CN117033726A - Report access duration prediction method and device, processor and electronic equipment - Google Patents

Report access duration prediction method and device, processor and electronic equipment Download PDF

Info

Publication number
CN117033726A
CN117033726A CN202311041798.XA CN202311041798A CN117033726A CN 117033726 A CN117033726 A CN 117033726A CN 202311041798 A CN202311041798 A CN 202311041798A CN 117033726 A CN117033726 A CN 117033726A
Authority
CN
China
Prior art keywords
model
report
target
information
access
Prior art date
Legal status (The legal status 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 status listed.)
Pending
Application number
CN202311041798.XA
Other languages
Chinese (zh)
Inventor
胡傈纹
郑凡奇
张宏兵
沈梦婷
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Industrial and Commercial Bank of China Ltd ICBC
Original Assignee
Industrial and Commercial Bank of China Ltd ICBC
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 Industrial and Commercial Bank of China Ltd ICBC filed Critical Industrial and Commercial Bank of China Ltd ICBC
Priority to CN202311041798.XA priority Critical patent/CN117033726A/en
Publication of CN117033726A publication Critical patent/CN117033726A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9035—Filtering based on additional data, e.g. user or group profiles
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/901—Indexing; Data structures therefor; Storage structures
    • G06F16/9014—Indexing; Data structures therefor; Storage structures hash tables

Abstract

The application discloses a method, a device, a processor and electronic equipment for predicting report access duration, wherein the method is applied to the field of artificial intelligence and comprises the following steps: obtaining access information of a target report form from log data of a server to obtain target log data; extracting report information of a target report, network information of an access target report, user role information of the access target report and access time from target log data to obtain target information; the target information is input into a target model to predict the access time length of the access target report form, so that the target time length is obtained; and adjusting the server according to the target time length. The application solves the problems of poor report service quality and lower user satisfaction caused by the fact that the user cannot open the report due to network blocking or insufficient server resources and the like in the related technology.

