WO2021164465A1 - 智能预警方法与*** - Google Patents
智能预警方法与*** Download PDFInfo
- Publication number
- WO2021164465A1 WO2021164465A1 PCT/CN2021/071217 CN2021071217W WO2021164465A1 WO 2021164465 A1 WO2021164465 A1 WO 2021164465A1 CN 2021071217 W CN2021071217 W CN 2021071217W WO 2021164465 A1 WO2021164465 A1 WO 2021164465A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- early warning
- warning information
- variable index
- preset
- variable
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/03—Credit; Loans; Processing thereof
Definitions
- the embodiments of the present application relate to the field of data monitoring, and in particular, to an intelligent early warning method and system.
- the current general warning schemes are roughly divided into two types: the first basically relies on the user to manually add, for example, setting the warning for business scenarios, the user needs to set the warning threshold for each business scenario, manually set multiple times, the warning is triggered Then it is pushed to the user client.
- setting the early warning needs to depend on the graph itself, such as limited to histogram, line graph, etc. This method relies heavily on industry experience, is cumbersome to operate, triggers passively, and cannot identify potential risks in time.
- the second type of intelligent early warning is biased towards public opinion monitoring, industry information and industry dynamics acquisition, using text mining, knowledge graphs and other technologies, focusing on changes in the overall big industry, public opinion trends, etc. It lacks monitoring of big data and cannot be based on specific business conditions. Provide targeted guidance.
- the purpose of the embodiments of the present application is to provide an intelligent early warning method and system, which can monitor variable index data in an all-round way, improve the efficiency of abnormal monitoring, and save time.
- an intelligent early warning method including:
- variable index parameter exceeds the index range interval, an early warning information is generated according to a preset natural language description template and the variable index parameter for the user to view.
- an embodiment of the present application also provides an intelligent early warning system, including:
- An obtaining module used to obtain business scenarios related to the target user and variable indicators related to the business scenarios
- the configuration module is used to configure the variable index parameters of the variable index
- the monitoring module is used to monitor whether the variable index parameter exceeds the preset index range interval
- the generating module is configured to generate early warning information according to a preset natural language description template and the variable index parameter if the variable index parameter exceeds the index range interval, for the user to view.
- an embodiment of the present application also provides a computer device, the computer device includes a memory and a processor, the memory stores a computer program that can be stored on the processor, and the computer program is The processor executes the intelligent early warning method as described above, and the intelligent early warning method includes the following steps:
- variable index parameter exceeds the index range interval, an early warning information is generated according to a preset natural language description template and the variable index parameter for the user to view.
- an embodiment of the present application also provides a computer-readable storage medium, and a computer program is stored in the computer-readable storage medium, and the computer program can be executed by at least one processor to enable the At least one processor executes the intelligent early warning method as described above, and the intelligent early warning method includes the following steps:
- variable index parameter exceeds the index range interval, an early warning information is generated according to a preset natural language description template and the variable index parameter for the user to view.
- variable index data corresponding to the variable index in the business scenario to determine whether the variable index is abnormal, and then generates early warning information.
- the variable index data can be monitored in an all-round way, which improves the efficiency of abnormal monitoring and saves time.
- Fig. 1 is a flowchart of Embodiment 1 of an intelligent early warning method of this application.
- FIG. 2 is the first flowchart of step S104 in the first embodiment of this application.
- FIG. 3 is the second flowchart of step S104 in FIG. 1 in the first embodiment of the application.
- FIG. 4 is the third flowchart of step S104 in the first embodiment of this application.
- FIG. 5 is a flowchart of step S106 in FIG. 1 according to the first embodiment of the application.
- Fig. 6 is a flowchart of early warning information processing according to an embodiment of the application.
- Fig. 7 is a schematic diagram of the program modules of the second embodiment of the intelligent early warning system of this application.
- FIG. 8 is a schematic diagram of the hardware structure of the third embodiment of the computer equipment of this application.
- variable indicators, indicator range intervals, templates, and/or early warning information, etc. can be stored in a database, or can be stored in a blockchain, such as distributed storage through a blockchain, this application Not limited.
- FIG. 1 there is shown a flow chart of the steps of the intelligent early warning method according to the first embodiment of the present application. It can be understood that the flowchart in this method embodiment is not used to limit the order of execution of the steps.
- the following exemplarily describes the computer device 2 as the execution subject. details as follows.
- Step S100 Obtain business scenarios related to the target user and variable indicators related to the business scenarios.
- the variable indicators are variable indicators related to the banking business.
- the loan business scenario includes variable indicators such as loan amount and monthly repayment amount.
- the Telescope application debugging tool the user's operation instructions and request instructions for the banking indicators are monitored, so as to obtain the relevant variable indicators. And the Telescope application debugging tool can mark the variable indicators and related operation data corresponding to each business scenario with business tags, so that users can directly query a series of data related to the business scenario.
- Step S102 Configure the variable index parameter of the variable index.
- the relevant variable index data is configured for each business scenario
- the variable index parameter is the preset value of the variable index
- the variable index data of each business scenario is stored In the corresponding bottom table.
- the business indicator parameter is the loan amount of a small enterprise
- the business indicator parameter is xx million
- the corresponding business indicator parameter is xx million.
- the user's bottom table has fields "number of defaulted users" () and “number of all users” (). Now you can configure a "proportion of defaulted users" field through SQL.
- the pseudo code of the SELECT part of the specific SQL is as follows:
- Step S104 monitoring whether the variable index parameter exceeds a preset index range interval.
- monitoring the variable index data includes real-time monitoring of the loan amount data of the loan business scenario in the business scenario, and regular and quantitative monitoring of the loan amount data, such as monitoring weekly, daily, hourly, etc.
- the updated variable index data monitors quantitative variable index data such as the total loan amount this month and the weekly loan amount.
- the step S104 includes:
- Step S104A0 Obtain the historical time series of the business scenario and the variable index.
- step S104A2 the historical time series is decomposed by a preset time series decomposition algorithm to obtain a residual component at each time point of the historical time series.
- STL is divided into inner loop and outer loop.
- the inner loop mainly calculates trend fitting and periodic component, and converges the remainder component through the outer loop.
- S-ESD Seasonal ESD
- X is the historical time series
- SX is the periodic component after STL decomposition
- X' is the median of X.
- step S104A4 the residual component of each time point is analyzed according to the preset abnormal value detection algorithm to obtain the abnormal time point.
- the step of constructing an outlier detection algorithm in advance includes: constructing N binary trees, and randomly sampling each tree without replacement to generate a training set.
- the binary tree is a random binary tree.
- the construction process of the binary tree can be described as follows: first randomly select an attribute A of the time series, and then randomly select a value V in the attribute A, Split the time series tree according to the value of attribute A, put records less than V on the left branch, and put records greater than V on the right branch, and then recursively build the tree according to the above process until the following conditions are met: Incoming The time series has only one record or multiple identical records; the height of the tree has reached a limit height.
- Prediction The process of prediction is to walk the time series down the corresponding branch on the binary tree until it reaches the leaf node, and record the path length c(n) passed in the process. Bring c(n) into the outlier scoring function to get the outlier score.
- the calculation formula is as follows:
- s(x,n) is to record the abnormal index of c(n) in a binary tree composed of n samples.
- the value range of s(x,n) is [0,1], and the closer to 1 means it is an abnormal point.
- the probability is high. The closer to 0, the higher the probability of being a normal point. If the s(x,n) of most of the time series is close to 0.5, it means that the entire time series has no obvious outliers.
- Step S104A6 judging whether the abnormal time point is within the index range interval.
- variable index parameters corresponding to the abnormal time point are monitored to query whether an abnormal event occurs at the abnormal time point.
- variable index parameter exceeds the index range interval
- generating early warning information according to a preset natural language description template and the variable index parameter for the user to view includes:
- the early warning information is generated according to a preset natural language description template and the abnormal time point for the user to view.
- variable index parameters corresponding to the abnormal cosine component of the abnormal index are generated according to the preset natural language description template to generate early warning information.
- the step S104 includes:
- Step S104B0 Obtain variable index parameters of the variable indexes of multiple target users.
- obtaining the variable index parameter of the variable index over a period of time may be a certain variable index of multiple target users, or multiple variable indexes of multiple target users. For example: in a loan business scenario, the consumption amount of multiple target users in a month.
