ENHANCED ONLINE USER-INTERACTION TRACKING
Disclosed are systems and methods for enhanced tracking of user interactions with online documents, such as, in accordance with various embodiments, interaction tracking on a sub-document level of granularity and/or interaction tracking that involves storing one or more interaction parameters (e.g., an identifier of the document or of the interacting user) for each interaction.
1 . A method comprising:
executing, by one or more computer processors, instructions stored in one or more machine-readable media to perform operations for recommending documents to a plurality of users of a web site, the operations comprising:
tracking online user interactions with a plurality of online documents and storing, for each of the interactions, a plurality of associated interaction parameters comprising at least an identifier of the document interacted with, an identifier of the interacting user, and a time of interaction; and,
for each of the plurality of users,
generating a document recommendation based on user-interaction metrics computed for the plurality of documents, each user-interaction metric comprising an interaction count aggregated over interactions within a specified time window preceding a time of computation and over interacting users within a defined user group to which the respective user for whom the recommendation is generated belongs; and
sending the document recommendation to a web page generator, the web page generator causing display of the document recommendations to the respective user.
2 . The method of claim 1 , wherein the user group comprises users with a common institutional affiliation.
3 . The method of claim 1 , wherein the user group comprises researchers sharing a common area of research.
4 . The method of claim 3 , wherein topics of the recommended document match interests of the user groups to which the respective users for whom the recommendations are generated belong.
5 . The method of claim 1 , wherein the interaction counts are weighted interaction counts, each interaction being weighted by one or more attributes of the respective interacting user.
6 . The method of claim 5 , wherein the one or more attributes comprise a reputation of the interaction user in a field of research to which the publication pertains.
7 . The method of claim 1 , wherein the interaction counts are based at least in part on interactions having an associated sentiment, interactions having an associated sentiment being weighted in the interaction count based on that sentiment.
8 . The method of claim 1 , wherein the recommendations are generated and displayed responsive to search requests received from the users.
9 . The method of claim 1 , wherein the recommendations are published on a web page broken down by research discipline.
10 . The method of claim 1 , further storing, for each of the interactions, a type of interaction, the interactions in the interaction counts being weighted based on the respective types of interaction.
11 . The method of claim 10 , further comprising analyzing user interactions and their associated interaction types across time to determine consumption patterns, and identifying hypes based on the consumption patterns.
12 . A system comprising:
one or more computer processors executing instructions stored in one or more machine-readable media to perform operations for recommending documents to a plurality of users of a web site, the operations comprising:
tracking online user interactions with a plurality of online documents and storing, for each of the interactions, a plurality of associated interaction parameters comprising at least an identifier of the document interacted with, an identifier of the interacting user, and a time of interaction; and,
for each of the plurality of users,
generating a document recommendation based on user-interaction metrics computed for the plurality of documents, each user-interaction metric comprising an interaction count aggregated over interactions within a specified time window preceding a time of computation and over interacting users within a defined user group to which the respective user for whom the recommendation is generated belongs; and
causing display of the document recommendations to the respective user.
13 . The system of claim 12 , wherein the interaction counts are weighted interaction counts, each interaction being weighted by one or more attributes of the respective interacting user.
14 . The system of claim 12 , wherein the interaction counts are based at least in part on interactions having an associated sentiment, interactions having an associated sentiment being weighted in the interaction count based on that sentiment.
15 . The system of claim 12 , wherein the recommendations are generated and displayed responsive to search requests received from the users.
16 . The system of claim 12 , the operations further comprising storing, for each of the interactions, a type of interaction, the interactions in the interaction counts being weighted based on the respective types of interaction.
17 . One or more machine-readable media storing instructions which, when executed by one or more processors of a machine, cause the one or more processors to perform operations for recommending documents to a plurality of users of a web site, the operations comprising:
tracking online user interactions with a plurality of online documents and storing, for each of the interactions, a plurality of associated interaction parameters comprising at least an identifier of the document interacted with, an identifier of the interacting user, and a time of interaction; and,
for each of the plurality of users,
generating a document recommendation based on user-interaction metrics computed for the plurality of documents, each user-interaction metric comprising an interaction count aggregated over interactions within a specified time window preceding a time of computation and over interacting users within a defined user group to which the respective user for whom the recommendation is generated belongs; and
causing display of the document recommendations to the respective user.
18 . The one or more machine-readable media of claim 17 , wherein the interaction counts are weighted interaction counts, each interaction being weighted by one or more attributes of the respective interacting user.
19 . The one or more machine-readable media of claim 17 , wherein the recommendations are generated and displayed responsive to search requests received from the users.
20 . The one or more machine-readable media of claim 17 , the operations further comprising storing, for each of the interactions, a type of interaction, the interactions in the interaction counts being weighted based on the respective types of interaction.