CN110598772A

CN110598772A - Operation data detection method and device, computer equipment and storage medium

Info

Publication number: CN110598772A
Application number: CN201910819032.7A
Authority: CN
Inventors: 张嘉成; 赵静
Original assignee: Zhejiang Dasou Vehicle Software Technology Co Ltd
Current assignee: Zhejiang Dasou Vehicle Software Technology Co Ltd
Priority date: 2019-08-30
Filing date: 2019-08-30
Publication date: 2019-12-20

Abstract

The application relates to an operation data detection method, an operation data detection device, computer equipment and a storage medium. The method comprises the following steps: and acquiring operation data, and analyzing the operation data to obtain data classification information. And calculating to obtain data characteristics according to the data classification information. And inputting the data characteristics to a linear model to obtain a grade probability corresponding to the operation data, and obtaining a first detection result according to the grade probability. And calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information. And comparing the first detection result with the second detection result to obtain an operation detection result. By adopting the method, the accuracy of the operation detection result can be effectively improved.

Description

Operation data detection method and device, computer equipment and storage medium

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for detecting operation data, a computer device, and a storage medium.

Background

With the development of economic society, automobiles become an essential part for people to go out daily as a travel tool, and the growth of automobile purchasing demand promotes the vigorous development of the automobile sales industry. In the automobile sales industry, in order to better detect sales performance and formulate a sales target, a seller analyzes operation data in a preset period to obtain a target value and an operation suggestion corresponding to the operation data in the next period.

Conventionally, when the operation data is analyzed, the operation data is often manually selected according to experience, so that the operation data manually selected is compared with historical similar operation data to obtain a detection result corresponding to the operation data, and the operation data selected by an individual according to experience can result in incomplete selection, and the detection result is obtained by only calculating according to the historical similar operation data corresponding to the selected operation data, so that the detection result of the operation data is inaccurate.

Disclosure of Invention

In view of the above, it is necessary to provide an operation data detection method, an apparatus, a computer device, and a storage medium capable of improving accuracy.

A method of operational data detection, the method comprising:

acquiring operation data, and analyzing the operation data to obtain data classification information;

calculating to obtain data characteristics according to the data classification information;

inputting the data characteristics to a linear model to obtain a grade probability corresponding to the operation data, and obtaining a first detection result according to the grade probability;

calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information;

and comparing the first detection result with the second detection result to obtain an operation detection result.

In one embodiment, the calculating the data feature according to the data classification information includes:

when the data classification information represents that the operation data is operation index data, inquiring the grade evaluation logic corresponding to the operation index data;

according to the grade evaluation logic, calculating an index grade corresponding to the operation index data, and inquiring a target code corresponding to the index grade;

and generating an index feature according to the target code as the data feature.

when the data classification information represents that the operation data is operation position data, acquiring a position data coefficient corresponding to the operation position data;

extracting a location data amount corresponding to the operation location data;

calculating according to the position data coefficient and the position data quantity to obtain a position characteristic value, and standardizing the position characteristic value to obtain a standardized position characteristic value;

and generating a position characteristic as a data characteristic according to the normalized position characteristic value.

In one embodiment, the normalizing the position feature value to obtain a normalized position feature value includes:

acquiring standard data, calculating a standard characteristic value according to the standard data, and calculating to obtain a maximum difference value corresponding to the standard characteristic value;

calculating according to the minimum value in the standard characteristic values and the position characteristic value to obtain a position characteristic value difference value;

and calculating through the maximum difference value of the standard characteristic values and the difference value of the position characteristic values to obtain the position characteristic values after standardization.

In one embodiment, after the inputting the data feature into a linear model and obtaining a level probability corresponding to the operation data, the method further includes:

obtaining an operation probability value corresponding to the operation index data according to the level probability;

calculating to obtain an operation detection score according to the operation probability value and the operation index data;

and sending the operation detection score to a terminal.

In one embodiment, the comparing the first detection result and the second detection result to obtain an operation detection result includes:

extracting a target grade corresponding to the first detection result, wherein the target grade is generated according to the grade probability;

acquiring a grade to be compared corresponding to the second detection result, wherein the grade to be compared is calculated according to the operation data;

when the target grade is lower than or equal to the grade to be compared, taking the target grade as the operation detection result;

and when the target grade is higher than the grade to be compared, inquiring a target value corresponding to the grade to be compared, and taking the target value as the operation detection result.

