Back to results
US 8,914,400 B2Granted
Adjusting results based on a drop point
AssigneeInternational Business Machines Corporation
InventorDennis D. Osentoski
Life of this patent
Estimating the term…
Abstract
Provided are techniques for providing entities. For a keyword, historical data is used to generate a list of entities, wherein a recommendation score is associated with each of the entities. The entities are ranked based on the recommendation score associated with each of the entities. A drop point that is associated with an entity whose recommendation score is low is determined. One or more additional entities are identified for at least one of the entities having a high recommendation score in the list. The one or more additional entities are inserted above the drop point in the list. The list is stored with the keyword.
Claims24
A computer-implemented method, with a processor coupled to memory, for providing entities, the method comprising:
- ranking, by the processor, entities in a first list of entities based on a recommendation score associated with each of the entities in the first list of entities;
- determining a drop point in the first list of entities by identifying a percentage drop that exceeds a threshold used to determine continued relevancy in the recommendation score of a first entity in the first list of entities and the recommendation score of a second entity in the first list of entities, wherein the drop point identifies the second entity as a point in the first list of entities above which a second list of entities are to be added;
- selecting an entity in the first list of entities that has a high recommendation score;
- using the historical data to generate the second list of entities having associated recommendation scores;
- adjusting a recommendation score associated with each of the entities in the second list of entities to fall between the recommendation score of the first entity and the recommendation score of the second entity;
- inserting entities from the second list of entities between the first entity and the second entity in the first list of entities based on the adjusted recommendation score associated with each of the entities in the second list of entities;
- and
- storing the first list of entities with the keyword.
5 more elements
A system for providing entities, comprising:
- a processor;
- and
- storage coupled to the processor, wherein the storage stores a computer program, and wherein the processor executes the computer program to perform operations, the operations comprising:
- ranking entities in a first list of entities based on a recommendation score associated with each of the entities in the first list of entities;
- determining a drop point in the first list of entities by identifying a percentage drop that exceeds a threshold used to determine continued relevancy in the recommendation score of a first entity in the first list of entities and the recommendation score of a second entity in the first list of entities, wherein the drop point identifies the second entity as a point in the first list of entities above which a second list of entities are to be added;
- selecting an entity in the first list of entities that has a high recommendation score;
- using the historical data to generate the second list of entities having associated recommendation scores;
- adjusting a recommendation score associated with each of the entities in the second list of entities to fall between the recommendation score of the first entity and the recommendation score of the second entity;
- inserting entities from the second list of entities between the first entity and the second entity in the first list of entities based on the adjusted recommendation score associated with each of the entities in the second list of entities;
- and
- storing the first list of entities with the keyword.
8 more elements
A computer program product for providing entities, the computer program product comprising:
- a computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising:
- computer readable program code, when executed by a processor of a computer, configured to perform:
- ranking entities in a first list of entities based on a recommendation score associated with each of the entities in the first list of entities;
- determining a drop point in the first list of entities by identifying a percentage drop that exceeds a threshold used to determine continued relevancy in the recommendation score of a first entity in the first list of entities and the recommendation score of a second entity in the first list of entities, wherein the drop point identifies the second entity as a point in the first list of entities above which a second list of entities are to be added;
- selecting an entity in the first list of entities that has a high recommendation score;
- using the historical data to generate the second list of entities having associated recommendation scores;
- adjusting a recommendation score associated with each of the entities in the second list of entities to fall between the recommendation score of the first entity and the recommendation score of the second entity;
- inserting entities from the second list of entities between the first entity and the second entity in the first list of entities based on the adjusted recommendation score associated with each of the entities in the second list of entities;
- and
- storing the first list of entities with the keyword.
7 more elements
Dependencies are inferred from the claim text.