System and method for inferring reporting relationships from a contact database

• US 9,477,698 B2
• Filed: 02/21/2013
• Issued: 10/25/2016
• Est. Priority Date: 02/22/2012
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Jagota, Arun Kumar
• Primary Examiner(s): Stevens, Robert
• Application Number:
  US13/773,221
• Publication Number:
  US 20130218904A1
• Time in Patent Office: 1,342 Days
• Field of Search: 707/748
• US Class Current: 1/1
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/2465 Query processing support for facilitating data mining operations in structured databases

Abstract

A system and method for inferring reporting relationships from contact records. Contact records from a single company are identified, and each record is ranked based on the title. A probabilistic analysis is used to compare the number of contacts on the current level with the number of contacts on a lower level, and make a guess as to reporting relationships between contacts on the different levels. If a confidence score of a guessed reporting relationship is high enough, the reporting relationship is accepted, and contact records updated.

Claims

1. A method for inferring reporting relationships in an organization from contact information, the contact information stored as a plurality of contact records in a database system, each contact record having a defined set of entities, including at least a name, a title, and an organization name, comprising:

selecting, by the database system, an organization name;

identifying, by the database system, a plurality of contact records having the selected organization name;

assigning, by the database system, a rank to each of the identified contact records, the rank corresponding with a title on the corresponding contact record;

inferring, by the database system, a reporting relationship from a first contact associated with a rank to a second contact associated with a different rank, based on a probabilistic analysis calculating a confidence score associated with the reporting relationship between the first contact record and the second contact record of the identified contact records;

determining, by the database system, whether the confidence score associated with the reporting relationship exceeds a confidence threshold; and

storing, by the database system, the reporting relationship in at least one of metadata associated with the identified contact records and a field associated with the identified contact records in response to a determination that the confidence score associated with the reporting relationship exceeds the confidence threshold.

2. The method of claim 1, the step of assigning a rank further comprising:

identifying a plurality of organizational levels in the organization, ranging from a lowest level to a highest level; and

associating an increasing rank with each level starting with the lowest level to the highest level.

3. The method of claim 2, the step of determining reporting relationships further comprising:

for each level which comprises at least one contact at one rank below the current rank, counting the number of contacts at the current rank and the number of contacts at one rank below the current rank;

for each level, determining an average reporting ratio as the number of contacts at one rank below the current rank divided by the number of contacts at the current rank;

for each level, determining the confidence score based on the probability distribution divided by the average reporting ratio divided by the number of contacts at the current rank;

for each level, if the confidence score is acceptable, for each contact A at one rank below the current rank, and for each contact B at the current rank, identifying that contact A reports to contact B.

4. The method of claim 3, further comprising:

updating contact records for contact A and contact B.

5. The method of claim 3, further comprising:

performing the step of determining reporting relationships iteratively through each level of the organization.

6. The method of claim 5, further comprising:

performing the step of determining reporting relationships iteratively starting with the highest level down to the lowest level.

7. The method of claim 3, further comprising:

move to the next level if the number of contacts at the current rank is zero, and the current rank is a rank for which reporting relationships are determined.

8. The method of claim 3, further comprising:

counting the number of contacts at two ranks below the current rank;

determining the average reporting ratio as the number of contacts at two ranks below the current rank divided by the number of contacts at the current rank, if the number of contacts at the current rank is not zero, and the number of contacts at one rank below the current rank is zero; and

if the confidence score is acceptable, for each contact A at two ranks below the current rank, and for each contact B at the current rank, identifying that contact A reports to contact B.

9. The method of claim 1, the defined set of entities further including a location, further comprising:

identifying at least one chief executive level position in a title field of the identified contact records, and assigning the chief executive level position a top rank;

identifying a department for each identified contact record, the department based on the title;

sorting the identified contact records by department and location thereby creating a plurality of record subsets, each subset identifying a specific department/location pair;

identifying, for each department/location subset, the contact records having the highest rank below the top rank; and

determining reporting relationships between the contact record having a top rank and the contact records having the highest rank below the top rank using a probabilistic analysis.

10. The method of claim 9, further comprising:

determining an average reporting ratio as the union of all contacts having the highest rank below the top rank divided by the number of contacts at the top rank;

determining the confidence score based on the probability distribution divided by the average reporting ratio divided by the number of contacts at the top rank;

if the confidence score is acceptable, for each pair of contacts (A, B) in the set having the intersection of the union of all contacts having the highest rank below the top rank with the number of contacts at the top rank, identifying that contact A reports to contact B.