In one embodiment, the generation manner of the linear model includes:

acquiring sample characteristics corresponding to sample data, inputting the sample characteristics into an initial model to obtain a sample result corresponding to the sample characteristics, and acquiring a target result corresponding to the sample data;

and when the sample result is the same as the target result, taking the initial model as the linear model.

An operational data detection apparatus, the apparatus comprising:

the acquisition module is used for acquiring operation data of a user and analyzing the operation data to obtain data classification information;

the calculation module is used for calculating to obtain data characteristics according to the data classification information;

the first detection module is used for inputting the data characteristics to a linear model to obtain a grade probability corresponding to the operation data and obtain a first detection result according to the grade probability;

the second detection module is used for calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information;

and the comparison module is used for comparing the first detection result with the second detection result to obtain an operation detection result.

A computer device comprising a memory and a processor, the memory storing a computer program, the processor implementing the following steps when executing the computer program:

A computer-readable storage medium, on which a computer program is stored which, when executed by a processor, carries out the steps of:

According to the operation data detection method, the operation data detection device, the computer equipment and the storage medium, the operation data sent by the receiving terminal are carried with the data classification information. And calculating to obtain data characteristics according to the data classification information. And acquiring a linear model, inputting the data characteristics into the linear model, acquiring the grade probability corresponding to the operation data, and acquiring a first detection result according to the grade probability. And acquiring a second detection result corresponding to the operation data. And comparing the first detection result with the second detection result to obtain an operation detection result, and sending the operation detection result to the terminal. In the operation data processing process, all operation data are detected through the linear model, corresponding operation detection results are obtained, the influence of personal factors on the operation detection results is avoided in the whole process, meanwhile, the operation data are analyzed without comparing historical operation data of the same type, and the accuracy of the operation detection results can be effectively improved.

Drawings

Fig. 1 is an application scenario diagram of an operation data detection method in an embodiment;

FIG. 2 is a flow diagram illustrating a method for operational data detection in one embodiment;

FIG. 3 is a flow chart illustrating the operation data preprocessing step according to an embodiment;

FIG. 4 is a flowchart illustrating a location feature value normalization step in one embodiment;

FIG. 5 is a flowchart illustrating the operation detection score generation step in one embodiment;

FIG. 6 is a flowchart illustrating the operation detection result generation step in one embodiment;

FIG. 7 is a block diagram showing the structure of an operation data detecting apparatus according to an embodiment;

FIG. 8 is a diagram illustrating an internal structure of a computer device according to an embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.

The operation data detection method provided by the application can be applied to the application environment shown in fig. 1. Wherein the server 104 communicates with the terminal 102 via a network. After receiving the operation data sent by the terminal 102, the server 104 obtains data characteristics according to the preprocessing logic calculation. The server 104 inputs the data characteristics into a pre-stored linear model to obtain a level probability corresponding to the operation data, obtains a first detection result according to the level probability, and obtains a second detection result corresponding to the operation data. The server 104 compares the first detection result with the second detection result to obtain an operation detection result, and sends the operation detection result to the terminal 102. The terminal 102 may be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, and portable wearable devices, and the server 104 may be implemented by an independent server or a server cluster formed by a plurality of servers.

In one embodiment, as shown in fig. 2, an operation data detection method is provided, which is described by taking the application of the method to the server in fig. 1 as an example, and includes the following steps:

step 202, obtaining operation data, and analyzing the operation data to obtain data classification information.

Specifically, the operation data refers to data to be detected sent by the terminal to the server, each piece of operation data includes a data item and a true value corresponding to the data item, the data item may be an operation index or an operation position corresponding to the operation data, and the true value corresponding to the data item may be a specific numerical value in the operation data. The operation data may be operation location data including a location item and a true location value corresponding to the location item, for example, the location item may be the number of 4s stores in a range of 3 kilometers from a store, the true location value corresponding to the location item may be 2, and the operation data may also be operation index data including an index item and a true index value corresponding to the index item, for example, the index item may be the number of visitors to the store, and the true index value corresponding to the index item may be 30. The data classification information is information obtained from the operation data and used for representing a category corresponding to the operation data, and may be information used for instructing the server to calculate data characteristics corresponding to the operation data, or may be a flag used for instructing the server to query a data item corresponding to the operation data, where the category may be a location category or an index category. Specifically, the terminal is connected with the server through the network, the server receives the operation data sent by the terminal, and obtains data classification information corresponding to the operation data by analyzing the operation data, wherein the data classification information is used for identifying the category and the data item of the operation data.

