WO2020232899A1 - 数据分析系统会诊方法及相关装置 - Google Patents

数据分析系统会诊方法及相关装置 Download PDF

Info

Publication number
WO2020232899A1
WO2020232899A1 PCT/CN2019/103442 CN2019103442W WO2020232899A1 WO 2020232899 A1 WO2020232899 A1 WO 2020232899A1 CN 2019103442 W CN2019103442 W CN 2019103442W WO 2020232899 A1 WO2020232899 A1 WO 2020232899A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
adjustment
test
test result
machine learning
Prior art date
Application number
PCT/CN2019/103442
Other languages
English (en)
French (fr)
Inventor
陈家荣
Original Assignee
平安科技(深圳)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 平安科技(深圳)有限公司 filed Critical 平安科技(深圳)有限公司
Publication of WO2020232899A1 publication Critical patent/WO2020232899A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F11/00Error detection; Error correction; Monitoring
    • G06F11/36Preventing errors by testing or debugging software
    • G06F11/3668Software testing
    • G06F11/3672Test management
    • G06F11/3692Test management for test results analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06NCOMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00Machine learning

Definitions

  • This application relates to the technical field of machine learning applications, and in particular to a data analysis system consultation method and related devices.
  • the data analysis system is a system that processes and organizes the data information of various indicators through the data analysis system, calculates various analysis indicators, and transforms them into information forms that are easy to be accepted by people, and can store the processed information.
  • the data analysis system needs to be tested during the process from development to putting into use, or upgrading, to find the problems in these systems; this test process is generally carried out through the test system, and the test system will get the corresponding test when testing the software system. Test Results.
  • the tester does not need to adjust the data analysis system; however, the inventor of the present application realizes that if the test result of the test system fails, the tester needs to make continuous modifications, Retrieval and find out how to adjust the plan to make the test result pass, the adjustment efficiency is very low.
  • an object of the present application is to provide a data analysis system consultation method and related devices.
  • a data analysis system consultation method includes: when a test instruction of a data analysis system for analyzing target index data is received, performing a real-time test on the data analysis system, and outputting the test result; judging; Whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is failed; if the test result is not passed, the test result is obtained as fail
  • the predetermined upper limit value and the target limit value in the predetermined lower limit value corresponding to the passage, the difference between the test result exceeding the target limit value is calculated; the data of the target index and the test result ,
  • the difference value and the target limit value are input into a pre-trained machine learning model, and a test adjustment scheme is output, wherein the test adjustment scheme is used to indicate how to adjust to make the test result pass.
  • a data analysis system consultation device is characterized by comprising: a test module, which is used to perform a test on the data analysis system when receiving a test instruction for the data analysis system for analyzing target index data. Perform a real-time test and output the test result; a judging module for judging whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed; obtain Module for obtaining the target limit in the predetermined upper limit value and the predetermined lower limit value corresponding to the test result as failing, and calculating that the test result exceeds all The difference of the target limit; an output module for inputting the data of the target indicator, the test result, the difference, and the target limit into a pre-trained machine learning model, and outputting a test adjustment plan , Wherein the test adjustment scheme is used to indicate how to adjust to make the test result pass.
  • a data analysis system consultation device includes: a processor; and a memory for storing a data analysis system consultation program of the processor; wherein, the processor is configured to perform consultation through the data analysis system Program to execute the above-mentioned data analysis system consultation method.
  • a computer non-volatile readable storage medium stores a data analysis system consultation program, characterized in that the data analysis system consultation program is executed by a processor to realize the data analysis system as described above Consultation method.
  • Fig. 1 schematically shows a flow chart of a consultation method for a data analysis system.
  • Figure 2 schematically shows an example diagram of an application scenario of a data analysis system consultation method.
  • Fig. 3 schematically shows a flow chart of a method for obtaining the correct rate of a test adjustment scheme.
  • Fig. 4 schematically shows a block diagram of a consultation device of a data analysis system.
  • Fig. 5 shows a block diagram of an electronic device for implementing the aforementioned data analysis system consultation method according to an exemplary embodiment.
  • Fig. 6 shows a schematic diagram of a computer non-volatile readable storage medium for realizing the aforementioned data analysis system consultation method according to an exemplary embodiment.
  • This example embodiment first provides a data analysis system consultation method.
  • the data analysis system consultation method can be run on a server, a server cluster or a cloud server, etc. Of course, those skilled in the art can also run on other platforms as required
  • the method of the present application is not specifically limited in this exemplary embodiment.
  • the consultation method of the data analysis system may include the following steps:
  • Step S110 When a test instruction of the data analysis system for analyzing the target index data is received, the data analysis system is tested in real time, and the test result is output;
  • Step S120 determining whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed;
  • Step S130 If the test result is not passed, obtain the target limit in the predetermined upper limit and the predetermined lower limit corresponding to the test result as the fail, and calculate that the test result exceeds the The difference in the target limit.
  • Step S140 Input the data of the target indicator, the test result, the difference and the target limit together into a pre-trained machine learning model, and output a test adjustment plan, wherein the test adjustment plan is used for Indicate how to adjust to make the test result pass.
  • the data analysis system when receiving a test instruction from a data analysis system that analyzes the target index data, the data analysis system is tested in real time and the test results are output; the data analysis system is The data of the types of indicators are analyzed to obtain the processing result.
  • the test system performs real-time testing during the processing process, and obtains the test result according to the calculated difference between the processing result and the predetermined benchmark result, and realizes the real-time test.
  • test result is passed, where if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed; if the test result exceeds the predetermined reasonable predetermined threshold range, it means that there is an internal data analysis system In order to solve the problem, the normal processing will not get the index value beyond the predetermined range. If the test result is not passed, the test result fails to pass the corresponding predetermined upper limit and the target limit in the predetermined lower limit, and the difference between the test result exceeding the target limit; the test result fails to indicate that there is an internal data analysis system Errors, the test results need to be consulted.
  • test results exceed the predetermined threshold range, different internal errors can be accurately indicated by exceeding the predetermined upper limit or the predetermined upper limit of the predetermined upper limit and the predetermined lower limit.
  • the difference between the predetermined lower limit value can be combined with other data to accurately analyze the index to improve the accuracy of the consultation.
  • the data of the target index, the test result, the difference, and the target limit are input into the pre-trained machine learning model, and the test adjustment plan is output.
  • the test adjustment plan is used to indicate how to adjust the test result to pass;
  • the input of factors in the consultation The machine learning model trained according to the sample of the consultation factors can automatically, accurately and efficiently obtain the plan of how to adjust the test results to pass, effectively improving the efficiency of problem solving.
  • step S110 when a test instruction of the data analysis system for analyzing the target index data is received, the data analysis system is tested in real time, and the test result is output.
  • the server 201 receives the test instruction of the data analysis system for analyzing the target index data sent by the server 202, and then the test system in the server 201 performs real-time testing of the data analysis system , Output the test result.
  • the server 201 can be any device with processing capability, for example, a computer, a microprocessor, etc., which is not specifically limited here, and the server 202 can be any device with the ability to issue orders, such as a mobile phone, a computer, etc. Special restrictions.
  • the data analysis system analyzes and processes the data of one indicator, and obtains the processing result, which is the entire data analysis process; use the test system to test the data analysis system during the entire processing process, and you will get the data analysis system for one indicator data
  • the test result in the process of analysis and processing; the test result is determined by the difference between the processing result of the data analysis system and the reference value by the test system.
  • the processing result of the data analysis system for an indicator usually has a data that can characterize this The reference value of the change of the result, and the change is maintained within a certain range, so that the test result can be used to conduct consultations on the internal problems of the data analysis system in the subsequent steps.
  • Step S120 It is determined whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is failed.
  • whether the test result exceeds the predetermined upper limit and the predetermined lower limit is used to determine whether the test result passes;
