Machined book detection

• US 9,372,850 B1
• Filed: 12/19/2012
• Issued: 06/21/2016
• Est. Priority Date: 12/19/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Takaki, Mitsuo, Leatham, David Gordon, Mahalingam, Divya, Azari, David Rezazadeh
• Primary Examiner(s): Guerra-Erazo, Edgar
• Application Number:
  US13/720,195
• Time in Patent Office: 1,280 Days
• Field of Search: 704/9, 704/3, 704/5, 704/8, 704/10, 704/277, 704/275, 704/251
• US Class Current: 1/1
• CPC Class Codes: G06F 16/35 Clustering; Classification
• CPC Class Codes: G06F 16/353 into predefined classes
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 40/253 Grammatical analysis; Style critique
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30

Abstract

A system and method for determining whether a textual work submitted for publishing is machine generated or non-machine generated by identifying and quantifying various aspects of the textual work and comparing those aspects to known works. For example, the system and method may identify aspects of a textual work, including, a relationship between the sentences within the textual work, a writing style of the author of the textual work, a grammatical structure of the sentences within the textual work, a quality of the textual work, and other aspects of the textual work. Upon determining that the textual work is machine generated the textual work may be rejected for publishing.

Claims

1. A computer-implemented method of identifying machine generated text, comprising:

analyzing a plurality of pre-determined machine generated works;

storing, in a database, pre-determined centers of mass corresponding to the pre-determined machine generated works;

storing, in the database, pre-determined shape descriptors corresponding to the pre-determined machine generated works;

analyzing a plurality of pre-determined non-machine generated works;

storing, in the database, pre-determined centers of mass corresponding to the pre-determined non-machine generated works;

storing, in the database, pre-determined shape descriptors corresponding to the pre-determined non-machine generated works;

receiving a textual work submitted for publishing;

generating an N-gram of N words of the textual work;

plotting the N-gram in an N-dimensional space;

generating a 2-dimensional plot based, at least in part, on the plot of the N-gram in N-dimensional space;

calculating a center of mass of the 2-dimensional plot;

calculating a shape descriptor of the 2-dimensional plot, wherein the shape descriptor includes one or more points defining a shape of the 2-dimensional plot;

comparing the center of mass to the pre-determined centers of mass and the shape descriptor to the pre-determined shape descriptors;

calculating, based at least in part on the comparison, a confidence score indicative of a correlation between the textual work and at least one of the pre-determined machine generated works;

determining, based at least in part on the confidence score, that the textual work is machine generated; and

rejecting the textual work for publishing based on the determination.

2. The computer-implemented method of claim 1, in which generating the N-gram of the N words of the textual work includes generating the N-gram including the N words identifying a number of times each of the N words appears in one or more sentences of the textual work.

3. The computer-implemented method of claim 1, further comprising:

parsing the textual work and identifying at least one of verbs, nouns, pronouns, adjectives, adverbs, prepositions, conjunctions, and interjections in each sentence of the textual work;

applying one or more grammatical rules to the parsed textual work; and

assigning a score for each sentence of the textual work based on each application of the one or more grammatical rules.

4. The computer-implemented method of claim 3, further comprising computing an overall score for the textual work based on the one or more sentence scores.

5. The computer-implemented method of claim 4, in which the calculating the confidence score includes calculating the confidence score based on the comparison and the overall score.

6. A computer-implemented method, comprising:

identifying pre-determined shape descriptors corresponding to pre-determined non-machine generated works;

generating an N-gram of N words of a textual work;

plotting the N-gram in an N-dimensional space;

generating a 2-dimensional plot based, at least in part, on the plot of the N-gram in N-dimensional space;

comparing a shape of the 2-dimensional plot of the textual work to the pre-determined shape descriptors; and

determining, based at least in part on the comparison, that the textual work is a desired work.

7. The computer-implemented method of claim 6, further comprising calculating a center of mass and the shape descriptor of the 2-dimensional plot.

8. The computer-implemented method of claim 7, in which the comparing includes comparing the center of mass to pre-determined centers of the pre-determined non-machine generated works.

9. The computer-implemented method of claim 6, further comprising calculating, based at least in part on the comparison, a confidence score indicative of a correlation between the textual work and at least one of the pre-determined non-machine generated works.

10. The computer-implemented method of claim 6, further comprising searching an index of a database of the pre-determined shape descriptors for pre-determined non-machine generated works having substantially a same shape as the shape of the textual work.

11. The computer-implemented method of claim 6, in which the plotting of the N-gram in an N-dimensional space includes identifying a frequency of word occurrence in each sentence in the textual work.

12. The computer-implemented method of claim 6, in which the generating of the 2-dimensional plot includes plotting points based on a frequency of word occurrence in each sentence in the textual work and inverse similarity to a center of mass of the N-dimensional plot.

13. The computer-implemented method of claim 6, further comprising applying one or more grammatical rules to each sentence of the textual work, and computing a score for the textual work based on the application of the one or more grammatical rules.

14. The computer-implemented method of claim 6, in which determining that the textual work is the desired work comprises determining that the desired work is a non-machine generated work.

15. A computing device, comprising:

at least one processor;

a memory device including instructions operable to be executed by the at least one processor to perform a set of actions, configuring the processor;

to identify pre-determined shape descriptors corresponding to pre-determined non-machine generated works;

to generate an N-gram of N words of a textual work;

to plot the N-gram in an N-dimensional space;

to generate a 2-dimensional plot based, at least in part, on the plot of the N-gram in N-dimensional space;

to compare a shape of the 2-dimensional plot of the textual work to the pre-determined shape descriptors; and

to determine, based at least in part on the comparison, that the textual work is a desired work.

16. The computing device of claim 15, further comprising configuring the processor to calculate a center of mass and the shape descriptor of the 2-dimensional plot.

17. The computing device of claim 15, in which the processor is configured to compare the center of mass to pre-determined centers of mass of the pre-determined non-machine generated works.

18. The computing device of claim 15, further comprising configuring the processor to calculate, based at least in part on the comparison, a confidence score indicative of a correlation between the textual work and at least one of the pre-determined non-machine generated works.

19. The computing device of claim 15, further comprising configuring the processor to search an index of a database of the pre-determined shape descriptors of sample textual works for pre-determined non-machine generated works having substantially a same shape as the shape of the textual work.

20. The computing device of claim 15, in which the plot of the N-gram in an N-dimensional space identifies a frequency of word occurrence in each sentence in the textual work.

21. The computing device of claim 15, in which the generation of the 2-dimensional plot includes plotting points based on a frequency of word occurrence in each sentence in the textual work and inverse similarity to a center of mass of the N-dimensional plot.

22. The computing device of claim 15, further comprising configuring the processor to apply one or more grammatical rules to each sentence of the textual work, and to compute a score for the textual work based on the application of the one or more grammatical rules.