- Step S104B2 arbitrarily select a variable index parameter of a target user from the variable index parameters of the plurality of target users as a clustering center.
- a variable index parameter of a target user is arbitrarily selected as the cluster center.
- Step S104B4 Calculate the distance from the variable index parameters of other users to the cluster center.
- the distance between the variable index parameters of other users and the cluster center can be calculated by Euler's formula, and the cluster center and the variable index parameters of other users can also be analyzed by the clustering algorithm.
- the clustering algorithm can be the DBSCAN algorithm, WSN algorithm, etc.
- Step S104B6 monitoring whether the distance exceeds a preset index range interval.
- a preset index range such as the amount of consumption
- the consumption amount of multiple target users in a month exceeds the preset range
- variable index parameter exceeds the index range interval
- generating early warning information according to a preset natural language description template and the variable index parameter for the user to view includes:
- variable index parameter If the distance is outside the preset range of the variable index parameter, the variable index parameter is used as abnormal data, and early warning information is generated according to the preset natural language description template and the abnormal time point for the user to view.
- the points are divided into core points/boundary points/noise points.
- the core point is a data point that contains at least m points within the distance e (it can be a point within the preset range), and the boundary point is the neighboring point within the distance e of the core point, but the number of points contained is less than m, and the rest are all.
- the noise point is the abnormal time point.
- the variable index parameters corresponding to the abnormal time point and the corresponding target user are generated according to the preset natural language description template to generate early warning information.
- the step S104 includes:
- Step S104C0 Obtain historical time series data of the variable index parameter of the variable index.
- the historical time series data is historical data obtained by variable index parameters in a time series.
- Step S104C checking the stationarity of the historical time series data.
- the stationarity of the historical time series data can be roughly obtained.
- Step S104C2 Perform difference calculation on the non-stationary historical time series data in the historical time series data to obtain a difference series.
- the non-stationary historical time series data is subjected to difference calculation to make it stable. If an original series is stable, we call it the I(0) process; if the original series is not stable but is stable after the first-order difference, we call the series I(1) process; the same series is stable after n-th difference, then it is called the series It is the I(n) process.
- Step S104C3 Calculate the autocorrelation coefficient and the average movement coefficient of the difference sequence.
- MA(q) which is equivalent to ARIMA(0,0,q).
- ⁇ 1 ,..., ⁇ p are called moving average coefficients.
- the autocorrelation coefficient and the average moving coefficient of the difference sequence are denoted as (p, q).
- Step S104C4 Perform residual and white noise processing on the difference sequence according to the autocorrelation coefficient and the average movement coefficient to obtain a time sequence correlation model.
- the useful information in the differential sequence is extracted, and the rest is all random disturbances, which cannot be predicted and used. If the differential sequence passes the white noise test, the modeling (time series correlation model) can be terminated. , Because there is no information to continue to extract. If the residual is not white noise, it means that there is still useful information in the residual, and the model needs to be modified or further extracted.
- Step S104C5 Predict the time series data to be tested through the preset time series correlation model to obtain forecast data.
- the preset time series correlation model may be a time series correlation model such as ARIMA, which uses historical time series data itself to build a model to study the law of the development of things. Input the processed difference sequence into a preset time series correlation model to obtain forecast data.
- ARIMA time series correlation model
- Step S104C6 monitoring whether the predicted data exceeds a preset index range interval.
- variable index data can be monitored. If the variable index data exceeds the index range interval of the predicted data, it indicates that abnormal variable index data has occurred.
- variable index parameter exceeds the index range interval
- generating early warning information according to a preset natural language description template and the variable index parameter for the user to view includes:
- variable index parameter If the distance is outside the preset range of the variable index parameter, the variable index parameter is used as abnormal data, and early warning information is generated according to the preset natural language description template and the abnormal time point for the user to view.
- the corresponding natural language description template is found in the preset natural language description template, and the predicted data is input to the corresponding natural language description template to generate early warning information.
- Step S106 If the variable index parameter exceeds the index range interval, generate early warning information according to a preset natural language description template and the variable index parameter for the user to view.
- the natural language description templates generated according to different types of early warning variable indicators are different.
- the template can be formulated as: XX variable indicators have risen in the past XX days/weeks/months/quarters/years , Rose XX%.
- step S106 further includes:
- Step S106A Substituting the variable index parameters into a preset natural language description template to generate the early warning information.
- the warning information is generated, if the abnormal variable index parameters are processed first, then the processed variable index parameters are substituted into the preset natural language Language description template.
- the natural language description template is an early warning template preset according to different business scenarios. It is used according to different early warning situations. When using it, you only need to input the business indicator parameters corresponding to the business indicators. For example: when the variable index parameters of the loan amount of the loan business scenario change, and the small and medium-sized enterprise's quota is allocated to the large enterprise, that is, the template: xx enterprise, the loan amount is abnormal, and the loan amount that should be issued is xx. Among them, the actual loan amount is xxx, and the actual loan amount is the changed loan amount. Later, give the warning information a color warning reminder, so that the user can view it immediately.
- step S106B the early warning information is divided into positive early warning information, negative relevant early warning information and neutral early warning information.
- variable index parameter changes of different variable indicators are different.
- Positive early warning information refers to the situation where the content of the early warning information is good, such as a significant increase in sales; negative related early warning information refers to whether the content of the early warning information is bad. For example, the rate of bad debts has increased sharply; the neutral warning information situation is where the impact of the event cannot be directly judged, such as the discovery of abnormal points.
- the abnormal points may be fraudulent users, or they may be very potential buying customers.
- Step S106C Use corresponding colors to distinguish the positively related early warning information, the negatively related early warning information and the neutral early warning information for the user to view.
- different colors can be used to distinguish positive-related early-warning information, negative-related early-warning information, and neutral early-warning information.
- the user can know what kind of early-warning information is based on the color, and then choose whether to deal with it immediately.
- the method further includes the step of processing early warning information:
- Step S108 When the target user clicks on the early warning information, monitor the processing operation process of the target user on the early warning information.
- the user can click the warning details to see the specific content of the warning information. If the user feels that this warning message is valuable or helpful, click the "Valuable Warning” button, otherwise, you can click the "Valueless Warning”.
- the early warning information is optimized, and the early warning information that is useful to the user will continue to be paid attention to, and the early warning information that is of no value to the user will be filtered and released and placed in the early warning box (the box saves the used early warning information, The user can view, but the warning prompt will not pop up).
- Step S109 Generate a log file according to the processing operation process, and store the log file.
- FIG. 7 shows a schematic diagram of the program modules of the second embodiment of the intelligent early warning system of the present application.
- the intelligent early warning system 20 may include or be divided into one or more program modules, and the one or more program modules are stored in a storage medium and executed by one or more processors to complete the task. Apply, and realize the above-mentioned intelligent early warning method.
- the program module referred to in the embodiments of the present application refers to a series of computer program instruction segments capable of completing specific functions, and is more suitable for describing the execution process of the intelligent early warning system 20 in the storage medium than the program itself. The following description will specifically introduce the functions of each program module in this embodiment:
- the obtaining module 200 is configured to obtain business scenarios related to the target user and variable indicators related to the business scenarios.
- the variable indicators are variable indicators related to the banking business.
- the loan business scenario includes variable indicators such as loan amount and monthly repayment amount.
- the Telescope application debugging tool the user's operation instructions and request instructions for the banking indicators are monitored, so as to obtain the relevant variable indicators. And the Telescope application debugging tool can mark the variable indicators and related operation data corresponding to each business scenario with business tags, so that users can directly query a series of data related to the business scenario.
- the configuration module 202 is used to configure the variable index parameters of the variable index.
- the relevant variable index data is configured for each business scenario
- the variable index parameter is the preset value of the variable index
- the variable index data of each business scenario is stored In the corresponding bottom table.
- the business indicator parameter is the loan amount of a small enterprise
- the business indicator parameter is xx million
- the corresponding business indicator parameter is xx million.
- the user's bottom table has fields "number of defaulted users" () and “number of all users” (). Now you can configure a "proportion of defaulted users" field through SQL.
- the pseudo code of the SELECT part of the specific SQL is as follows:
- the monitoring module 204 is used to monitor whether the variable index parameter exceeds a preset index range interval.
- monitoring the variable index data includes real-time monitoring of the loan amount data of the loan business scenario in the business scenario, and regular and quantitative monitoring of the loan amount data, such as monitoring weekly, daily, hourly, etc.