For example, a shop operation management APP (Application) is installed on the terminal, the terminal sends shop operation data to the server through the APP, the current operation data needs to be detected for detecting the current operation condition of the shop and formulating a later-stage operation target, the detection result is displayed on an interface corresponding to the APP through the terminal, and meanwhile, an operation suggestion for the shop is output according to the detection result. The terminal is connected with the server through the network and sends the operation data to the server. If the number of visiting persons in the store is 25, the category corresponding to the number of visiting persons in the store, which is obtained by analyzing the operation data and is the number of visiting persons in the store, is the index class, the data item is the number of visiting persons in the store, the true value is 25, the data classification information corresponding to the operation data is the operation index data, and the number of visiting persons in the store is 25.

And step 204, calculating to obtain data characteristics according to the data classification information.

Specifically, the data feature refers to information that is resolvable by the linear model and can represent the data item of the operation data and the true value corresponding to the data item. The server may obtain the preprocessing logic, preprocess the operation data by using the preprocessing logic to obtain the corresponding data feature, obtain the data item of the operation data according to the data feature, and obtain a preprocessing result calculated according to the true value corresponding to the data item, where the preprocessing result may be understood as an evaluation result of the server on the operation data. The data characteristics correspond to operation data, and the data characteristics can be divided into position characteristics and index characteristics, wherein the position characteristics comprise position items and corresponding preprocessing results, for example, the number of 4s shops of an automobile within 3 kilometers of a shop is 2, the position items are the 4s shops of the automobile within 3 kilometers of the shop, the preprocessing results are obtained by calculation according to 2, for example, 0.9, the index characteristics comprise the index items and corresponding preprocessing results, for example, the number of visit people to the shop is 25, the index items are the number of visit people to the shop, the preprocessing results are codes obtained by calculation according to 25 people, and the grades are high-grade corresponding codes.

Specifically, after receiving the operation data sent by the terminal, the server analyzes the operation data to obtain corresponding data classification information, queries the category and the data item of the operation data according to the data classification information, further queries the preprocessing logic corresponding to the operation data according to the category and the data item, preprocesses the operation data according to the preprocessing logic to obtain a preprocessing result, and generates corresponding data characteristics according to the data classification information and the preprocessing result. It should be noted that the preprocessing logic is a data processing rule pre-stored in the server, and different preprocessing logics are required for preprocessing different types of operation data. The server supports the configuration of a plurality of preprocessing logics, so that preprocessing of various data is realized. In a specific implementation, the preprocessing logic may be a preprocessing logic corresponding to the operation location data, or may be a preprocessing logic corresponding to the operation index data.

For example, the server receives that the number of visiting persons to the store sent by the terminal is 25, analyzes the operation data to obtain data classification information, wherein the corresponding category is an index category, the corresponding data item is the number of visiting persons to the store, inquires about preprocessing logic corresponding to the number of visiting persons to the store of 25 according to the index category and the number of visiting persons to the store, preprocesses the number of visiting persons to the store of 25 according to the preprocessing logic to obtain a preprocessing result of the number of visiting persons to the store, and further generates corresponding data characteristics according to the data classification information and the preprocessing result.

And step 206, inputting the data characteristics into the linear model to obtain a grade probability corresponding to the operation data, and obtaining a first detection result according to the grade probability.

Specifically, the linear model is a model that is stored in the server in advance and can be calculated by data characteristics to obtain a detection result corresponding to the operation data, and the linear model may be at least one of a logistic regression model, a least square model, a linear regression model, a decision tree, and the like that are obtained by training using sample data. In a specific implementation, the server obtains sample data, which may be positive sample data and negative sample data, performs machine learning on sample characteristics corresponding to the sample data and sample results corresponding to the sample data to obtain an initial model, and optimizes the initial model to obtain a linear model. The server detects the operation data through a linear model according to the data characteristics corresponding to the operation data, wherein the position characteristics contained in the data characteristics are only used as fixed variables influencing the detection process, the index characteristics contained in the data characteristics are used as variables influencing the detection process, and the index characteristics form an index range of the detection result output by the model through corresponding index items. The level probability refers to a probability that different processing results corresponding to the operation data appear in a preset period, and the level probability may be a level probability corresponding to the operation index data, such as a level probability of the number of visiting persons in a store or a level probability of sales volume. The first detection result is a target result expected to be achieved by the operation data, and may be a target level obtained according to a level probability corresponding to the operation data, wherein the target level may be a level calculated according to the level probability of the operation data, for example, the target level of the number of visitors to the store is a high level.