Description

Report access duration prediction method and device, processor and electronic equipment
Technical Field
The application relates to the field of artificial intelligence, in particular to a method and a device for predicting report access time, a processor and electronic equipment.
Background
The report forms are important tools for enterprise management, and information such as business conditions, financial conditions, market conditions and the like of the enterprise can be reflected through the report forms, so that basis is provided for decision making of the enterprise. The online report is a report form based on Web technology, and a user can access the online report through a browser or mobile equipment to view and analyze report data in real time. The online report has the advantages of cross-platform, easy deployment, easy maintenance, easy expansion and the like, and is suitable for various business scenes and requirements.
The access time of the report refers to the time from the request of accessing the online report to the complete loading of the report and displaying the report on the user interface. Too long access time can affect user experience and efficiency, even resulting in users relinquishing use of online reports. The open time is affected by a variety of factors, such as network bandwidth, server performance, report design, data volume, etc. Therefore, for report server manager, timely monitoring and early warning report open time overlong condition is an important task for ensuring report service quality and user satisfaction.
The traditional report early warning system focuses on the quick response of the alarm, has the defect of insufficient message processing of information magnitude, and can only process a relatively fixed scene; in the prior art, some report servers can provide a report opening time statistics function, but the functions can only generally count reports on a single server, and cannot comprehensively analyze and compare the reports on a plurality of servers. In addition, the functions cannot dynamically predict and adjust the report opening time according to factors such as different time periods, different user groups, different network environments and the like, and cannot pre-warn the report on different servers in real time according to the prediction results.
Aiming at the problems that in the related art, a user cannot open a report due to network blocking or insufficient server resources and the like, so that the report has poor service quality and low user satisfaction, no effective solution is proposed at present.
Disclosure of Invention
The application mainly aims to provide a method, a device, a processor and electronic equipment for predicting report access time, which are used for solving the problems that in the related art, a user cannot open a report due to network blocking or insufficient server resources and the like, so that the service quality of the report is poor and the user satisfaction is low.
In order to achieve the above object, according to one aspect of the present application, there is provided a report access duration prediction method, including: obtaining access information of a target report form from log data of a server to obtain target log data; extracting report information of the target report, network information of the target report, user role information of the target report and access time from the target log data to obtain target information; the target information is input into a target model to predict the access time length of accessing the target report, so as to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of accessing the report and training a linear regression model according to the first data set, the second data set and the third data set; and adjusting the server according to the target duration.
Further, the target model is obtained by the following steps: acquiring access information related to a report form from log data of the server, and determining the first data set, the second data set and the third data set according to access time of accessing the report form in the access information; training a linear regression model by adopting the first data set to obtain a first model; training a linear regression model by adopting the second data set, and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; training the local weighted linear regression model by adopting the third data set to obtain a third model; fusing the first model, the second model and the third model to obtain a fourth model; and optimizing the fourth model by adopting an R-square index to obtain the target model.
Further, fusing the first model, the second model and the third model to obtain a fourth model includes: respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; determining a loss function according to the first model, the weight of the second model, the weight of the third model and the weight of the third model; adjusting and optimizing the weight of the first model according to the loss function after the preset time length is passed, and the weight of the second model and the weight of the third model; and determining the fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
Further, obtaining access information related to the report form from the log data of the server, and determining the first data set, the second data set and the third data set according to the access time of accessing the report form in the access information includes: extracting the access information from the log data to obtain the first data set; and determining the second data set and the third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in the preset time period.
Further, extracting the access information from the log data, the obtaining the first data set includes: preprocessing the log data to obtain preprocessed information; determining report information, user role information, network information, access time and access time length when accessing the report each time according to the preprocessed information; and determining the first data set according to the report information, the user role information, the network information, the access time and the access time length when the report is accessed each time.
Further, after preprocessing the log data to obtain preprocessed information, the method further includes: determining query sentences when accessing the report forms each time in the preprocessed information; converting the query statement into a word vector by adopting a word2Vec algorithm to obtain a target word vector; and adding the target word vector corresponding to each report access to the first data set.
Further, adjusting the server according to the target duration includes: generating alarm information when the time length of the detected target is greater than a preset threshold value; and sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
In order to achieve the above object, according to another aspect of the present application, there is provided a report access duration prediction apparatus, including: the acquisition unit is used for acquiring access information to the target report form from the log data of the server to obtain target log data; the extraction unit is used for extracting report information of the target report, network information of accessing the target report, user role information of accessing the target report and access time from the target log data to obtain target information; the prediction unit is used for predicting the access time length of the access to the target report form by inputting the target information into a target model to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of the access report form and training a linear regression model according to the first data set, the second data set and the third data set; and the adjusting unit is used for adjusting the server according to the target duration.
Further, the prediction unit includes: the acquisition subunit is used for acquiring access information related to the report form from the log data of the server, and determining the first data set, the second data set and the third data set according to the access time of accessing the report form in the access information; the first training subunit is used for training the linear regression model by adopting the first data set to obtain a first model; the second training subunit is used for training the linear regression model by adopting the second data set and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; the third training subunit is used for training the local weighted linear regression model by adopting the third data set to obtain a third model; the fusion subunit is used for fusing the first model, the second model and the third model to obtain a fourth model; and the optimization subunit is used for optimizing the fourth model by adopting an R-square index to obtain the target model.
Further, the fusion subunit includes: the configuration module is used for respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; a first determining module, configured to determine a loss function according to the first model, the weight of the second model, the weight of the third model, and the weight of the third model; the adjusting module is used for adjusting and optimizing the weight of the first model according to the loss function every time a preset duration passes, and the weight of the second model and the weight of the third model; and the second determining module is used for determining the fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
Further, the acquiring subunit includes: the extraction module is used for extracting the access information from the log data to obtain the first data set; and a third determining module, configured to determine, in the first data set, the second data set and the third data set according to an access time in the access information, where the second data set includes access information in a day before a current date, and the third data set includes access information in a preset time period.
Further, the extraction module includes: the processing sub-module is used for preprocessing the log data to obtain preprocessed information; the first determining submodule is used for determining report information, user role information, network information, access time and access time length when the report is accessed each time according to the preprocessed information; and the second determining submodule is used for determining the first data set according to the report information, the user role information, the network information, the access time and the access time length when the report is accessed each time.
Further, the extraction module further comprises: the determining submodule is used for determining query sentences when the report is accessed each time in the preprocessed information after preprocessing the log data to obtain the preprocessed information; the conversion sub-module is used for converting the query statement into a word vector by adopting a word2Vec algorithm to obtain a target word vector; and the adding sub-module is used for adding the target word vector corresponding to each report accessing to the first data set.
Further, the adjusting unit includes: the generation subunit is used for generating alarm information when the target time length is detected to be greater than a preset threshold value; and the sending subunit is used for sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
In order to achieve the above object, according to one aspect of the present application, there is provided a processor for running a program, wherein the program runs to execute the method for predicting the report access duration according to any one of the above.
