System and method for data extraction and searching

1. A non-transitory computer-readable medium having computer executable program code embodied thereon, the computer executable program code configured to cause a computer system to:

perform image pre-processing on a document image;

perform optical character recognition on the document image;

create a textual data representation of the document image;

apply processing rules to the textual data representation;

perform one or more extractions based on the processing rules to extract metadata from the textual data representation and contextualize the extracted metadata;

perform automated validation of the contextualization of the extracted metadata;

store the extracted metadata and the textual data representation in a full text index database;

transfer the extracted metadata and the textual data representation from the full text index database to a search engine platform, the search engine platform indexing and storing the transferred extracted metadata to allow for searching of the indexed, extracted metadata, the indexed, extracted metadata having been correlated to the textual data representation, wherein the search engine platform allows for the selection of extracted metadata stored in full text index database that is transferred to the search engine platform;

perform additional validation of the extracted metadata upon transfer of the extracted metadata to the search engine platform by correlating at least one of the extracted metadata and the textual data representation with at least one additional reference document relevant to the at least one of the extracted metadata and the textual data representation;

display at the search engine platform a plurality of facets for filtering of search results; and

cause the search engine platform to learn from facets utilized in prior searches how to access data from the full text index database.

2. The non-transitory computer-readable medium of claim 1 , wherein the computer executable program code is configured to further cause the computer system to receive pre-processing rules feedback for performance of the image pre-processing based upon optical character recognition correction information obtained from correction of the textual data representation subsequent to the performance of the optical character recognition.

3. The non-transitory computer-readable medium of claim 1 , wherein the computer executable program code for performing the image pre-processing is configured to cause the computer system to apply a preferred pre-processing profile.

4. The non-transitory computer-readable medium of claim 3 , wherein the computer executable program code is configured to further cause the computer system to determine if an output from performance of the optical character recognition meets a desired accuracy threshold.

5. The non-transitory computer-readable medium of claim 4 , wherein the computer executable program code is configured to further cause the computer system to output the textual data representation for performance of the one or more extractions upon determining that the output from performance of the optical character recognition meets the desired accuracy threshold.

6. The non-transitory computer-readable medium of claim 4 , wherein the computer executable program code is configured to further cause the computer system to apply multiple pre-processing profiles upon which performance of the optical character recognition is based, select the most accurate output from performance of the optical character recognition, and output the textual data representation resulting from the most accurate output from performance of the optical character recognition.

7. The non-transitory computer-readable medium of claim 1 , wherein each of the multiple pre-processing profiles comprises at least one combination of image enhancement processing, automatically applied pre-processing, erosion processing, and dilation processing.

8. The non-transitory computer-readable medium of claim 1 , wherein the computer executable program code is configured to further cause the computer system to receive research and training feedback for application of the processing rules based upon extraction error correction information obtained from correction of the extracted metadata subsequent to the performance of the one or more extractions.

9. The non-transitory computer-readable medium of claim 1 , wherein the one or more extractions comprise at least one of natural language processing extraction, coordinate-based extraction, quadrant-based extraction, and regular expression extraction.

10. The non-transitory computer-readable medium of claim 1 , wherein the computer executable program code configured to perform the automated validation of the extracted metadata further causes the computer system to access existing collected data with which to compare the extracted metadata to validate the contextualization of the extracted metadata.

11. The non-transitory computer-readable medium of claim 1 , wherein the document image comprises a scanned image of a real estate document.

12. The non-transitory computer-readable medium of claim 11 , wherein the extracted metadata comprises at least one of a title of the real estate document, state information associated with the real estate document, county information associated with the real estate document, address of a property relevant to the real estate document, an assessor's parcel number associated with the property, any party name indicated in the real estate document, a document number associated with the real estate document, a recording date associated with the real estate document, any reference document related to the real estate document, and any legal description of the property.

13. The computer program product of claim 1 , wherein the computer executable program code is configured to further cause the search engine platform to learn from prior searches conducted on the search engine platform how to access data from the full text index database.

14. The non-transitory computer-readable medium of claim 1 , wherein the search engine facets comprise a type of real estate-related document, a county associated property at issue in a real estate-related document, or a date range associated with a real-estate related document.

15. The non-transitory computer-readable medium of claim 14 , wherein the search engine facets comprise a type of real estate-related document.

16. The non-transitory computer-readable medium of claim 14 , wherein the search engine facets comprise a county associated property at issue in a real estate-related document.

17. A computer-implemented method, comprising:

receiving a plurality of scanned document images representative of real estate-related documents;

performing optical character recognition on the plurality of scanned document images to obtain a plurality of textual data representations of the real estate-related documents;

extracting metadata from the textual data representations of the real estate-related documents, and contextualizing the extracted metadata in a real estate-related context;

performing automated validation of the contextualization of the extracted metadata;

storing the extracted metadata and the textual data representation in a full text index database;

transferring a subset of the extracted metadata and the textual data representation from the full text index database to a search engine platform, wherein the search engine platform selects the subset of extracted metadata stored in the full text index database that is transferred to the search engine platform, the search engine platform indexing and storing the transferred subset of the extracted metadata to allow for searching of the indexed, subset of extracted metadata, the indexed, subset of extracted metadata having been correlated to the textual data representation;

performing validation of the extracted metadata by correlating the extracted metadata with at least one additional reference document relevant to the extracted metadata;

providing a link to the at least one additional reference document and one or more elements of the extracted subset of metadata and the textual data representations as search results based on one or more searches for real estate-related information using the search engine platform;

displaying at the search engine platform a plurality of facets for filtering of the search results; and

causing the search engine platform to learn from facets utilized in prior searches how to access data from the full text index database.

18. The computer-implemented method of claim 17 , wherein the plurality of displayed facets comprise a plurality of real estate-related facets, the method further comprising: providing the one or more elements of the extracted subset of metadata and the textual data representations as search results categorized in accordance with the plurality of real-estate related facets.

19. A non-transitory computer-readable medium having computer executable program code embodied thereon, the computer executable program code configured to cause a computer system to:

perform optical character recognition on a plurality of scanned document images to obtain a plurality of textual data representations of the real estate-related documents;

extract metadata from the textual data representations of the real estate-related documents, and contextualize the extracted metadata in a real estate-related context;

store the extracted metadata and the textual data representation in a full text index database;

transfer a subset of the extracted metadata and the textual data representation from the full text index database to a search engine platform, wherein the search engine platform selects the subset of extracted metadata stored in the full text index database that is transferred to the search engine platform, the search engine platform indexing and storing the transferred subset of the extracted metadata to allow for searching of the indexed, subset of extracted metadata, the indexed, subset of extracted metadata having been correlated to the textual data representation;

perform validation of the extracted metadata by correlating the extracted metadata with at least one additional reference document relevant to the extracted metadata;

provide data from the at least one additional reference document and at least one of the extracted metadata and the textual data representations as search results based on one or more searches for real estate-related information performed via the search engine platform; and

display at the search engine platform a plurality of real-estate related facets for filtering of the search results; and

cause the search engine platform to learn from real-estate related facets utilized in prior searches how to access real estate-related data from the full text index database.