System and method to facilitate the association of structured content in a structured document with unstructured content in an unstructured document

• US 9,684,691 B1
• Filed: 08/25/2015
• Issued: 06/20/2017
• Est. Priority Date: 08/30/2012
• Status: Active Grant

• Current Assignee: Ez-XBRL Solutions, Inc.
• Original Assignee: Ez-XBRL Solutions, Inc.
• Inventors: Bediako, George, Murty, Srinivas
• Primary Examiner(s): OBERLY, VAN HONG
• Application Number:
  US14/834,869
• Time in Patent Office: 665 Days
• Field of Search: 707811
• CPC Class Codes: G06F 16/245 Query processing
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93
• CPC Class Codes: G06F 16/33 Querying
• CPC Class Codes: G06F 16/84 Mapping; Conversion
• CPC Class Codes: G06F 16/93 Document management systems

Abstract

This disclosure relates to facilitating the association of structured content in a structured document with unstructured content in an unstructured document. The system described herein may be configured to facilitate the association by linking numeric instances in the structured document to corresponding numeric instances in the unstructured document. In some implementations, the system may be configured to link the numeric instances in the structured document to the corresponding numeric instances the unstructured document based on a uniqueness of the numeric instances in the structured document, structural information assigned to non-unique numeric instances, structural information assigned to unique numeric instances related to the non-unique numeric instances, unstructured contextual information related to non-unique numeric instances, and/or other information. In some implementations, the system may include one or more of one or more processors, a user interface, a display, electronic storage, and/or other components.

Claims

1. A system configured to facilitate associating structured content in a structured document with unstructured content in an unstructured document, the system comprising:

one or more processors configured to;

obtain an unstructured document, and generate a structured document from the unstructured document, wherein the structured document has content related to the unstructured document, wherein the structured document is generated from the unstructured document by;

analyzing human-readable textual content in the unstructured document,

segmenting the human-readable textual content into individual unstructured content fragments based on the analysis, the unstructured content fragments being contiguous sections of textual content in the unstructured document,

applying tags to the unstructured content fragments, and

generating the structured document based on the tags and the unstructured content fragments such that the segmented unstructured content fragments from the unstructured document correspond to the structured content fragments in the structured document;

identify numeric instances present in the structured document and the unstructured document, wherein the numeric instances include a first set of numeric instances present in the structured document and a second set of numeric instances present in the unstructured document, wherein the first set of numeric instances includes a first numeric instance appearing in the structured document within a first structured content fragment;

determine uniqueness of the individual ones of the first set of numeric instances present in the structured document, wherein a unique numeric instance expresses a unique first number; and

correlate structured content fragments in the structured document with unstructured content fragments in the unstructured document responsive to the first numeric instance being unique, such that the first numeric instance in the structured document is correlated with an unstructured content fragment in the unstructured document.

2. The system of claim 1, wherein the one or more processors are further configured to:

obtain differentiated data sets associated with non-unique numeric instances in the structured document that differentiate the non-unique numeric instances from other non-unique numeric instances with the same numeric value such that, responsive to the first numeric instance being non-unique, a differentiated data set is associated with the first numeric instance, the differentiated data set including structured contextual information assigned to the first numeric instance; and

correlate structured content fragments in the structured document with unstructured content fragments in the unstructured document, such that, responsive to obtaining the differentiated data set associated with the first numeric instance in the structured document, the first structured content fragment is correlated with an unstructured content fragment containing a second numeric instance from the second set of numeric instances in the unstructured document subsequent to an identification of the second numeric instance that corresponds to the first numeric instance.

3. The system of claim 2, wherein the differentiated data set includes information related to structured contextual information assigned to unique numeric instances related to the first numeric instance.

4. The system of claim 2, wherein the one or more processors are further configured to determine the structured contextual information for the first numeric instance.

5. The system of claim 4, wherein the structured contextual information is related to one or more of a cluster, a semantic label, or a tag.

6. The system of claim 4, wherein the one or more processors are further configured to associate the structured contextual information with unstructured contextual information for correlated numeric instances, and wherein, responsive to correlation of the first numeric instance with the second numeric instance, the structured contextual information for the first numeric instance is associated with unstructured contextual information for the second numeric instance.

7. The system of claim 2, wherein the one or more processors are further configured to differentiate individual ones of the non-unique numeric instances based on unstructured contextual information in the unstructured document that is related to the non-unique numeric instances in the structured document,

wherein the related unstructured contextual information includes a first piece of unstructured contextual information associated with the first numeric instance, such that,

responsive to a failure to obtain a differentiated data set associated with the non-unique first numeric instance, the first piece of unstructured contextual information is determined that is associated with the first numeric instance;

wherein the first piece of unstructured contextual information is related to the first numeric instance,

wherein the first numeric instance is correlated to the second numeric instance based on the first piece of unstructured contextual information.