In order to achieve the above object, according to one aspect of the present application, there is provided an electronic device including one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method for predicting a report access duration according to any one of the above.
According to the application, the following steps are adopted: obtaining access information of a target report form from log data of a server to obtain target log data; extracting report information of the target report, network information of the target report, user role information of the target report and access time from the target log data to obtain target information; the target information is input into a target model to predict the access time length of accessing the target report, so as to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of accessing the report and training a linear regression model according to the first data set, the second data set and the third data set; the server is adjusted according to the target duration, so that the problems that in the related art, due to network blocking, insufficient server resources and the like, a user cannot open a report, the service quality of the report is poor, and the user satisfaction is low are solved. By collecting log data of the access report at different moments, a model for predicting the access time length can be trained according to the access information at different moments, so that the predicted result of the access time length of the real-time predicted report is more fit with the actual production condition, the accuracy of the predicted result is improved, the access time length of the target report can be processed and optimized according to the predicted result, the effect of improving the service quality of the report is achieved, and the effect of improving the satisfaction degree of a user using the report is further achieved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:
FIG. 1 is a flowchart of a method for predicting report access duration according to a first embodiment of the present application;
FIG. 2 is a schematic diagram of an alternative report access duration prediction method according to a first embodiment of the present application;
FIG. 3 is a schematic diagram of a report access duration prediction apparatus according to a second embodiment of the present application;
fig. 4 is a schematic diagram of an electronic device for predicting report access duration according to a fifth embodiment of the present application.
Detailed Description
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
It should be noted that, the user information (including, but not limited to, user equipment information, user personal information, information input by a user, etc.) and the data (including, but not limited to, data used for analysis, stored data, displayed data, data recorded in a log, etc.) related to the present application are information and data authorized by the user or fully authorized by each party, and the collection, use and processing of the related data need to comply with the related laws and regulations and standards of the related country and region, and a corresponding operation entry is provided for the user to select authorization or rejection.
In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the application herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
The present application will be described with reference to preferred implementation steps, and fig. 1 is a flowchart of a method for predicting report access duration according to an embodiment of the present application, and as shown in fig. 1, the method includes the following steps:
step S101, access information of a target report is obtained from log data of a server, and target log data is obtained.
In the first embodiment, the target report may be any report in a server (i.e., the server described above) on which the report service is installed. In order to avoid longer access time for accessing the target report in actual production activities, the user has poor report opening experience, and access information (such as network condition at the current moment, the number of times of accessing the target report within 10 minutes before the current moment, and the like) about the target report needs to be acquired from the server in real time.
Step S102, extracting report information of the target report, network information of the access target report, user role information of the access target report and access time from the target log data to obtain the target information.
In the first embodiment, the key information of the access target report (i.e., the report information of the target report, the network information of the access target report, the user role information of the access target report, and the access time) is extracted from the target log data, then the key information is used as an independent variable, and the access time of the access target report is used as a dependent variable, so that the access time of the access target report is obtained by predicting the dependent variable according to the independent variable. The access duration of accessing the report refers to the time taken from the user to initiate an access instruction to access the report to the report for presentation to the user.
Step S103, the target information is input into a target model to predict the access time length of an access target report, so as to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of the access report and training a linear regression model according to the first data set, the second data set and the third data set.
After the target model is trained in the first embodiment, the trained target model and related parameters are required to be transmitted to a server carrying report service, then the acquired real-time access information (i.e. the target information) of the target report is input into the target model contained in the server, and the access duration of the target report is predicted through the target model, so as to obtain the target duration.
Step S104, the server is adjusted according to the target duration.
In the first embodiment, in order to avoid that the access duration of the access report is long, resulting in poor customer experience of accessing the target report, the resource allocation condition in the server needs to be adjusted through the target duration obtained by prediction. For example, more resources are allocated in the server for the cluster nodes providing the report service, so as to increase the loading speed of the target report.
In summary, according to the report access duration prediction method provided by the first embodiment of the present application, the target log data is obtained by obtaining the access information of the target report from the log data of the server; extracting report information of a target report, network information of an access target report, user role information of the access target report and access time from target log data to obtain target information; the method comprises the steps of inputting target information into a target model to predict access time length of an access target report, and obtaining target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to access time of the access report and training a linear regression model according to the first data set, the second data set and the third data set; the server is adjusted according to the target duration, so that the problems of poor report service quality and low user satisfaction degree caused by the fact that a user cannot open a report due to network blocking or insufficient server resources in the related technology are solved. By collecting log data of the access report at different moments, a model for predicting the access time length can be trained according to the access information at different moments, so that the predicted result of the access time length of the real-time predicted report is more fit with the actual production condition, the accuracy of the predicted result is improved, the access time length of the target report can be processed and optimized according to the predicted result, the effect of improving the service quality of the report is achieved, and the effect of improving the satisfaction degree of a user using the report is further achieved.
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, the target model is obtained by: acquiring access information related to the report form from log data of a server, and determining a first data set, a second data set and a third data set according to access time of accessing the report form in the access information; training a linear regression model by adopting a first data set to obtain a first model; training a linear regression model by adopting a second data set, and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; training the local weighted linear regression model by adopting a third data set to obtain a third model; fusing the first model, the second model and the third model to obtain a fourth model; and optimizing the fourth model by adopting the R-square index to obtain a target model.
In the first embodiment, in order to improve accuracy of predicting a prediction result of a real-time access target report, access information of different access moments may be acquired to obtain a first data set, a second data set and a third data set, then the access information when the report is accessed is used as an independent variable of a linear regression model, the access duration of the access report is used as a dependent variable of the linear regression model, the linear regression model is trained by respectively adopting the first data set, the second data set and the third data set to obtain a linear relationship between the independent variable and the dependent variable (i.e., the target model), the trained models are fused, and finally the target model for predicting the access duration is obtained, so that the access duration of the access report is predicted.
After access information related to the report is acquired from log data of the server, the access information is divided according to access time in the access information, and a first data set, a second data set and a third data set are obtained.
Specifically, a first data set is composed of the total amount of access information, a second data set is composed of the access information in the day before the current date, and a third data set is composed of the access information in the month before the current date.
Then, for a first dataset containing full access information, a simple linear regression model may be trained using the first dataset to yield a first model of general accuracy. The linear regression model may be represented by the following formula:
y＝a+b 1 x 1 +b 2 x 2 +b 3 x 3 +......+b k x k +e
wherein y is the access time length of the access report, and x is 1 To x k Is access information, a, b 1 To b k Is the regression coefficient and e is the error term.
For the second data set containing the access information in the day before the current date, because the access time of the access information in the second data set is closer to the current date, the access information in the second data set has a larger influence on the prediction result, and the second data set can be used for training the linear regression model to obtain a second model. Meanwhile, in order to avoid the situation that the second model is over-fitted due to less access information in the second data set, a Ridge Regression (Ridge Regression) algorithm can be adopted to estimate the obtained second model so as to reduce the error of the prediction result of the second model. Ridge regression is a technique for multiple co-linearity (i.e., the access time length of an accessed report has a strong correlation with the access information of the accessed report) data that reduces standard error by adding a penalty to the regression coefficients.
