Model Entity Network for Analyzing a Real Entity Network
AssigneeSap SE
InventorAlexey Soshin
Life of this patent
Estimating the term…
Abstract
Systems and techniques that can be used for analyzing a social network or any other type of entity networks. In an effort to preserve the privacy rights of individuals, a model of a real entity network can be generated that is a balanced representation of the entity network, and various tests can be performed on metadata in the model. For example, the model network can be generated based on only two data portions: the total number of nodes in the network and the number of relations per node.
Claims20
A computer-implemented method for analyzing an entity network, the method comprising:
- receiving input including a first number (E) corresponding to a number of entities in a real entity network to be analyzed, and a second number (R) corresponding to a number of relations per entity in the real entity network;
- in response to the input, generating (i) E number of entity entries in an entities table, each entity entry including at least an entity identifier and entity metadata, and (ii) a calculations table including the E number of entity identifiers, each associated with a level value of zero;
- incrementing every Rth level value in the calculations table by one;
- determining, using one or more processors, whether more than R number of the level values were incremented in a most recent incrementing step, and if so again performing the incrementing step on those level values, wherein such incrementing and determination are repeated until the determination finds at most R number of recently incremented level values;
- after the incrementing is finished, creating a model entity network that have the E number of entity entries organized according to their respective level values, the model entity network created by generating a relations table with associations between the E number of entity entries;
- performing one or more tests on the generated model entity network that involve at least some of the metadata;
- and
- modifying the real entity network based on an outcome of the performed test.
A computer program product tangibly embodied in a computer-readable storage medium and comprising instructions that when executed by a processor perform a method comprising:
- receiving input including a first number (E) corresponding to a number of entities in a real entity network to be analyzed, and a second number (R) corresponding to a number of relations per entity in the real entity network;
- in response to the input, generating (i) E number of entity entries in an entities table, each entity entry including at least an entity identifier and entity metadata, and (ii) a calculations table including the E number of entity identifiers, each associated with a level value of zero;
- incrementing every Rth level value in the calculations table by one;
- determining, using one or more processors, whether more than R number of the level values were incremented in a most recent incrementing step, and if so again performing the incrementing step on those level values, wherein such incrementing and determination are repeated until the determination finds at most R number of recently incremented level values;
- after the incrementing is finished, creating a model entity network that have the E number of entity entries organized according to their respective level values, the model entity network created by generating a relations table with associations between the E number of entity entries;
- performing one or more tests on the generated model entity network that involve at least some of the metadata;
- and
- modifying the real entity network based on an outcome of the performed test.
A system comprising:
- one or more processors;
- and
- a computer program product comprising instructions that when executed perform a method comprising:
- receiving input including a first number (E) corresponding to a number of entities in a real entity network to be analyzed, and a second number (R) corresponding to a number of relations per entity in the real entity network;
- in response to the input, generating (i) E number of entity entries in an entities table, each entity entry including at least an entity identifier and entity metadata, and (ii) a calculations table including the E number of entity identifiers, each associated with a level value of zero;
- incrementing every Rth level value in the calculations table by one;
- determining, using one or more processors, whether more than R number of the level values were incremented in a most recent incrementing step, and if so again performing the incrementing step on those level values, wherein such incrementing and determination are repeated until the determination finds at most R number of recently incremented level values;
- after the incrementing is finished, creating a model entity network that have the E number of entity entries organized according to their respective level values, the model entity network created by generating a relations table with associations between the E number of entity entries;
- performing one or more tests on the generated model entity network that involve at least some of the metadata;
- and
- modifying the real entity network based on an outcome of the performed test.
Dependencies are inferred from the claim text.