8. The system of claim 7, wherein related unstructured contextual information includes one or more of information related to a unit reference associated with the numeric instances, or information related to an absolute value of the numeric instances.

9. The system of claim 1, wherein responsive to the determination that the first numeric instance in the structured document expresses a unique first number,

the one or more processors are further configured to identify a corresponding second numeric instance present in the second set of numeric instances in the unstructured document that expresses the first unique number and correlates the first numeric instance to the second numeric instance.

10. The system of claim 1, wherein the structured document is an XBRL document.

11. The system of claim 1, wherein the unstructured document is a corporate financial filing.

12. A method to facilitate associating structured content in a structured document with unstructured content in an unstructured document, the method comprising:

obtaining an unstructured document, and generate a structured document from the unstructured document, wherein the structured document has content related to the unstructured document, wherein the structured document is generated from the unstructured document by;

analyzing human-readable textual content in the unstructured document,

segmenting the human-readable textual content into individual unstructured content fragments based on the analysis, the unstructured content fragments being contiguous sections of textual content in the unstructured document,

applying tags to the unstructured content fragments, and

generating the structured document based on the tags and the unstructured content fragments such that the segmented unstructured content fragments from the unstructured document correspond to the structured content fragments in the structured document;

identifying numeric instances present in the structured document and the unstructured document, wherein the numeric instances include a first set of numeric instances present in the structured document and a second set of numeric instances present in the unstructured document, wherein the first set of numeric instances includes a first numeric instance appearing in the structured document within a first structured content fragment;

determining uniqueness of the individual ones of the first set of numeric instances present in the structured document, wherein a unique numeric instance expresses a unique first number; and

correlating structured content fragments in the structured document with unstructured content fragments in the unstructured document responsive to the first numeric instance being unique, such that the first numeric instance in the structured document is correlated with an unstructured content fragment in the unstructured document.

13. The method of claim 12, further comprising:

obtaining differentiated data sets associated with non-unique numeric instances in the structured document that differentiate the non-unique numeric instances from other non-unique numeric instances with the same numeric value such that, responsive to the first numeric instance being non-unique, a differentiated data set is associated with the first numeric instance, the differentiated data set including structured contextual information assigned to the first numeric instance; and

correlating structured content fragments in the structured document with unstructured content fragments in the unstructured document, such that, responsive to obtaining the differentiated data set associated with the first numeric instance in the structured document, the first structured content fragment is correlated with an unstructured content fragment containing a second numeric instance from the second set of numeric instances in the unstructured document subsequent to an identification of the second numeric instance that corresponds to the first numeric instance.

14. The method of claim 13, wherein the differentiated data set includes information related to structured contextual information assigned to unique numeric instances related to the first numeric instance.

15. The method of claim 13, further comprising determining structured contextual information for the individual numeric instances in the structured document such that structured contextual information is determined for the first numeric instance.

16. The method of claim 15, wherein the structured contextual information is related to one or more of a cluster, a semantic label, or a tag.

17. The method of claim 15, further comprising further comprising correlating individual numeric instances in the structured document with individual numeric instances in the unstructured document and associating the structured contextual information with unstructured contextual information for correlated numeric instances such that, responsive to correlation of the first numeric instance with the second numeric instance, the structured contextual information for the first numeric instance is associated with unstructured contextual information for the second numeric instance.

18. The method of claim 13, further comprising differentiating individual ones of the non-unique numeric instances based on unstructured contextual information in the unstructured document that is related to the non-unique numeric instances in the structured document,

wherein the unstructured contextual information includes a first piece of unstructured contextual information associated with the first numeric instance, such that, responsive to being unable to obtain a differentiated data set associated with the non-unique first numeric instance:

the first piece of unstructured contextual information associated with the first numeric instance is determined; and

the first piece of unstructured contextual information is related to the first numeric instance,

wherein the first numeric instance is correlated to the second numeric instance based on the first piece of unstructured contextual information.

19. The method of claim 18, wherein related unstructured contextual information includes one or more of information related to a unit reference associated with the numeric instances, or information related to an absolute value of the numeric instances.

20. The method of claim 13, wherein responsive to determining that the first numeric instance in the structured document expresses a unique first number,

wherein the method further comprises identifying a corresponding second numeric instance present in the second set of numeric instances in the unstructured document that expresses the first unique number and correlating the first numeric instance to the second numeric instance.

21. The method of claim 13, wherein the structured document is an XBRL document.

22. The method of claim 13, wherein the unstructured document is a corporate financial filing.