For the third data set of the access information in the preset time period, because the access information used for training is the access information in the preset time period, the trained model is easy to be under-fitted, and the third data set can be used for training a local weighted linear regression (Locally Weighted Linear Regression, LWLR) model to obtain the third model. The local weighted linear regression model is a non-parametric regression algorithm that predicts continuous dependent variables (i.e., access times) based on local characteristics of the sample data (i.e., access information over a predetermined period of time). In addition, considering the performance problem of the local weighted linear regression model, the access information in the third data set may be limited to 10000 or so in addition to the log data of 30 days in the last day of each report.
After the first model, the second model and the third model are obtained through training, the first model, the second model and the third model can be fused to obtain a final target model for prediction.
Finally, the fitting degree of the fused target model can be evaluated by adopting an R-square index so as to gradually optimize the target model. For example, if the R-square value is closer to 1, it indicates that the fitting degree of the target model is higher. In addition, other evaluation algorithms may be used to evaluate the fitting degree of the target model, and the first embodiment is not particularly limited. For example, the F-test algorithm or the T-test algorithm is used to test whether the coefficients of the target model are significantly non-zero, i.e., whether the independent variable (i.e., access information) has a significant effect on the dependent variable (i.e., access duration).
By collecting data sets at different access moments, selecting a proper model for training according to a time period to which the access moments belong, fitting access information with different characteristics can be carried out, and models (namely the first model, the second model and the third model) with different characteristic information are obtained, so that generalization capability of the models is improved, and accuracy of predicting a prediction result of a real-time access target report is improved.
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, fusing the first model, the second model and the third model to obtain the fourth model includes: respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; determining a loss function according to the first model, the weight of the second model, the weight of the third model and the weight of the third model; adjusting and optimizing the weight of the first model, the weight of the second model and the weight of the third model according to the loss function after the preset time length; and determining a fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
In the first embodiment, in order to avoid errors caused by a single model, weights may be configured for the first model, the second model, and the third model, so that the fused model (i.e., the fourth model described above) is the sum of the prediction result of each model multiplied by the weight.
Specifically, weights w are configured for the first model, the second model, and the third model, respectively 1 、w 2 And w 3 And continuously optimizing the weights w by constructing a loss function 1 、w 2 And w 3 . The formula for the loss function is shown below:
wherein Loss represents the Loss value of the target model, w 1 、w 2 And w 3 Respectively representing the weight of the first model, the weight of the second model and the weight of the third model, p 1 、p 2 And p 3 And r represents the actual access time length of the access report.
By fusing the first model, the second model and the third model, errors or deviations of the single model can be reduced, and therefore robustness and stability of the target model are improved.
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, obtaining access information related to a report from log data of a server, and determining, according to access time of accessing the report in the access information, a first data set, a second data set, and a third data set includes: extracting access information from the log data to obtain a first data set; and determining a second data set and a third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in a preset time period.
In the first embodiment, the report refers to a document or a chart for showing information such as data, statistical data, and data analysis results. In order to acquire the log data training model, a log acquisition module is required to be installed on each server (i.e. the server) on which the report service is installed, and the log acquisition module is used for recording relevant information when a user accesses the report each time.
Because the correlation of the access information at different moments to the access time length of the current moment access report is different, the models are required to be trained respectively according to the access information at different moments, and the three models obtained through training are fused to obtain the target model. For example, since the project a is developed in 1 month and a plurality of workers often access the data table A1 of the project a, the access time period for accessing the data table A1 in 1 month is long. When the research and development of the project B is carried out in 2 months, the data table A1 is not accessed or the times of accessing the data table A1 are greatly reduced, so that the access time for accessing the data table A1 in 2 months is shorter. Therefore, in the first embodiment, training can be performed based on access information at different times.
In addition, in order to reduce the burden of the server on which the report service is installed, the access information collected by the log collection module may be periodically transmitted to other servers in the form of log files to train and generate a model through the other servers. For example, the server with the report service packages the acquired access information into json packets every one hour, and transmits the json packets to other servers through http transmission. And in the sending process, if other servers do not receive the complete json packet, setting the server carrying the report service to resend the json packet after 5 minutes.
Specifically, after the log data is collected in the server, the access information may be extracted from the log data, to obtain a first data set containing a full amount of log data. The access information includes, but is not limited to, report information of a report accessed in log data, network information (e.g., bandwidth information of a server, delay information, the number of network connections, packet loss rate, etc.), access time, etc. information when the report is accessed.
And then, screening access information of which the access time is one day before the current date from the first data set according to the access time in the access information to obtain a second data set, and screening access information of a preset time period (for example, the access time is one week before the current date, the access time is one month before the current date, and the like) from the first data set to obtain a third data set.
The three data sets are obtained by screening the access information in the full log data according to the access time, so that the training of the target model according to the actual production condition is facilitated, and the effect of improving the accuracy of the prediction result of the target model is achieved.
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, extracting access information from log data to obtain the first data set includes: preprocessing log data to obtain preprocessed information; determining report information, user role information, network information, access time and access time length when accessing the report each time according to the preprocessed information; and determining a first data set according to the report information, the user role information, the network information, the access time and the access time length when each report is accessed.
Specifically, after log data is collected in the server, the log data may be preprocessed using Pandas. For example, null values, repeated values and abnormal values in the log data are removed, and normalization processing and other operations are performed on the log data.
Then, access information is extracted from the preprocessed log data. The access information includes, but is not limited to, report information (e.g., report name, database information of the report, etc.) of the report accessed in the log data, user role information (e.g., color information such as developer, tester, project manager, etc.) of the report accessed, network information (e.g., information such as bandwidth information, delay information, network connection number, packet loss rate, etc.) of the report accessed, access time of the report accessed, and the like. In addition, other access information when accessing the report may be determined according to the service requirement, which is not particularly limited in the first embodiment.
Finally, the report name, the user role information, the network information, the access date and the access time in the access information are used as independent variables of the linear regression model, and the access time length is used as the dependent variable of the linear regression model and is stored in a database (for example, a database such as a MySql database) so as to use and maintain the access information. Table one is a specific example of access information in a database. As shown in table one, "report_name" is a report name, that is, the report information described above; "user_id" is a user identification, that is, the above-mentioned user role information; "network_status" is the network condition, i.e., the above-mentioned network information; "query_time" is the access time; "open_time" is the access duration.
The data set is obtained by determining to preprocess the log data, so that the data set training model is adopted, meanwhile, the access information is determined and extracted according to the service requirement, the access information can be determined according to the actual production condition, the independent variable and the dependent variable in the linear regression model are determined, and the dependent variable is further predicted according to the independent variable.
Access information in a table-database
Variable name Type(s) Description of the application
report_name Independent variable Report name
user_id Independent variable User identification
network_status Independent variable Network situation
query_time Independent variable Access time
open_time Dependent variable Access duration
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, after preprocessing log data to obtain preprocessed information, the method further includes: determining query sentences when accessing the report forms each time in the preprocessed information; converting the query sentence into a word vector by using a word2Vec algorithm to obtain a target word vector; and adding the corresponding target word vector to the first data set when accessing the report.
