US2013024185A1PendingUtilityA1

Automatic Dynamic Contextual Date Entry Completion

Assignee: PARIKH PRASHANTPriority: Jan 21, 2005Filed: Sep 20, 2012Published: Jan 24, 2013
Est. expiryJan 21, 2025(expired)· nominal 20-yr term from priority
Inventors:Prashant Parikh
G06F 3/0237G06F 40/284G06F 40/274
53
PatentIndex Score
0
Cited by
0
References
0
Claims

Abstract

A method performed in a computer device having associated therewith a plurality of unstructured documents having words therein, the method involves accessing at least some of the plurality of unstructured documents, extracting a multiset of words, forming a matrix from the documents in which each word in the multiset is represented in a column and each document from which the words came is represented in a row, treating each document as a vector in a multidimensional Euclidean space, uniquely pairing the unique documents, measuring the similarity between the pairs as a cosine of the angle between vectors, comparing the cosines to a specified threshold to determine relatedness among the documents, and based upon the relatedness, when an input is received by the computer device representing a string of a threshold number of characters, the computer device will provide at least one word that would complete the character string.

Claims

exact text as granted — not AI-modified
1 . A method performed in a computer device having associated therewith a plurality of unstructured documents having words therein, the unstructured documents being stored in non-transient memory accessible to the computer device, the method comprising:
 accessing at least some of the plurality of unstructured documents;   extracting from each document making up the at least some of the plurality of unstructured documents, a multiset of words;   forming a matrix from the documents in which each word in the multiset is represented in a column and each document from which the words came is represented in a row;   treating each document as a vector in a multidimensional Euclidean space with the number of dimensions being equal to one of the number of columns or number of rows in the matrix;   uniquely pairing the unique documents making up the at least some of the plurality of unstructured documents;   measuring the similarity between the pairs of unique documents making up the at least some of the plurality of unstructured documents as a cosine of the angle between vectors;   comparing the cosines to a specified threshold to determine relatedness among the documents; and   based upon the relatedness, when an input is received by the computer device representing a string of a threshold number of characters, the computer device will provide at least one word that would complete the character string.

Join the waitlist — get patent alerts

Track US2013024185A1 — get alerts on status changes and closely related new filings.

We store only your email — no account needed. See our privacy policy.