Error correction method for computer systems and time-series data

Abstract

To provide a computer system for applying an optimal correction program among a plurality of correction programs for correcting errors in time series data processed by an inference system.SOLUTION: A computer system is connected to an inference system that performs inference using time series data, and holds correction program management information for managing a correction program that corrects errors in the time series data. Each of the plurality of correction programs managed by the correction program management information performs different correction processing. When an error in the time series data is detected, the computer system selects an applicable correction program, corrects the error in the time series data based on the selected correction program, performs an evaluation process that evaluates the influence on inference by an inference system using the corrected time series data, and sets a correction program selected based on a result of the evaluation of the influence on inference to enable execution by the inference system.SELECTED DRAWING: Figure 9

Description

Technical Field

[0001] The present invention relates to a system and method for correcting errors in time-series data in a system having an inference function for making inferences using time-series data.

Background Art

[0002] With the expansion of the application fields of machine learning technologies, machine learning technologies are being utilized to improve the efficiency of operations or enforce safe behavior by using work logs and worker behavior logs in factories and warehouses.

[0003] Many of the data used in machine learning, such as work logs and worker behavior logs in factories and warehouses, are time-series data. Errors in time-series data may occur due to sensor malfunctions, communication failures between sensors and the hubs that collect data, and worker carelessness, etc., such as partial or complete loss of time-series data, or the mixing in of abnormal data (data outside the assumed range or data with different units) into the time-series data.

[0004] Errors in time-series data cannot be comprehensively grasped during the design of the inference system and often occur after actual operation, so it is difficult to implement a mechanism for correcting time-series data in the inference system in advance.

[0005] To address errors in time-series data, it is also conceivable to stop the inference system and investigate the cause. However, stopping the inference system will result in stopping the collection of time-series data related to operations, leading to a longer period during which time-series data cannot be collected, and having a great impact on operations.

[0006] Therefore, in an inference system that uses time-series data, a technique for correcting time-series data collected while operating the inference system is required. In contrast, for example, the technique described in Patent Document 1 is known.

[0007] Patent Document 1 discloses a method for automatically filling in power values ​​collected at a factory if a gap occurs in the power values ​​collected, by changing the method of filling in the gap according to the length of the gap period, thereby minimizing the impact on operations. [Prior art documents] [Patent Documents]

[0008] [Patent Document 1] Japanese Patent Publication No. 2019-7882 [Overview of the Initiative] [Problems that the invention aims to solve]

[0009] In inference systems that perform inference using time-series data, failure to properly correct errors in the time-series data, such as missing or abnormal data, will affect the inference. Here, the impact on inference refers to the output of incorrect inference results and the cost required for inference. The correction method described in Patent Document 1 has the problem that it cannot be applied in environments where the type and content of errors in the time-series data are unknown in advance.

[0010] The present invention aims to apply the optimal correction method from among multiple error correction methods when correcting errors in time-series data detected in an inference system that performs inference using time-series data. [Means for solving the problem]

[0011] A representative example of the invention disclosed in this application is as follows: a computer system including at least one computer, connected to an inference system that performs inference using time-series data, and a correction program management information for managing a correction program that corrects errors in the time-series data. And, detection program management information for managing a detection program that detects errors in the aforementioned time-series data, Each of the multiple correction programs managed by the correction program management information maintains the above and executes a different correction process. Each of the multiple detection programs managed by the aforementioned detection program management information executes a process to detect a different error, and the identification information of the detection program and the identification information of the correction program are managed to correspond to each other. The aforementioned computer system, The detection program is executed, and it is determined whether or not an error has occurred in the time series data corresponding to the detection program. If it is determined that an error has occurred in the time series data corresponding to the executed detection program, the applicable correction program is the one associated with the detection program.The process involves selecting a correction program, correcting errors in the time series data based on the selected correction program, performing an evaluation process to assess the impact of the corrected time series data on the inference performed by the inference system, and setting the correction program to be selected based on the results of the evaluation of the impact on the inference so that the inference system can be executed. [Effects of the Invention]

