Data processing: database and file management or data structures – Database design – Data structure types
Reexamination Certificate
2001-01-19
2003-12-23
Shah, Sanjiv (Department: 2176)
Data processing: database and file management or data structures
Database design
Data structure types
C707S793000, C707S793000
Reexamination Certificate
active
06668256
ABSTRACT:
BACKGROUND OF THE INVENTION
A general problem that arises in many systems for automating the handling of documents is the need to assign a particular document to a class or category dependent on the subject matter discussed in its content.
Most commercial systems for achieving this rely on a human creating lists of terms which are used to categorize the document. That is, if the terms appear in a document, then this fact is taken as evidence that the document should be assigned to a particular category. Such approaches are inaccurate and require a lot of manual work in setting up and maintaining the term lists and associated boolean operator combinations.
There is need for accurate, self-maintaining and automatically created categorization systems.
Although there have been attempts to create self structuring in other fields these have usually been at the expense of unimodality and have required adaptation algorithms of very high computational loads such as simulated annealing to remove even the simplest redundancies. See Qiuzhen Xue, Yu Hen Hu & Paul Milenkovic, Analysis of the hidden units of the multi-layer preceptron and its application in acoustic-to-articulatory mapping. Proc. ICASSP90 April 1990. Other singular value decomposition (SVD) approaches have been applied but have only removed a subset of the redundant terms. See Schetzen, The Volterra and Weiner Theories of Non-linear systems New York, N.Y.: John Wiley 1980. These methods utilise SVD to reduce the recognition space to a more compact form by use of the singular values.
REFERENCES
(1) P. Rayner & M. R. Lynch, a new connectionist model based on a non-linear adaptive filter proc. ICASSP89 April 1989
(2) M. J. D. Powell, Radial Basis Function approximations to polynomials Proc. Department of applied mathematics and theorectical Physics.
(3) A. Ivankhnenko, Heuristic self-organisation problems of engineering cybernetics automatica. Vol 6. 1970, pp.207-209
(4) Qiuzhen Xue, Yu Hen Hu & Paul Milenkovic, Analysis of the hidden units of the multi-layer preceptron and its application in acoustic-to-articulatory mapping. Proc. ICASSP90 April 1990
(5) Schetzen, The Volterra and Weiner Theories of Non-linear systems New York, N.Y.: John Wiley 1980.
(6) S. Haykin, Adaptive Filter Theroy Englewood Cliffs, N.J.: Prentice-Hall 1986
(7) V. Klema & A. Laub, The singular Valve Decomposition: Its computation and some applications. IEEE Trans. AC,vol. AC-25, No 2. April 1980, pp 164-176
SUMMARY OF THE INVENTION
A method and apparatus for document categorization are described. In one embodiment, the method comprises automatically selecting one or more discriminant term combinations and using the one or more discriminant term combinations for document categorization.
REFERENCES:
patent: 6006221 (1999-12-01), Liddy et al.
patent: 6137911 (2000-10-01), Zhilyaev
patent: 6477524 (2002-11-01), Taskiran et al.
patent: 6523026 (2003-02-01), Gillis
Autonomy, Technology White Paper, Sep. 1997, pp. 1-17.
Autonomy, Technology White Paper, Nov. 2000, pp. 1-9.
Autonomy—Adding Intelligence to the XML, Nov. 2000, pp. 1-3.
Autonomy, First to Fully Automate XML Tagging, Jun. 1998, pp. 1-2.
P.J. Rayner and M.R. Lynch, ICASSP 90 volume 2 S2VA, Speech Processing 2 VLSI Audio and Electroacoustics, Complexity Reduction in Volterra Connectionist Modelling by Consideration of Output Mapping, Apr. 3-6, 1990, pp. 885-888.
M.R. Lynch, P.J. Rayner and S.B. Holden, ICASSP 91 vol. 3D, Digital Signal Processing, Removal of Degeneracy in Adaptive Volterra Networks by Dynamic Structuring, May 14-17, 1991, pp. 2069-2072.
Haykin, S.,Adaptive Filter Theory, Englewood Cliffs, N.J.: Prentice-Hall, 1986, Chapter 5 and 8.
Ivankhnenko, A. “Heuristic Self-Organization Problems of Engineering Cybernetics Automatica,” vol. 6, Mar., 1970, pp. 207-219.
Klema, V., et al., “The Singular Valve Decomposition: Its Computation and Some Applications,” IEEE Trans. AC, vol. AC-25, No. 2, Apr. 1980, pp. 164-176.
Powell, M.J.D.,Radial Basis Function Approximations to Polynomials, Proc. Department of Applied Mathematics and Theoretical Physics,Numerical Analysis, 1987.
Rayner, P. et al., “A New Connectionist Model Based on a Non-Linear Adaptive Filer Process,” ICASSP89, Apr., 1989.
Schetzen, Martin,The Volterra and Weiner Theories on Non-Linear Systems, New York, New York: John Wiley, 1980, Chapter 5.
Xue, Q., et al., “Analysis of the Hidden Units of the Multi-Layer Preceptron and its Application in Acoustic-To-Articulatory Mapping,” Proc. ICASSP90, Apr. 1990.
Autonomy Corporation Ltd
Blakely , Sokoloff, Taylor & Zafman LLP
Shah Sanjiv
LandOfFree
Algorithm for automatic selection of discriminant term... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Algorithm for automatic selection of discriminant term..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Algorithm for automatic selection of discriminant term... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3174068