  • the data analysis usually focuses on different indicators, such as what is the core concern of a product? Another example is the sales growth rate and market share? Sales growth rate and market share are an indicator; during data analysis, the data of one of the indicators is input into the data analysis system, and a result is obtained by analyzing and processing according to a predetermined algorithm, and this result is under normal circumstances.
  • the test system passes the real-time test, and compares the output result of the test system with the normal threshold according to predetermined rules, such as the difference, etc., if the output result is within the certain threshold range If the difference is large, it is judged that the test result is not passed, and vice versa. For example, after the data analysis system analyzes an index, the result is 50, the reference value is 60, the difference is 10, and the given predetermined threshold range is- 2-5, the test result exceeds the upper limit of 5, and the out of range is 5. The test result fails to indicate that there is an error in the data analysis system, which causes the processing result to exceed the predetermined range.
  • predetermined rules such as the difference, etc.
  • step S130 if the test result is not passed, the predetermined upper limit value and the target limit in the predetermined lower limit value corresponding to when the test result is not passed are obtained, and the test result exceeds The difference between the target limit.
  • the test result of the test system on the data analysis system is not passed, indicating that there is a problem in the data analysis system, which causes the test result to be failed.
  • the data of an indicator is analyzed by the data analysis system When processing, it is executed by the corresponding module or corresponding part of the program in the data analysis system, so the test results of different indicators do not pass the corresponding solutions, and the specific test results exceed the predetermined upper limit of the predetermined threshold range.
  • the difference between the limit value or the predetermined lower limit value can reflect further problems within the function or formula.
  • a coefficient that is too large may cause the test result to exceed the upper limit value, while a coefficient that is too small may cause the test
  • the result is lower than the lower limit; at the same time, the size of the difference between the test results exceeding the limit can accurately reflect the severity of the problem, for example, when the coefficient of a certain function exceeds a certain threshold, the difference will exceed the corresponding The threshold. Therefore, the specific test result obtained through this step exceeds the predetermined upper limit or the predetermined lower limit of the predetermined threshold range including the predetermined upper limit and the predetermined lower limit, which can accurately guide the detailed solution and effectively improve Accuracy of consultation.
  • step S140 the data of the target indicator, the test result, the difference value, and the target limit value are input into a pre-trained machine learning model to output a test adjustment scheme, wherein the test adjustment scheme Used to indicate how to adjust to make the test result pass.
  • a machine learning model is trained in advance, and the machine learning model is based on the expert’s historical data that has been based on the target index, the test result, the difference and the predetermined limit corresponding to the difference (target Limit) Find how to adjust the test results to pass the program training; so the target indicator data, the test results, the difference and the target are input into the machine learning model together as consultation factors, which can be accurate, Efficiently predict how to adjust the plan to make the test result. This can avoid the low efficiency and low accuracy problems caused by manual search.
  • machine learning models corresponding to different types of target indicators are trained respectively, then the data of the target indicators, the test results, and the difference The value and the target limit are input to the pre-trained machine learning model for training, and the test adjustment plan is output, including:
  • the data of the target indicator, the test result, the difference, and the target limit are input together into a pre-trained machine learning model corresponding to the type of the target indicator, and a test adjustment scheme is output.
  • train multiple machine learning models corresponding to different types of target indicators so that the training can make the machine learning model pertinent, and how to adjust the output to make the test result pass the program more accurate, and train machine learning The model is more efficient.
  • a machine learning model corresponding to all types of target indicators is trained according to all types of target indicators, then the data of the target indicators, the test results, and the difference And the target limit value is input to the pre-trained machine learning model for training, and the test adjustment plan is output, including:
  • the data of the target indicator, the test result, the difference, and the target limit are input together into a pre-trained machine learning model corresponding to the type of the target indicator, and a test adjustment scheme is output.
  • the target indicator data, the indicator value, the difference value, and the predetermined upper limit value corresponding to the difference value or The lower limit value is input to the machine learning model, and the output is how to adjust the plan to make the test result pass, which can effectively save costs, and only one machine learning model is used.
  • the training method of the machine learning model includes:
  • the output of the machine learning model is consistent with the test adjustment scheme corresponding to the sample, and the training of the machine learning model ends.
  • the target index data, the test result, the difference and the target limit as the consultation elements can accurately guide the problems in the internal data processing of the system, and the corresponding solutions have been based on these consultation elements as input samples in history.
  • the training machine learning model accurately outputs the sample label according to the input sample, which can effectively ensure the accuracy of the machine model.
  • the determining whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed including:
  • the test result is passed, wherein if the test result is outside the predetermined upper limit value and the predetermined lower limit value, the test result is not passed.
  • the preset limit table records the data range of all index data and the predetermined upper limit value and the predetermined lower limit value corresponding to the data range. In this way, according to the data range in which the size of the target index data corresponding to the test result is located, the predetermined upper limit value and the predetermined lower limit value corresponding to the target index data can be accurately found from the preset limit table. Then, according to the predetermined upper limit value and the predetermined lower limit value, it is determined whether the test result is outside the range of the predetermined upper limit value and the predetermined lower limit value, and then when the test result is within the predetermined upper limit value and the predetermined lower limit value Otherwise, it can be accurately judged that the test result is not passed. Through the preset limit table, the size of the predetermined upper limit value and the predetermined lower limit value in the limit table can be conveniently and accurately adjusted according to the specific situation.
  • the method further includes:
  • the adjustment plan record table records data such as the number of times each test adjustment plan is selected by the user.
  • the test adjustment scheme output by the machine learning model may be at least one, for example, the output includes 5 possible sub-adjustment schemes.
  • the historical selection rate of each sub-adjustment scheme is obtained by dividing the number of times each possible sub-adjustment scheme is selected by the number of times all possible sub-adjustment schemes are selected.
  • the high selection rate indicates that the solution is adopted with a high frequency, that is, a solution with a high user acceptance rate. Outputting a solution with a high selection rate can effectively help users quickly locate problems and improve test efficiency.
  • the method further includes:
  • Step 310 Obtain the first adjustment target of the current version of the data analysis system for the previous version before the current version;
  • Step 320 Obtain a second adjustment target of the test adjustment scheme
  • Step 330 Acquire the correctness probability of the test adjustment scheme according to the first adjustment target and the second adjustment target.
  • the first adjustment target of the current version with respect to the previous version before the current version is the improvement or change effect that the current version of the data analysis system needs to obtain after improvement compared to the previous version.
  • This effect is generally composed of multiple sub-modifications, so the first adjustment target may include multiple first sub-adjustment targets.
  • the test adjustment plan is also a plan for improving the data analysis system, and the improved effect will also be obtained, which is the second adjustment target. In this way, according to the degree of deviation or promotion between the first adjustment target and the second adjustment target, the correctness probability of the test adjustment scheme can be reflected.
  • the first target score of the adjustment target can be represented by calibrating the first adjustment target in advance
  • the second target score of the adjustment target can be represented by calibrating the second adjustment target in advance.
  • the ratio of the difference between the two target scores to the sum of the first target score and the second target score is normalized to a value between 1 and 10 to obtain the correctness probability of the test adjustment scheme.
  • the obtaining the correctness probability of the test adjustment scheme according to the first adjustment target and the second adjustment target includes: obtaining the adjustment influence corresponding to the first adjustment target Function; obtain the adjustment coefficient of each adjustment variable in the adjustment function according to the second adjustment target; obtain the correctness probability of the test adjustment scheme according to the adjustment coefficient and the adjustment influence function.