- the updated variable index data monitors quantitative variable index data such as the total loan amount this month and the weekly loan amount.
- the generating module 206 is configured to generate early warning information according to a preset natural language description template and the variable index parameter if the variable index parameter exceeds the index range interval, for the user to view.
- the natural language description templates generated according to the different types of early warning variable indicators are different.
- the template can be formulated as: XX variable indicators have risen in the past XX days/weeks/months/quarters/years , Rose XX%.
- the generating module 206 is further configured to:
- variable index parameters Substituting the variable index parameters into a preset natural language description template to generate the early warning information.
- the warning information is generated, if the abnormal variable index parameters are processed first, then the processed variable index parameters are substituted into the preset natural language Language description template.
- the natural language description template is an early warning template preset according to different business scenarios. It is used according to different early warning situations. When using it, you only need to input the business indicator parameters corresponding to the business indicators. For example: when the variable index parameters of the loan amount of the loan business scenario change, and the small and medium-sized enterprise's quota is allocated to the large enterprise, that is, the template: xx enterprise, the loan amount is abnormal, and the loan amount that should be issued is xx. Among them, the actual loan amount issued is xxx, and the actual loan amount is the changed loan amount. Later, the warning information will be given a color warning reminder, so that the user can view it immediately.
- the early warning information is divided into positive related early warning information, negative related early warning information and neutral early warning information.
- variable index parameter changes of different variable indicators are different.
- Positive early warning information refers to the situation where the content of the early warning information is good, such as a significant increase in sales; negative related early warning information refers to whether the content of the early warning information is bad. For example, the rate of bad debts has increased sharply; the neutral warning information situation is where the impact of the event cannot be directly judged, such as the discovery of abnormal points.
- the abnormal points may be fraudulent users, or they may be very potential buying customers.
- the corresponding colors are used to distinguish the positively related early warning information, the negatively related early warning information and the neutral early warning information for the user to view.
- different colors can be used to distinguish positive-related early-warning information, negative-related early-warning information, and neutral early-warning information.
- the user can know what kind of early-warning information is based on the color, and then choose whether to deal with it immediately.
- the computer device 2 is a device that can automatically perform numerical calculation and/or information processing according to pre-set or stored instructions.
- the computer device 2 may be a rack server, a blade server, a tower server, or a cabinet server (including an independent server or a server cluster composed of multiple servers).
- the computer device 2 at least includes, but is not limited to, a memory and a processor.
- the computer equipment may also include a network interface and/or an intelligent early warning system.
- the computer device 2 may include a memory 21, a processor 22, a network interface 23, and an intelligent warning system 20.
- the memory 21, the processor 22, the network interface 23, and the intelligent warning system 20 can be connected to each other through a system bus. in:
- the memory 21 includes at least one type of computer-readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory ( RAM), static random access memory (SRAM), read only memory (ROM), electrically erasable programmable read only memory (EEPROM), programmable read only memory (PROM), magnetic memory, magnetic disks, optical disks, etc.
- the memory 21 may be an internal storage unit of the computer device 2, for example, a hard disk or a memory of the computer device 2.
- the memory 21 may also be an external storage device of the computer device 2, such as a plug-in hard disk, a smart media card (SMC), and a secure digital (Secure Digital, SMC) equipped on the computer device 2. SD) card, flash card (Flash Card), etc.
- the memory 21 may also include both the internal storage unit of the computer device 2 and its external storage device.
- the memory 21 is generally used to store the operating system and various application software installed in the computer device 2, such as the program code of the intelligent early warning system 20 in the second embodiment.
- the memory 21 can also be used to temporarily store various types of data that have been output or will be output.
