# US 2009/0222389 A1 · Change analysis system, method and program

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2009/0222389 A1
- Document ID: US20090222389A1
- Status: Published application
- Kind: A1
- Application number: 12/372,545
- Filed: 2009-02-17
- Published: 2009-09-03
- Earliest claimed: 2008-02-29
- Assignees: International Business Machines Corporation
- Inventors: SHOHEI HIDO; TSUYOSHI IDE; HISASHI KASHIMA; HARUNOBU KUBO; HIROFUMI MATSUZAWA
- Web page: https://patentgrep.com/patent/US20090222389A1

## CPC

- G06N 20/00: Machine learning
- G06F 18/2113: by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation
- G06N 20/10: using kernel methods, e.g. support vector machines [SVM]
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

Different virtual labels, for example, like +1 and −1, are assigned to two data sets. A change analysis problem for the two data sets is reduced to a supervised learning problem by using the virtual labels. Specifically, a classifier such as logical regression, decision tree and SVM is prepared and is trained by use of a data set obtained by merging the two data sets assigned the virtual labels. A feature selection function of the resultant classifier is used to rank and output both every attribute contributing to classification and its contribution rate.

## Claims

### Claim 1

A change analysis system for performing, by means of a computer function, a change analysis on two data sets to be compared to find a change therebetween, the change analysis system comprising:
a virtual labeling section for preparing a labeled data set by assigning a first virtual label to each data sample in a first data set, and for assigning a second virtual label to each data sample in a second data set, the second virtual label being different from the first virtual label;
a classifier for classifying the data samples in the labeled data set by use of the virtual labels; and
a change analysis section for training the classifier with the labeled data set and outputting a contribution rate of every attribute contributing to classification of the classifier.

### Claim 2 (depends on 1)

The system according to claim 1, wherein the classifier is a logistic regression.

### Claim 3 (depends on 1)

The system according to claim 1, wherein the classifier is support vector machine.

### Claim 4 (depends on 1)

The system according to claim 1, wherein the classifier is decision tree.

### Claim 5 (depends on 1)

The system according to claim 1, wherein the first and second virtual labels are +1 and −1, respectively.

### Claim 6

A change analysis system for performing, by means of a computer function, a change analysis on two data sets to be compared to find a change therebetween, the change analysis system comprising:
a virtual labeling section for preparing a labeled data set by assigning a first virtual label to each data sample in a first data set, and for assigning a second virtual label to each data sample in a second data set, the second virtual label being different from the first virtual label;
a change detection section for detecting a significant change in the labeled data set;
a classifier for classifying the data samples in the labeled data set by use of the virtual labels; and
a change analysis section for training the classifier with the labeled data set and outputting a contribution rate of every attribute contributing to classification of the classifier, in response to an event in which the change detection section detects the significant change.

### Claim 7 (depends on 6)

The system according to claim 6, wherein the classifier is a logistic regression.

### Claim 8 (depends on 6)

The system according to claim 6, wherein the classifier is a support vector machine.

### Claim 9 (depends on 6)

The system according to claim 6, wherein the classifier is a decision tree.

### Claim 10 (depends on 6)

The system according to claim 6, wherein the first and second virtual labels are +1 and −1, respectively.

### Claim 11

A method for performing, by means of a computer function, a change analysis on two data sets to be compared to find a change therebetween, the change analysis method comprising the steps of:
preparing a labeled data set by assigning a first virtual label to each data sample in a first data set, and assigning a second virtual label to each data sample in a second data set, the second virtual label being different from the first virtual label;
detecting a significant change in the labeled data set; and
training the classifier with the labeled data set and outputting a contribution rate of every attribute contributing to classification of the classifier, in response to an event in which the change detection section detects the significant change.

### Claim 12 (depends on 11)

The method according to claim 11, wherein the classifier is logistic regression.

### Claim 13 (depends on 11)

The method according to claim 11, wherein the classifier is support vector machine.

### Claim 14 (depends on 11)

The method according to claim 11, wherein the classifier is a decision tree.

### Claim 15 (depends on 11)

The method according to claim 11, wherein the first and second virtual labels are +1 and −1, respectively.

### Claim 16

A change analysis program for performing, by means of a computer function, a change analysis on two data sets to be compared to find a change therebetween, the change analysis program causing the computer to execute the steps of:
preparing a labeled data set by assigning a first virtual label to each data sample in a first data set, and assigning a second virtual label to each data sample in a second data set, the second virtual label being different from the first virtual label;
detecting a significant change in the labeled data set; and
training the classifier with the labeled data set and outputting a contribution rate of every attribute contributing to classification of the classifier, in response to an event in which the change detection section detects the significant change.

### Claim 17 (depends on 16)

The program according to claim 16, wherein the classifier is a logistic regression.

### Claim 18 (depends on 16)

The program according to claim 16, wherein the classifier is a support vector machine.

### Claim 19 (depends on 16)

The program according to claim 16, wherein the classifier is a decision tree.

### Claim 20 (depends on 16)

The program according to claim 16, wherein the first and second virtual labels are +1 and −1, respectively.