[0012] According to the present invention, the optimal error correction method can be applied to the inference system from among multiple error correction methods. Other issues, configurations, and effects will be clarified by the following description of the embodiments. [Brief explanation of the drawing]

[0013] [Figure 1] This is a diagram showing an example of the system configuration of Example 1. [Figure 2] This figure shows an example of the data structure of element data stored in the time-series data database of Example 1. [Figure 3] This figure shows an example of the data structure of the error detection program management information in Example 1. [Figure 4] This figure shows an example of the data structure of the correction program management information in Example 1. [Figure 5] This figure shows an example of the data structure of the evaluation program management information in Example 1. [Figure 6] This figure shows an example of the data structure of the setting control information in Example 1. [Figure 7] This figure shows an example of the data structure of the similar data definition information in Example 1. [Figure 8] This figure shows an example of the data structure of the error reproduction program management information in Example 1. [Figure 9] This is a sequence diagram illustrating the flow of the correction program setting process performed in the system of Example 1. [Figure 10] This flowchart illustrates an example of the correction program selection process performed by the correction program evaluation system of Example 1. [Figure 11] It is a sequence diagram for explaining the flow of correction program setting processing executed in the system of Example 2. [Figure 12] It is a diagram showing an example of the screen presented by the terminal of Example 2.

Mode for Carrying Out the Invention

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the present invention is not to be construed as being limited to the description of the embodiments shown below. It will be easily understood by those skilled in the art that the specific configuration can be changed without departing from the spirit or gist of the present invention.

[0015] In the configuration of the invention described below, the same or similar configurations or functions are denoted by the same reference numerals, and redundant descriptions are omitted.

[0016] The notations such as "first", "second", "third", etc. in this specification and the like are for identifying components, and do not necessarily limit the number or order.

[0017] The positions, sizes, shapes, and ranges of each configuration shown in the drawings and the like may not represent the actual positions, sizes, shapes, and ranges in order to facilitate understanding of the invention. Therefore, in the present invention, it is not limited to the positions, sizes, shapes, and ranges disclosed in the drawings and the like.

Examples

[0018] FIG. 1 is a diagram showing a configuration example of the system of Example 1.

[0019] The system consists of a time-series data acquisition system 100, an inference system 110, a correction program evaluation system 120, a correction program setting support system 130, an error reproduction system 140, and a terminal 150. The time-series data acquisition system 100, the inference system 110, the correction program evaluation system 120, the correction program setting support system 130, the error reproduction system 140, and the terminal 150 are connected via a network 160. The network 160 can be, for example, a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, or a dedicated line.

[0020] Each of the time-series data acquisition system 100, inference system 110, correction program evaluation system 120, correction program setting support system 130, and error reproduction system 140 consists of at least one computer having a processor, memory, and network interface.

[0021] The databases and information managed by the system or device are implemented using, for example, source code management services such as GitHub (registered trademark, hereinafter the same), device drivers, file systems, and DBMS (Database Management Systems) such as relational databases and NoSQL.

[0022] The functional components of a system or device may be implemented using one or more computers, or they may be implemented using virtual machines and containers, etc.

[0023] The time-series data acquisition system 100 collects data including time and one or more metrics from information processing devices and sensors, and stores it as time-series data in the time-series data DB 101. Information processing devices include computers, smartphones, and robots. Sensors include environmental sensors such as temperature sensors and humidity sensors, and biosensors such as heart rate monitors and pedometers.

[0024] Possible communication methods between the time-series data acquisition system 100 and the information processing device and sensors include Zigbee (registered trademark, hereinafter the same), BLE (Bluetooth Low Energy) (BLE is a registered trademark, hereinafter the same), WiFi (registered trademark, hereinafter the same), cellular networks, IEEE 802.1 standard, Ant+, etc.