- the processor 22 may be a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, a microprocessor, or other data processing chips.
- the processor 22 is generally used to control the overall operation of the computer device 2.
- the processor 22 is used to run the program code or process data stored in the memory 21, for example, to run the intelligent early warning system 20, so as to implement the intelligent early warning method of the first embodiment.
- the network interface 23 may include a wireless network interface or a wired network interface, and the network interface 23 is generally used to establish a communication connection between the server 2 and other electronic devices.
- the network interface 23 is used to connect the server 2 to an external terminal through a network, and to establish a data transmission channel and a communication connection between the server 2 and the external terminal.
- the network may be Intranet, Internet, Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G Network, Bluetooth (Bluetooth), Wi-Fi and other wireless or wired networks.
- GSM Global System of Mobile communication
- WCDMA Wideband Code Division Multiple Access
- 4G network Fifth Generation
- 5G Network Fifth Generation
- Bluetooth Bluetooth
- Wi-Fi Wireless Fidelity
- the intelligent early warning system 20 stored in the memory 21 can also be divided into one or more program modules, and the one or more program modules are stored in the memory 21 and composed of one or more program modules. It is executed by two processors (the processor 22 in this embodiment) to complete the application.
- FIG. 7 shows a schematic diagram of program modules for implementing the second embodiment of the intelligent early warning system 20.
- the intelligent early warning system 20 can be divided into an acquisition module 200, a configuration module 202, a monitoring module 204, and a generation module.
- Module 206 the program module referred to in the present application refers to a series of computer program instruction segments that can complete specific functions, and is more suitable than a program to describe the execution process of the intelligent early warning system 20 in the computer device 2.
- the specific functions of the program modules 200-206 have been described in detail in the second embodiment, and will not be repeated here.
- This embodiment also provides a computer-readable storage medium, such as flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), only Readable memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disks, optical disks, servers, App application malls, etc., on which computer programs are stored, The corresponding function is realized when the program is executed by the processor.
- the computer-readable storage medium of this embodiment is used to store the intelligent early warning system 20, and when executed by a processor, the intelligent early warning method of the first embodiment is implemented.
- the storage medium involved in this application such as a computer-readable storage medium, may be non-volatile or volatile.
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Economics (AREA)
- General Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Technology Law (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
一种智能预警方法,包括:获取目标用户相关的业务场景及与所述业务场景相关的变量指标(S100);配置所述变量指标的变量指标参数(S102);监测所述变量指标参数是否超出预设的指标范围区间(S104);若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看(S106)。所述方法可以对变量指标数据进行全方位的监测,提高了异常监测的效率,进而节省了时间。
Description
本申请要求于2020年2月20日提交中国专利局、申请号为202010103422.7,发明名称为“智能预警方法与***”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本申请实施例涉及数据监控领域,尤其涉及一种智能预警方法与***。
各行各业运转时间越久,各种数据与日俱增,如何及时对大数据进行监控,进而做出反馈,是急需解决的问题。发明人发现,目前通用预警方案大致分为两种:第一种基本上依赖于用户手动添加,例如设置业务场景的预警,用户需对每一个业务场景设置预警阈值,手动设置多次,预警触发后则推送至用户客户端。同时设置预警需要依赖于图形本身,如局限于柱状图、折线图等。这种方法大量依赖于行业经验,操作繁琐,触发被动,无法及时识别潜在风险。第二种智能预警则偏向于舆情监听、行业资讯及行业动态获取,采用文本挖掘、知识图谱等技术,着眼于整体大行业变化、舆论倾向等,缺少对大数据的监控,不能根据具体业务情况进行针对性指导。
发明内容
有鉴于此,本申请实施例的目的是提供一种智能预警方法与***,可以对变量指标数据进行全方位的监测,提高了异常监测的效率,进而节省了时间。
为实现上述目的,本申请实施例提供了一种智能预警方法,包括:
获取目标用户相关的业务场景及与所述业务场景相关的变量指标;
配置所述变量指标的变量指标参数;
监测所述变量指标参数是否超出预设的指标范围区间;
若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
为实现上述目的,本申请实施例还提供了一种智能预警***,包括:
获取模块,用于获取目标用户相关的业务场景及与所述业务场景相关的变量指标;
配置模块,用于配置所述变量指标的变量指标参数;
监测模块,用于监测所述变量指标参数是否超出预设的指标范围区间;
生成模块,用于若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板及所述变量指标参数生成预警信息,以供用户进行查看。
为实现上述目的,本申请实施例还提供了一种计算机设备,所述计算机设备包括存储器、处理器,所述存储器上存储有可在所述处理器上计算机程序,所述计算机程序被所述处理器执行时实现如上所述的智能预警方法,该智能预警方法包括以下步骤:
获取目标用户相关的业务场景及与所述业务场景相关的变量指标;
配置所述变量指标的变量指标参数;
监测所述变量指标参数是否超出预设的指标范围区间;
若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
为实现上述目的,本申请实施例还提供了一种计算机可读存储介质,所述计算机可读存储介质内存储有计算机程序,所述计算机程序可被至少一个处理器所执行,以使所述至少一个处理器执行如上所述的智能预警方法,该智能预警方法包括以下步骤:
获取目标用户相关的业务场景及与所述业务场景相关的变量指标;
配置所述变量指标的变量指标参数;
监测所述变量指标参数是否超出预设的指标范围区间;
若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
本申请通过对业务场景下的变量指标对应的变量指标数据进行监测,判断变量指标是否出现异常,进而生成预警信息。可以对变量指标数据进行全方位的监测,提高了异常监测的效率,进而节省了时间。
图1为本申请智能预警方法实施例一的流程图。
图2为本申请实施例一中步骤S104的流程图一。
图3为本申请实施例一图1中步骤S104的流程图二。
图4为本申请实施例一中步骤S104的流程图三。
图5为本申请实施例一图1中步骤S106的流程图。
图6为本申请实施例预警信息处理的流程图。
图7为本申请智能预警***实施例二的程序模块示意图。
图8为本申请计算机设备实施例三的硬件结构示意图。
为了使本申请的目的、技术方案及优点更加清楚明白,以下结合附图及实施例,对本申请进行进一步详细说明。应当理解,此处所描述的具体实施例仅用以解释本申请,并不用于限定本申请。基于本申请中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本申请保护的范围。
