Comparing text pages using image features based on word positions

A non-transitory computer-readable storage medium storing executable computer program instructions for generating a signature for a page of text, wherein the signature serves as an identifier of the text page, the instructions performing steps comprising:

• determining positions of a plurality of words in the text page;
• for a first word of the plurality of words having a first position in the text page, determining positions of a plurality of second words in the text page relative to the first word position;
• generating a signature value that describes the second word positions relative to the first word position;
• generating additional signature values for the text page, each signature value describing positions of other words in the text page relative to a word in the text page for which the signature value is being generated;
• comparing a first set of signature values for the text page to a second set of signature values for a second text page, wherein the first set of signature values comprises the signature value that describes the second word positions relative to the first word position and the additional signature values;
• and
• generating a measure of similarity that describes a result of the comparison.

A computer system for generating a signature for a page of text, wherein the signature serves as an identifier of the text page, comprising:

• at least one non-transitory computer-readable storage medium storing executable computer program instructions comprising instructions for:
• determining positions of a plurality of words in the text page;
• for a first word of the plurality of words having a first position in the text page, determining positions of a plurality of second words in the text page relative to the first word position;
• generating a signature value that describes the second word positions relative to the first word position;
• generating additional signature values for the text page, each signature value describing positions of other words in the text page relative to a word in the text page for which the signature value is being generated;
• comparing a first set of signature values for the text page to a second set of signature values for a second text page, wherein the first set of signature values comprises the signature value that describes the second word positions relative to the first word position and the additional signature values;
• and
• generating a measure of similarity that describes a result of the comparison;
• and
• a processor for executing the computer program instructions.