Data processing apparatus and method for imaging time series data of multiple wavelengths
A data processing apparatus comprises processing circuitry configured to generate normalized data by normalizing time series data of multiple wavelengths, measured from a substrate, by using predetermined reference data; divide the normalized data into a plurality of regions for each predetermined time range and predetermined wavelength range; for each respective region of the plurality of regions, extract an outlier value in the respective region as a representative value; and convert the representative values from the plurality of regions into color data to generate image data from the color data.
1 . A data processing apparatus, comprising:
processing circuitry configured to
generate normalized data by normalizing time series data of multiple wavelengths, measured from a substrate, by using predetermined reference data;
divide the normalized data into a plurality of regions for each predetermined time range and predetermined wavelength range;
for each respective region of the plurality of regions;
calculate an average value of the normalized data included in the respective region; and
extract an outlier value, from the normalized data included in the respective region, that has a largest difference from the average value as a representative value for the respective region; and
convert the representative values from the plurality of regions into color data to generate image data from the color data.
2 . The data processing apparatus according to claim 1 , wherein the processing circuitry normalizes the time series data of the multiple wavelengths by using, as the predetermined reference data, time series data of the multiple wavelengths when the substrate serving as a reference is processed in a semiconductor manufacturing process.
3 . The data processing apparatus according to claim 1 , wherein the time series data of the multiple wavelengths is data measured by a light emission spectroscopy analyzer or a mass spectrometer during processing of the substrate in a semiconductor manufacturing process.
4 . The data processing apparatus according to claim 3 , wherein the processing circuitry is further configured to input the image data into a model, which is trained using a correspondence relationship with a processing result of processing the substrate, to infer the processing result of processing the substrate.
5 . The data processing apparatus according to claim 4 , wherein the processing circuitry infers that the substrate includes an error and a cause of the error of processing the substrate.
6 . The data processing apparatus according to claim 4 , wherein the processing circuitry is further configured to perform learning processing on the model such that an output, obtained when the image data generated based on the time series data of the multiple wavelengths measured during the processing of the substrate is input into the model, approaches the processing result of processing the substrate.
7 . A data processing system, comprising:
a plurality of the data processing apparatuses according to claim 4 ; and
a server apparatus configured to receive, from one or more data processing apparatuses of the plurality of data processing apparatuses, any of
the image data generated based on the time series data of the multiple wavelengths measured during processing of the substrate, or
the model trained using the correspondence relationship between the image data and the processing result of processing the substrate.
8 . A data processing system, comprising:
the data processing apparatus according to claim 4 ; and
a server apparatus including learning circuitry configured to perform learning processing on the model such that an output, obtained when image data generated based on the time series data of the multiple wavelengths measured during processing of the substrate is input into the model, approaches the processing result of processing the substrate.
9 . A data processing method, comprising:
generating normalized data by normalizing time series data of multiple wavelengths, measured from a substrate, by using predetermined reference data;
dividing the normalized data into a plurality of regions for each predetermined time range and predetermined wavelength range;
for each respective region of the plurality of regions:
calculating an average value of the normalized data included in the respective region; and
extracting an outlier value, from the normalized data included in the respective region, that has a largest difference from the average value as a representative value for the respective region; and
converting the representative values from the plurality of regions into color data to generate image data from the color data.
10 . The data processing method according to claim 9 , wherein in the time series data of the multiple wavelengths is normalized by using, as the predetermined reference data, time series data of the multiple wavelengths when the substrate serving as a reference is processed in a semiconductor manufacturing process.
11 . The data processing method according to claim 9 , wherein the time series data of the multiple wavelengths is data measured by a light emission spectroscopy analyzer or a mass spectrometer during processing of the substrate in a semiconductor manufacturing process.
12 . The data processing method according to claim 11 , further comprising inputting the image data into a model, which is trained using a correspondence relationship with a processing result of processing the substrate, to infer the processing result of processing the substrate.
13 . The data processing method according to claim 12 , wherein the processing result indicates whether an error occurred and a cause of the error of processing the substrate.
14 . The data processing method according to claim 12 , further comprising performing learning processing on the model such that an output, obtained when the image data generated based on the time series data of the multiple wavelengths measured during the processing of the substrate is input into the model, approaches the processing result of processing the substrate.
15 . A non-transitory computer readable medium storing computer executable instructions which, when executed by processing circuitry, cause the processing circuitry to:
generate normalized data by normalizing time series data of multiple wavelengths, measured from a substrate, by using predetermined reference data;
divide the normalized data into a plurality of regions for each predetermined time range and predetermined wavelength range;
for each respective region of the plurality of regions:
calculate an average value of the normalized data included in the respective region; and
extract an outlier value, from the normalized data included in the respective region, that has a largest difference from the average value as a representative value for the respective region; and
convert the representative values from the plurality of regions into color data to generate image data from the color data.
16 . The non-transitory computer readable medium according to claim 15 , wherein the processing circuitry is further caused to normalize the time series data of the multiple wavelengths by using, as the predetermined reference data, time series data of the multiple wavelengths when the substrate serving as a reference is processed in a semiconductor manufacturing process.
17 . The non-transitory computer readable medium according to claim 15 , wherein the time series data of the multiple wavelengths is data measured by a light emission spectroscopy analyzer or a mass spectrometer during processing of the substrate in a semiconductor manufacturing process.