[0025] The inference system 110 performs inference using time-series data. The inference system 110 includes a data acquisition unit 111, a preprocessing unit 112, an inference unit 113, and a postprocessing unit 114.

[0026] The data acquisition unit 111 acquires time-series data from the time-series data collection system 100 and transmits it to the preprocessing unit 112. The preprocessing unit 112 performs preprocessing on the time-series data and converts it into time-series data that can be inferred. The preprocessing unit 112 transmits the converted time-series data to the inference unit 113. The inference unit 113 obtains an output (inference result) by inputting the converted time-series data into a model generated by machine learning. The inference unit 113 transmits the inference result to the postprocessing unit 114. The model is managed as information in a file format such as pickle. The postprocessing unit 114 converts the inference result into an output that can be understood by humans. The inference system 110 may have a display function that presents the output of the postprocessing unit 114.

[0027] The correction program evaluation system 120 detects errors in time-series data and evaluates the impact of correction of time-series data errors according to the correction program on inference. The correction program evaluation system 120 has a control unit 121 and also holds error detection program management information 122, correction program management information 123, and evaluation program management information 124.

[0028] In this embodiment, "errors in time-series data" refers to missing metrics in the elemental data that constitute the time-series data, column misalignment caused by missing metrics in the elemental data, the presence of elemental data that stores outliers, etc.

[0029] The correction program setting support system 130 assists in setting the correction program for the inference system 110. The correction program setting support system 130 has a correction program setting unit 131 and also holds setting control information 132.

[0030] The error reproduction system 140 extracts time-series data similar to the time-series data in which the error was detected, and uses the extracted time-series data to generate time-series data in which an error similar to the detected error is reproduced. The error reproduction system 140 has a similar data extraction unit 141 and an error reproduction unit 142, and also holds similar data definition information 143 and error reproduction program management information 144. The processing related to the error reproduction system 140 will be explained in Example 2.

[0031] Terminal 150 is a terminal used to perform various operations on the system and to refer to various information within the system.

[0032] The correction program evaluation system 120, the correction program setting support system 130, and the error reproduction system 140 may be implemented as a single system.

[0033] Figure 2 shows an example of the data structure of element data stored in the time-series data DB101 of Example 1.

[0034] Element data 200 includes time and one or more metrics. The present invention is not limited to the number of metrics included in element data 200.

[0035] Figure 3 shows an example of the data structure of the error detection program management information 122 in Example 1.

[0036] Error detection program management information 122 is information for managing error detection programs for detecting errors in time-series data, and stores entries including error detection program ID 301, error detection program 302, execution timing 303, and target 304. There is one entry for each error detection program. Note that the columns included in the entry are not limited to those mentioned above. It may not include any of the aforementioned columns, or it may include other columns.

[0037] Error detection program ID 301 is a column that stores the identification information of the error detection program. The identification information of the error detection program is also used as identification information for associating it with the correction program and the evaluation program.

[0038] Error detection program 302 is a column that stores the actual error detection program, a path indicating the location where the program is stored, and a link to access the location where the program is stored. The error detection program is a program implemented in a programming language such as Python.

[0039] The execution timing 303 is a column that stores the execution timing of the error detection program. For example, it stores the execution cycle, the number of element data included in the time series data, and the number of times the time series data has been received.

[0040] Target 304 is a column that stores the target of the error detection program. Figure 3 shows that the time-series data transmitted from the data acquisition unit 111 is the target of the program.

[0041] Figure 4 shows an example of the data structure of the correction program management information 123 in Example 1.

[0042] The correction program management information 123 is information for managing correction programs for correcting errors in time-series data, and stores entries including error detection program ID 401 and correction program 402. There is one entry for each correction program. Note that the columns included in the entry are not limited to those mentioned above. It may not include any of the aforementioned columns, or it may include other columns.

[0043] Error detection program ID 401 is the same column as error detection program ID 301. Correction program 402 is a column that stores the actual correction program, the path indicating the location where the actual program is stored, and a link to access the location where the actual program is stored.