本申请的技术方案涉及人工智能技术领域,可应用于金融业务场景,以实现智能预警。可选的,本申请涉及的数据如变量指标、指标范围区间、模板和/或预警信息等可存储于数据库中,或者可以存储于区块链中,比如通过区块链分布式存储,本申请不做限定。
实施例一
参阅图1,示出了本申请实施例一之智能预警方法的步骤流程图。可以理解,本方法实施例中的流程图不用于对执行步骤的顺序进行限定。下面以计算机设备2为执行主体进行示例性描述。具体如下。
步骤S100,获取目标用户相关的业务场景及与所述业务场景相关的变量指标。
具体的,获取用户提供的业务场景及与业务场景相关的指标集合,变量指标是与银行业务相关的变量指标,例如贷款业务场景下包括贷款量、每月还款量等变量指标。通过Telescope应用调试工具,监控用户对银行业务的指标的进行的操作指令和请求指令,从而得到相关的变量指标。并且Telescope应用调试工具可以将各个业务场景对应的变量指标及相关的操作数据进行业务标签标注,以便用户可以直接查询到业务场景相关的一系列数据。
步骤S102,配置所述变量指标的变量指标参数。
具体地,根据用户提供的业务场景及与业务场景相关的指标集合,为每个业务场景配置相关的变量指标数据,变量指标参数为变量指标的预设值,每个业务场景的变量指标数据存储于相应的底表里。例如:业务场景为贷款时,业务指标参数为小企业贷款金额,业务指标参数即为xx万;若业务指标为大企业贷款金额,则相应的业务指标参数为xx千万。通过调用SQL语句在底表上进行SELECT,选择与业务场景相关的变量指标的字段获取变量指标参数。例如用户的底表有字段“违约用户数”()和“所有用户数”(),现在可以通过SQL配置一个“违约用户比例”的字段。具体SQL的SELECT部分伪代码如下:
SELECT“违约用户数”/“所有用户数”AS“违约用户比例”FROM Table。
步骤S104,监测所述变量指标参数是否超出预设的指标范围区间。
具体的,监测所述变量指标数据包括实时监测所述业务场景中的放款业务场景的放款量数据,对所述放款量数据进行定时与定量的监测,例如监测每周、每天、每时等定时更 新的变量指标数据,监测本月总放款量、周放款量等定量的变量指标数据。
示例性地,在一实施方式中,参阅图2,所述步骤S104包括:
步骤S104A0,获取所述业务场景及所述变量指标的历史时间序列。
具体地,获取每个业务场景下,各个变量指标的历史时间序列
步骤S104A2,通过预设的时间序列分解算法将所述历史时间序列进行分解,得到所述历史时间序列每个时间点的余项分量。
具体地,时间序列分解算法(Seasonal-Trend decomposition procedure based on Loess,STL)的计算公式为:Yv=Tv+Sv+Rv;v=1,…,N;其中,Yv表示预设时间N内的还款时间序列,Tv、Sv、Rv为时间序列分解算法的趋势分量、周期分量与余项分量。STL分为内循环与外循环,其中内循环主要做了趋势拟合与周期分量的计算,通过外循环对余项分量进行收敛计算。异常值检测算法进行分析之前,利用S-ESD(Seasonal ESD)算法,通过用中位数替换掉趋势分量;余项计算公式如下:
RX=X-SX-X′;
其中,X为历史时间序列,SX为STL分解后的周期分量,X′为X的中位数。
步骤S104A4,根据预设的异常值检测算法分析所述每个时间点的余项分量,以得到异常时间点。
具体地,预先构建异常值检测算法的步骤包括:构建N棵二叉树,为每棵树随机做无放回采样生成训练集。二叉树是随机二叉树,给定时间序列,假设时间序列内所有属性都是连续型变量,二叉树构造过程可以描述为:首先随机选择时间序列的一个属性A,然后随机选择属性A中的一个值V,按照属性A的值对时间序列进行树的***,将小于V的记录放在左分支上,把大于V的记录放在右分支上,然后按上述过程递归构建树,直到满足如下条件:传入的时间序列只有一条记录或者多条一样的记录;树的高度达到了限定高度。进行预测:预测的过程就是把时间序列在二叉树上沿对应的分支往下走,走到达到叶子节点,并记录着过程中经过的路径长度c(n)。将c(n)带入到异常值评分函数中,得到异常值分数,计算公式如下:
c(n)=2H(n-1)-(2(n-1)/n),其中H(k)=ln(k)+§,§为欧拉常数,
s(x,n)就是记录c(n)在由n个样本的时间序列构成的二叉树的异常指数,s(x,n)取值范围为[0,1],越接近1表示是异常点的可能性高,越接近0表示是正常点的可能性比较高,如果大部分的时间序列的s(x,n)都接近于0.5,说明整个时间序列都没有明显的异常值。
步骤S104A6,判断所述异常时间点是否在所述指标范围区间内。
具体地,对异常时间点对应的变量指标参数进行监控,以查询异常时间点是否发生异常事件。
示例性地,所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:
若所述异常时间点超出所述指标范围区间,则根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
具体地,若异常时间点在所述指标范围区间内,说明无异常,若异常时间点超出所述指标范围区间,说明异常。将异常指数异常的余弦分量对应的变量指标参数,按照预设的自然语言描述模板生成预警信息。
示例性地,在另一实施方式中,参阅图3,所述步骤S104包括:
步骤S104B0,获取多个目标用户的变量指标的变量指标参数。
具体地,获取变量指标在一段时间内的变量指标参数,可以是多个目标用户的某个变 量指标,也可以是多个目标用户的多个变量指标。例如:借贷业务场景下,多个目标用户在一个月内的消费金额。
步骤S104B2,从所述多个目标用户的变量指标参数中任意选取一个目标用户的变量指标参数作为聚类中心。
具体地,任意选择一个目标用户的变量指标参数作为聚类中心。
步骤S104B4,计算其他用户的变量指标参数到所述聚类中心的距离。
具体地,可以通过欧拉公式计算其他用户的变量指标参数到聚类中心的距离,也可通过聚类算法对聚类中心及其他用户的变量指标参数进行分析,聚类算法可以为DBSCAN算法、WSN算法等。
步骤S104B6,监测所述距离是否超出预设的指标范围区间。
具体地,监测距离是否超出预设的指标范围区间,例如消费金额,若多个有目标用户在一个月内的消费金额超出预设范围,说明该目标用户存在骗贷行为。
示例性地,所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:
若所述距离在预设范围外的变量指标参数,则将变量指标参数作为异常数据,根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
具体地,对聚类中心点及其他用户的变量指标参数进行判断后,将点分为核心点/边界点/噪声点。核心点为距离e内至少包含m个点数的数据点(可以为预设范围内的点),边界点是核心点距离e内的邻近点,但包含的点数小于m,其他剩下的都是噪声点,即异常时间点。将异常时间点对应的变量指标参数及对应的目标用户,按照预设的自然语言描述模板生成预警信息。
示例性地,在另一实施方式中,参阅图4,所述步骤S104包括:
步骤S104C0,获取所述变量指标的变量指标参数的历史时间序列数据。
具体地,历史时间序列数据为变量指标参数按照时间序列得到的历史数据。
步骤S104C1,检验所述历史时间序列数据的平稳性。
具体地,通过检验历史时间序列数据的图形,可以粗略得到历史时间序列数据的平稳性。
步骤S104C2,对所述历史时间序列数据中非平稳的历史时间序列数据做差分计算,得到差分序列。
具体地,将非平稳的历史时间序列数据进行差分计算,使其平稳。如果一个原始系列平稳,我们称之为I(0)过程;如果原始系列不平稳但经过一阶差分平稳,我们称系列为I(1)过程;同样系列经过n次差分才平稳,则称系列为I(n)过程。
步骤S104C3,计算所述差分序列的自相关系数和平均移动系数。
具体地,自相关系数和平均移动系数用于验证差分序列的数据的平稳性。由于差分序列可以是随机变量序列,对于任意t,E(ε
t)=0,
则称差分序列{y
t}服从p阶自回归模型,记为AR(p),等同于ARIMA(p,0,0)。
称为自回归系数。如果差分序列{y
t}满足:y
t=ε
t-θ
1ε
t-1-…-θ
qε
t-q,则称差分序列{y
t}服从q阶移动平均模型,记为MA(q),等同于ARIMA(0,0,q)。θ
1,…,θ
p称为移动平均系数。将差分序列的自相关系数和平均移动系数记为(p,q)。
步骤S104C4,根据所述自相关系数和所述平均移动系数对所述差分序列进行残 差和白噪声处理,以得到时间序列相关模型。
具体地,将差分序列中有用的信息进行提取,剩下的全是随机扰动,是无法预测和使用的,差分序列如果通过了白噪声检验,则建模(时间序列相关模型)就可以终止了,因为没有信息可以继续提取。如果残差不是白噪声,就说明残差中还有有用的信息,需要修改模型或者进一步提取。
步骤S104C5,通过预设的时间序列相关模型对待测时间序列数据进行预测,得到预测数据。
具体地,预设的时间序列相关模型可以为ARIMA等时间序列相关模型,利用历史时间序列数据本身建立模型,以研究事物发展自身的规律。将处理后的差分序列输入至预设的时间序列相关模型,得到预测数据。
步骤S104C6,监测所述预测数据是否超出预设的指标范围区间。
具体地,可以对变量指标数据进行监测,若变量指标数据超出预测数据的指标范围区间,则表示出现了异常的变量指标数据。
示例性地,所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:
若所述距离在预设范围外的变量指标参数,则将变量指标参数作为异常数据,根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
具体地,预测数据超出指标范围区间,则在预设的自然语言描述模板中找到对应的自然语言描述模板,将预测数据输入至对应的自然语言描述模板以生成预警信息。
步骤S106,若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
具体地,根据预警的变量指标的类型不同生成的自然语言描述模板不同,例如对于时间序列的连续型变量,可以制定模版为:XX变量指标在过去XX日/周/月/季度/年中上涨,上涨XX%。
示例性地,参阅图5,步骤S106还包括:
步骤S106A,将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息。
具体地,根据异常的变量指标参数代入至预设的自然语言描述模版,生成所述预警信息,若要先对异常的变量指标参数处理,再将处理后的变量指标参数代入至预设的自然语言描述模版。自然语言描述模版是根据不同的业务场景预设的预警模版,根据不同的预警情况进行使用,使用时,只需要将业务指标对应的业务指标参数输入即可。例如:当放款业务场景的放款量的变量指标参数出现变化,给中小企业放了大企业的额度,即调用模版:xx企业,放款量异常,应发放的放款量为xx。其中,实际发放的放款量为xxx,实际放款量为变化后的放款量。后续再给该预警信息进行颜色警示提醒,以使用户立即查看。
步骤S106B,将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息。
具体地,不同变量指标的变量指标参数变化是不同的,正相关预警信息指的是预警信息的内容是好的情况,例如销售额大增;负相关预警信息指的是预警信息的内容是坏的情况,例如坏账率大增;中立预警信息情况就是一些无法直接判断事件发生影响好坏的,例如异常点发现,异常点可能是欺诈用户,也可能是极具潜力的购买客户。
步骤S106C,采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
具体地,可以用不同的颜色区分正相关预警信息、负相关预警信息与中立预警信息,用户可以根据颜色知道是那种预警信息,进而选择是否马上处理。
示例性地,参阅图6,所述方法还包括预警信息处理的步骤:
步骤S108,当所述目标用户点击所述预警信息时,监控所述目标用户对所述预警信息的处理操作过程。
