# US 2019/0318266 A1 · Two-class classification method for predicting class to which specific item belongs, and computing device using same

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

- Number: US 2019/0318266 A1
- Document ID: US20190318266A1
- Status: Published application
- Kind: A1
- Application number: 16/348,455
- Filed: 2017-11-09
- Published: 2019-10-17
- Earliest claimed: 2016-11-09
- Assignees: BioInfra Life Science Inc.
- Applicants: Bioinfra Life Science Inc.
- Inventors: Chul Woo KIM; Yong Dai KIM; Kyung Nam KANG; Oh Ran KWON
- Web page: https://patentgrep.com/patent/US20190318266A1

## CPC

- G06N 20/20: Ensemble learning
- G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)}
- G06F 18/23: Clustering techniques
- G06F 18/24323: {Tree-organised classifiers}
- G06N 5/01: Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- H04L 7/00: Arrangements for synchronising receiver with transmitter {(synchronisation of generators of electric oscillations or pulses H03L7/00)}
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)
- H04L: TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION (arrangements common to telegraphic and telephonic communication H04M)

## Abstract

A computing device of the present invention estimates an unknown parameter β of a model formula when distributed sample data is acquired, wherein when an estimated quantity of β is acquired, a function g is estimated using a random forest model, such that when an estimated quantity of g is acquired, the estimated quantity of g and the estimated quantity of β are used so as estimate a function G as a prediction formula for new data corresponding to a specific item such that an estimated quantity of G is acquired, and thus new data xnew is received, thereby enabling the class of the specific item to be classified from the calculated value.

## Claims

### Claim 1

A method for 2-class classification to estimate a specific class to which a specific item belongs, comprising steps of:
(a) a computing device acquiring sample data (Y1, X1), . . . , (Yn, Xn) which is independently and identically distributed, wherein n is the number of the sample data, Xi=(1, Xi1, . . . , Xid)T∈X⊂Rd is a d-dimensional vector, Yi has or is manipulated to have a value of −1 or 1, and −1 and 1 are set as respectively corresponding to two of classes;
(b) the computing device, if the sample data is acquired, estimating or supporting another device to estimate an unknown parameter β of a model equation
G

(
x
)
=


[
ℙ

(
Y

X
)
]
=
ℙ

(
Y
=
1

X
)
=
exp

{
2

X
T

β
+
2

g

(
X
)
}
1
+
exp


{
2

X
T

β
+
2

g

(
X
)
}
,
and
1
2


log


ℙ

(
Y
=
1

X
)
ℙ

(
Y
=
-
1

X
)
=
X
T

β
+
g

(
X
)
wherein β=(β0, β1, . . . , βd), X is an independent variable, Y is a dependent variable, and the function g takes a form of a random forest model;
(c) the computing device, if an estimator {circumflex over (β)} of the unknown parameter β is acquired, estimating or supporting another device to estimate the function g by using the random forest model; and
(d) the computing device, if an estimator ĝ of the function g is acquired, estimating or supporting another device to estimate the function G as an estimating equation for new data corresponding to the specific item by using the estimator {circumflex over (β)} of the unknown parameter β and the estimator ĝ of the function g, to thereby acquire an estimator Ĝ of the function G.

### Claim 2 (depends on 1)

The method of claim 1, further comprising a step of:
(e) the computing device, if the estimator Ĝ of the function G is acquired and new data Xnew is inputted, calculating a value of Ĝ(Xnew) and determining or supporting another device to determine the specific class of the specific item by referring to the value of the calculated Ĝ(Xnew).

### Claim 3 (depends on 1)

The method of claim 1, wherein, at the step of (b), the estimator {circumflex over (β)} of the unknown parameter β is calculated by using
β
^
=

argmin
β

∑
i
=
1
N

L
(
y
i
,
X
i
T

β
)
=

argmin
β

∑
i
=
1
N


log

[
1
+
exp


(
-
2

y
i

X
i
T

β
)
]
,
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data,
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data, and
D is the number of independent variables including the j-th independent variable.

### Claim 4 (depends on 1)

The method of claim 1, wherein, at the step of (c), the estimator ĝ
g
^
=

argmin
g

∑
i
=
1
N

L
(
y
i
,
X
i
T

β
^
+
g

(
X
i
)
)
=

argmin
g

∑
i
=
1
N


log

[
1
+
exp


{
-
2

y
i

(
X
i
T

β
^
+
g

(
X
i
)
)
}
]
of the function g is calculated by using a negative slope and a tuning parameter λ≥0, and
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data, and
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data.

### Claim 5 (depends on 1)

The method of claim 1, wherein the step of (c) includes steps of:
(c1) the computing device calculating
r
i
=
2

y
i


•


exp


(
-
2

y
i


•


X
i
T

β
^
)
1
+
exp


(
-
2

y
i


•


X
i
T

β
^
)
,
i
=
1
,
…


…

..
,
n
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data,
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data, and
D is the number of independent variables including the j-th independent variable; and
(c2) the computing device, if ri is calculated, calculating a function RF(X) resulting from data {(ri,Xi)}i=1n fitted by the random forest model, and estimating or supporting another device to estimate the function g from g(X)=λ·RF(X) with respect to a tuning parameter λ.

### Claim 6 (depends on 4, 5)

The method of claim 4 or 5, wherein a value of the tuning parameter λ is 1.

### Claim 7 (depends on 4, 5)

The method of claim 4 or 5, wherein the tuning parameter λ is calculated by using a logistic regression model without intercept.

### Claim 8 (depends on 1)

The method of claim 1, wherein the two of the classes are set as respectively corresponding to being a patient of a specific disease and to not being the patient of the specific disease, and
each independent variable Xij of the sample data is (i) a physical quantity of a specific material included in a biological sample acquired from an individual patient or (ii) at least one demographical variable of the individual patient.

### Claim 9 (depends on 8)

The method of claim 8, wherein the specific disease is breast cancer or stomach cancer.

### Claim 10

A computing device for 2-class classification to estimate a specific class to which a specific item belongs, comprising:
a communication part for acquiring sample data (Y1, X1), . . . , (Yn, Xn) which is independently and identically distributed, wherein n is the number of the sample data, Xi=(1, Xi1, . . . , Xid)T∈X⊂Rd is a d-dimensional vector, Yi has or is manipulated to have a value of −1 or 1, and −1 and 1 are set as respectively corresponding to two of classes; and
a processor for, (i) if the sample data is acquired, estimating or supporting another device to estimate an unknown parameter β of a model equation
G

(
x
)
=


[
ℙ

(
Y

X
)
]
=
ℙ

(
Y
=
1

X
)
=
exp

{
2

X
T

β
+
2

g

(
X
)
}
1
+
exp


{
2

X
T

β
+
2

g

(
X
)
}
,
and
1
2


log


ℙ

(
Y
=
1

X
)
ℙ

(
Y
=
-
1

X
)
=
X
T

β
+
g

(
X
)
,
wherein β=(β0, β1, . . . , βd), X is an independent variable, Y is a dependent variable, and the function g takes a form of a random forest model, (ii) if an estimator {circumflex over (β)} of the unknown parameter β is acquired, estimating or supporting another device to estimate the function g by using the random forest model, and (iii) if an estimator ĝ of the function g is acquired, estimating or supporting another device to estimate the function G as an estimating equation for new data corresponding to the specific item by using the estimator {circumflex over (β)} of the unknown parameter β and the estimator ĝ of the function g, to thereby acquire an estimator Ĝ of the function G.

### Claim 11 (depends on 10)

The computing device of claim 10, further comprising a process of:
(iv) if the estimator Ĝ of the function G is acquired and new data Xnew is inputted, calculating a value of Ĝ(Xnew) and determining or supporting another device to determine the specific class of the specific item by referring to the value of the calculated Ĝ(Xnew).

### Claim 12 (depends on 10)

The computing device of claim 10, wherein, at the process of (i), the estimator {circumflex over (β)} of the unknown parameter β is calculated by using
β
^
=

argmin
β

∑
i
=
1
N

L
(
y
i
,
X
i
T

β
)
=

argmin
β

∑
i
=
1
N


log

[
1
+
exp


(
-
2

y
i

X
i
T

β
)
]
,
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data,
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data, and
D is the number of independent variables including the j-th independent variable.

### Claim 13 (depends on 10)

The computing device of claim 10, wherein, at the process of (ii), the estimator ĝ
g
^
=

argmin
g

∑
i
=
1
N

L
(
y
i
,
X
i
T

β
^
+
g

(
X
i
)
)
=

argmin
g

∑
i
=
1
N


log

[
1
+
exp


{
-
2

y
i

(
X
i
T

β
^
+
g

(
X
i
)
)
}
]
of the function g is calculated by using a negative slope and a tuning parameter λ≥0, and
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data, and
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data.

### Claim 14 (depends on 10)

The computing device of claim 10, wherein the process of (ii) includes steps of:
(ii1) calculating
r
i
=
2

y
i


•


exp


(
-
2

y
i


•


X
i
T

β
^
)
1
+
exp


(
-
2

y
i


•


X
i
T

β
^
)
,
i
=
1
,
…


…

..
,
n
wherein yi is an actually measured Y value of i-th sample data,
Xi=(1, Xi1, . . . , XiD)T is an actually measured X value of the i-th sample data,
Xij is a value of a j-th independent variable of the actually measured X value of the i-th sample data, and
D is the number of independent variables including the j-th independent variable; and
(ii2) if ri is calculated, calculating a function RF(X) resulting from data {(ri,Xi)}i=1n fitted by the random forest model, and estimating or supporting another device to estimate the function g from g(X)=λ·RF(X) with respect to a tuning parameter λ.

### Claim 15 (depends on 13, 14)

The computing device of claim 13 or 14, wherein a value of the tuning parameter λ is 1.

### Claim 16 (depends on 13, 14)

The computing device of claim 13 or 14, wherein the tuning parameter λ is calculated by using a logistic regression model without intercept.

### Claim 17 (depends on 10)

The computing device of claim 10, wherein the two of the classes are set as respectively corresponding to being a patient of a specific disease and to not being the patient of the specific disease, and
each independent variable Xij of the sample data is (i) a physical quantity of a specific material included in a biological sample acquired from an individual patient or (ii) at least one demographical variable of the individual patient.

### Claim 18 (depends on 17)

The computing device of claim 17, wherein the specific disease is breast cancer or stomach cancer.