[0044] The correction program is a program implemented in a programming language such as Python. For example, the correction program may insert a zero into missing metrics in element data, or insert a value obtained from the metrics of preceding and succeeding element data in the time series. Each correction program managed in the correction program management information 123 will perform a different correction process.

[0045] The correspondence between time-series data errors and correction programs may be pre-configured by the user, or it may be configured based on the history of actual correction operations.

[0046] Figure 5 shows an example of the data structure of the evaluation program management information 124 in Example 1.

[0047] The evaluation program management information 124 is information for managing evaluation programs used to evaluate the impact of corrections on inference, and stores entries including error detection program ID 501 and evaluation program 502. There is one entry for each correction program. Note that the columns included in the entries are not limited to those mentioned above. It may not include any of the aforementioned columns, or it may include other columns.

[0048] Error detection program ID 501 is the same column as error detection program ID 301. Evaluation program 502 is a column that stores the actual evaluation program, the path indicating the location where the actual program is stored, and a link to access the location where the actual program is stored.

[0049] The evaluation program is a program implemented in a programming language such as Python. For example, the evaluation program performs inference using corrected time-series data and evaluates the obtained inference results. If a correct inference result can be obtained, the evaluation program performs the evaluation by comparing the obtained evaluation result with the correct inference result. It is also possible to evaluate the time required for inference.

[0050] Figure 6 shows an example of the data structure of the setting control information 132 in Example 1.

[0051] The setting control information 132 is information for controlling the setting of a correction program to the inference system 110, and stores an entry that includes the correction program 601 and the setting timing 602. There is one entry for each correction program set to the inference system 110. Note that the columns included in the entry are not limited to those described above. It may not include any of the aforementioned columns, or it may include other columns.

[0052] Correction program 601 is the same column as correction program 402. Setting timing 602 is a column that stores the timing for setting the correction program to the inference system 110. Possible timings for setting the correction program include, for example, when the inference system 110 is not running, 5 minutes from the current time, or immediately. The timing for setting the correction program may be set manually or automatically based on an arbitrary setting rule.

[0053] Figure 7 shows an example of the data structure of the similar data definition information 143 in Example 1.

[0054] The similar data definition information 143 is information for managing definitions regarding the similarity of time series data, and stores entries that include the time series data type 701 and the extraction conditions 702. There is one entry for each metric. Note that the columns included in the entry are not limited to those mentioned above. It may not include any of the aforementioned columns, or it may include other columns.

[0055] Time series data type 701 is a column that stores the type of time series data. Extraction condition 702 is a column that stores the conditions for extracting similar time series data when an error is detected in time series data of the type corresponding to time series data type 701.

[0056] In the example shown in Figure 7, time series data composed of acceleration information collected from acceleration sensors of factory workers are defined as being similar when the time series data consists of the same worker, production line, and day of the week.

[0057] Figure 8 shows an example of the data structure of the error reproduction program management information 144 in Example 1.

[0058] Error reproduction program management information 144 is information for managing error reproduction programs that reproduce errors in time-series data using similar time-series data, and stores entries including error detection program ID 801 and error reproduction program 802. There is one entry for each error reproduction program. Note that the columns included in the entry are not limited to those mentioned above. It may not include any of the aforementioned columns, or it may include other columns.

[0059] Error detection program ID 801 is the same column as error detection program ID 301. Error reproduction program 802 is a column that stores the actual error reproduction program, the path indicating the location where the actual program is stored, and a link to access the location where the actual program is stored. The error reproduction program is a program implemented in a programming language such as Python.

[0060] Figure 9 is a sequence diagram illustrating the flow of the correction program setting process performed in the system of Example 1.

[0061] The inference system 110 transmits inference information to the correction program evaluation system 120 (step S101). The inference information includes time-series data acquired by the inference system 110, processing data output from each functional unit, and the inference result. Note that only time-series data, processing data, or the inference result may be included. The transmission of the inference information may be done manually or automatically. When transmitting the inference information automatically, the transmission timing can be set arbitrarily.