In the first embodiment, each time a report is accessed, because of different query conditions (for example, accessing report data within one year, accessing report data within a national range, etc.), there is a great difference in the amount of query data, which results in a great fluctuation in the access time length of accessing the report, thereby affecting the accuracy of the target model prediction result. Therefore, the query condition when the report is accessed can be used as an independent variable of the linear regression model, and the word vector corresponding to the query condition can be added into the access information.
Specifically, the query conditions in the access request may be combined with the granularity of the report to obtain the character string. Because the target model is used for predicting the access time length of the access report in real time without considering context semantic information among character strings, the Word2Vec model can be used for converting the character strings into Word vectors and adding the Word vectors into the access information. Table two is an example of access information containing query conditions. As shown in table two, "args_list" is a word vector corresponding to the query condition, i.e., the above-described target word vector.
By modeling the query conditions in the access request by using the Word2Vec model, a Word vector is obtained, and compared with the method that the feature value is obtained by directly carrying out hash operation on the query conditions, the Word vector can retain semantic information among the query conditions, is beneficial to more accurately fitting a linear regression model, and improves the accuracy of the model obtained by training.
Table II contains access information for query conditions
Variable name Type(s) Description of the application
report_name Independent variable Report name
user_id Independent variable User identification
network_status Independent variable Network situation
query_time Independent variable Access time
args_list Independent variable Word vector corresponding to query condition
open_time Dependent variable Access duration
Optionally, in the method for predicting report access duration provided in the first embodiment of the present application, adjusting the server according to the target duration includes: generating alarm information when the time length of the detected target is greater than a preset threshold value; and sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
Specifically, during a high load period of report access, the preset threshold may be set to 1.5 times the historical average access duration, and during a low load period of report access, the preset threshold may be set to 1.2 times the historical average access duration. And if the access time length predicted by the target model in real time exceeds a preset threshold value, sending alarm information to an administrator of the report server. The administrator can allocate additional cluster resources according to the actual generation situation server so as to ensure that the server can process the access request of the report as soon as possible and display the report to the user.
When the access time predicted by the target model exceeds a preset threshold, alarm information is sent to the target object, so that the target object can timely adjust a server running report service, the problem of poor user experience caused by long access time of a report is avoided, the effect of improving the service quality of the report is achieved, and the effect of improving the user satisfaction is further achieved.
Alternatively, in the first embodiment, the process of predicting the report access duration according to the present embodiment may be as shown in fig. 2. Firstly, acquiring log data when accessing a report in a server, performing preprocessing operations such as data cleaning and feature extraction on the log data, and dividing the processed data into a first data set, a second data set and a third data set according to the access time of each access report, wherein the first data set contains total log data, the second data set contains data of a day before the current date, and the third data set contains data of a month before the current date. Then training a linear regression model by adopting a first data set to obtain a first model; training a linear regression model by adopting a second data set, and carrying out regression by adopting a ridge regression algorithm to obtain a second model; and training the local weighted linear regression model by adopting a third data set to obtain a third model. And secondly, fusing the first model, the second model and the third model to obtain a target model. And finally, inputting the target report information generated in real time into a target model to predict the access time length of the target report accessed in real time, notifying a target object when the access time length is greater than a preset threshold value, and timely processing the report service to ensure that a user normally uses the report service.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
Example two
The second embodiment of the present application also provides a report access duration prediction apparatus, and it should be noted that the report access duration prediction apparatus of the second embodiment of the present application may be used to execute the report access duration prediction method provided by the first embodiment of the present application. The following describes a report access duration prediction device provided in the second embodiment of the present application.
Fig. 3 is a schematic diagram of a report access duration prediction apparatus according to a second embodiment of the present application. As shown in fig. 3, the apparatus includes: an acquisition unit 301, an extraction unit 302, a prediction unit 303, and an adjustment unit 304.
Specifically, the obtaining unit 301 is configured to obtain, from log data of a server, access information to a target report, and obtain target log data.
And the extracting unit 302 is configured to extract report information of the target report, network information of the access target report, user role information of the access target report, and access time from the target log data, so as to obtain the target information.
The predicting unit 303 is configured to obtain a target duration by inputting target information into a target model to predict an access duration of an access target report, where the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set, and a third data set according to an access time of the access report, and training a linear regression model according to the first data set, the second data set, and the third data set.
The adjusting unit 304 is configured to adjust the server according to the target duration.
According to the report access duration prediction device provided by the second embodiment of the application, the access information of the target report is obtained from the log data of the server through the obtaining unit 301, so as to obtain the target log data; the extracting unit 302 extracts report information of the target report, network information of the access target report, user role information of the access target report and access time from the target log data to obtain target information; the prediction unit 303 predicts the access time length of the access target report form by inputting the target information into a target model, so as to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of the access report form and training a linear regression model according to the first data set, the second data set and the third data set; the adjustment unit 304 adjusts the server according to the target duration, so as to solve the problems of poor report service quality and low user satisfaction caused by the fact that the user cannot open the report due to network blocking or insufficient server resources in the related art. By collecting log data of the access report at different moments, a model for predicting the access time length can be trained according to the access information at different moments, so that the predicted result of the access time length of the real-time predicted report is more fit with the actual production condition, the accuracy of the predicted result is improved, the access time length of the target report can be processed and optimized according to the predicted result, the effect of improving the service quality of the report is achieved, and the effect of improving the satisfaction degree of a user using the report is further achieved.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the prediction unit 303 includes: the acquisition subunit is used for acquiring access information related to the report form from the log data of the server and determining a first data set, a second data set and a third data set according to the access time of accessing the report form in the access information; the first training subunit is used for training the linear regression model by adopting the first data set to obtain a first model; the second training subunit is used for training the linear regression model by adopting a second data set and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; the third training subunit is used for training the local weighted linear regression model by adopting a third data set to obtain a third model; the fusion subunit is used for fusing the first model, the second model and the third model to obtain a fourth model; and the optimizing subunit is used for optimizing the fourth model by adopting the R-square index to obtain a target model.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the above-mentioned fusion subunit includes: the configuration module is used for respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; the first determining module is used for determining a loss function adjusting module according to the first model, the weight of the second model, the weight of the third model and the weight of the third model, and adjusting and optimizing the weight of the first model, the weight of the second model and the weight of the third model according to the loss function after a preset time length; the second determining module is used for determining a fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the obtaining subunit includes: the extraction module is used for extracting access information from the log data to obtain a first data set; and the third determining module is used for determining a second data set and a third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in the preset time period.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the extracting module includes: the processing sub-module is used for preprocessing the log data to obtain preprocessed information; the first determining submodule is used for determining report information, user role information, network information, access time and access time length when the report is accessed each time according to the preprocessed information; the second determining sub-module is used for determining the first data set according to the report information, the user role information, the network information, the access time and the access time length when the report is accessed each time.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the extracting module further includes: the determining submodule is used for determining query sentences when the report is accessed each time in the preprocessed information after preprocessing the log data to obtain the preprocessed information; the conversion sub-module is used for converting the query statement into a word vector by adopting a word2Vec algorithm to obtain a target word vector; and the adding sub-module is used for adding the corresponding target word vector to the first data set when the report is accessed each time.