  • the adjustment influence function corresponding to the first adjustment target is the adjustment influence function that represents each improvement or change adjustment variable and the unknown coefficient of each variable when the current version of the data analysis system needs to be improved compared to the previous version.
  • the adjustment variable representing each improvement or change can be obtained from a preset adjustment variable table corresponding to the impact of each change stored in the data analysis system.
  • the unknown coefficient of each variable can be obtained from the second adjustment target according to the second adjustment target from the second adjustment target when the part of the first adjustment target is changed. , Which is the adjustment coefficient for some variables. Then, by assigning the adjustment coefficient to the corresponding variable part in the adjustment influence function, the correctness probability of the test adjustment scheme can be accurately obtained.
  • the application also provides a data analysis system consultation device.
  • the consultation device of the data analysis system may include a test module 410, a judgment module 420, an acquisition module 430 and an output module 440. among them:
  • the test module 410 may be used to perform a real-time test on the data analysis system when receiving a test instruction of the data analysis system for analyzing the target index data, and output the test result;
  • the judging module 420 can be used to judge whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed;
  • the obtaining module 430 may be configured to obtain the predetermined upper limit value and the target limit value in the predetermined lower limit value corresponding to the predetermined upper limit value and the predetermined lower limit value when the test result is not passed, and calculate the test result The difference exceeding the stated target limit;
  • the output module 440 may be used to input the data of the target index, the test result, the difference value, and the target limit value into a pre-trained machine learning model to output a test adjustment plan, wherein the test adjustment The scheme is used to indicate how to adjust to make the test result pass.
  • the exemplary embodiments described herein can be implemented by software, or can be implemented by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , Including several instructions to make a computing device (which can be a personal computer, server, mobile terminal, or network device, etc.) execute the method according to the embodiment of the present application.
  • a non-volatile storage medium can be a CD-ROM, U disk, mobile hard disk, etc.
  • Including several instructions to make a computing device which can be a personal computer, server, mobile terminal, or network device, etc.
  • the electronic device 500 according to this embodiment of the present application will be described below with reference to FIG. 5.
  • the electronic device 500 shown in FIG. 5 is only an example, and should not bring any limitation to the function and scope of use of the embodiments of the present application.
  • the electronic device 500 is represented in the form of a general-purpose computing device.
  • the components of the electronic device 500 may include, but are not limited to: the aforementioned at least one processing unit 510, the aforementioned at least one storage unit 520, and a bus 530 connecting different system components (including the storage unit 520 and the processing unit 510).
  • the storage unit stores program code, and the program code can be executed by the processing unit 510, so that the processing unit 510 executes the various exemplary methods described in the “exemplary method” section of this specification.
  • the processing unit 510 may perform step S110 as shown in FIG.
  • test adjustment plan when receiving a test instruction of a data analysis system for analyzing target index data, perform a real-time test on the data analysis system, And output the test result;
  • S120 determine whether the test result is passed, wherein if the test result is outside the predetermined upper limit and the predetermined lower limit, the test result is not passed;
  • step S130 if the test result If the test result is not passed, obtain the target limit in the predetermined upper limit value and the predetermined lower limit value corresponding to the test result that is not passed, and calculate the difference between the test result exceeding the target limit value;
  • S140 Input the data of the target index, the test result, the difference value, and the target limit together into a pre-trained machine learning model, and output a test adjustment plan, where the test adjustment plan is used to indicate How to adjust to make the test result pass.
  • the storage unit 520 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 5201 and/or a cache storage unit 5202, and may further include a read-only storage unit (ROM) 5203.
  • RAM random access storage unit
  • ROM read-only storage unit
  • the storage unit 520 may also include a program/utility tool 5204 having a set (at least one) program module 5205.
  • program module 5205 includes but is not limited to: an operating system, one or more application programs, other program modules, and program data, Each of these examples or some combination may include the implementation of a network environment.
  • the bus 530 may represent one or more of several types of bus structures, including a storage unit bus or a storage unit controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area using any bus structure among multiple bus structures. bus.
  • the electronic device 500 may also communicate with one or more external devices 700 (such as keyboards, pointing devices, Bluetooth devices, etc.), and may also communicate with one or more devices that enable customers to interact with the electronic device 500, and/or communicate with Any device (such as a router, modem, etc.) that enables the electronic device 500 to communicate with one or more other computing devices. This communication can be performed through an input/output (I/O) interface 550.
  • the electronic device 500 may also communicate with one or more networks (for example, a local area network (LAN), a wide area network (WAN), and/or a public network, such as the Internet) through the network adapter 560. As shown in the figure, the network adapter 560 communicates with other modules of the electronic device 500 through the bus 530.
  • LAN local area network
  • WAN wide area network
  • public network such as the Internet
  • the exemplary embodiments described herein can be implemented by software, or can be implemented by combining software with necessary hardware. Therefore, the technical solution according to the embodiments of the present application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (can be a CD-ROM, U disk, mobile hard disk, etc.) or on the network , Including several instructions to make a computing device (which may be a personal computer, server, terminal device, or network device, etc.) execute the method according to the embodiment of the present application.
  • a non-volatile storage medium can be a CD-ROM, U disk, mobile hard disk, etc.
  • Including several instructions to make a computing device which may be a personal computer, server, terminal device, or network device, etc.
  • a computer non-volatile readable storage medium on which is stored a program product capable of implementing the above method in this specification.
  • various aspects of the present application can also be implemented in the form of a program product, which includes program code.
  • the program product runs on a terminal device, the program code is used to enable the The terminal device executes the steps according to various exemplary embodiments of the present application described in the above-mentioned "Exemplary Method" section of this specification.
  • a program product 600 for implementing the above method according to an embodiment of the present application is described. It can adopt a portable compact disk read-only memory (CD-ROM) and include program code, and can be installed in a terminal device, For example, running on a personal computer.
  • CD-ROM compact disk read-only memory
  • the program product of this application is not limited to this.
  • the readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or combined with an instruction execution system, device, or device.
  • the program product can use any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Type programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • the computer-readable signal medium may include a data signal propagated in baseband or as a part of a carrier wave, and readable program code is carried therein.
  • This propagated data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the readable signal medium may also be any readable medium other than a readable storage medium, and the readable medium may send, propagate, or transmit a program for use by or in combination with the instruction execution system, apparatus, or device.
  • the program code contained on the readable medium can be transmitted by any suitable medium, including but not limited to wireless, wired, optical cable, RF, etc., or any suitable combination of the foregoing.
  • the program code used to perform the operations of this application can be written in any combination of one or more programming languages.
  • the programming languages include object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural Programming language-such as "C" language or similar programming language.
  • the program code can be executed entirely on the client computing device, partly executed on the client device, executed as a stand-alone software package, partly executed on the client computing device and partly executed on the remote computing device, or entirely on the remote computing device or server Executed on.
  • the remote computing device can be connected to the client computing device through any kind of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (for example, using Internet service providers) Business to connect via the Internet).
  • LAN local area network
  • WAN wide area network
  • Internet service providers Internet service providers
  • FIG. 1 the above-mentioned drawings are only schematic illustrations of the processing included in the method according to the exemplary embodiments of the present application, and are not intended for limitation. It is easy to understand that the processing shown in the above drawings does not indicate or limit the time sequence of these processings. In addition, it is easy to understand that these processes can be executed synchronously or asynchronously in multiple modules, for example.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Quality & Reliability (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Computer Hardware Design (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Debugging And Monitoring (AREA)