[0062] The correction program evaluation system 120 performs error detection processing using inference information (step S102). Specifically, the following processing is performed.

[0063] (S102-1) The correction program evaluation system 120 refers to the error detection program management information 122 and obtains one error detection program. The correction program evaluation system 120 executes the obtained error detection program on the inference information.

[0064] (S102-2) If no errors are detected in the time-series data, the correction program evaluation system 120 returns to S102-1 and performs the same process.

[0065] (S102-3) If an error is detected in the time series data, or if no errors are detected in the time series data after executing all error detection programs, the correction program evaluation system 120 terminates the error detection process. This concludes the explanation of the process in step S102.

[0066] If an error is detected in the time-series data as a result of the error detection process, the correction program evaluation system 120 executes the correction program evaluation process (step S103). Details of the correction program selection process will be explained using Figure 10.

[0067] The correction program evaluation system 120 selects the correction program with the smallest influence of inference based on the results of the correction program evaluation process (step S104).

[0068] The correction program evaluation system 120 transmits information about the selected correction program (correction program information) to the correction program setting support system 130 (step S105). The correction program information includes the actual correction program, a path indicating the storage location of the actual program, or a link to access the storage location of the actual program.

[0069] When the correction program setting support system 130 receives correction program information, it updates the setting control information 132 (step S106).

[0070] Specifically, the correction program setting support system 130 adds an entry to the setting control information 132, sets the value included in the correction program information to the correction program 601 of the added entry, and sets the value to the setting timing 602.

[0071] When the correction program setting support system 130 detects the setting timing, it configures the inference system 110 to execute the correction program (step S107).

[0072] In this way, the correction program setting support system 130 can set the correction program without stopping the inference system 110 by adjusting the timing of setting the correction program to the inference system 110.

[0073] Figure 10 is a flowchart illustrating an example of the correction program selection process performed by the correction program evaluation system 120 of Example 1.

[0074] The control unit 121 extracts an applicable correction program from the correction program management information 123 (step S201).

[0075] Specifically, the control unit 121 refers to the correction program management information 123 and searches for an entry in the error detection program ID 401 where the identification information of the error detection program that detected an error in the time-series data is stored. The control unit 121 retrieves the value stored in the correction program 402 of the retrieved entry.

[0076] The control unit 121 starts loop processing of the correction program (step S202). Specifically, the control unit 121 selects one correction program from the extracted correction programs.

[0077] The control unit 121 corrects errors in the time-series data using the selected correction program (step S203). At this time, the control unit 121 stores the correction program and the correction result in a work area of ​​memory (not shown) in association with each other.

[0078] The control unit 121 determines whether processing has been completed for all extracted correction programs (step S204).

[0079] If processing is not complete for all extracted correction programs, the control unit 121 returns to step S202 and selects a new correction program.

[0080] Once processing is complete for all extracted correction programs, the control unit 121 extracts an applicable evaluation program from the evaluation program management information 124 (step S205).

[0081] Specifically, the control unit 121 refers to the evaluation program management information 124 and searches for an entry in the error detection program ID 501 where the identification information of the error detection program that detected an error in the time-series data is stored. The control unit 121 retrieves the value stored in the evaluation program 502 of the retrieved entry.

[0082] The control unit 121 starts loop processing of the evaluation program (step S206). Specifically, the control unit 121 selects one evaluation program from the extracted evaluation programs.

[0083] The control unit 121 evaluates each of the correction results of multiple time-series data using the selected evaluation program (step S207). At this time, the control unit 121 stores the evaluation program and the evaluation results of each correction result in a work area of ​​memory (not shown) in association with each other.

[0084] The control unit 121 determines whether processing has been completed for all extracted evaluation programs (step S208).

[0085] If processing is not complete for all extracted evaluation programs, the control unit 121 returns to step S206 and selects a new evaluation program.

