Text analysis devices, articles of manufacture, and text analysis methods

• US 8,996,993 B2
• Filed: 09/15/2006
• Issued: 03/31/2015
• Est. Priority Date: 09/15/2006
• Status: Active Grant

• Current Assignee: Battelle Memorial Institute
• Original Assignee: Battelle Memorial Institute
• Inventors: Turner, Alan E., Hetzler, Elizabeth G., Nakamura, Grant C.
• Primary Examiner(s): Baderman, Scott
• Assistant Examiner(s): AMIN, MUSTAFA A
• Application Number:
  US11/522,177
• Publication Number:
  US 20080071762A1
• Time in Patent Office: 3,119 Days
• Field of Search: 715/256
• US Class Current: 715/256
• CPC Class Codes: G06F 16/313 Selection or weighting of terms for indexing
• CPC Class Codes: G06F 16/3347 using vector based model

Abstract

Text analysis devices, articles of manufacture, and text analysis methods are described according to some aspects. In one aspect, a text analysis device includes a display configured to depict visible images, and processing circuitry coupled with the display and wherein the processing circuitry is configured to access a first vector of a text item and which comprises a plurality of components, to access a second vector of the text item and which comprises a plurality of components, to weight the components of the first vector providing a plurality of weighted values, to weight the components of the second vector providing a plurality of weighted values, and to combine the weighted values of the first vector with the weighted values of the second vector to provide a third vector.

Claims

1. A text analysis method comprising:

using processing circuitry, generating a first representation of a text item using a first measurement basis;

using the processing circuitry, generating a second representation of the text item using a second measurement basis different than the first measurement basis, wherein the second representation is different than the first representation;

using the processing circuitry, analyzing the text item using the first representation and the second representation;

using the processing circuitry, steering the generating of at least one of the first and second representations according to a perspective of interest of a user;

using the processing circuitry, accessing least one text pattern of interest to the user;

using the processing circuitry, generating one of the first measurement basis and the second measurement basis using the at least one text pattern of interest to the user; and

wherein the generating the one of the first measurement basis and the second measurement basis comprises generating the one of the first measurement basis and the second measurement basis to comprise associations of a plurality of measurement features in the form of text patterns with a plurality of dimension anchors which comprise different topics of textual content.

2. The method of claim 1 further comprising combining the first and second representations before the analyzing.

3. The method of claim 1 wherein the first and second representations comprise vectors.

4. The method of claim 3 further comprising weighting a plurality of components of the vectors.

5. The method of claim 4 further comprising concatenating the vectors before the analyzing.

6. The method of claim 1 further comprising weighting a plurality of values of at least one of the first and second representations before the analyzing.

7. The method of claim 1 wherein the steering comprises guiding the generatings to include information regarding the at least one text pattern of interest to the user in respective ones of the first and second representations.

8. The method of claim 1 further comprising displaying information regarding the text item in a visualization using the analysis of the first and second representations.

9. The method of claim 8 further comprising accessing user input, generating a third representation of the text item responsive to the user input, and displaying information regarding the text item in another visualization using the third representation.

10. The method of claim 1 wherein the analyzing comprises associating the text item with another text item using the first and second representations.

11. The method of claim 1 wherein the first and second representations individually comprise a mathematical signature comprising a plurality of values corresponding to the dimension anchors.

12. The method of claim 1 wherein the steering comprises steering according to the at least one text pattern of interest to the user.

13. The method of claim 1 wherein the generating the one of the first measurement basis and the second measurement basis comprises generating the one of the first measurement basis and the second measurement basis wherein the at least one text pattern comprises one of the measurement features and one of the dimension anchors.

14. The method of claim 1 further comprising generating the other of the first measurement basis and the second measurement basis, and wherein the accessing comprises accessing a plurality of text patterns of interest to the user, and wherein the generating the first measurement basis and the generating the second measurement basis comprise generating using respective different ones of the text patterns of interest to the user.