Optionally, in the report access duration prediction apparatus provided in the second embodiment of the present application, the adjustment unit 304 includes: the generation subunit is used for generating alarm information when the time length of detecting the target is greater than a preset threshold value; and the sending subunit is used for sending the alarm information to the target object, wherein the target object adjusts the server according to the alarm information.
The report access duration prediction device includes a processor and a memory, where the above-mentioned obtaining unit 301, extracting unit 302, predicting unit 303, and adjusting unit 304 are all stored as program units in the memory, and the processor executes the above-mentioned program units stored in the memory to implement corresponding functions.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one, and the accuracy of predicting the access time length of the access report is improved by adjusting the kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
The third embodiment of the invention provides a computer readable storage medium, on which a program is stored, which when executed by a processor, implements a method for predicting report access duration.
The fourth embodiment of the invention provides a processor, which is used for running a program, wherein the program runs to execute a method for predicting the report access duration.
As shown in fig. 4, a fifth embodiment of the present invention provides an electronic device, where the device includes a processor, a memory, and a program stored in the memory and executable on the processor, and the processor implements the following steps when executing the program: obtaining target log data by obtaining access information to a target report form from the log data of a server; extracting report information of a target report, network information of an access target report, user role information of the access target report and access time from target log data to obtain target information; the method comprises the steps of inputting target information into a target model to predict access time length of an access target report, and obtaining target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to access time of the access report and training a linear regression model according to the first data set, the second data set and the third data set; and adjusting the server according to the target time length.
The processor also realizes the following steps when executing the program: the target model is obtained by the following steps: acquiring access information related to the report form from log data of a server, and determining a first data set, a second data set and a third data set according to access time of accessing the report form in the access information; training a linear regression model by adopting a first data set to obtain a first model; training a linear regression model by adopting a second data set, and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; training the local weighted linear regression model by adopting a third data set to obtain a third model; fusing the first model, the second model and the third model to obtain a fourth model; and optimizing the fourth model by adopting the R-square index to obtain a target model.
The processor also realizes the following steps when executing the program: fusing the first model, the second model and the third model to obtain a fourth model comprises: respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; determining a loss function according to the first model, the weight of the second model, the weight of the third model and the weight of the third model; adjusting and optimizing the weight of the first model, the weight of the second model and the weight of the third model according to the loss function after the preset time length; and determining a fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
The processor also realizes the following steps when executing the program: acquiring access information related to the report form from log data of the server, and determining the first data set, the second data set and the third data set according to access time of accessing the report form in the access information comprises: extracting access information from the log data to obtain a first data set; and determining a second data set and a third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in a preset time period.
The processor also realizes the following steps when executing the program: extracting access information from the log data, the obtaining a first data set comprising: preprocessing log data to obtain preprocessed information; determining report information, user role information, network information, access time and access time length when accessing the report each time according to the preprocessed information; and determining a first data set according to the report information, the user role information, the network information, the access time and the access time length when each report is accessed.
The processor also realizes the following steps when executing the program: after preprocessing the log data to obtain preprocessed information, the method further comprises the following steps: determining query sentences when accessing the report forms each time in the preprocessed information; converting the query sentence into a word vector by using a word2Vec algorithm to obtain a target word vector; and adding the corresponding target word vector to the first data set when accessing the report.
The processor also realizes the following steps when executing the program: the adjusting of the server according to the target duration comprises the following steps: generating alarm information when the time length of the detected target is greater than a preset threshold value; and sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
The device herein may be a server, PC, PAD, cell phone, etc.
The application also provides a computer program product adapted to perform, when executed on a data processing device, a program initialized with the method steps of: obtaining target log data by obtaining access information to a target report form from the log data of a server; extracting report information of a target report, network information of an access target report, user role information of the access target report and access time from target log data to obtain target information; the method comprises the steps of inputting target information into a target model to predict access time length of an access target report, and obtaining target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to access time of the access report and training a linear regression model according to the first data set, the second data set and the third data set; and adjusting the server according to the target time length.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: the target model is obtained by the following steps: acquiring access information related to the report form from log data of a server, and determining a first data set, a second data set and a third data set according to access time of accessing the report form in the access information; training a linear regression model by adopting a first data set to obtain a first model; training a linear regression model by adopting a second data set, and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model; training the local weighted linear regression model by adopting a third data set to obtain a third model; fusing the first model, the second model and the third model to obtain a fourth model; and optimizing the fourth model by adopting the R-square index to obtain a target model.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: fusing the first model, the second model and the third model to obtain a fourth model comprises: respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model; determining a loss function according to the first model, the weight of the second model, the weight of the third model and the weight of the third model; adjusting and optimizing the weight of the first model, the weight of the second model and the weight of the third model according to the loss function after the preset time length; and determining a fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: acquiring access information related to the report form from log data of the server, and determining the first data set, the second data set and the third data set according to access time of accessing the report form in the access information comprises: extracting access information from the log data to obtain a first data set; and determining a second data set and a third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in a preset time period.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: extracting access information from the log data, the obtaining a first data set comprising: preprocessing log data to obtain preprocessed information; determining report information, user role information, network information, access time and access time length when accessing the report each time according to the preprocessed information; and determining a first data set according to the report information, the user role information, the network information, the access time and the access time length when each report is accessed.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: after preprocessing the log data to obtain preprocessed information, the method further comprises the following steps: determining query sentences when accessing the report forms each time in the preprocessed information; converting the query sentence into a word vector by using a word2Vec algorithm to obtain a target word vector; and adding the corresponding target word vector to the first data set when accessing the report.
When executed on a data processing device, is further adapted to carry out a program initialized with the method steps of: the adjusting of the server according to the target duration comprises the following steps: generating alarm information when the time length of the detected target is greater than a preset threshold value; and sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, etc., such as Read Only Memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.
Computer readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of storage media for a computer include, but are not limited to, phase change memory (PRAM), static Random Access Memory (SRAM), dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), read Only Memory (ROM), electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transmission medium, which can be used to store information that can be accessed by a computing device. Computer-readable media, as defined herein, does not include transitory computer-readable media (transmission media), such as modulated data signals and carrier waves.
It should also be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or apparatus that comprises an element.
It will be appreciated by those skilled in the art that embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims (10)