[0086] Once processing is complete for all extracted evaluation programs, the control unit 121 terminates the correction program selection process.

[0087] In step S104, for example, if the evaluation result is a numerical value, the control unit 121 calculates the sum, average, maximum, or minimum value of the evaluation results as an evaluation index for each correction result. Based on the evaluation index, the control unit 121 selects the correction program that has the smallest impact on the inference.

[0088] According to Example 1, when the correction program evaluation system 120 detects an error in time-series data, it can evaluate the impact of applying multiple correction programs to the inference system 110 on the inference process and select a correction program to apply based on the evaluation results.

[0089] Furthermore, by reusing error detection programs, correction programs, and evaluation programs when building a new inference system, the man-hours required for error design can be reduced. [Examples]

[0090] The correction program evaluation system 120 of Example 2 reproduces the error using time series data similar to the time series data in which the error was detected, and then evaluates the correction program.

[0091] Figure 11 is a sequence diagram illustrating the flow of the correction program setting process performed in the system of Example 2. Figure 12 is a diagram showing an example of a screen displayed by terminal 150 in Example 2.

[0092] The inference system 110 transmits inference information to the correction program evaluation system 120 (step S301). The processing in step S301 is the same as in step S101.

[0093] The correction program evaluation system 120 performs error detection processing using inference information (step S302). The processing in step S302 is the same as in step S102.

[0094] If the correction program evaluation system 120 detects an error in the time series data as a result of executing the error detection process, it transmits the time series data in which the error was detected to the error reproduction system 140 (step S303). At this time, the correction program evaluation system 120 also transmits identification information of the error detection program that detected the error.

[0095] The similar data extraction unit 141 of the error reproduction system 140 extracts time series data (similar time series data) that are similar to the time series data in which the error was detected (step S304). Specifically, the following processing is performed.

[0096] (S304-1) The similar data extraction unit 141 analyzes the type of time series data and the type of error in the time series data, and selects extraction conditions by referring to the similar data definition information 143 based on the analysis results.

[0097] (S304-2) The similar data extraction unit 141 acquires similar time-series data from the time-series data collection system 100 based on the extraction conditions. One possible method for acquiring time-series data is to send a query to the database. This concludes the explanation of the process in step S304.

[0098] The error reproduction unit 142 of the error reproduction system 140 generates error reproduction time series data that reproduces an error similar to the detected error using similar time series data (step S305). Specifically, the following processes are performed.

[0099] (S305-1) The error reproduction unit 142 refers to the error reproduction program management information 144 and searches for an entry in the error detection program ID 801 in which the identification information of the error detection program that detected the error in the time-series data is stored. The control unit 121 obtains the value stored in the error reproduction program 802 of the searched entry.

[0100] (S305-2) The error reproduction unit 142 generates error reproduction time series data by executing an error reproduction program on similar time series data. This concludes the explanation of the process in step S305.

[0101] The error reproduction unit 142 of the error reproduction system 140 transmits the generated error reproduction time series data to the correction program evaluation system 120 (step S306).

[0102] When the correction program evaluation system 120 receives error reproduction time series data, it executes the correction program evaluation process (step S307). The flow of the correction program evaluation process in Example 2 is the same as in Example 1. However, in step S203, correction is performed on both the time series data in which the error was detected and the error reproduction time series data, which is different from Example 1.

[0103] The correction program evaluation system 120 transmits evaluation information, including the correction result and evaluation result, to the terminal 150 (step S308).

[0104] When terminal 150 receives evaluation information, it displays a screen 1200 as shown in Figure 12. Screen 1200 includes a time-series data display area 1201 and an evaluation result display area 1202.

[0105] The time-series data display area 1201 is a field for displaying time-series data in which errors have been detected and time-series data in which errors have been reproduced. All of the time-series data may be displayed, or only element data with missing data or element data containing outliers may be displayed. The time-series data display area 1201 may also display the details of the error. In addition, the time-series data display area 1201 may also display data output by each functional unit of the inference system 110.