Specifically, the server obtains a linear model, inputs data characteristics into the linear model, obtains a level probability corresponding to the operation data through calculation, and obtains a target result of the operation index data according to the level probability as a first detection result.

For example, the data item of one piece of operation data is the number of visiting persons in the store, the server acquires a linear model, data features corresponding to all the operation data are input into the linear model, the level probability corresponding to the number of visiting persons in the store is obtained through calculation, a target result of the number of visiting persons in the store is obtained according to the level probability, and the target result is used as a first detection result of the number of visiting persons in the store.

It should be noted that the server may count the operation data of the preset period, and perform processing respectively to obtain the data characteristics of the preset period, so as to obtain the corresponding first detection result according to the linear model.

And 208, calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information.

Specifically, the conversion logic is a data conversion rule pre-stored in the server, and different conversion logics are required for the conversion process of different types of operation data. The server supports the configuration of a plurality of conversion logics, so that the conversion of various data is realized. In a particular implementation, the conversion logic may be conversion logic corresponding to the operation index data. The second detection result is an actual conversion result obtained according to the operation data, and may be an actual grade of the index item obtained according to the operation index data. Specifically, the server generates a corresponding data feature as a second detection result according to an actual conversion result obtained by the operation data. The server may obtain the operation index data according to the data classification information, obtain conversion logic corresponding to the operation index data according to the data classification information, and obtain an actual conversion result corresponding to the operation index data according to the conversion logic. It should be noted that the conversion logic is a data processing rule pre-stored in the server. In the specific implementation, the operation index data is obtained according to the conversion logic, the actual values corresponding to the operation index data are obtained according to the preset period, the actual values are summed to obtain the accumulated actual value, the conversion threshold value is obtained from the threshold database, and the conversion threshold value is compared with the accumulated actual value to obtain the actual conversion result corresponding to the operation index data.

For example, one of the pieces of operation data is the number of visiting persons to the store, and the server converts the number of visiting persons to the store within 30 days according to conversion logic corresponding to the number of visiting persons to the store, and obtains a conversion result of the number of visiting persons to the store within 30 days as a second detection result. When the preset period is 30 days, the server respectively obtains the real values of the number of visiting persons of the arriving store every day through the operation data, and the number of visiting persons of the arriving store 5 persons in the first day is the real value of the number of visiting persons of the arriving store in the first day, and the real values of the number of visiting persons of the arriving store every day are respectively obtained and summed to obtain the high grade of the number of visiting persons of the arriving store in 30 days, so that the number of visiting persons of the arriving store and the high grade are obtained. As a second detection of the number of visits to the store within 30 days.

And step 210, comparing the first detection result with the second detection result to obtain an operation detection result.

Specifically, the operation detection result refers to a result obtained after detecting the operation data, and the operation detection result may be a preprocessing result corresponding to the first detection result or a target value corresponding to an actual conversion result in the second detection result, for example, the target value corresponding to the second detection result obtained by the server according to the operation index data of the preset period.

Specifically, after receiving the first detection result and the second detection result, the server compares the first detection result and the second detection result to obtain an operation detection result, sends a conversion result corresponding to the first detection result as the operation detection result to the terminal when the first detection result and the second detection result are the same, and sends a target value corresponding to an actual conversion result in the second detection result as the operation detection result to the terminal when the first detection result and the second detection result are different.

The operation data detection method is used for detecting all operation data, the method achieves coverage of all operation data, in addition, data characteristics corresponding to the operation data are obtained through data classification information calculation, the grade probability of operation index data is obtained through a linear model, a first detection result is obtained through the grade probability, and the first detection result is compared with a second detection result corresponding to the operation data, so that the operation detection result is obtained, the influence of human factors on the detection process is avoided, and the accuracy of the operation detection result is improved.

