item listings to a threshold value to determine that the particular item listing is to be removed.
10. The computer-implemented method of claim 8 , wherein the identifying of the plurality of candidate item listings from among the plurality of item listings based on the quality score of the plurality of item listings further comprises:
comparing the quality score of each item listing of the plurality of item listings to a threshold; and
filtering out each item listing having a quality score less than the threshold.
11. The computer-implemented method of claim 8 , wherein the plurality of candidate item listings is identified before the client device accesses the website hosting the plurality of item listings.
12. The computer-implemented method of claim 8 , wherein the contextual information includes a device type of the client device.
13. The computer-implemented method of claim 8 , wherein the item freshness indicates how recently each item listing was put on sale or posted for sale.
14. The computer-implemented method of claim 8 , wherein the machine-learning model comprises a random forest classifier.
15. A non-transitory machine-readable storage medium storing instructions that, when executed by one or more hardware processors of a machine, cause the machine to perform operations comprising:
determining, using a machine-learning model, a quality score for each item listing of a plurality of item listings, the quality score being based on an item listing attribute including item freshness and a user attribute including a user engagement, via user interfaces of client devices, with one or more item listings included in a collection of item listings, wherein the item freshness indicates how recently a corresponding item listing was posted for sale;
identifying, based on the determined quality scores of the plurality of item listings, a plurality of candidate item listings from among the plurality of item listings;
determining a browser type of a plurality of browser types, the determined browser type used by a client device to access a website hosting the plurality of item listings;
generating contextual information for the client device that indicates one or more device parameters, the one or more device parameters indicating the determined browser type;
selecting one or more item listings from among the plurality of candidate item listings by filtering the plurality of candidate item listings, wherein the filtering includes:
determining whether user attributes are available;
responsive to determining that user attributes are not available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and then filtering based on the contextual information; and
responsive to determining that user attributes are available, causing the one or more item listings to be selected after filtering the plurality of candidate item listings based on the determined quality scores and the user attributes, and then filtering based on the contextual information, wherein the contextual information indicates the determined browser type and the user attributes include a user interest relationship for the one or more item listings indicated by the determined browser type; and
causing display of the one or more selected item listings on a user interface displayed on the client device.
16. The non-transitory machine-readable storage medium of claim 15 , wherein the filtering further includes at least one of:
ranking and removing of a particular item listing using a random forest classifier; or
comparing ranks of the plurality of item listings to a threshold value to determine that the particular item listing is to be removed.
17. The non-transitory machine-readable storage medium of claim 15 , wherein the operations further comprise:
comparing the quality score of each item listing of the plurality of item listings to a threshold; and
filtering out each item listing whose quality score is less than the threshold.
18. The non-transitory machine-readable storage medium of claim 15 , wherein the plurality of candidate item listings is identified before the client device accesses the website hosting the plurality of item listings.
19. The non-transitory machine-readable storage medium of claim 15 , wherein the contextual information includes a device type of the client device.
20. The non-transitory machine-readable storage medium of claim 15 , wherein the machine-learning model comprises a random forest classifier.