US 2003/0191726 A1 · Machine decisions based on preferential voting techniques

Abstract

A method and apparatus for computing an overall or aggregate decision based on intermediate decisions as to which of a set of alternatives best characterize an object. The alternatives are partitioned into at least two series of preferences corresponding to at least two intermediate rankings. Various embodiments may base the intermediate rankings on: a machine learning technique; a decision tree; a belief network; a neural network; a static model; a program; or an evolutionary training method. Based on the preferences, a weak alternative is selected and removed from the series. The selection of the weak alternative may include identifying which alternatives lose pairwise to the other alternatives, are excluded from the first preferences, are included in the last preferences, or have a lowest average preference ranking. The selecting and removing continue until the remaining alternatives are the aggregate decision. Various embodiments may be applied to classification problems, prediction problems or selection problems.

Drawings

Drawings are temporarily unavailable. Open the PDF at USPTO

Drawings

Claims30

Dependent claims

A method of computing an aggregate decision about an object, the method comprising:

obtaining at least two intermediate rankings, each of which ranks at least three alternatives with respect to the object;
initializing at least two series of preferences, wherein each series corresponds to one of the intermediate rankings, each series is ordered from a first preference to a last preference, each series initially groups each alternative into one of the preferences selected based on the corresponding intermediate ranking, and each preference includes at least one of the alternatives;
selecting, based on the preferences, at least one of the alternatives as a weak alternative;
removing at least one weak alternative from each series;
then
re-grouping each series into preferences based on the intermediate ranking corresponding to the series;
deciding, when each series contains less than a first threshold number of the alternatives, that the aggregate decision includes the alternatives remaining in the series;
and
repeating selecting, removing, re-grouping, and deciding until an aggregate decision is made.

6 more elements

A computer readable media containing instructions that, when the instructions are read and executed by a system, control a process comprising:

obtaining at least two intermediate rankings, each of which ranks at least three alternatives with respect to the object;
initializing least two series of preferences, wherein each series corresponds to one of the intermediate rankings, each series is ordered from a first preference to a last preference, each series initially groups each alternative into one of the preferences selected based on the corresponding intermediate ranking, and each preference includes at least one of the alternatives;
selecting, based on the preferences, at least one of the alternatives as a weak alternative;
removing at least one weak alternative from each series;
then
re-grouping each series into preferences based on the intermediate ranking corresponding to the series;
deciding, when each series contains less than a first threshold number of the alternatives, that the aggregate decision includes the alternatives remaining in the series;
and
repeating selecting, removing, re-grouping, and deciding until an aggregate decision is made.

6 more elements

Dependencies are inferred from the claim text.

Description10,563 words

Summary1,943 words

#1. Field of the Invention

#The invention relates generally to decision making and, alternatively, to learning, using a computing machine.

#2. Discussion of the Related Art

Brief description of the drawings8 figures

#The drawing illustrates technologies related to the invention, shows example embodiments of the invention, and gives examples of using the invention. The objects, features and advantages of the invention will become more apparent to those skilled in the art from the following detailed description, when read in conjunction with the accompanying drawing, wherein:

#FIG. 1 shows the basic steps used, in some embodiments of the invention, to compute an aggregate decision about an object;

#FIG. 2 shows an example of how, in some embodiments of the invention, machine learning models can be trained, thereby producing evaluators;

Detailed description8,304 words

#The descriptions and discussions herein illustrate technologies related to the invention, show examples of the invention and give examples of using the invention. Known methods, procedures, systems, circuits or components may be discussed without giving details, so as to avoid obscuring the principles of the invention. On the other hand, numerous details of specific examples of the invention may be described, even though such details may not apply to other embodiments of the invention. Details are included and omitted so as to better explain the invention and so as to aid in understanding the invention.

#The invention is not to be understood as being limited to or defined by what is discussed herein; rather, the invention may be practiced without the specific details described herein. One skilled in the art will realize that numerous modifications, variations, selections among alternatives, changes in form, and improvements can be made without departing from the principles, intention or legal scope of the invention.

#Some descriptions herein use abstract or general terms including among others: symbol, number, bit, procedure, step or block. Those skilled in the art use such terms as a convenient nomenclature for data and operations within a computer, digital device or electromechanical system. Such data and operations are represented by physical properties of actual objects including electronic voltage, magnetic field, optical reflectivity or other properties. Similarly, perceptive or mental terms including, among others, detect, sense, recall, present, compare, process, manipulate, analyze, decide or determine, may also refer to such data, such operations or to manipulations of such physical properties.

Facts

Application

10/116,835Filed 5 Apr 2002

Published

9 Oct 2003

Term ends

Not estimatedThe source records don’t support an estimate.

Classification4 codes

G06N 3/086
Learning methods: using evolutionary algorithms, e.g. genetic algorithms or genetic programming
G06F 18/254
Fusion techniques: of classification results, e.g. of results related to same input data
G06N
AI and machine learning
G06F
Computing and data processing

Timeline

5 Apr 2002Application filedUS 10/116,835
15 Oct 2002Assigned to Hewlett-Packard Companyassignment of assignor's interest
18 Jun 2003Assigned to Hewlett-Packard Development Company, L.P.assignment of assignor's interest
9 Oct 2003Published
9 Nov 2015Assigned to Hewlett Packard Enterprise Development LPassignment of assignor's interest
9 Jun 2017Assigned to Entit Software LLCassignment of assignor's interest
11 Oct 2017Assigned to Jpmorgan Chase Bank, N.A.security interest
11 Oct 2017Assigned to Jpmorgan Chase Bank, N.A.security interest
8 Aug 2019Assigned to Micro Focus LLCchange of name
2 Feb 2023Assigned to Micro Focus LLC (F/K/A Entit Software LLC), Borland Software Corporation, Micro Focus (US), Inc., Serena Software, Inc, Attachmate Corporation, Micro Focus Software Inc. (F/K/A Novell, Inc.), Netiq Corporationrelease of security interest reel/frame 044183/0718
2 Feb 2023Assigned to Micro Focus LLC (F/K/A Entit Software LLC)release of security interest reel/frame 044183/0577

Machine decisions based on preferential voting techniques

Life of this patent

Machine decisions based on preferential voting techniques

Life of this patent

Life of this patent

Abstract

Drawings

Drawings

Claims30

Description10,563 words

Drawings

Classification4 codes

Life of this patent

Description10,563 words