具体地,当预警信息出现后,用户点击预警详情,可以看到预警信息的具体内容。若用户觉得这条预警信息是有价值或者用帮助的,点击“有价值的预警”按钮,反之可以点击“无价值的预警”。接受用户的处理操作过程后,对预警信息进行优化,对用户觉得有用的预警信息会持续关注,对用户觉得无价值的预警信息会过滤放行,放在预警盒子中(盒子中保存所用预警信息,用户可以查看,但是不会弹出预警提示)。
步骤S109,根据所述处理操作过程生成日志文件,并存储所述日志文件。
具体地,将处理操作过程生成日志文件可以方便知道,以前针对该情况怎样进行操作,作为参考。存储日志文件后,方便后续进行处理。
实施例二
请继续参阅图7,示出了本申请智能预警***实施例二的程序模块示意图。在本实施例中,智能预警***20可以包括或被分割成一个或多个程序模块,一个或者多个程序模块被存储于存储介质中,并由一个或多个处理器所执行,以完成本申请,并可实现上述智能预警方法。本申请实施例所称的程序模块是指能够完成特定功能的一系列计算机程序指令段,比程序本身更适合于描述智能预警***20在存储介质中的执行过程。以下描述将具体介绍本实施例各程序模块的功能:
获取模块200,用于获取目标用户相关的业务场景及与所述业务场景相关的变量指标。
具体的,获取用户提供的业务场景及与业务场景相关的指标集合,变量指标是与银行业务相关的变量指标,例如贷款业务场景下包括贷款量、每月还款量等变量指标。通过Telescope应用调试工具,监控用户对银行业务的指标的进行的操作指令和请求指令,从而得到相关的变量指标。并且Telescope应用调试工具可以将各个业务场景对应的变量指标及相关的操作数据进行业务标签标注,以便用户可以直接查询到业务场景相关的一系列数据。
配置模块202,用于配置所述变量指标的变量指标参数。
具体的,根据用户提供的业务场景及与业务场景相关的指标集合,为每个业务场景配置相关的变量指标数据,变量指标参数为变量指标的预设值,每个业务场景的变量指标数据存储于相应的底表里。例如:业务场景为贷款时,业务指标参数为小企业贷款金额,业务指标参数即为xx万;若业务指标为大企业贷款金额,则相应的业务指标参数为xx千万。通过调用SQL语句在底表上进行SELECT,选择与业务场景相关的变量指标的字段获取变量指标参数。例如用户的底表有字段“违约用户数”()和“所有用户数”(),现在可以通过SQL配置一个“违约用户比例”的字段。具体SQL的SELECT部分伪代码如下:
SELECT“违约用户数”/“所有用户数”AS“违约用户比例”FROM Table。
监测模块204,用于监测所述变量指标参数是否超出预设的指标范围区间。
具体的,监测所述变量指标数据包括实时监测所述业务场景中的放款业务场景的放款量数据,对所述放款量数据进行定时与定量的监测,例如监测每周、每天、每时等定时更新的变量指标数据,监测本月总放款量、周放款量等定量的变量指标数据。
生成模块206,用于若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板及所述变量指标参数生成预警信息,以供用户进行查看。
具体地,根据预警的变量指标的类型不同生成的自然语言描述模板不同,例如对于时间序列的连续型变量,可以制定模版为:XX变量指标在过去XX日/周/月/季度/年中上涨,上涨XX%。
示例性地,所述生成模块206还用于:
将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息。
具体地,根据异常的变量指标参数代入至预设的自然语言描述模版,生成所述预警信 息,若要先对异常的变量指标参数处理,再将处理后的变量指标参数代入至预设的自然语言描述模版。自然语言描述模版是根据不同的业务场景预设的预警模版,根据不同的预警情况进行使用,使用时,只需要将业务指标对应的业务指标参数输入即可。例如:当放款业务场景的放款量的变量指标参数出现变化,给中小企业放了大企业的额度,即调用模版:xx企业,放款量异常,应发放的放款量为xx。其中,实际发放的放款量为xxx,实际放款量为变化后的放款量。后续再给该预警信息进行颜色警示提醒,以使用户立即查看。
将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息。
具体地,不同变量指标的变量指标参数变化是不同的,正相关预警信息指的是预警信息的内容是好的情况,例如销售额大增;负相关预警信息指的是预警信息的内容是坏的情况,例如坏账率大增;中立预警信息情况就是一些无法直接判断事件发生影响好坏的,例如异常点发现,异常点可能是欺诈用户,也可能是极具潜力的购买客户。
采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
具体地,可以用不同的颜色区分正相关预警信息、负相关预警信息与中立预警信息,用户可以根据颜色知道是那种预警信息,进而选择是否马上处理。
实施例三
参阅图8,是本申请实施例三之计算机设备的硬件架构示意图。本实施例中,所述计算机设备2是一种能够按照事先设定或者存储的指令,自动进行数值计算和/或信息处理的设备。该计算机设备2可以是机架式服务器、刀片式服务器、塔式服务器或机柜式服务器(包括独立的服务器,或者多个服务器所组成的服务器集群)等。如图8所示,所述计算机设备2至少包括,但不限于,存储器和处理器。可选的,该计算机设备还可包括网络接口和/或智能预警***。例如,计算机设备2可包括存储器21、处理器22、网络接口23、以及智能预警***20,如可通过***总线相互通信连接存储器21、处理器22、网络接口23、以及智能预警***20。其中:
本实施例中,存储器21至少包括一种类型的计算机可读存储介质,所述可读存储介质包括闪存、硬盘、多媒体卡、卡型存储器(例如,SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘等。在一些实施例中,存储器21可以是计算机设备2的内部存储单元,例如该计算机设备2的硬盘或内存。在另一些实施例中,存储器21也可以是计算机设备2的外部存储设备,例如该计算机设备2上配备的插接式硬盘,智能存储卡(Smart Media Card,SMC),安全数字(Secure Digital,SD)卡,闪存卡(Flash Card)等。当然,存储器21还可以既包括计算机设备2的内部存储单元也包括其外部存储设备。本实施例中,存储器21通常用于存储安装于计算机设备2的操作***和各类应用软件,例如实施例二的智能预警***20的程序代码等。此外,存储器21还可以用于暂时地存储已经输出或者将要输出的各类数据。
处理器22在一些实施例中可以是中央处理器(Central Processing Unit,CPU)、控制器、微控制器、微处理器、或其他数据处理芯片。该处理器22通常用于控制计算机设备2的总体操作。本实施例中,处理器22用于运行存储器21中存储的程序代码或者处理数据,例如运行智能预警***20,以实现实施例一的智能预警方法。
所述网络接口23可包括无线网络接口或有线网络接口,该网络接口23通常用于在所述服务器2与其他电子装置之间建立通信连接。例如,所述网络接口23用于通过网络将所述服务器2与外部终端相连,在所述服务器2与外部终端之间的建立数据传输通道和通信连接等。所述网络可以是企业内部网(Intranet)、互联网(Internet)、全球移动通讯***(Global System of Mobile communication,GSM)、宽带码分多址(Wideband Code Division Multiple Access,WCDMA)、4G网络、5G网络、蓝牙(Bluetooth)、Wi-Fi等无线或有线网络。需要指出的是,图8仅示出了具有部件20-23的计算机设备2,但是应理解的是,并不要求实施所有示出的部件,可以替代的实施更多或者更少的部件。
在本实施例中,存储于存储器21中的所述智能预警***20还可以被分割为一个或者多个程序模块,所述一个或者多个程序模块被存储于存储器21中,并由一个或多个处理器(本实施例为处理器22)所执行,以完成本申请。
例如,图7示出了所述实现智能预警***20实施例二的程序模块示意图,该实施例中,所述智能预警***20可以被划分为获取模块200、配置模块202、监测模块204与生成模块206。其中,本申请所称的程序模块是指能够完成特定功能的一系列计算机程序指令段,比程序更适合于描述所述智能预警***20在所述计算机设备2中的执行过程。所述程序模块200-206的具体功能在实施例二中已有详细描述,在此不再赘述。
实施例四
本实施例还提供一种计算机可读存储介质,如闪存、硬盘、多媒体卡、卡型存储器(例如,SD或DX存储器等)、随机访问存储器(RAM)、静态随机访问存储器(SRAM)、只读存储器(ROM)、电可擦除可编程只读存储器(EEPROM)、可编程只读存储器(PROM)、磁性存储器、磁盘、光盘、服务器、App应用商城等等,其上存储有计算机程序,程序被处理器执行时实现相应功能。本实施例的计算机可读存储介质用于存储智能预警***20,被处理器执行时实现实施例一的智能预警方法。
可选的,本申请涉及的存储介质如计算机可读存储介质可以是非易失性的,也可以是易失性的。
上述本申请实施例序号仅仅为了描述,不代表实施例的优劣。
通过以上的实施方式的描述,本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现,当然也可以通过硬件,但很多情况下前者是更佳的实施方式。
以上仅为本申请的优选实施例,并非因此限制本申请的专利范围,凡是利用本申请说明书及附图内容所作的等效结构或等效流程变换,或直接或间接运用在其他相关的技术领域,均同理包括在本申请的专利保护范围内。
Claims (20)
- 一种智能预警方法,其中,包括:获取目标用户相关的业务场景及与所述业务场景相关的变量指标;配置所述变量指标的变量指标参数;监测所述变量指标参数是否超出预设的指标范围区间;若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
- 根据权利要求1所述的智能预警方法,其中,所述监测所述变量指标参数是否超出预设的指标范围区间包括:获取所述业务场景及所述变量指标的历史时间序列;通过预设的时间序列分解算法将所述历史时间序列进行分解,得到所述历史时间序列每个时间点的余项分量;根据预设的异常值检测算法分析所述每个时间点的余项分量,以得到异常时间点;判断所述异常时间点是否在所述指标范围区间内;所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:若所述异常时间点超出所述指标范围区间,则根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求1所述的智能预警方法,其中,所述监测所述变量指标参数是否超出预设的指标范围区间包括:获取多个目标用户的变量指标的变量指标参数;从所述多个目标用户的变量指标参数中任意选取一个目标用户的变量指标参数作为聚类中心;计算其他用户的变量指标参数到所述聚类中心的距离;监测所述距离是否超出预设的指标范围区间;所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:若所述距离在预设范围外的变量指标参数,则将变量指标参数作为异常数据,根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求1所述的智能预警方法,其中,所述监测所述变量指标参数是否超出预设的指标范围区间包括:获取所述变量指标的变量指标参数的历史时间序列数据;检验所述历史时间序列数据的平稳性;对所述历史时间序列数据中非平稳的历史时间序列数据做差分计算,得到差分序列;计算所述差分序列的自相关系数和平均移动系数;根据所述自相关系数和所述平均移动系数对所述差分序列进行残差和白噪声处理,以得到时间序列相关模型;通过所述时间序列相关模型对待测时间序列数据进行预测,得到预测数据;监测所述预测数据是否超出预设的指标范围区间;所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看包括:若所述预测数据超出所述指标范围区间,则根据预设的自然语言描述模板以及所述预测数据生成预警信息,以供用户进行查看。
- 根据权利要求1所述的智能预警方法,其中,所述根据预设的自然语言描述模版以 及所述变量指标参数并生成预警信息,以供用户进行查看包括:将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息;将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息;采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
