Automatic human-emulative document analysis

1. A computer-implemented method of generating a user interface showing results of human-emulative document analysis, the method comprising:

receiving documents containing text, individual ones of the documents associated with a generated document identification number;

pre-processing the documents to generate one or more processed document portions from the individual ones of the documents, individual ones of the processed document portions associated with one of the generated document identification numbers and comprising text parsed into separate words with stop words, duplicate words, and punctuation removed;

detecting anomalies based on anomalous words and normative words in the processed document portions;

generating a word count for the individual ones of the processed document portions by counting the number of separate words in the individual ones of the processed document portions;

identifying a referential word count;

calculating a word count ratio for the individual ones of the processed document portions by dividing the referential word count by the word count for the individual ones of the processed document portions;

determining a word frequency for individual words included in the document portions, the word frequency being a number of instances that a word is found in the document portions prior to removal of duplicate words;

generating a commonness score for the individual ones of the processed document portions by taking the square root of the sum of the squares of the inverse of the word frequency for the individual ones of the separate words in the individual ones of the processed document portions;

identifying a highest commonness score of the processed document portions;

calculating a commonness score ratio for the individual ones of the processed document portions by dividing the highest commonness score by the commonness score for the individual ones of the processed document portions;

calculating an overall score for individual ones of the processed document portions based on a normalization of the square root of the sum of the square of the word count ratio and the square of the commonness score ratio for the individual ones of the processed document portions; and

generating the user interface including at least one overall score for one of the processed document portions in proximity to the generated document identification number associated with the one of the processed document portions and an indicia indicating a result from the detecting anomalies for the one of the processed document portions.

2. The method of claim 1 , wherein the pre-processing: divides one of the plurality of documents into two processed document portions at a location following a period and preceding a numeral.

3. The method of claim 1 , wherein the pre-processing replaces abbreviations and acronyms with corresponding full words.

4. The method of claim 1 , wherein at least part of the stop words are specific to a document classification of the plurality of documents.

5. The method of claim 1 , further comprising:

determining that one of the processed document portions contains a word on an anomalous word list; and

excluding the one of the document portions from further analysis.

6. The method of claim 1 , further comprising determining that one of the plurality of documents is associated with a plurality of associated, processed document portions following pre-processing;

identifying the one of the associated, processed document portions having a highest overall score as compared to other associated, processed document portions; and

assigning the highest overall score as the score of the one of the plurality of documents.

7. The method of claim 1 , wherein the pre-processing, the detecting anomalies, the generating the word count, the identifying the referential word count, the calculating the word count ratio, the determining the word frequency, the generating the commonness score, the identifying the highest commonness score, the calculating the commonness score ratio, and the calculating the overall score are performed at a rate that exceeds one document per minute.

8. The method of claim 1 , wherein the plurality of documents containing text comprise patents, the unique document identification number comprises a patent number, and the processed document portions comprise patent claims.

9. A method for automatically assigning a claim breadth score to a patent claim, the method comprising:

obtaining a data file including a corpus of patent claims;

obtaining a first set of rules that defines an anomalous patent claim, the first set of rules comprising a first rule for identifying at least one of a dependent patent claim, a deleted patent claim, a means-plus-function patent claim, or a patent claim containing normative language;

generating an indicium marking the patent claim by applying the first set of rules;

obtaining a second set of rules that define a word count score for a patent claim as a function of a word count in the patent claim;

obtaining a third set of rules that define a commonness score for the patent claim as a function of the frequency with which words in the patent claim are present in the corpus of patent claims, wherein:

the third set of rules comprises a third rule defining the commonness score as based on a per-claim commonness score;

the per-claim commonness score is calculated by a square root of a sum of, for each word in the patent claim following pre-processing, the square of the inverse of a global word count for each word; and

the commonness score is a ratio of (i) the per-claim commonness score of a high-commonality patent claim in the corpus of patent claims having a largest per-claim commonness score to (ii) the per-claim commonness score of the patent claim;

generating the word count score and the commonness score for the patent claim by evaluating the patent claim against the second set of rules and the third set of rules;

generating a claim breadth score for the patent claim from a combination of the word count score and the commonness score; and

applying the claim breadth score to the patent claim to produce a ranking with respect to a plurality of other patent claims from the corpus of patent claims.

10. The method of claim 9 , wherein when the first set of rules comprises the first rule for identifying the dependent patent claim and the deleted patent claim and generating an ignore list for patent claims included in the corpus of patent claims by applying the first set of rules, wherein the word count score and the commonness score are not generated for patent claims included in the ignore list.

11. The method of claim 9 , wherein the second set of rules comprises a second rule defining the word count score as based on a number of words in the patent claim following pre-processing, wherein pre-processing comprises stemming, removal of duplicate words, and removal of stop words.

12. The method of claim 11 , wherein the word count score is a ratio of (i) a number of words in a high-word count patent claim in the corpus of patent claims having a largest number of words following pre-processing to (ii) the number of words in the patent claim following pre-processing.

13. The method of claim 9 , wherein the claim breadth score is calculated by a square root of the sum of both the square of the word count score and the square of the commonness score.

14. The method of claim 9 , further comprising generating a user interface that displays the ranking showing the claim breadth score of the patent claim, claim breadth scores of the other patent claims, and patent numbers associated with the individual patent claims.

15. One or more computing devices for automatically analyzing a corpus of patent documents, the computing devices comprising:

one or more processing units;

one or more memories coupled to the one or more processing units;

a pre-processing component coupled to the one or more processing units and configured to process claim sections of the corpus of patent documents, delimit individual claims, stem words in the individual claims to root forms, remove duplicate root forms from the individual claims, and remove stop words from the individual claims;

an anomaly detection component coupled to the one or more processing units and configured to detect and mark dependent claims and deleted claims in the individual claims;

a claim breadth calculation component coupled to the one or more processing units and configured to calculate a claim breadth score for each of individual claims that are not marked, the claim breadth score based on a word count score of a claim and a commonness score of the claim; and

a ranking component coupled to the one or more processing units and configured to:

assign the claim breadth score for a highest ranked claim in a patent document as the claim breadth score for the patent document;

rank order the patent documents in the corpus of patent documents based on claim breadth score; and

bin the claim breadth scores of the patent documents into a predetermined number of bins.

16. The one or more computing devices of claim 15 , wherein the claim breadth calculation component comprises a word count score calculation component configured to determine a word count score for the claim based on a word count for the claim and a maximum word count for a claim from the corpus of patent documents having a highest word count.

17. The one or more computing devices of claim 15 , wherein the claim breadth calculation component comprises a commonness score calculation component configured to determine a commonness score for the claim based on the frequencies individual words in the claim occur throughout all of the claims in the corpus of patent documents.