1. The method for predicting the report access time length is characterized by comprising the following steps:
obtaining access information of a target report form from log data of a server to obtain target log data;
extracting report information of the target report, network information of the target report, user role information of the target report and access time from the target log data to obtain target information;
the target information is input into a target model to predict the access time length of accessing the target report, so as to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of accessing the report and training a linear regression model according to the first data set, the second data set and the third data set;
and adjusting the server according to the target duration.
2. The method according to claim 1, wherein the object model is obtained by:
acquiring access information related to a report form from log data of the server, and determining the first data set, the second data set and the third data set according to access time of accessing the report form in the access information;
Training a linear regression model by adopting the first data set to obtain a first model;
training a linear regression model by adopting the second data set, and optimizing model parameters by adopting a ridge regression algorithm to obtain a second model;
training the local weighted linear regression model by adopting the third data set to obtain a third model;
fusing the first model, the second model and the third model to obtain a fourth model;
and optimizing the fourth model by adopting an R-square index to obtain the target model.
3. The method of claim 2, wherein fusing the first model, the second model, and the third model to obtain a fourth model comprises:
respectively configuring weights for the first model, the second model and the third model to obtain the weight of the first model, the weight of the second model and the weight of the third model;
determining a loss function according to the first model, the weight of the second model, the weight of the third model and the weight of the third model;
adjusting and optimizing the weight of the first model according to the loss function after the preset time length is passed, and the weight of the second model and the weight of the third model;
And determining the fourth model according to the first model, the weight after the optimization of the first model, the second model, the weight after the optimization of the second model, the third model and the weight after the optimization of the third model.
4. The method of claim 2, wherein obtaining access information related to a report from log data of the server, and determining the first data set, the second data set, and the third data set according to access times of accessing the report in the access information comprises:
extracting the access information from the log data to obtain the first data set;
and determining the second data set and the third data set in the first data set according to the access time in the access information, wherein the second data set contains the access information in the day before the current date, and the third data set contains the access information in the preset time period.
5. The method of claim 4, wherein extracting the access information from the log data to obtain the first data set comprises:
preprocessing the log data to obtain preprocessed information;
Determining report information, user role information, network information, access time and access time length when accessing the report each time according to the preprocessed information;
and determining the first data set according to the report information, the user role information, the network information, the access time and the access time length when the report is accessed each time.
6. The method of claim 5, wherein after preprocessing the log data to obtain preprocessed information, the method further comprises:
determining query sentences when accessing the report forms each time in the preprocessed information;
converting the query statement into a word vector by adopting a word2Vec algorithm to obtain a target word vector;
and adding the target word vector corresponding to each report access to the first data set.
7. The method of claim 1, wherein adjusting the server in accordance with the target time period comprises:
generating alarm information when the time length of the detected target is greater than a preset threshold value;
and sending the alarm information to a target object, wherein the target object adjusts the server according to the alarm information.
8. The report access duration prediction device is characterized by comprising the following steps:
the acquisition unit is used for acquiring access information to the target report form from the log data of the server to obtain target log data;
the extraction unit is used for extracting report information of the target report, network information of accessing the target report, user role information of accessing the target report and access time from the target log data to obtain target information;
the prediction unit is used for predicting the access time length of the access to the target report form by inputting the target information into a target model to obtain the target time length, wherein the target model is a model obtained by dividing log data related to the access information into a first data set, a second data set and a third data set according to the access time of the access report form and training a linear regression model according to the first data set, the second data set and the third data set;
and the adjusting unit is used for adjusting the server according to the target duration.
9. A processor, wherein the processor is configured to run a program, and wherein the program runs to perform the method for predicting report access time according to any one of claims 1 to 7.
10. An electronic device comprising one or more processors and a memory for storing one or more programs, wherein the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of predicting report access duration of any one of claims 1 to 7.
CN202311041798.XA 2023-08-17 2023-08-17 Report access duration prediction method and device, processor and electronic equipment Pending CN117033726A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311041798.XA CN117033726A (en) 2023-08-17 2023-08-17 Report access duration prediction method and device, processor and electronic equipment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311041798.XA CN117033726A (en) 2023-08-17 2023-08-17 Report access duration prediction method and device, processor and electronic equipment