Abstract

本申请提供了一种数据分析系统会诊方法及相关装置,属于机器学习应用技术领域,该方法包括:当接收到数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;判断所述测试结果是否通过;若所述测试结果为未通过,获取所述测试结果未通过对应的所述预定上限值及所述上限值中的目标限值,以及所述测试结果超出所述目标限值的差值;将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案。本申请通过训练机器学习模型,根据会诊要素自动、准确、快速的进行会诊得到会诊方案,进而有效提高系统的调整效率。

Description

数据分析系统会诊方法及相关装置 技术领域
本申请要求2019年05月23日递交、发明名称为“数据分析系统会诊方法及相关装置”的中国专利申请201910436034.8的优先权,在此通过引用将其全部内容合并于此。
本申请涉及机器学习应用技术领域,尤其涉及一种数据分析系统会诊方法及相关装置。
背景技术
数据分析系统是通过数据分析系统对各种指标的数据信息进行加工、整理,计算得到各种分析指标,转变为易于被人们所接受的信息形式,并可以将处理后的信息进行贮存的系统。数据分析系统在从开发到投入使用,或者升级等过程都需要进行测试,发现这些系统中存在的问题;这个测试过程一般是通过测试系统进行的,测试系统在对软件系统测试时会得到相应的测试结果。然后,如果测试系统的测试结果通过,则不需要测试人员进行调整数据分析系统;但是,本申请的发明人意识到,如果测试系统的测试结果没有通过,则需要测试人员进行通过不断的修改、检索、查找如何调整使得测试结果通过的方案,调整的效率非常低。
技术问题
所以需要一种可以自动根据测试系统的测试结果进行会诊,得到如何调整使得测试结果通过的方案的方法;可以准确、快速预测输出调整方案,进而有效提高调整效率。
技术解决方案
为了解决上述技术问题,本申请的一个目的在于提供一种数据分析系统会诊方法及相关装置。
其中,本申请所采用的技术方案为:
一方面,一种数据分析系统会诊方法,包括:当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
另一方面,一种数据分析系统会诊装置,其特征在于,包括:测试模块,用于当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;判断模块,用于判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;获取模块,用于若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;输出模块,用于将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
另一方面,一种数据分析系统会诊装置,包括:处理器;以及存储器,用于存储所述处理器的数据分析系统会诊程序;其中,所述处理器配置为经由执行所述数据分析系统会诊程序来执行如上所述的数据分析系统会诊方法。
另一方面,一种计算机非易失性可读存储介质,其上存储有数据分析系统会诊程序,其特征在于,所述数据分析系统会诊程序被处理器执行时实现如上所述的数据分析系统会诊方法。
有益效果
在上述技术方案中,通过将用于会诊的因素输入根据会诊因素样本训练好的机器学习模型,可以自动、准确和高效的得到如何调整使得测试结果通过的方案,有效提高解决问题的效率。
应当理解的是,以上的一般描述和后文的细节描述仅是示例性和解释性的,并不能限制本申请。
附图说明
此处的附图被并入说明书中并构成本说明书的一部分,示出了符合本申请的实施例,并于说明书一起用于解释本申请的原理。
图1示意性示出一种数据分析系统会诊方法的流程图。
图2示意性示出一种数据分析系统会诊方法的应用场景示例图。
图3示意性示出一种测试调整方案正确率获取方法流程图。
图4示意性示出一种数据分析系统会诊装置的方框图。
图5示出根据示例性实施例的用于实现上述数据分析系统会诊方法的电子设备的框图。
图6示出根据示例性实施例的用于实现上述数据分析系统会诊方法的计算机非易失性可读存储介质的示意图。
通过上述附图,已示出本申请明确的实施例,后文中将有更详细的描述,这些附图和文字描述并不是为了通过任何方式限制本申请构思的范围,而是通过参考特定实施例为本领域技术人员说明本申请的概念。
本发明的实施方式
这里将详细地对示例性实施例执行说明,其示例表示在附图中。下面的描述涉及附图时,除非另有表示,不同附图中的相同数字表示相同或相似的要素。以下示例性实施例中所描述的实施方式并不代表与本申请相一致的所有实施方式。相反,它们仅是与如所附权利要求书中所详述的、本申请的一些方面相一致的装置和方法的例子。
现在将参考附图更全面地描述示例实施方式。然而,示例实施方式能够以多种形式实施,且不应被理解为限于在此阐述的范例;相反,提供这些实施方式使得本申请将更加全面和完整,并将示例实施方式的构思全面地传达给本领域的技术人员。所描述的特征、结构或特性可以以任何合适的方式结合在一个或更多实施方式中。
本示例实施方式中首先提供了数据分析系统会诊方法,该数据分析系统会诊方法可以运行于服务器,也可以运行于服务器集群或云服务器等,当然,本领域技术人员也可以根据需求在其他平台运行本申请的方法,本示例性实施例中对此不做特殊限定。参考图1所示,该数据分析系统会诊方法可以包括以下步骤:
步骤S110,当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;
步骤S120,判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;
步骤S130,若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值。
步骤S140. 将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
上述数据分析系统会诊方法中,首先,当接收到在对目标指标的数据进行分析的数据分析系统的测试指令,对所述数据分析系统进行实时测试,并输出测试结果;数据分析系统对一种类型的指标的数据进行分析得到处理结果,测试系统在处理过程实时测试,根据处理结果和预定的基准结果的计算差值得到测试结果,实现实时测试。判断测试结果是否通过,其中,若测试结果在预定上限值和预定下限值之外则测试结果为未通过;测试结果如果超出预先给定的合理的预定阈值范围这样说明数据分析系统内部出现了问题,正常的处理是不会得到超出该预定范围的指标值。若测试结果为未通过,获取测试结果未通过对应的预定上限值及预定下限值中的目标限值,以及测试结果超出目标限值的差值;测试结果未通过说明数据分析系统内部存在错误,需要对测试结果进行会诊,测试结果超出预定阈值范围的情况可以准确指引不同的内部错误,通过超过所述包括预定上限值和预定下限值的预定限值范围的预定上限值或者预定下限值的差值,可以结合其它数据准确的针对该指标进行分析,提高会诊的准确率。将目标指标的数据、测试结果、差值以及目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,测试调整方案用于指示如何调整使得测试结果为通过;通过将上述用于会诊的因素输入根据会诊因素样本训练好的机器学习模型,可以自动、准确和高效的得到如何调整使得测试结果通过的方案,有效提高解决问题的效率。
下面,将结合附图对本示例实施方式中上述数据分析系统会诊方法中的各步骤进行详细的解释以及说明。
在步骤S110中, 当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果。
本示例的实施方式中,参考图2所示,服务器201接收到服务器202发出的对目标指标的数据进行分析的数据分析系统的测试指令,然后服务器201中的测试系统对数据分析系统进行实时测试,输出测试结果。其中,服务器201可以是任何具有处理能力的设备,例如,电脑、微处理器等,在此不做特殊限定,服务器202可以是任何具有发令能力的设备,例如手机、电脑等,在此不做特殊限定。数据分析系统对一种指标的数据进行分析处理,得到处理结果,就是整个数据分析的过程;利用测试系统在整个处理过程中对数据分析系统进行测试,会得到数据分析系统对一种指标的数据在分析处理过程中的测试结果;测试结果通过测试系统对数据分析系统的处理结果和基准值求差得到差值确定,数据分析系统对一种指标的数据的处理结果通常是存在一个可以表征这个结果变化情况的基准值,而且变化时维持在一定范围的,这样就可以利用该测试结果在后续步骤中进行数据分析系统内部问题的会诊。
步骤S120, 判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过。
