Context accumulation based on properties of entity features
According to one embodiment of the present invention, a system for determining related entities includes at least one processor. The system identifies one or more candidate entities for a data object associated with an entity based on one or more features. The entity and candidate entities are each associated with one or more features and each feature is associated with one or more behaviors. The identified candidate entities related to the entity are determined based on the associated behaviors of the features of the identified candidate entities and the entity. Embodiments of the present invention further include a method and computer program product for determining related entities in substantially the same manner described above.
1. A computer-implemented method for conjoining entities comprising:
identifying a candidate entity associated with one or more features based on one or more features associated with an entity, wherein each of the one or more features of the candidate entity and each of the one or more features of the entity is associated with behaviors consisting of a frequency behavior indicating a quantity of entities of a plurality of entities having a corresponding feature and an exclusivity behavior indicating a quantity of the corresponding feature entities of the plurality of entities have;
determining, for each feature of the candidate entity, a feature property score that represents a degree of similarity between the behaviors of that feature of the candidate entity and the behaviors of that feature of the entity;
applying rules to each feature property score of the candidate entity, wherein each rule of the rules specifies a threshold of a particular degree of similarity in conjunction with one or more behaviors; and
in response to applying the rules, conjoining the candidate entity and the entity to form a conjoined entity by assigning a common entity identifier to each of one or more data records associated with the candidate entity and to each of one or more data records associated with the entity.
2. The computer-implemented method of claim 1 , further comprising:
modifying a value of one of the behaviors of the entity based on statistics.
3. The computer-implemented method of claim 1 , wherein the behaviors further consist of a stability behavior indicating whether a value for the corresponding feature changes for the entity.
4. The computer-implemented method of claim 1 , further comprising:
combining two or more features of the entity to produce a new feature with a new behavior for the entity.
5. The computer-implemented method of claim 1 , wherein the degree of similarity is any one of: same, close, likely, plausible, unlikely, and no chance.
6. A system for conjoining comprising:
at least one processor configured to:
identifying a candidate entity associated with one or more features based on one or more features associated with an entity, wherein each of the one or more features of the candidate entity and each of the one or more features of the entity is associated with behaviors consisting of a frequency behavior indicating a quantity of entities of a plurality of entities having a corresponding feature and an exclusivity behavior indicating a quantity of the corresponding feature entities of the plurality of entities have;
determining, for each feature of the candidate entity, a feature property score that represents a degree of similarity between the behaviors of that feature of the candidate entity and the behaviors of that feature of the entity;
applying rules to each feature property score of the candidate entity, wherein each rule of the rules specifies a threshold of a particular degree of similarity in conjunction with one or more behaviors; and
in response to applying the rules, conjoining the candidate entity and the entity to form a conjoined entity by assigning a common entity identifier to each of one or more data records associated with the candidate entity and to each of one or more data records associated with the entity.
7. The system of claim 6 , wherein the at least one processor is further configured to:
modify a value of one of the behaviors of the entity based on statistics.
8. The system of claim 6 , wherein the behaviors further consist of a stability behavior indicating whether a value for the corresponding feature changes for the entity.
9. The system of claim 6 , further comprising:
combining two or more features of the entity to produce a new feature with a new behavior for the entity.
10. The system of claim 6 , wherein the degree of similarity is any one of: same, close, likely, plausible, unlikely, and no chance.
11. A computer program product for conjoining entities comprising:
a non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code comprising computer readable program code configured to:
identifying a candidate entity associated with one or more features based on one or more features associated with an entity, wherein each of the one or more features of the candidate entity and each of the one or more features of the entity is associated with behaviors consisting of a frequency behavior indicating a quantity of entities of a plurality of entities having a corresponding feature and an exclusivity behavior indicating a quantity of the corresponding feature entities of the plurality of entities have;
determining, for each feature of the candidate entity, a feature property score that represents a degree of similarity between the behaviors of that feature of the candidate entity and the behaviors of that feature of the entity;
applying rules to each feature property score of the candidate entity, wherein each rule of the rules specifies a threshold of a particular degree of similarity in conjunction with one or more behaviors; and
in response to applying the rules, conjoining the candidate entity and the entity to form a conjoined entity by assigning a common entity identifier to each of one or more data records associated with the candidate entity and to each of one or more data records associated with the entity.
12. The computer program product of claim 11 , wherein the computer readable program code further comprises computer readable program code configured to:
modify a value of one of the behaviors of the entity based on statistics.
13. The computer program product of claim 11 , wherein the behaviors further consist of a stability behavior indicating whether a value for the corresponding feature changes for the entity.
14. The computer program product of claim 11 , further comprising:
combining two or more features of the entity to produce a new feature with a new behavior for the entity.
15. The computer program product of claim 11 , wherein the degree of similarity is any one of: same, close, likely, plausible, unlikely, and no chance.