Determining navigation patterns associated with a social networking system to provide content associated with a destination page on a starting page
Systems, methods, and non-transitory computer readable media can obtain user navigation data associated with transitions by users between one or more pages associated with a system. Reduced dimensionality user navigation data can be generated based on the user navigation data. A plurality of clusters can be generated based on the reduced dimensionality user navigation data, wherein each cluster of the plurality of clusters corresponds to a user navigation pattern associated with the system.
1. A computer-implemented method comprising:
obtaining, by a computing system, user navigation data associated with transitions by users between a plurality of pages associated with a system;
generating, by the computing system, reduced dimensionality user navigation data based on the user navigation data;
generating, by the computing system, a plurality of clusters based on the reduced dimensionality user navigation data, wherein each cluster of the plurality of clusters corresponds to a user navigation pattern associated with the system;
identifying, by the computing system, a starting page from the plurality of pages for a first cluster of the plurality of clusters, wherein the starting page is identified from the plurality of pages based on a frequency by which users associated with the first cluster transition to other pages from the starting page; and
providing, by the computing system, content associated with a destination page of the first cluster to the users associated with the first cluster via the starting page for the first cluster.
2. The computer-implemented method of claim 1 , wherein a transition includes a navigation from a first page of the plurality of pages to a second page of the plurality of pages.
3. The computer-implemented method of claim 1 , wherein a number of possible types of transitions included in the transitions is equal to a difference between a square of a number of the plurality of pages and the number of the plurality of pages.
4. The computer-implemented method of claim 3 , wherein the obtaining the user navigation data comprises generating a frequency vector for each of the users, wherein the frequency vector includes a frequency associated with each type of transition of the possible types of transitions.
5. The computer-implemented method of claim 4 , wherein the generating the reduced dimensionality user navigation data comprises determining a singular value decomposition (SVD) of the frequency vectors for the users.
6. The computer-implemented method of claim 5 , wherein the singular value decomposition of the frequency vectors for the users includes a matrix of reduced user vectors, a matrix of eigenvectors, and a matrix of eigenvalues.
7. The computer-implemented method of claim 6 , wherein the generating the plurality of clusters comprises clustering the matrix of reduced user vectors.
8. The computer-implemented method of claim 1 , wherein the generating the plurality of clusters is based on k-means clustering.
9. The computer-implemented method of claim 1 , wherein the user navigation pattern indicated by each cluster of the plurality of clusters is characterized by a centroid of the cluster.
10. The computer-implemented method of claim 1 , wherein the plurality of pages includes at least one of: a feed page that provides one or more content items, a notification page that provides notifications, a profile page that provides user information, a content page that provides details of a specific content item, a video page that provides one or more videos, a photo page that provides one or more photos, a web view page that provides a view of content as presented in a web browser, or a feedback page that allows users to provide feedback associated with an application.
11. A system comprising:
at least one processor; and
a memory storing instructions that, when executed by the at least one processor, cause the system to perform:
obtaining user navigation data associated with transitions by users between a plurality of pages associated with a system;
generating reduced dimensionality user navigation data based on the user navigation data;
generating a plurality of clusters based on the reduced dimensionality user navigation data, wherein each cluster of the plurality of clusters corresponds to a user navigation pattern associated with the system;
identifying a starting page from the plurality of pages for a first cluster of the plurality of clusters, wherein the starting page is identified from the plurality of pages based on a frequency by which users associated with the first cluster transition to other pages from the starting page; and
providing content associated with a destination page of the first cluster to the users associated with the first cluster via the starting page for the first cluster.
12. The system of claim 11 , wherein a number of possible types of transitions included in the transitions is equal to a difference between a square of a number of the plurality of pages and the number of the plurality of pages.
13. The system of claim 12 , wherein the obtaining the user navigation data comprises generating a frequency vector for each of the users, wherein the frequency vector includes a frequency associated with each type of transition of the possible types of transitions.
14. The system of claim 13 , wherein the generating the reduced dimensionality user navigation data comprises determining a singular value decomposition (SVD) of the frequency vectors for the users.
15. The system of claim 14 , wherein the singular value decomposition of the frequency vectors for the users includes a matrix of reduced user vectors, a matrix of eigenvectors, and a matrix of eigenvalues.
16. A non-transitory computer readable medium including instructions that, when executed by at least one hardware processor of a computing system, cause the computing system to perform a method comprising:
obtaining user navigation data associated with transitions by users between a plurality of pages associated with a system;
generating reduced dimensionality user navigation data based on the user navigation data;
generating a plurality of clusters based on the reduced dimensionality user navigation data, wherein each cluster of the plurality of clusters corresponds to a user navigation pattern associated with the system;
identifying a starting page from the plurality of pages for a first cluster of the plurality of clusters, wherein the starting page is identified from the plurality of pages based on a frequency by which users associated with the first cluster transition to other pages from the starting page; and
providing content associated with a destination page of the first cluster to the users associated with the first cluster via the starting page for the first cluster.
17. The non-transitory computer readable medium of claim 16 , wherein a number of possible types of transitions included in the transitions is equal to a difference between a square of a number of the plurality of pages and the number of the plurality of pages.
18. The non-transitory computer readable medium of claim 17 , wherein the obtaining the user navigation data comprises generating a frequency vector for each of the users, wherein the frequency vector includes a frequency associated with each type of transition of the possible types of transitions.
19. The non-transitory computer readable medium of claim 18 , wherein the generating the reduced dimensionality user navigation data comprises determining a singular value decomposition (SVD) of the frequency vectors for the users.
20. The non-transitory computer readable medium of claim 19 , wherein the singular value decomposition of the frequency vectors for the users includes a matrix of reduced user vectors, a matrix of eigenvectors, and a matrix of eigenvalues.