本示例的实施方式中,利用测试结果的是否超过预定上限值和预定下限值判断测试结果是否通过;进行数据分析时通常会针对不同的指标,例如一个产品核心关注的是什么?又比如是销售增长率、市场占有率?销售增长率、市场占有率就是一种指标;数据分析时将其中一种的指标的指标的数据输入数据分析系统,按照预定的算法进行分析处理得到一个结果,而这个结果正常情况下是在一定的范围内,也就是一定的阈值范围内,测试系统通过实时测试,将测试系统输出的结果与正常阈值按照预定的规则进行比较,例如求差等,如果输出的结果与所述一定的阈值范围相差较大,判断测试结果不通过,反之则通过,例如数据分析系统对一种指标的数据分析后得到的结果为50,基准值60,差值为10,而给定的预定阈值范围为-2-5,则测试结果超过上限值5,超出范围为5,测试结果未通过说明数据分析系统系统内部存在错误,导致处理结果超出预定范围。
在步骤S130中, 若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值。
本示例的实施方式中,测试系统对数据分析系统的测试结果为未通过,说明数据分析系统内部存在问题,导致测试结果为未通过,而通常情况下,一种指标的数据由数据分析系统分析处理时,在数据分析系统内部通过相应的模块或者相应的部分的程序来执行,所以不同的指标的测试结果不通过具有相应的解决方案,而具体的测试结果超过所述预定阈值范围的预定上限值或者预定下限值的差值的情况,可以反映出更进一步的函数或者公式内部的问题,例如,某个系数过大可能导致测试结果超过上限值,而系数过小有可能导致测试结果低于下限值;同时,测试结果超过所述限制的差值的大小可以准确的反映出问题的严重程度,例如,某个函数的系数超过一定阈值时就会导致所述差值超过对应的阈值。所以通过这一步获取具体的测试结果超过所述包括预定上限值和预定下限值的预定阈值范围的预定上限值或者预定下线值的差值可以准确的指引详细的解决方案,有效提高会诊的准确性。
在步骤S140中, 将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
本示例的实施方式中,通过事先训练一个机器学习模型,该机器学习模型是根据专家历史上已经根据目标指标的数据、测试结果、所述差值以及所述差值对应的预定限值(目标限值)查找到的如何调整使得测试结果通过的方案训练得到的;所以将目标指标的数据、所述测试结果、所述差值以及所述目标作为会诊因素一起输入机器学习模型,可以准确、高效预测出如何调整使得测试结果方案。这样可以避免人工查找带来的低效率、低准确率问题。
在本示例的一种实施方式中,按照不同类型的目标指标分别训练有对应于不同类型的目标指标的机器学习模型,则所述将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型进行训练,输出测试调整方案,包括:
获取所述目标指标的类型对应的预先训练好的机器学习模型;
将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
按照不同类型的目标指标,训练多个对应于不同的类型的目标指标的机器学习模型,这样训练可以使得机器学习模型具有针对性,输出的如何调整使得测试结果通过的方案更加准确,训练机器学习模型的效率更高。
在本示例的一种实施方式中,按照所有类型的目标指标训练有对应于所有类型的目标指标的机器学习模型,则所述将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型进行训练,输出测试调整方案,包括:
获取所述对应于所有类型的目标指标的机器学习模型;
将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
通过训练适用于所有类型的目标指标的机器学习模型,当测试结果不通过时,可以直接将目标指标的数据、所述指标值、所述差值以及所述差值对应的预定上限值或者下限值输入机器学习模型,输出如何调整使得测试结果通过的方案,这样可以有效节省成本,只用一个机器学习模型。
在本示例的一种实施方式中,所述机器学习模型的训练方法包括:
收集事先标记了如何调整使得测试结果为通过的测试调整方案的包括目标指标的数据、所述测试结果、所述差值以及所述目标限值的样本的集合;
将所述样本的集合中每个样本分别输入机器学习模型,调整机器学习模型输出每个所述样本对应的测试调整方案;
如果存在有所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案不一致,则调整机器学习模型的系数直到一致;
当所有的所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案一致,机器学习模型的训练结束。
目标指标的数据、所述测试结果、所述差值以及目标限值作为会诊要素可以准确的指引系统内部数据处理出现的问题,通过历史上已经根据这些会诊要素作为输入样本,对应的解决的方案作为样本标记,训练机器学习模型根据输入的样本准确输出样本标记,可以有效保证机器模型的准确度。
在本示例的一种实施方式中,所述判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外则测试结果为未通过,包括:
从预设限值表中,获取所述测试结果对应的所述目标指标的数据对应的所述预定上限值和所述预定下限值;
根据所述预定上限值和所述预定下限值,判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外则测试结果为未通过。
预设限值表中记录了所有的指标的数据的数据范围及与数据范围对应的预定上限值和预定下限值。这样根据测试结果对应的目标指标的数据的大小所在的数据范围可以准确的从预设限值表中查找到目标指标的数据对应的预定上限值和预定下限值。然后,根据预定上限值和所述预定下限值,判断测试结果是否位于预定上限值和所述预定下限值的范围之外,进而当测试结果在预定上限值和预定下限值之外则可以准确判断到测试结果为未通过。通过预设限值表可以根据具体情况,便捷、准确地调整限值表中预定上限值和预定下限值等数据的大小。
在本示例的一种实施方式中,所述方法还包括:
从调整方案记录表中,获取机器学习模型输出的测试调整方案中每个可能的子调整方案的历史选择率;向用户输出历史选择率超过预定阈值的多个所述子调整方案。
调整方案记录表中记录了每个测试调整方案被用户选择的次数等数据。机器学习模型输出的测试调整方案可以为至少一个,例如输出包含5个可能的子调整方案的测试调整方案。然后,通过每个可能的子调整方案被选择的次数除以所有可能的子调整方案被选择的次数得到每个子调整方案的历史选择率。选择率高说明该解决方案被采纳的频率高,也就是用户认可率高的解决方案,将选择率高的方案输出,可以有效帮助用户快速定位问题,提高测试效率。
在本示例的一种实施方式中,参考图3所示,所述方法还包括:
步骤310,获取所述数据分析系统的当前版本对于所述当前版本之前的上一版本的第一调整目标;
步骤320,获取所述测试调整方案的第二调整目标;
步骤330,根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率。
当前版本对于当前版本之前的上一版本的第一调整目标,就是当前版本的数据分析系统相对于上一版本需要在改进后获得的提升或者改变的效果。该效果一般是由多个子改动共同组成的,所以第一调整目标可以包括多个第一子调整目标。同时,测试调整方案也是对数据分析系统进行改进的方案,也会获得改进后的效果,也就是第二调整目标。这样,根据第一调整目标及第二调整目标之间向背离或者促进的程度,可以反映出测试调整方案的正确性概率。一种示例中,可以通过事先为第一调整目标标定可以代表调整目标的第一目标分数,同时事先为第二调整目标标定可以代表调整目标的第二目标分数,然后通过第一目标分数与第二目标分数之间的差值与第一目标分数与第二目标分数之和的比值,进行归一化到1到10之间的数值,获得测试调整方案的正确性概率。输出所有如何调整使得测试结果通过的方案的同时输出如何调整使得测试结果通过的方案的正确性概率,这样可以清楚的帮助测试人员了解所有的方案及方案的正确率情况,用户可以根据正确率结合经验进行选择解决方案。
在本示例的一种实施方式中,所述根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率,包括:获取所述第一调整目标对应的调整影响函数;根据所述第二调整目标获取所述调整函数中每个调整变量的调整系数;根据所述调整系数及所述调整影响函数,获取所述测试调整方案的正确性概率。
第一调整目标对应的调整影响函数就是当前版本的数据分析系统相对于上一版本需要在改进时,代表每个改进或者变动的调整变量和每个变量的未知系数组成的调整影响函数。代表每个改进或者变动的调整变量可以根据预设的存储有数据分析系统的每个改动的影响对应的调整变量表中获取。每个变量的未知系数则可以根据第二调整目标,从存储有对第一调整目标中的部分进行变动得到第二调整目标时,发生的二次变动对第一次变动的影响系数表中获取,也就是对部分变量的调整系数。然后,通过将调整系数赋值到调整影响函数中对应的变量部分,可以准确地获取到测试调整方案的正确性概率。