[0106] The evaluation result display area 1202 is the area where the evaluation results are displayed. In Figure 12, a table is displayed showing the combination of the correction program and the evaluation index. The user selects the correction program displayed in the evaluation result display area 1202 and presses the select button.

[0107] The correction program evaluation system 120 may also display screen 1200. Furthermore, similar to Example 1, the correction program evaluation system 120 may select a correction program.

[0108] Terminal 150 transmits information about the correction program selected by the user (correction program information) to the correction program setting support system 130 (step S309).

[0109] When the correction program setting support system 130 receives correction program information, it updates the setting control information 132 (step S310). The process in step S310 is the same as the process in step S105.

[0110] When the correction program setting support system 130 detects the setting timing, it sets the correction program in the inference system 110 (step S311). The process in step S311 is the same as the process in step S106.

[0111] According to Example 2, by applying corrections to multiple time-series data, the impact of the corrections on inference can be evaluated more accurately. Furthermore, by presenting the correction method and evaluation results to the user, the user can select a correction program to apply with confidence.

[0112] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. Furthermore, for example, the embodiments described above are detailed explanations of the configuration in order to clearly illustrate the present invention, and are not necessarily limited to those having all the configurations described. In addition, some of the configurations in each embodiment can be added to, deleted from, or replaced with other configurations.

[0113] Furthermore, each of the above-mentioned configurations, functions, processing units, processing means, etc., may be implemented in hardware, in whole or in part, for example, by designing them as integrated circuits. The present invention can also be implemented by software program code that realizes the functions of the embodiment. In this case, a storage medium on which the program code is recorded is provided to a computer, and the processor of that computer reads the program code stored in the storage medium. In this case, the program code read from the storage medium itself realizes the functions of the embodiment described above, and the program code itself and the storage medium on which it is stored constitute the present invention. Examples of storage media used to supply such program code include flexible disks, CD-ROMs, DVD-ROMs, hard disks, SSDs (Solid State Drives), optical disks, magneto-optical disks, CD-Rs, magnetic tapes, non-volatile memory cards, ROMs, and the like.

[0114] Furthermore, the program code that implements the functions described in this embodiment can be implemented in a wide range of programming or scripting languages, such as assembler, C / C++, Perl, Shell, PHP, Python, and Java (registered trademark).

[0115] Furthermore, the program code for the software that implements the functions of the embodiment may be distributed via a network and stored in a storage means such as a computer's hard disk or memory, or in a storage medium such as a CD-RW or CD-R, and the computer's processor may read and execute the program code stored in the storage means or storage medium.

[0116] In the above-described embodiment, the control lines and information lines shown are those deemed necessary for explanation and do not necessarily represent all control lines and information lines in the actual product. All components may be interconnected. [Explanation of symbols]

[0117] 100 Time-Series Data Acquisition Systems 101 Time Series Data Database 110 Inference Systems 111 Data Acquisition Unit 112 Pre-processing 113 Reasoning Department 114 Post-processing 120 Correction Program Evaluation System 121 Control Unit 122 Error Detection Program Management Information 123 Correction Program Management Information 124 Evaluation Program Management Information 130 Correction Program Setting Support System 131 Correction Program Setting Section 132 Setting Control Information 140 Error Reproduction System 141 Similar Data Extraction Unit 142 Error Reproduction Section 143 Similar Data Definition Information 144 Error Reproduction Program Management Information 150 devices 160 Networks 200-element data 1200 screens

Claims

1. A computer system comprising at least one computer, Connect to an inference system that performs inference using time series data, It maintains correction program management information for managing a correction program that corrects errors in the time series data, and detection program management information for managing a detection program that detects errors in the time series data. Each of the multiple correction programs managed by the correction program management information executes a different correction process. Each of the multiple detection programs managed by the aforementioned detection program management information executes a process to detect a different error. The identification information of the detection program and the identification information of the correction program are managed to correspond to each other. The aforementioned computer system, The detection program is executed, and it is determined whether or not an error has occurred in the time-series data corresponding to the detection program. If it is determined that an error has occurred in the time-series data corresponding to the executed detection program, the correction program associated with the detection program is selected as the applicable correction program. Based on the selected correction program, an evaluation process is performed to correct errors in the time series data and to evaluate the impact of the corrected time series data on the inference performed by the inference system. A computer system characterized by setting the correction program, which is selected based on the results of an evaluation of the impact on the inference, so that the inference system can be executed.

