Method and apparatus for generating recommendations from descriptive information
Inventors
Lamere, Paul B. • Green, Stephen J. • Alexander, Jeffrey H. • Maillet, Francois • Eck, Douglas
Assignees
Interested in licensing this patent?
MTEC can help explore whether this patent might be available for licensing for your application.
Publication Number
US-10373079-B2
Publication Date
2019-08-06
Expiration Date
Abstract
Meaningful words or phrases may be extracted from the information and used as tags. Weights may be determined for the tags, and tag clouds may be generated for the items. The tag clouds may be stored to a data store. Information specifying a tag cloud may be received. Recommended items for which the tag clouds most closely match the specified tag cloud may be identified. Standard vector space distance calculations, for example the cosine distance between the tag clouds, may be used to determine cloud similarity. The results may be filtered to optimize relevance, novelty and familiarity in accordance with preferences of the user. The recommended items may be displayed to a user interface. Users may interact with the user interface to steer the recommendations towards more relevant content.
Core Innovation
The invention provides a steerable recommender user interface that displays a tag cloud comprising one or more tags, where each tag specifies a descriptive word or phrase and a weight is associated with each tag to indicate relative importance in the tag cloud. The interface also displays a list of recommended items of content from a collection of items of content, and dynamic interactive input to the displayed tag cloud drives updates to the displayed list of recommended items.
After the tag cloud and the recommended items list are displayed, the system receives interactive input to the displayed tag cloud to make one or more modifications that increase or decrease the weight of one or more tags without eliminating the one or more tags from the tag cloud. In response, the system alters the graphic or textual appearance of the displayed tag cloud according to the modifications such that the altered appearance reflects the increase or decrease to the weight while keeping the tags in the tag cloud after the change.
After each modification, the system determines a similarity value for each item of content with respect to the displayed modified tag cloud. The similarity value for each item is determined independent of collaborative filtering, and the system dynamically updates the displayed recommended items list according to a subset of items whose similarity value is above a specified threshold.
Each item of content in the collection is associated with its own respective tag cloud comprising one or more respective tags that specify descriptive words or phrases related to the respective item, and the determined similarity values indicate the strength of similarity between the respective item’s tag cloud and the one or more weighted tags in the displayed modified tag cloud.
Claims Coverage
The document includes three independent claims: clm-00001 (computer-implemented method), clm-00007 (system), and clm-00014 (non-transitory computer-readable storage medium). Across these independent claims, there are four inventive features.
Steerable weighted tag cloud driving dynamic recommendations
Displaying a tag cloud to a steerable recommender user interface, where the tag cloud graphically or textually illustrates that a weight is associated with each tag indicating relative importance of the respective tag; displaying, to the steerable recommender user interface, a list of recommended items from a collection of items of content, wherein dynamic interactive input to the displayed cloud drives updates to the list of recommended items.
Interactive tag-weight modifications without removing tags
Receiving interactive input to the displayed tag cloud subsequent to said display to make one or more modifications including input that increases or decreases the weight of one or more tags in the tag cloud without eliminating the one or more tags from the tag cloud, and altering a graphic or textual appearance of the displayed tag cloud to reflect the increase or decrease to the weight according to the interactive input while the one or more tags remain in the tag cloud.
Similarity independent of collaborative filtering using a modified tag cloud
In response to each modification, determining a similarity value for each item of content with respect to the displayed modified tag cloud, where the similarity value is determined independent of collaborative filtering and indicates a strength of similarity of the item’s associated tag cloud for the respective item to the one or more weighted tags in the displayed modified tag cloud that was modified via the interactive input.
Dynamic recommendation updating based on similarity threshold after each tag-weight change
Dynamically updating the displayed list of recommended items according to a subset of the items for which the determined similarity value with respect to the modified tag cloud is above a specified threshold, wherein dynamically updating includes updating the displayed list after each interactive input to the displayed tag cloud that changes a weight of one or more tags in the displayed tag cloud.
All independent claims focus on a steerable recommender UI with a weighted tag cloud, interactive modification of tag weights without removing tags and corresponding updates to the cloud’s appearance, computation of item similarity values relative to the modified tag cloud independent of collaborative filtering, and dynamic updating of recommended items based on a similarity threshold after each tag-weight change.
Stated Advantages
Not explicitly described in patent.
Documented Applications
Not explicitly described in patent.
Interested in licensing this patent?