In another embodiment, as shown in fig. 3, a flow chart of the data feature obtaining step is provided, where the data feature obtaining step, that is, the step of obtaining the data features according to the data classification information by calculation, includes:

and 304, when the data classification information represents that the operation data is the operation index data, inquiring the grade evaluation logic corresponding to the operation index data.

In particular, the rating logic refers to a rule that ranks the operation index data of the same index item. The ranking logic can be a rule for ranking the number of visitors to the store, for example, the shop with the number of visitors to the store ranked 25% is high, the shop with the number of visitors to the store ranked 25% -50% is medium, and the shop with the number of visitors to the store ranked 50% is low. The ranking logic may be a rule for ranking the follow-up rate of the customer, such as that the pre-stored follow-up rate is 50%, when the follow-up rate reaches 50%, the follow-up rate of the shop customer is at a high ranking, and when the follow-up rate does not reach 50%, the follow-up rate of the shop customer is at a low ranking. And obtaining a grade threshold corresponding to the grade according to the grade evaluation logic, wherein the grade threshold refers to the minimum value in the real values corresponding to the operation index data at a certain grade, and the minimum value required to be reached when the operation index data reaches the grade is represented. The grade threshold value can be a grade threshold value corresponding to the grade logic of the number of visitors to the store, for example, the number of visitors to the store is 30 persons at a high grade, 15 persons at a medium grade and 2 persons at a low grade. The rating threshold may also be a pre-stored value for rating in the rating logic, such as 50% of the rating for the customer's follow-up rate.

Specifically, after receiving the operation data sent by the terminal, the server identifies the type and item of the operation data according to the corresponding data classification information. When the data classification information represents that the operation data is the operation index data, the server inquires a grade evaluation logic corresponding to the operation data, and the grade of the operation data can be determined through the grade evaluation logic.

For example, if the number of visiting persons to the store is n, the data classification information corresponding to the operation data represents that the operation data is operation index data, and the ranking logic corresponding to the number of visiting persons to the store in the pre-stored server database is inquired. The ranking rule in the ranking logic may be that shops 25% of the number of visitors to the store are ranked high, shops 25% -50% of the number of visitors to the store are ranked medium, and shops 50% of the number of visitors to the store are ranked low. If the total number of shops is 40, the number of shops before 10 shops are ranked high, the number of shops 11-20 shops is ranked medium, and the number of shops after 20 shops are ranked low, wherein the number of visitors to the store from the day to the 10 th shop is 30, the number of visitors to the store from the day to the 20 th shop is 15, and the number of visitors to the store from the day to the 40 th shop is 2, then the number of visitors to the store from the day to the shop is 30, 15 and 2 in sequence according to the ranking threshold values.

And step 306, calculating the index grade corresponding to the operation index data according to the grade evaluation logic, and inquiring the target code corresponding to the index grade.

Specifically, the target code is a code representing a storage state corresponding to the index level of the operation index data, and the target code may be (100) at a high level, (010) at a medium level, or (001) at a low level.

Specifically, the server calculates an index grade corresponding to the operation data according to the inquired grade rating logic, and inquires a target code corresponding to the index grade.

For example, the number of visiting persons of the shop A is n, and the corresponding ranking logic of the number of visiting persons of the shop A is as follows: the shops 25% of the number of visiting people in the store are in a high grade, the shops 25% -50% of the number of visiting people in the store are in a medium grade, and the shops 50% of the number of visiting people in the store are in a low grade; and according to the grade evaluation logic, combining the number of visitors to the store on the day of all shops to obtain the number of visitors to the store on the day, wherein the high-grade segmentation threshold value is 30, the middle-grade segmentation threshold value is 15 and the low-grade segmentation threshold value is 2. If the number of the shop-to-shop visitors is 25% of the rank, if n is 50, the number of the shop-to-shop visitors is high, and the corresponding target code is (100); if the number of the shop-to-shop visitors is ranked at 25% -50%, if n is 20, the number of the shop-to-shop visitors is in a middle level, and the corresponding target code is (010); if the number of the shop-to-shop visitors is 50% after ranking, if n is 8, the number of the shop-to-shop visitors is low, and the corresponding target code is (001).

And 308, generating index characteristics according to the target codes as data characteristics.

