Customized today module
A method and apparatus for customizing content presented to individual users or user segments is provided. There may be three components, a web portal and toolbar component, a modeling component, and a scoring component. The web portal and toolbar component presents content items and collects data. The web portal and toolbar component generates user event data based on the user actions. The user event data is forwarded to the modeling component. The modeling component generates content scoring functions based on user event data and attributes of content items. Content scoring functions may be unique to individual user segments. The content scoring functions based on content features generate probability a content item will be viewed. The scoring component decides which content items are placed in a portal. The scoring component uses the scoring functions generated by the modeling component to rank content items in real time.
1. A method comprising performing a machine-executed operation involving instructions, wherein said instructions are instructions which, when executed by one or more processors, cause the one or more processors to perform certain steps comprising:
assembling user event data from a plurality of content segments on a portal;
wherein the user event data encodes components of user interaction with first content items within each of the plurality of content segments on the portal;
wherein the encoding of the components of user interaction includes, at least in part, encoding a position of a particular content item of the first content items within the particular content item's corresponding segment on the portal;
for each of a plurality of user segments, generating one or more ranking functions based on the user event data and based on said position of said particular content item within said portal;
selecting a ranking function from the one or more ranking functions based on a first user segment of the plurality of user segments;
for each of a plurality of second content items for which there is no user event data for any user, extracting a set of features for each second content item;
with the selected ranking function, generating a rank for the second content items based on the set of features for each second content item;
providing, based on the rank, at least one of the second content items.
2. The method of claim 1 wherein one of the components of user interaction being encoded is a time of the user interaction with the first content items.
3. The method of claim 1 wherein the step of selecting a ranking function based on the first user segment further includes selecting a ranking function based on context, wherein the context is, at least in part, a time of day when the selection is occurring.
4. The method of claim 1 wherein the step of, with the selected ranking function, generating a rank for the second content items based on the set of features for each second content item further comprises computing a probability of a user clicking on a particular second content item of the second content items.
5. A system comprising:
one or more processors; and
a hard disk drive storing a plurality of modules comprising:
a first module that is operable to assemble user event data from a plurality of content segments on a portal, wherein the user event data encodes components of user interaction with first content items within each of the plurality of content segments on the portal;
wherein the first module is further operable to encode a component of user interaction that is a position of a particular content item of the first content items within the particular content item's corresponding segment on the portal;
a second module that is operable to, for each of a plurality of user segments, generate one or more ranking functions based on the user event data and based on said position of said particular content item within said portal;
a third module that is operable to extract a set of features for each second content item of a plurality of second content items for which there is no user event data for any user;
a fourth module that is operable to generate, based on a first of the ranking functions, a rank for the second content items based on the set of features for each second content item;
a fifth module that is operable to provide, based on the rank, at least one of the second content items.
6. The system of claim 5 , wherein the first module is further operable to assemble user event data which encodes a component of user interaction which is a time of the user interaction with the first content items.
7. The system of claim 5 wherein the fourth module further comprises being operable to generate the ranking based at least in part on a probability of a user clicking on a particular second content item of the second content items.
8. The method of claim 1 wherein generating the one or more ranking function is based, at least in part, on which social networks the first user segment is associated with.