PROBABILISTIC TREE-STRUCTURED LEARNING SYSTEM FOR EXTRACTING CONTACT DATA FROM QUOTES

• US 20120066160A1
• Filed: 02/24/2011
• Published: 03/15/2012
• Est. Priority Date: 09/10/2010
• Status: Active Grant

• Current Assignee: Salesforce.com, Inc.
• Original Assignee: Salesforce.com, Inc.
• Inventors: Jagota, Arun Kumar
• Granted Patent: US 9,619,534 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 16/23 Updating
• CPC Class Codes: G06F 16/254 Extract, transform and load [ETL] procedures, e.g. ETL data flows in data warehouses
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

Systems and methods for updating data stored in a database, such as contact information. An input string is obtained through a search for timely material associated with the stored contact. The input string is parsed using probabilistic tendencies to extract entities corresponding to those stored with the contact. Secondary entities are used to assist in the identification of the primary entities. The contact is then updated (or added if new) using the extracted primary entities.

Claims

1. A method for creating or updating a data set stored as a record in a database, wherein a plurality of data sets are stored in the database, wherein each data set is defined to include a plurality of fields corresponding to a plurality of predefined entities, the method comprising:

searching through documents for current information about a data set;

upon locating a document containing current information about the data set, copying and storing a data string from the document;

extracting one or more entities from the data string using probabilistic techniques; and

creating and storing a new record for the data set if none exists, or updating an existing record for the data set, using the extracted entities.

2. The method of claim 1, wherein the extracting step includes:

identifying primary entities and second entities in the data string; and

aligning the entities into a sequence;

and wherein the creating and storing step includes;

creating the new record or updating the existing record using the primary entities.

3. The method of claim 1, wherein the data sets store contact information including one or more primary entities, and wherein the searching step includes:

searching through documents for current information instances of the primary entities.

4. The method of claim 1, wherein each data set is a contact, wherein the contact is configured to have one or more defined entity fields having stored values associated with the contact, wherein the extracting step includes:

parsing the data string using probabilistic tendencies to extract a sequence of tokens corresponding to the data string, wherein the tokens represent entity values; and

aligning the sequence of tokens with a sequence of the defined entity fields using probabilistic tendencies.

5. The method of claim 4, wherein the step of parsing the data string includes:

removing a trailing period if present;

adding a space before an apostrophe or a comma; and

splitting the data string into a plurality of entities at each space.

6. The method of claim 5, wherein the extracting step further includes:

identifying primary entities and secondary entities in the sequence of tokens;

evaluating the alignment of the sequence of tokens using the secondary entities; and

providing the entity values for the primary entities to the updating step.

7. The method of claim 6, wherein the probabilistic tendencies for the steps of parsing the data string and aligning the sequence of tokens are determined by analyzing a plurality of training sets of data strings to extract an accurate alignment of primary and secondary entities.

8. The method of claim 7, wherein the analyzing step includes:

emulating linguistic rules using the secondary entities.

9. The method of claim 8, wherein the step of aligning entities includes:

emulating the linguistic rules to obtain an alignment of tokens representing a defined entity sequence.

10. The method of claim 7, wherein the step of analyzing training sets includes:

identifying the probabilities that a given sequence of tokens corresponds to an actual sequence of defined entities in the training set; and

storing the probabilities in a tree-like structure, wherein different sequence probabilities are represented by separate branches, each branch having one or more nodes, each node representing an entity sequence, each node in a branch connected by an arc to a subsequent node, each arc representing a transition probability from one entity sequence to another entity.

11. A machine-readable medium carrying one or more sequences of instructions for updating information associated with a contact stored in a multi-tenant database system, which instructions, when executed by one or more processors, cause the one or more processors to carry out the steps of:

obtaining and storing an input string from a search for quoted material associated with the contact;

extracting one or more entities from the input string using probabilistic techniques; and

creating and storing a new record for the contact if none exists, or updating an existing record for the contact, using the extracted entities.

12. The machine-readable medium of claim 11, wherein the instructions for carrying out the step of extracting entities includes:

parsing the input string using probabilistic tendencies to extract a sequence of tokens corresponding to the input string, wherein the tokens represent entity values; and

aligning the sequence of tokens with a sequence of predefined entity fields using probabilistic tendencies.

13. The machine-readable medium of claim 12, wherein the instructions for carrying out the step of extracting entities includes:

identifying primary entities and secondary entities in the sequence of tokens;

evaluating the alignment of the sequence of tokens using the secondary entities; and

providing the entity values for the primary entities to the updating step.

14. The machine-readable medium of claim 12, wherein the instructions for carrying out the step of parsing the input string includes:

removing a trailing period if present;

adding a space before an apostrophe or a comma; and

splitting the data string into a plurality of entities at each space.

15. The machine-readable medium of claim 12, wherein the instructions for carrying out the step of aligning the sequence of tokens includes:

emulating linguistic rules learned from training sets of input strings.

16. The machine-readable medium of claim 14, wherein the instructions for carrying out the step of aligning the sequence of tokens includes:

applying stored probabilities, said stored probabilities learned from training sets of input strings.

17. An apparatus for extracting contact data from quotes, wherein a plurality of contacts are stored in a multi-tenant database, the apparatus 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;

obtaining and storing an input string from a search for quoted material associated with a contact;

extracting one or more entities from the input string using probabilistic tendencies; and

creating and storing a new record for the contact if none exists, or updating an existing record for the contact, using the extracted entities.

18. The apparatus of claim 17, wherein the probabilistic tendencies are learned from training sets of input strings.

19. A method for transmitting code for extracting contact data from quotes in a multi-tenant database system on a transmission medium, the method comprising:

transmitting code to obtain and store an input string from a search for quoted material associated with a contact;

transmitting code to extract one or more entities from the input string using probabilistic tendencies; and

transmitting code to create and store a new record for the contact if none exists, or updating an existing record for the contact, using the extracted entities.

20. The method of claim 19, wherein the probabilistic tendencies are learned from training sets of input strings.