- 根据权利要求1所述的智能预警方法,其中,所述方法还包括:当所述目标用户点击所述预警信息时,监控所述目标用户对所述预警信息的处理操作过程;根据所述处理操作过程生成日志文件,并存储所述日志文件。
- 一种智能预警***,其中,包括:获取模块,用于获取目标用户相关的业务场景及与所述业务场景相关的变量指标;配置模块,用于配置所述变量指标的变量指标参数;监测模块,用于监测所述变量指标参数是否超出预设的指标范围区间;生成模块,用于若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板及所述变量指标参数生成预警信息,以供用户进行查看。
- 根据权利要求7所述的智能预警***,其中,所述生成模块还用于:将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息;将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息;采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
- 一种计算机设备,其中,所述计算机设备包括存储器、处理器,所述存储器上存储有可在所述处理器上运行的计算机程序,所述计算机程序被所述处理器执行时实现智能预警方法,所述智能预警方法包括以下步骤:获取目标用户相关的业务场景及与所述业务场景相关的变量指标;配置所述变量指标的变量指标参数;监测所述变量指标参数是否超出预设的指标范围区间;若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
- 根据权利要求9所述的计算机设备,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取所述业务场景及所述变量指标的历史时间序列;通过预设的时间序列分解算法将所述历史时间序列进行分解,得到所述历史时间序列每个时间点的余项分量;根据预设的异常值检测算法分析所述每个时间点的余项分量,以得到异常时间点;判断所述异常时间点是否在所述指标范围区间内;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述异常时间点超出所述指标范围区间,则根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求9所述的计算机设备,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取多个目标用户的变量指标的变量指标参数;从所述多个目标用户的变量指标参数中任意选取一个目标用户的变量指标参数作为聚类中心;计算其他用户的变量指标参数到所述聚类中心的距离;监测所述距离是否超出预设的指标范围区间;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述距离在预设范围外的变量指标参数,则将变量指标参数作为异常数据,根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求9所述的计算机设备,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取所述变量指标的变量指标参数的历史时间序列数据;检验所述历史时间序列数据的平稳性;对所述历史时间序列数据中非平稳的历史时间序列数据做差分计算,得到差分序列;计算所述差分序列的自相关系数和平均移动系数;根据所述自相关系数和所述平均移动系数对所述差分序列进行残差和白噪声处理,以得到时间序列相关模型;通过所述时间序列相关模型对待测时间序列数据进行预测,得到预测数据;监测所述预测数据是否超出预设的指标范围区间;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述预测数据超出所述指标范围区间,则根据预设的自然语言描述模板以及所述预测数据生成预警信息,以供用户进行查看。
- 根据权利要求9所述的计算机设备,其中,执行所述根据预设的自然语言描述模版以及所述变量指标参数并生成预警信息,以供用户进行查看,包括:将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息;将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息;采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
- 根据权利要求9所述的计算机设备,其中,所述处理器执行所述智能预警方法时,还包括:当所述目标用户点击所述预警信息时,监控所述目标用户对所述预警信息的处理操作过程;根据所述处理操作过程生成日志文件,并存储所述日志文件。
- 一种计算机可读存储介质,其中,所述计算机可读存储介质内存储有计算机程序,所述计算机程序可被至少一个处理器所执行,以使所述至少一个处理器执行智能预警方法,所述智能预警方法包括以下步骤:获取目标用户相关的业务场景及与所述业务场景相关的变量指标;配置所述变量指标的变量指标参数;监测所述变量指标参数是否超出预设的指标范围区间;若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看。
- 根据权利要求15所述的计算机可读存储介质,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取所述业务场景及所述变量指标的历史时间序列;通过预设的时间序列分解算法将所述历史时间序列进行分解,得到所述历史时间序列每个时间点的余项分量;根据预设的异常值检测算法分析所述每个时间点的余项分量,以得到异常时间点;判断所述异常时间点是否在所述指标范围区间内;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述异常时间点超出所述指标范围区间,则根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求15所述的计算机可读存储介质,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取多个目标用户的变量指标的变量指标参数;从所述多个目标用户的变量指标参数中任意选取一个目标用户的变量指标参数作为聚类中心;计算其他用户的变量指标参数到所述聚类中心的距离;监测所述距离是否超出预设的指标范围区间;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述距离在预设范围外的变量指标参数,则将变量指标参数作为异常数据,根据预设的自然语言描述模板以及所述异常时间点生成预警信息,以供用户进行查看。
- 根据权利要求15所述的计算机可读存储介质,其中,执行所述监测所述变量指标参数是否超出预设的指标范围区间,包括:获取所述变量指标的变量指标参数的历史时间序列数据;检验所述历史时间序列数据的平稳性;对所述历史时间序列数据中非平稳的历史时间序列数据做差分计算,得到差分序列;计算所述差分序列的自相关系数和平均移动系数;根据所述自相关系数和所述平均移动系数对所述差分序列进行残差和白噪声处理,以得到时间序列相关模型;通过所述时间序列相关模型对待测时间序列数据进行预测,得到预测数据;监测所述预测数据是否超出预设的指标范围区间;执行所述若所述变量指标参数超出所述指标范围区间,则根据预设的自然语言描述模板以及所述变量指标参数生成预警信息,以供用户进行查看,包括:若所述预测数据超出所述指标范围区间,则根据预设的自然语言描述模板以及所述预测数据生成预警信息,以供用户进行查看。
- 根据权利要求15所述的计算机可读存储介质,其中,执行所述根据预设的自然语言描述模版以及所述变量指标参数并生成预警信息,以供用户进行查看,包括:将所述变量指标参数代入至预设的自然语言描述模版,以生成所述预警信息;将所述预警信息分成正相关预警信息、负相关预警信息与中立预警信息;采用对应的颜色对所述正相关预警信息、所述负相关预警信息与所述中立预警信息进行区分,以供用户进行查看。
- 根据权利要求15所述的计算机可读存储介质,其中,所述处理器执行所述智能预警方法时,还包括:当所述目标用户点击所述预警信息时,监控所述目标用户对所述预警信息的处理操作过程;根据所述处理操作过程生成日志文件,并存储所述日志文件。
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010103422.7 | 2020-02-20 | ||
CN202010103422.7A CN111325466A (zh) | 2020-02-20 | 2020-02-20 | 智能预警方法与*** |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2021164465A1 true WO2021164465A1 (zh) | 2021-08-26 |
Family
ID=71168889
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2021/071217 WO2021164465A1 (zh) | 2020-02-20 | 2021-01-12 | 智能预警方法与*** |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN111325466A (zh) |
WO (1) | WO2021164465A1 (zh) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114675754A (zh) * | 2022-01-27 | 2022-06-28 | 浪潮(山东)计算机科技有限公司 | 一种鼠标控制方法、装置、设备及存储介质 |
CN115423385A (zh) * | 2022-11-04 | 2022-12-02 | 碳管家智能云平台有限公司 | 一种能耗双控管理方法、设备及介质 |
CN115471141A (zh) * | 2022-11-02 | 2022-12-13 | 成都飞机工业(集团)有限责任公司 | 一种业务流程周期管控方法、装置、设备及介质 |
CN115499289A (zh) * | 2022-08-17 | 2022-12-20 | 华电电力科学研究院有限公司 | 一种设备状态评估预警方法和*** |
CN115907830A (zh) * | 2022-12-22 | 2023-04-04 | 北京领雁科技股份有限公司 | 基于指标预警的策略执行方法、装置、设备及可读介质 |
CN116520755A (zh) * | 2023-06-29 | 2023-08-01 | 深圳东原电子有限公司 | 一种音响的自动化产线监测预警方法及*** |
CN116840600A (zh) * | 2023-07-05 | 2023-10-03 | 河北久维电子科技有限公司 | 设备异常告警方法及变电站辅助***综合监控联动平台 |
CN116860562A (zh) * | 2023-09-04 | 2023-10-10 | 湖南中青能科技有限公司 | 一种用于数据中台数据质量的监控方法及*** |
CN117791626A (zh) * | 2024-02-23 | 2024-03-29 | 广东佰林电气设备厂有限公司 | 一种智能综合电力箱电力供给优化方法 |
CN117827684A (zh) * | 2024-03-05 | 2024-04-05 | 数翊科技(北京)有限公司武汉分公司 | 数据计算作业的预警方法、装置、电子设备及存储介质 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111325466A (zh) * | 2020-02-20 | 2020-06-23 | 深圳壹账通智能科技有限公司 | 智能预警方法与*** |
CN111754123B (zh) * | 2020-06-28 | 2023-06-02 | 深圳壹账通智能科技有限公司 | 数据监控方法、装置、计算机设备及存储介质 |