11. A non-transitory computer-readable medium encoded with instructions for inferring reporting relationships in an organization from contact information, the contact information stored as a plurality of contact records in a database system, each contact record having a defined set of entities, including at least a name, a title, and an organization name, the instructions when executed by a processor cause the processor to carry out the steps of:

selecting, with the database system, an organization name;

identifying a plurality of contact records in the database system having the selected organization name;

assigning, by the database system, a rank to each of the identified contact records, the rank corresponding with a title on the corresponding contact record; and

inferring, by the database system, a reporting relationship from a first contact associated with a rank to a second contact associated with a different rank, based on a probabilistic analysis calculating a confidence score associated with the reporting relationship between the first contact record and the second contact record of the identified contact records;

determining whether the confidence score associated with the reporting relationship exceeds a confidence threshold; and

storing the reporting relationship in at least one of metadata associated with the identified contact records and a field associated with the identified contact records in response to a determination that the confidence score associated with the reporting relationship exceeds the confidence threshold.

12. The computer-readable medium of claim 11, the step of assigning a rank further comprising:

identifying a plurality of organizational levels in the organization, ranging from a lowest level to a highest level; and

associating an increasing rank with each level starting with the lowest level to the highest level.

13. The computer-readable medium of claim 12, the step of determining reporting relationships further comprising:

for each level which comprises at least one contact at one rank below the current rank, counting the number of contacts at the current rank and the number of contacts at one rank below the current rank;

for each level, determining the average reporting ratio as the number of contacts at one rank below the current rank divided by the number of contacts at the current rank;

for each level, determining the confidence score based on the probability distribution divided by the average reporting ratio divided by the number of contacts at the current rank;

for each level, if the confidence score is acceptable, for each contact A at one rank below the current rank, and for each contact B at the current rank, identifying that contact A reports to contact B.

14. The computer-readable medium of claim 13, further comprising:

updating contact records for contact A and contact B.

15. The computer-readable medium of claim 13, further comprising:

performing the step of determining reporting relationships iteratively through each level of the organization.

16. The computer-readable medium of claim 15, further comprising:

performing the step of determining reporting relationships iteratively starting with the highest level down to the lowest level.

17. The computer-readable medium of claim 13, further comprising:

move to the next level if the number of contacts at the current rank is zero, and the current rank is a rank for which reporting relationships are determined.

18. The computer-readable medium of claim 13, further comprising:

counting the number of contacts at two ranks below the current rank;

determining the average reporting ratio as the number of contacts at two ranks below the current rank divided by the number of contacts at the current rank, if the number of contacts at the current rank is not zero, and the number of contacts at one rank below the current rank is zero; and

if the confidence score is acceptable, for each contact A at two ranks below the current rank, and for each contact B at the current rank, identifying that contact A reports to contact B.

19. The computer-readable medium of claim 11, the defined set of entities further including a location, further comprising:

identifying at least one chief executive level position in a title field of the identified contact records, and assigning the chief executive level position a top rank;

identifying a department for each identified contact record, the department based on the title;

sorting the identified contact records by department and location thereby creating a plurality of record subsets, each subset identifying a specific department/location pair;

identifying, for each department/location subset, the contact records having the highest rank below the top rank; and

determining reporting relationships between the contact record having a top rank and the contact records having the highest rank below the top rank using a probabilistic analysis.

20. The computer-readable medium of claim 19, further comprising:

determining an average reporting ratio as the union of all contacts having the highest rank below the top rank divided by the number of contacts at the top rank;

determining the confidence score based on the probability distribution divided by the average reporting ratio divided by the number of contacts at the top rank;

if the confidence score is acceptable, for each pair of contacts (A, B) in the set having the intersection of the union of all contacts having the highest rank below the top rank with the number of contacts at the top rank, identifying that contact A reports to contact B.

21. An apparatus for inferring reporting relationships in an organization from contact information, the contact information stored as a plurality of contact records in a database system, each contact record having a defined set of entities, including at least a name, a title, and an organization name, comprising:

a processor; and

one or more stored sequences of instructions which, when executed by the processor, cause the processor to carry out the steps of;

selecting, with the database system, an organization name;

identifying a plurality of contact records in the database system having the selected organization name;

assigning, by the database system, a rank to each of the identified contact records, the rank corresponding with a title on the corresponding contact record; and

inferring, by the database system, a reporting relationship from a first contact associated with a rank to a second contact associated with a different rank, based on a probabilistic analysis calculating a confidence score associated with the reporting relationship between the first contact record and the second contact record of the identified contact records;

determining whether the confidence score associated with the reporting relationship exceeds a confidence threshold; and

storing the reporting relationship in at least one of metadata associated with the identified contact records and a field associated with the identified contact records in response to a determination that the confidence score associated with the reporting relationship exceeds the confidence threshold.