本申请还提供了一种数据分析系统会诊装置。参考图4所示,该数据分析系统会诊装置可以包括测试模块410、判断模块420、获取模块430以及输出模块440。其中:
测试模块410可以用于当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;
判断模块420可以用于判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;
获取模块430可以用于若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;
输出模块440可以用于将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
上述数据分析系统会诊装置中各模块的具体细节已经在对应的数据分析系统会诊方法中进行了详细的描述,因此此处不再赘述。
应当注意,尽管在上文详细描述中提及了用于动作执行的设备的若干模块或者单元,但是这种划分并非强制性的。实际上,根据本申请的实施方式,上文描述的两个或更多模块或者单元的特征和功能可以在一个模块或者单元中具体化。反之,上文描述的一个模块或者单元的特征和功能可以进一步划分为由多个模块或者单元来具体化。 此外,尽管在附图中以特定顺序描述了本申请中方法的各个步骤,但是,这并非要求或者暗示必须按照该特定顺序来执行这些步骤,或是必须执行全部所示的步骤才能实现期望的结果。附加的或备选的,可以省略某些步骤,将多个步骤合并为一个步骤执行,以及/或者将一个步骤分解为多个步骤执行等。
通过以上的实施方式的描述,本领域的技术人员易于理解,这里描述的示例实施方式可以通过软件实现,也可以通过软件结合必要的硬件的方式来实现。因此,根据本申请实施方式的技术方案可以以软件产品的形式体现出来,该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM,U盘,移动硬盘等)中或网络上,包括若干指令以使得一台计算设备(可以是个人计算机、服务器、移动终端、或者网络设备等)执行根据本申请实施方式的方法。
在本申请的示例性实施例中,还提供了一种能够实现上述方法的电子设备。
所属技术领域的技术人员能够理解,本申请的各个方面可以实现为系统、方法或程序产品。因此,本申请的各个方面可以具体实现为以下形式,即:完全的硬件实施方式、完全的软件实施方式(包括固件、微代码等),或硬件和软件方面结合的实施方式,这里可以统称为“电路”、“模块”或“系统”。
下面参照图5来描述根据本申请的这种实施方式的电子设备500。图5显示的电子设备500仅仅是一个示例,不应对本申请实施例的功能和使用范围带来任何限制。
如图5所示,电子设备500以通用计算设备的形式表现。电子设备500的组件可以包括但不限于:上述至少一个处理单元510、上述至少一个存储单元520、连接不同系统组件(包括存储单元520和处理单元510)的总线530。
其中,所述存储单元存储有程序代码,所述程序代码可以被所述处理单元510执行,使得所述处理单元510执行本说明书上述“示例性方法”部分中描述的根据本申请各种示例性实施方式的步骤。例如,所述处理单元510可以执行如图1中所示的步骤S110:当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;S120:判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;步骤S130:若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;步骤S140:将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
存储单元520可以包括易失性存储单元形式的可读介质,例如随机存取存储单元(RAM)5201和/或高速缓存存储单元5202,还可以进一步包括只读存储单元(ROM)5203。
存储单元520还可以包括具有一组(至少一个)程序模块5205的程序/实用工具5204,这样的程序模块5205包括但不限于:操作系统、一个或者多个应用程序、其它程序模块以及程序数据,这些示例中的每一个或某种组合中可能包括网络环境的实现。
总线530可以为表示几类总线结构中的一种或多种,包括存储单元总线或者存储单元控制器、外围总线、图形加速端口、处理单元或者使用多种总线结构中的任意总线结构的局域总线。
电子设备500也可以与一个或多个外部设备700(例如键盘、指向设备、蓝牙设备等)通信,还可与一个或者多个使得客户能与该电子设备500交互的设备通信,和/或与使得该电子设备500能与一个或多个其它计算设备进行通信的任何设备(例如路由器、调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口550进行。并且,电子设备500还可以通过网络适配器560与一个或者多个网络(例如局域网(LAN),广域网(WAN)和/或公共网络,例如因特网)通信。如图所示,网络适配器560通过总线530与电子设备500的其它模块通信。应当明白,尽管图中未示出,可以结合电子设备500使用其它硬件和/或软件模块,包括但不限于:微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID系统、磁带驱动器以及数据备份存储系统等。
通过以上的实施方式的描述,本领域的技术人员易于理解,这里描述的示例实施方式可以通过软件实现,也可以通过软件结合必要的硬件的方式来实现。因此,根据本申请实施方式的技术方案可以以软件产品的形式体现出来,该软件产品可以存储在一个非易失性存储介质(可以是CD-ROM,U盘,移动硬盘等)中或网络上,包括若干指令以使得一台计算设备(可以是个人计算机、服务器、终端装置、或者网络设备等)执行根据本申请实施方式的方法。
在本申请的示例性实施例中,还提供了一种计算机非易失性可读存储介质,其上存储有能够实现本说明书上述方法的程序产品。在一些可能的实施方式中,本申请的各个方面还可以实现为一种程序产品的形式,其包括程序代码,当所述程序产品在终端设备上运行时,所述程序代码用于使所述终端设备执行本说明书上述“示例性方法”部分中描述的根据本申请各种示例性实施方式的步骤。
参考图6所示,描述了根据本申请的实施方式的用于实现上述方法的程序产品600,其可以采用便携式紧凑盘只读存储器(CD-ROM)并包括程序代码,并可以在终端设备,例如个人电脑上运行。然而,本申请的程序产品不限于此,在本文件中,可读存储介质可以是任何包含或存储程序的有形介质,该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。
所述程序产品可以采用一个或多个可读介质的任意组合。可读介质可以是可读信号介质或者可读存储介质。可读存储介质例如可以为但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件,或者任意以上的组合。可读存储介质的更具体的例子(非穷举的列表)包括:具有一个或多个导线的电连接、便携式盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。
计算机可读信号介质可以包括在基带中或者作为载波一部分传播的数据信号,其中承载了可读程序代码。这种传播的数据信号可以采用多种形式,包括但不限于电磁信号、光信号或上述的任意合适的组合。可读信号介质还可以是可读存储介质以外的任何可读介质,该可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。可读介质上包含的程序代码可以用任何适当的介质传输,包括但不限于无线、有线、光缆、RF等等,或者上述的任意合适的组合。
可以以一种或多种程序设计语言的任意组合来编写用于执行本申请操作的程序代码,所述程序设计语言包括面向对象的程序设计语言—诸如Java、C++等,还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在客户计算设备上执行、部分地在客户设备上执行、作为一个独立的软件包执行、部分在客户计算设备上部分在远程计算设备上执行、或者完全在远程计算设备或服务器上执行。在涉及远程计算设备的情形中,远程计算设备可以通过任意种类的网络,包括局域网(LAN)或广域网(WAN),连接到客户计算设备,或者,可以连接到外部计算设备(例如利用因特网服务提供商来通过因特网连接)。此外,上述附图仅是根据本申请示例性实施例的方法所包括的处理的示意性说明,而不是限制目的。易于理解,上述附图所示的处理并不表明或限制这些处理的时间顺序。另外,也易于理解,这些处理可以是例如在多个模块中同步或异步执行的。
本领域技术人员在考虑说明书及实践这里公开的申请后,将容易想到本申请的其他实施例。本申请旨在涵盖本申请的任何变型、用途或者适应性变化,这些变型、用途或者适应性变化遵循本申请的一般性原理并包括本申请未公开的本技术领域中的公知常识或惯用技术手段。说明书和实施例仅被视为示例性的,本申请的真正范围和精神由权利要求指出。