CN113821416A (zh) * | 2021-09-18 | 2021-12-21 | 中国电信股份有限公司 | 监测告警方法、装置、存储介质及电子设备 |
CN114647555B (zh) * | 2022-05-13 | 2022-09-02 | 太平金融科技服务(上海)有限公司 | 基于多业务***的数据预警方法、装置、设备和介质 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106844138A (zh) * | 2016-12-14 | 2017-06-13 | 北京奇艺世纪科技有限公司 | 运维报警***及方法 |
CN108197011A (zh) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | 一种基于人工智能大数据平台的单指标预测和预警方法 |
CN108595300A (zh) * | 2018-03-21 | 2018-09-28 | 北京奇艺世纪科技有限公司 | 一种可配置的监控和报警的方法及装置 |
CN109191283A (zh) * | 2018-08-30 | 2019-01-11 | 成都数联铭品科技有限公司 | 风险预警方法及*** |
CN110083803A (zh) * | 2019-04-22 | 2019-08-02 | 水利部信息中心 | 基于时间序列arima模型取水异常检测方法与*** |
CN110309125A (zh) * | 2019-06-24 | 2019-10-08 | 招商局金融科技有限公司 | 数据校验方法、电子装置及存储介质 |
WO2020007198A1 (zh) * | 2018-07-03 | 2020-01-09 | 南京邮电大学 | 一种车联网中基于d2d通信的紧急消息传输方法 |
CN111325466A (zh) * | 2020-02-20 | 2020-06-23 | 深圳壹账通智能科技有限公司 | 智能预警方法与*** |
-
2020
- 2020-02-20 CN CN202010103422.7A patent/CN111325466A/zh active Pending
-
2021
- 2021-01-12 WO PCT/CN2021/071217 patent/WO2021164465A1/zh active Application Filing
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106844138A (zh) * | 2016-12-14 | 2017-06-13 | 北京奇艺世纪科技有限公司 | 运维报警***及方法 |
CN108197011A (zh) * | 2018-01-29 | 2018-06-22 | 上海洞识信息科技有限公司 | 一种基于人工智能大数据平台的单指标预测和预警方法 |
CN108595300A (zh) * | 2018-03-21 | 2018-09-28 | 北京奇艺世纪科技有限公司 | 一种可配置的监控和报警的方法及装置 |
WO2020007198A1 (zh) * | 2018-07-03 | 2020-01-09 | 南京邮电大学 | 一种车联网中基于d2d通信的紧急消息传输方法 |
CN109191283A (zh) * | 2018-08-30 | 2019-01-11 | 成都数联铭品科技有限公司 | 风险预警方法及*** |
CN110083803A (zh) * | 2019-04-22 | 2019-08-02 | 水利部信息中心 | 基于时间序列arima模型取水异常检测方法与*** |
CN110309125A (zh) * | 2019-06-24 | 2019-10-08 | 招商局金融科技有限公司 | 数据校验方法、电子装置及存储介质 |
CN111325466A (zh) * | 2020-02-20 | 2020-06-23 | 深圳壹账通智能科技有限公司 | 智能预警方法与*** |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114675754A (zh) * | 2022-01-27 | 2022-06-28 | 浪潮(山东)计算机科技有限公司 | 一种鼠标控制方法、装置、设备及存储介质 |
CN115499289A (zh) * | 2022-08-17 | 2022-12-20 | 华电电力科学研究院有限公司 | 一种设备状态评估预警方法和*** |
CN115499289B (zh) * | 2022-08-17 | 2023-08-25 | 华电电力科学研究院有限公司 | 一种设备状态评估预警方法和*** |
CN115471141A (zh) * | 2022-11-02 | 2022-12-13 | 成都飞机工业(集团)有限责任公司 | 一种业务流程周期管控方法、装置、设备及介质 |
CN115471141B (zh) * | 2022-11-02 | 2023-03-24 | 成都飞机工业(集团)有限责任公司 | 一种业务流程周期管控方法、装置、设备及介质 |
CN115423385A (zh) * | 2022-11-04 | 2022-12-02 | 碳管家智能云平台有限公司 | 一种能耗双控管理方法、设备及介质 |
CN115907830A (zh) * | 2022-12-22 | 2023-04-04 | 北京领雁科技股份有限公司 | 基于指标预警的策略执行方法、装置、设备及可读介质 |
CN115907830B (zh) * | 2022-12-22 | 2023-10-13 | 北京领雁科技股份有限公司 | 基于指标预警的策略执行方法、装置、设备及可读介质 |
CN116520755B (zh) * | 2023-06-29 | 2023-09-26 | 深圳东原电子有限公司 | 一种音响的自动化产线监测预警方法及*** |
CN116520755A (zh) * | 2023-06-29 | 2023-08-01 | 深圳东原电子有限公司 | 一种音响的自动化产线监测预警方法及*** |
CN116840600A (zh) * | 2023-07-05 | 2023-10-03 | 河北久维电子科技有限公司 | 设备异常告警方法及变电站辅助***综合监控联动平台 |
CN116840600B (zh) * | 2023-07-05 | 2024-01-16 | 河北久维电子科技有限公司 | 设备异常告警方法及变电站辅助***综合监控联动平台 |
CN116860562A (zh) * | 2023-09-04 | 2023-10-10 | 湖南中青能科技有限公司 | 一种用于数据中台数据质量的监控方法及*** |
CN116860562B (zh) * | 2023-09-04 | 2023-11-24 | 湖南中青能科技有限公司 | 一种用于数据中台数据质量的监控方法及*** |
CN117791626A (zh) * | 2024-02-23 | 2024-03-29 | 广东佰林电气设备厂有限公司 | 一种智能综合电力箱电力供给优化方法 |
CN117791626B (zh) * | 2024-02-23 | 2024-04-30 | 广东佰林电气设备厂有限公司 | 一种智能综合电力箱电力供给优化方法 |
CN117827684A (zh) * | 2024-03-05 | 2024-04-05 | 数翊科技(北京)有限公司武汉分公司 | 数据计算作业的预警方法、装置、电子设备及存储介质 |
CN117827684B (zh) * | 2024-03-05 | 2024-05-24 | 数翊科技(北京)有限公司武汉分公司 | 数据计算作业的预警方法、装置、电子设备及存储介质 |
Also Published As
Publication number | Publication date |
---|---|
CN111325466A (zh) | 2020-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2021164465A1 (zh) | 智能预警方法与*** | |
US11847574B2 (en) | Systems and methods for enriching modeling tools and infrastructure with semantics | |
CN110033314B (zh) | 广告数据处理方法及装置 | |
US8014983B2 (en) | Computer-implemented system and method for storing data analysis models | |
US9047559B2 (en) | Computer-implemented systems and methods for testing large scale automatic forecast combinations | |
CN111694718A (zh) | 内网用户异常行为识别方法、装置、计算机设备及可读存储介质 | |
CN106293891B (zh) | 多维投资指标监督方法 | |
CN112070564B (zh) | 广告拉取方法、装置、***与电子设备 | |
CN112036631B (zh) | 采购量确定方法、装置、设备及存储介质 | |
CN115755614A (zh) | 基于碳排监测的用能优化调控方法及装置 | |
CN114969191A (zh) | 基于大数据的数据分析方法、***、装置及存储介质 | |
WO2019095569A1 (zh) | 基于微博财经事件的金融分析方法、应用服务器及计算机可读存储介质 | |
CN117472511A (zh) | 容器资源监控方法、装置、计算机设备和存储介质 | |
CN117312657A (zh) | 金融应用的异常功能定位方法、装置、计算机设备和介质 | |
CN111950860A (zh) | 一种企业舆情风险指数的监控方法及装置 | |
CN114495137B (zh) | 票据异常检测模型生成方法与票据异常检测方法 | |
CN116126859A (zh) | 一种数据管理方法、装置、电子设备及存储介质 | |
CN109218062B (zh) | 基于置信区间的互联网业务报警方法和装置 | |
CN115408186A (zh) | 根因定位方法、装置、计算机设备和存储介质 | |
KAMLEY et al. | Multiple regression: A data mining approach for predicting the stock market trends based on open, close and high price of the month | |
CN116432975B (zh) | 页岩气多场站平台的开采管理方法、装置、设备及介质 | |
CN115269879B (zh) | 知识结构数据的生成方法、数据搜索方法和风险告警方法 | |
CN114647555B (zh) | 基于多业务***的数据预警方法、装置、设备和介质 | |
CN113902496B (zh) | 数据分析方法、装置及电子设备 | |
US11520680B2 (en) | Predicting statistics gathering times for database activity data using machine learning techniques |
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: 21757143 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 08/12/2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 21757143 Country of ref document: EP Kind code of ref document: A1 |