2. A computer system according to claim 1, A computer system characterized by selecting the correction program that has the least impact on the inference, based on the results of an evaluation of the impact on the inference.

3. A computer system according to claim 2, Check the state of the inference system, A computer system characterized by setting the selected correction program so that the inference system can be executed when the inference system is in a standby state.

4. The computer system according to claim 2, Connects to a time-series data collection system that collects and manages the aforementioned time-series data. If an error is detected in the aforementioned time series data, similar time series data similar to the time series data is obtained from the time series collection system. By generating an error similar to the detected error in the aforementioned similar time series data, error reproduction time series data is generated. A computer system characterized by performing the evaluation process using the time series data in which the error was detected based on the selected correction program and the error in the time series data in which the error was reproduced.

5. The computer system according to Claim 1, A computer system characterized by displaying the results of an evaluation of the impact on the inference and providing a user interface for accepting the selection of the correction program to be applied.

6. A method for correcting errors in time-series data performed by a computer system, The aforementioned computer system, Includes at least one computer, Connect to an inference system that performs inference using time series data, It maintains correction program management information for managing a correction program that corrects errors in the time series data, and detection program management information for managing a detection program that detects errors in the time series data. Each of the multiple correction programs managed by the correction program management information executes a different correction process. Each of the multiple detection programs managed by the aforementioned detection program management information executes a process to detect a different error. The identification information of the detection program and the identification information of the correction program are managed to correspond to each other. The error correction method for the aforementioned time-series data is: The first step is for at least one computer to execute the detection program and determine whether or not an error has occurred in the time-series data corresponding to the detection program. If at least one of the computers determines that an error has occurred in the time-series data corresponding to the detection program it has executed, the second step is to select the correction program associated with the detection program as the applicable correction program. A third step in which at least one computer corrects errors in the time series data based on the selected correction program and performs an evaluation process to evaluate the impact of the inference system on the corrected time series data; A method for correcting errors in time series data, comprising a fourth step of setting up the correction program, which is selected based on the results of an evaluation of the impact on the inference, so that the at least one computer can execute the inference system.

7. An error correction method for time series data according to Claim 6, The fourth step is a method for correcting errors in time series data, characterized in that at least one computer selects the correction program that has the smallest impact on the inference based on the results of an evaluation of the impact on the inference.

8. A method for correcting errors in time-series data according to Claim 7, The fourth step described above is: The steps include: the at least one computer checking the state of the inference system; A method for correcting errors in time-series data, comprising the step of setting a selected correction program so that the inference system can be executed when the inference system is in a standby state, on at least one computer.

9. A method for correcting errors in time-series data according to Claim 7, Connects to a time-series data collection system that collects and manages the aforementioned time-series data. The second step described above is: If at least one of the computers detects an error in the time series data, it takes the step of obtaining similar time series data from the time series collection system that is similar to the time series data in question. The steps include: generating error reproduction time series data by having at least one computer generate an error similar to the detected error on the similar time series data; The third step is a method for correcting errors in time series data, characterized in that at least one computer performs the evaluation process using the time series data in which the error was detected and the time series data in which the error was reproduced, based on the selected correction program.

10. An error correction method for time series data according to Claim 6, A method for correcting errors in time series data, characterized in that at least one computer displays the results of an evaluation of the impact on the inference and provides a user interface for accepting the selection of the correction program to be applied.

Images