Claims (20)

  1. 一种数据分析系统会诊方法,其特征在于,包括:
    当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;
    判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;
    若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
  2. 根据权利要求1所述的方法,其特征在于,按照不同类型的目标指标分别训练有对应于不同类型的目标指标的机器学习模型,则所述将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型进行训练,输出测试调整方案,包括:
    获取所述目标指标的类型对应的预先训练好的机器学习模型;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
  3. 根据权利要求1所述的方法,其特征在于,按照所有类型的目标指标训练有对应于所有类型的目标指标的机器学习模型,则所述将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型进行训练,输出测试调整方案,包括:
    获取所述对应于所有类型的目标指标的机器学习模型;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
  4. 根据权利要求1所述的方法,其特征在于,所述机器学习模型的训练方法包括:
    收集事先标记了如何调整使得测试结果为通过的测试调整方案的包括目标指标的数据、所述测试结果、所述差值以及所述目标限值的样本的集合;
    将所述样本的集合中每个样本分别输入机器学习模型,调整机器学习模型输出每个所述样本对应的测试调整方案;
    如果存在有所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案不一致,则调整机器学习模型的系数直到一致;
    当所有的所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案一致,机器学习模型的训练结束。
  5. 根据权利要求1所述的方法,其特征在于,所述判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过,包括:
    从预设限值表中,获取所述测试结果对应的所述目标指标的数据对应的所述预定上限值和所述预定下限值;
    根据所述预定上限值和所述预定下限值,判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过。
  6. 根据权利要求1所述的方法,其特征在于,所述方法还包括:
    从调整方案记录表中,获取机器学习模型输出的测试调整方案中每个子调整方案的历史选择率;
    向用户输出历史选择率超过预定阈值的多个所述子调整方案。
  7. 根据权利要求1所述的方法,其特征在于,所述方法还包括:
    获取所述数据分析系统的当前版本对于所述当前版本之前的上一版本的第一调整目标;
    获取所述测试调整方案的第二调整目标;
    根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率。
  8. 根据权利要求7所述的方法,其特征在于,所述根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率,包括:
    获取所述第一调整目标对应的调整影响函数;
    根据所述第二调整目标获取所述调整影响函数中每个调整变量的调整系数;
    根据所述调整系数及所述调整影响函数,获取所述测试调整方案的正确性概率。
  9. 根据权利要求7所述的方法,其特征在于,事先为第一调整目标标定代表调整目标的第一目标分数,同时事先为第二调整目标标定代表调整目标的第二目标分数,所述根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率,包括:
    计算所述第一目标分数与所述第二目标分数之间的差值;
    计算所述第一目标分数与所述第二目标分数的总和;
    获取所述差值与所述总和的比值,作为所述测试调整方案的正确性概率。
  10. 一种数据分析系统会诊装置,其特征在于,包括:
    测试模块,用于当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;
    判断模块,用于判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外则测试结果为未通过;
    获取模块,用于若所述测试结果为未通过,获取所述测试结果未通过对应的所述预定上限值及所述预定下限值中的目标限值,以及所述测试结果超出所述目标限值的差值;
    输出模块,用于将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
  11. 根据权利要求10所述的装置,按照不同类型的目标指标分别训练有对应于不同类型的目标指标的机器学习模型,所述输出模块被配置为:
    获取所述目标指标的类型对应的预先训练好的机器学习模型;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
  12. 根据权利要求10所述的装置,按照所有类型的目标指标训练有对应于所有类型的目标指标的机器学习模型,所述输出模块被配置为:
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
  13. 根据权利要求10所述的装置,所述输出模块还被配置为:
    收集事先标记了如何调整使得测试结果为通过的测试调整方案的包括目标指标的数据、所述测试结果、所述差值以及所述目标限值的样本的集合;
    将所述样本的集合中每个样本分别输入机器学习模型,调整机器学习模型输出每个所述样本对应的测试调整方案;
    如果存在有所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案不一致,则调整机器学习模型的系数直到一致;
    当所有的所述样本输入机器学习模型后,机器学习模型的输出与所述样本对应的测试调整方案一致,机器学习模型的训练结束。
  14. 根据权利要求10所述的装置,所述判断模块被配置为:
    从预设限值表中,获取所述测试结果对应的所述目标指标的数据对应的所述预定上限值和所述预定下限值;
    根据所述预定上限值和所述预定下限值,判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过。
  15. 根据权利要求10所述的装置,所述输出模块还被配置为:
    从调整方案记录表中,获取机器学习模型输出的测试调整方案中每个子调整方案的历史选择率;
    向用户输出历史选择率超过预定阈值的多个所述子调整方案。
  16. 根据权利要求10所述的装置,所述输出模块还被配置为:
    获取所述数据分析系统的当前版本对于所述当前版本之前的上一版本的第一调整目标;
    获取所述测试调整方案的第二调整目标;
    根据所述第一调整目标及所述第二调整目标获取所述测试调整方案的正确性概率。
  17. 根据权利要求16所述的装置,所述输出模块还被配置为:
    获取所述第一调整目标对应的调整影响函数;
    根据所述第二调整目标获取所述调整函数中每个调整变量的调整系数;
    根据所述调整系数及所述调整影响函数,获取所述测试调整方案的正确性概率。
  18. 一种数据分析系统会诊装置,其特征在于,包括:处理器及存储器,存储所述处理器的数据分析系统会诊程序;其中,所述处理器配置为经由执行所述数据分析系统会诊程序来执行以下处理:
    当接收到用于对目标指标的数据进行分析的数据分析系统的测试指令时,对所述数据分析系统进行实时测试,并输出测试结果;
    判断所述测试结果是否通过,其中,若所述测试结果在预定上限值和预定下限值之外,则测试结果为未通过;
    若所述测试结果为未通过,获取所述测试结果为未通过时对应的所述预定上限值及所述预定下限值中的目标限值,计算所述测试结果超出所述目标限值的差值;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型,输出测试调整方案,其中,所述测试调整方案用于指示如何调整使得测试结果为通过。
  19. 根据权利要求18所述的装置,其特征在于,按照不同类型的目标指标分别训练有对应于不同类型的目标指标的机器学习模型,则所述将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入预先训练好的机器学习模型进行训练,输出测试调整方案,包括:
    获取所述目标指标的类型对应的预先训练好的机器学习模型;
    将所述目标指标的数据、所述测试结果、所述差值以及所述目标限值一起输入所述目标指标的类型对应的预先训练好的机器学习模型,输出测试调整方案。
  20. 一种计算机非易失性可读存储介质,其上存储有数据分析系统会诊程序,其特征在于,所述数据分析系统会诊程序被处理器执行时实现上述1-9任一项所述的方法。
PCT/CN2019/103442 2019-05-23 2019-08-29 数据分析系统会诊方法及相关装置 WO2020232899A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910436034.8A CN110347583A (zh) 2019-05-23 2019-05-23 一种数据分析系统会诊方法及相关装置
CN201910436034.8 2019-05-23