Publications (1)

Publication Number Publication Date
CN117033726A true CN117033726A (en) 2023-11-10

Family

ID=88633343

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311041798.XA Pending CN117033726A (en) 2023-08-17 2023-08-17 Report access duration prediction method and device, processor and electronic equipment

Country Status (1)

Country Link
CN (1) CN117033726A (en)

Similar Documents

Publication Publication Date Title
US20230325724A1 (en) Updating attribute data structures to indicate trends in attribute data provided to automated modelling systems
US20200012584A1 (en) Fully automated machine learning system which generates and optimizes solutions given a dataset and a desired outcome
CN109144724A (en) A kind of micro services resource scheduling system and method
CN109561052B (en) Method and device for detecting abnormal flow of website
CN105718490A (en) Method and device for updating classifying model
US11082509B1 (en) Determining session intent
US11016730B2 (en) Transforming a transactional data set to generate forecasting and prediction insights
CN113900921B (en) Court information system running state evaluation method, device, equipment and storage medium
CN111222994A (en) Client risk assessment method, device, medium and electronic equipment
CN110399268A (en) A kind of method, device and equipment of anomaly data detection
CN111176565A (en) Method and device for determining storage load of application
CN111242430A (en) Power equipment supplier evaluation method and device
CN115757035A (en) Data model monitoring method and device, processor and electronic equipment
CN111800807A (en) Method and device for alarming number of base station users
CN112101692B (en) Identification method and device for mobile internet bad quality users
CN107844496B (en) Statistical information output method and device
CN116629612A (en) Risk prediction method and device, storage medium and electronic equipment
CN116225958A (en) Fault prediction method and device, storage medium and electronic equipment
CN117033726A (en) Report access duration prediction method and device, processor and electronic equipment
CN115994029A (en) Container resource scheduling method and device
CN117853221A (en) Method, device and system for monitoring consumption credit model
CN114358641A (en) Prediction method, device, equipment and medium for budget approval
CN116433020A (en) Method and equipment for resource main body risk early warning
CN117453369A (en) Third party application call volume allocation method, system and equipment
CN116881787A (en) Data sample classification method and device, processor and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination