System and method for automatic electronic document identification

A method of automatically identifying an electronic publication, said method comprising:

• receiving from an electronic device said electronic publication local to the electronic device;
• extracting text from said electronic publication;
• determining, using circuitry, a first signature of said electronic publication based on said text of said electronic publication by calculating a frequency of a term in said text of said electronic document;
• comparing said first signature with each of a plurality of signatures of a library, said comparing being performed by determining whether an approximation of a cosine similarity value computed using Hamming distance function between said first signature and a second signature of said plurality of signatures in the signature library is greater than a predetermined threshold to determine a match;
• in response to said match of said first signature with said second signature of said plurality of signatures, determining an identifier of said electronic publication, the identifier being different from the first signature and the second signature;
• transmitting to said electronic device said identifier of said electronic publication over a communication channel;
• transmitting to said electronic device a recommendation based on the identifier of said electronic document, the recommendation being specific to the electronic document and different from the electronic document;
• and
• updating a data store based on said identifier of said electronic publication.

A system for automatically identifying an electronic publication, said system comprising:

• circuitry configured to:
• receive from an electronic device said electronic publication local to the electronic device;
• extract text from said electronic publication;
• generate a first signature of said electronic publication based on said text of said electronic publication by calculating a frequency of a term in said text of said electronic document;
• determine a match of said first signature of said electronic publication with a second signature of a plurality of signatures, said match being determined by determining whether an approximation of a cosine similarity value computed using Hamming distance function between said first signature and said second signature is greater than a predetermined threshold;
• determine an identifier corresponding to said electronic publication based on said match of said first signature of said electronic publication and said second signature of said plurality of signatures, the identifier being different from the first signature and the second signature;
• transmit to said electronic device said identifier of said electronic publication over a communication channel;
• transmit to said electronic device a recommendation based on the identifier of said electronic document, the recommendation being specific to the electronic document and different from the electronic document;
• and
• update a data store based on said identifier of said electronic publication.