Publications (1)

Publication Number Publication Date
WO2020232899A1 true WO2020232899A1 (zh) 2020-11-26

Family

ID=68174261

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/103442 WO2020232899A1 (zh) 2019-05-23 2019-08-29 数据分析系统会诊方法及相关装置

Country Status (2)

Country Link
CN (1) CN110347583A (zh)
WO (1) WO2020232899A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114817064A (zh) * 2022-05-20 2022-07-29 国网数字科技控股有限公司 一种安全测试结果的自动化判定方法及装置

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1577759A2 (en) * 2004-03-15 2005-09-21 Ramco Systems Limited Method and system for providing documentation and training in software development activity
US20150088462A1 (en) * 2012-06-07 2015-03-26 Tencent Technology (Shenzhen) Company Limited Hardware performance evaluation method and server
CN105045713A (zh) * 2015-07-14 2015-11-11 腾讯科技(深圳)有限公司 数据处理方法及移动终端
CN107766299A (zh) * 2017-10-24 2018-03-06 携程旅游信息技术(上海)有限公司 数据指标异常的监控方法及其系统、存储介质、电子设备
CN109389329A (zh) * 2018-11-13 2019-02-26 平安科技(深圳)有限公司 测试系统的测试性能预测方法、装置、介质及电子设备
CN109426700A (zh) * 2017-08-28 2019-03-05 腾讯科技(北京)有限公司 数据处理方法、装置、存储介质和电子装置
CN109783332A (zh) * 2018-12-12 2019-05-21 平安科技(深圳)有限公司 基于数据建模的服务器告警方法及系统
CN110083452A (zh) * 2019-04-26 2019-08-02 深信服科技股份有限公司 一种资源管理方法、系统及电子设备和存储介质

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105893256B (zh) * 2016-03-30 2018-06-08 西北工业大学 基于机器学习算法的软件故障定位方法
US20180150742A1 (en) * 2016-11-28 2018-05-31 Microsoft Technology Licensing, Llc. Source code bug prediction
CN107195186A (zh) * 2017-06-07 2017-09-22 千寻位置网络有限公司 自动优化路口车辆通行速度的方法和系统
CN107822617B (zh) * 2017-10-23 2020-10-16 上海百芝龙网络科技有限公司 一种基于WiFi信号的心率异常检测方法
CN107885503B (zh) * 2017-11-11 2021-01-08 湖南大学 一种基于程序特征分析的迭代编译优化方法
CN108494602B (zh) * 2018-04-08 2020-11-06 上海鸿洛通信电子有限公司 Ota参数的调整方法、装置及智能终端
CN109222208B (zh) * 2018-10-30 2021-07-06 杭州安脉盛智能技术有限公司 面向卷烟生产指标控制的制丝工艺分析优化方法及系统

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1577759A2 (en) * 2004-03-15 2005-09-21 Ramco Systems Limited Method and system for providing documentation and training in software development activity
US20150088462A1 (en) * 2012-06-07 2015-03-26 Tencent Technology (Shenzhen) Company Limited Hardware performance evaluation method and server
CN105045713A (zh) * 2015-07-14 2015-11-11 腾讯科技(深圳)有限公司 数据处理方法及移动终端
CN109426700A (zh) * 2017-08-28 2019-03-05 腾讯科技(北京)有限公司 数据处理方法、装置、存储介质和电子装置
CN107766299A (zh) * 2017-10-24 2018-03-06 携程旅游信息技术(上海)有限公司 数据指标异常的监控方法及其系统、存储介质、电子设备
CN109389329A (zh) * 2018-11-13 2019-02-26 平安科技(深圳)有限公司 测试系统的测试性能预测方法、装置、介质及电子设备
CN109783332A (zh) * 2018-12-12 2019-05-21 平安科技(深圳)有限公司 基于数据建模的服务器告警方法及系统
CN110083452A (zh) * 2019-04-26 2019-08-02 深信服科技股份有限公司 一种资源管理方法、系统及电子设备和存储介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114817064A (zh) * 2022-05-20 2022-07-29 国网数字科技控股有限公司 一种安全测试结果的自动化判定方法及装置

Also Published As

Publication number Publication date
CN110347583A (zh) 2019-10-18

Similar Documents

Publication Publication Date Title
US20180261236A1 (en) Speaker recognition method and apparatus, computer device and computer-readable medium
CN109726763B (zh) 一种信息资产识别方法、装置、设备及介质
CN111143226B (zh) 自动化测试方法及装置、计算机可读存储介质、电子设备
CN109461023B (zh) 流失用户挽回方法及装置、电子设备、存储介质
US11403303B2 (en) Method and device for generating ranking model
CN109344257B (zh) 文本情感识别方法及装置、电子设备、存储介质
CN110674009B (zh) 应用服务器性能监测方法、装置、存储介质及电子设备
CN113837596B (zh) 一种故障确定方法、装置、电子设备及存储介质
WO2021121296A1 (zh) 习题测试数据生成方法以及装置
CN114444608B (zh) 数据集质量评估方法及装置、电子设备及存储介质
CN110348471B (zh) 异常对象识别方法、装置、介质及电子设备
WO2021047184A1 (zh) 系统故障分析处理方法、装置、存储介质及电子设备
CN110109824B (zh) 大数据自动回归测试方法、装置、计算机设备和存储介质
US10545972B2 (en) Identification and elimination of non-essential statistics for query optimization
CN114360581A (zh) 识别设备故障的方法、装置及电子设备
CN110580217B (zh) 软件代码健康度的检测方法、处理方法、装置及电子设备
CN117786158B (zh) 一种点歌一体机的性能评估方法及系统
CN110704390A (zh) 获取服务器维护脚本的方法、装置、电子设备及介质
WO2020232899A1 (zh) 数据分析系统会诊方法及相关装置
WO2018098670A1 (zh) 一种进行数据处理的方法和装置
CN113299278B (zh) 一种声学模型性能评估的方法、装置和电子设备
CN110826616A (zh) 信息处理方法及装置、电子设备、存储介质
CN113590484B (zh) 算法模型服务测试方法、系统、设备及存储介质
US11520831B2 (en) Accuracy metric for regular expression
CN109240992A (zh) 天线测试数据的处理方法、系统、计算机设备及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19929735

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19929735

Country of ref document: EP

Kind code of ref document: A1