Data processing apparatus, data processing method and semiconductor manufacturing apparatus

Patent metadata

Field	Value
Publication number	US-11531848-B2
Application number	US-201916959093-A
Country	US
Kind code	B2
Filing date	Jul 26, 2019
Priority date	Jul 26, 2019
Publication date	Dec 20, 2022
Grant date	Dec 20, 2022

Abstract

A data processing apparatus in which a trade-off between over-learning prevention and calculation load prevention is eliminated when creating a model formula is provided. The data processing apparatus includes: a recording unit that records electronic data; and a computing unit that performs computing using the electronic data, in which the computing unit includes a feature amount selection unit used for computing, and the feature amount selection unit performs feature amount selection including: a first step (S101) of ranking feature amounts and rearranging the feature amounts from top; a second step (S103) of creating a plurality of data groups using only a part of the feature amounts according to the order; a third step (S104) of calculating a value that is an index for evaluating prediction performance of a regression or classification problem using each of the data groups using only a part of the feature amounts; a fourth step (S105) of deleting feature amounts based on the calculated prediction performance index; and a fifth step (S106) of updating the order of the feature amounts, which are feature amounts other than the deleted feature amount, using the prediction performance index, in which the second step to the fifth steps are iterated (S102) until an optimal value of the prediction performance index calculated in the third step is no longer updated.

First claim

The invention claimed is:

1. A data processing apparatus that obtains a prediction model using a feature amount, the data processing apparatus comprising:
a computing device configured to execute
a first step of rearranging ranked N first feature amounts in an order from first to N-th, provided that N is a natural number,
a second step of creating N first data groups,
a third step of obtaining a first evaluation index value for evaluating prediction performance of each of the first data groups,
a fourth step of deleting a part of the first feature amounts based on the first evaluation index values,
a fifth step of updating an order of second feature amounts, which are feature amounts other than the feature amounts deleted from the first feature amounts, using the first evaluation index values,
a sixth step of creating second data groups whose number is the same as the number of the second feature amounts,
a seventh step of obtaining a second evaluation index value for evaluating prediction performance of each of the second data groups, and
an eighth step of obtaining a prediction model using the second feature amounts when a minimum value of the first evaluation index values is the same as a minimum value of the second evaluation index values, or deleting a part of the second feature amounts based on the second evaluation index values, and updating an order of third feature amounts, which are feature amounts other than the feature amounts deleted from the second feature amounts using the second evaluation index values when the minimum value of the first evaluation index values is different from the minimum value of the second evaluation index values, wherein
an N-th first data group has the first to N-th feature amounts of the first feature amounts, and
an M-th second data group has the first to M-th feature amounts of the second feature amounts, provided that M is the number of the second feature amounts.

2. The data processing apparatus according to claim 1 , wherein
the first evaluation index value is a value obtained using Akaike Information Criteria (AIC).

3. The data processing apparatus according to claim 1 , wherein
in the fifth step, the order of the second feature amounts is updated using a value obtained by subtracting the first evaluation index value corresponding to an (N−1)th first data group from the first evaluation index value corresponding to the N-th first data group.

4. A semiconductor manufacturing apparatus in which a processing result is predicted by a prediction model obtained using a feature amount, the semiconductor manufacturing apparatus comprising:
a control device configured to execute
a first step of rearranging ranked N first feature amounts in an order from first to N-th, provided that N is a natural number,
a second step of creating N first data groups, a third step of obtaining a first evaluation index value for evaluating prediction performance of each of the first data groups,
a fourth step of deleting a part of the first feature amounts based on the first evaluation index values,
a fifth step of updating an order of second feature amounts, which are feature amounts other than the feature amounts deleted from the first feature amounts, using the first evaluation index values,
a sixth step of creating second data groups whose number is the same as the number of the second feature amounts,
a seventh step of obtaining a second evaluation index value for evaluating prediction performance of each of the second data groups, and
an eighth step of obtaining a prediction model using the second feature amounts when a minimum value of the first evaluation index values is the same as a minimum value of the second evaluation index values, or deleting a part of the second feature amounts based on the second evaluation index values, and updating an order of third feature amounts, which are feature amounts other than the feature amounts deleted from the second feature amounts, using the second evaluation index values when the minimum value of the first evaluation index values is different from the minimum value of the second evaluation index values, wherein
an N-th first data group has the first to N-th feature amounts of the first feature amounts, and
an M-th second data group has the first to M-th feature amounts of the second feature amounts, provided that M is the number of the second feature amounts.

5. The semiconductor manufacturing apparatus according to claim 4 , wherein
the first evaluation index value is a value obtained using Akaike Information Criteria (AIC).

6. The semiconductor manufacturing apparatus according to claim 4 , wherein
in the fifth step, the order of the second feature amounts is updated using a value obtained by subtracting the first evaluation index value corresponding to an (N−1)th first data group from the first evaluation index value corresponding to the N-th first data group.

7. The semiconductor manufacturing apparatus according to claim 4 , wherein
the prediction model predicts a plasma processing result of a plasma processing apparatus.

8. A data processing method that obtains a prediction model using a feature amount, the data processing method comprising:
a first step of rearranging ranked N first feature amounts in an order from first to N-th, provided that N is a natural number;
a second step of creating N first data groups;
a third step of obtaining a first evaluation index value for evaluating prediction performance of each of the first data groups;
a fourth step of deleting a part of the first feature amounts based on the first evaluation index values;
a fifth step of updating an order of second feature amounts, which are feature amounts other than the feature amounts deleted from the first feature amounts, using the first evaluation index values;
a sixth step of creating second data groups whose number is the same as the number of the second feature amounts;
a seventh step of obtaining a second evaluation index value for evaluating prediction performance of each of the second data groups; and
an eighth step of obtaining a prediction model using the second feature amounts when a minimum value of the first evaluation index values is the same as a minimum value of the second evaluation index values, or deleting a part of the second feature amounts based on the second evaluation index values, and updating an order of third feature amounts, which are feature amounts other than the feature amounts deleted from the second feature amounts, using the second evaluation index values when the minimum value of the first evaluation index values is different from the minimum value of the second evaluation index values, wherein
an N-th first data group has the first to N-th feature amounts of the first feature amounts, and
an M-th second data group has the first to M-th feature amounts of the second feature amounts, provided that M is the number of the second feature amounts.

9. The data processing method according to claim 8 , wherein
the first evaluation index value is a value obtained using Akaike Information Criteria (AIC).

10. The data processing method according to claim 8 , wherein
in the fifth step, the order of the second feature amounts is updated using a value obtained by subtracting the first evaluation index value corresponding to an (N−1)th first data group from the first evaluation index value corresponding to the N-th first data group.

Assignees

• Hitachi High Tech Corp

Inventors

• Ishiguro Masaki

Classifications

• G06F18/217

  Validation; Performance evaluation; Active pattern learning techniques · CPC title

• G06F18/2113

  by ranking or filtering the set of features, e.g. using a measure of variance or of feature cross-correlation · CPC title

• G06F2119/18

  Manufacturability analysis or optimisation for manufacturability · CPC title

• G06N20/00

  Machine learning · CPC title

• Y02P90/02

  Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS] · CPC title