Specifically, the index feature refers to information which is obtained according to the target code and the data classification information, can represent an index item of the operation index data and a true value corresponding to the index item, and can be analyzed by the linear model. The index features comprise index items of the operation index data and target codes corresponding to index levels, if the level of the number of people visiting from the shop A to the shop is high, and the corresponding target code is (100), the number of people visiting from the shop A to the shop (100) is the index features of the number of people visiting from the shop A to the shop.

Specifically, the server generates an index feature according to the target code and the data classification information corresponding to the operation index data, where the index feature is a data feature of the operation index data.

For example, if the number of people visiting the store from a certain day is ranked high and the target code corresponding to the high ranking is (100), the index feature corresponding to the high ranking is the number of people visiting the store from the store a (100).

In this embodiment, when the data classification information represents that the operation data is the operation index data, the operation index data is processed according to the corresponding grade evaluation logic to obtain a corresponding grade, and the index feature is generated according to the index item and the target code by querying the target code corresponding to the grade. In the process of processing the operation index data, the operation index data is converted into codes under the unified standard, so that the operation index data can be conveniently processed by the linear model, and meanwhile, the accuracy of detecting the operation index data by the linear model can be ensured.

In another embodiment, as shown in fig. 3, a flow chart of the data feature obtaining step is provided, where the data feature obtaining step, that is, obtaining the data features by calculating according to the data classification information, includes:

and step 310, when the data classification information represents that the operation data is operation position data, acquiring a position data coefficient corresponding to the operation position data.

Specifically, the location data coefficient refers to a pre-stored coefficient related to a location type, and the coefficient represents the degree of influence of the operation location data on the operation condition, and may be that the larger the value of the location data coefficient is, the larger the degree of influence on the operation condition is, and the smaller the value of the location data coefficient is, the smaller the degree of influence on the operation condition is. It can be understood that the coefficient of the position data of the automobile 4s store within 3 kilometers from the shop is 0.5, and the coefficient of the position data of the second-hand automobile market within 3 kilometers from the shop can also be 0.4.

Specifically, after receiving the operation data sent by the terminal, the server identifies the data type and the data item according to the data classification information. And when the data category is a position category, acquiring position statistical logic corresponding to the operation position data, and acquiring a position data coefficient corresponding to the operation data according to the position statistical logic.

For example, the server receives that the number of the automobile 4s stores within 3 km of the terminal from the store is 3, acquires the statistical logic of the automobile 4s stores according to the location class and the number of the automobile 4s stores, and inquires that the location coefficient of the automobile 4s store is 0.5 according to the statistical logic of the automobile 4s store. If the position data coefficient of the used car market is 0.4, the influence of the car 4s store on the shop operation is large, and the influence of the used car market on the shop operation is small.

In step 312, the location data amount corresponding to the operation location data is extracted.

Specifically, the location data amount refers to a true value representing the number of such locations in the operational location data. The amount of location data may be the actual value in the operational location data representing the number of such locations. Specifically, the server extracts a true value corresponding to the operation location data as the location data amount.

For example, the operation location data is 4 cells within 3 km from the store, and the server extracts the real values of 4 cells in the operation data as the location data amount of the number of cells within 3 km from the store.

And step 314, calculating according to the position data coefficient and the position data quantity to obtain a position characteristic value, and normalizing the position characteristic value to obtain a normalized position characteristic value.

Specifically, the position feature value refers to a numerical value obtained by multiplying the position data coefficient and the position data amount. Specifically, the server calculates a position characteristic value according to a position data coefficient and a position data amount corresponding to the operation position data, and normalizes the position characteristic value to obtain a normalized position characteristic value.

For example, the server calculates the position data coefficients 0.5 and the position data amount 3 corresponding to 3 automobile 4s stores within 3 km from the shop, multiplies 1.5 by 0.5 to obtain 3, and normalizes the position feature value to obtain the normalized position feature value 0.5.

Step 316, generating a position feature as a data feature from the normalized position feature value.

Specifically, the location feature refers to a data feature obtained by normalizing and calculating the operation index data according to the location statistical logic, and the location feature may be a location item of the operation location data and a normalized location feature value, such as 0.5 of a 4s store of an automobile within 3 km from a shop.

Specifically, after obtaining the normalized position feature value, the server generates a position feature as a data feature corresponding to the operation position data in combination with the data classification information.

In this embodiment, when the operation data is operation location data, the operation location data is counted according to the location counting logic. In the operation position data preprocessing process, the operation position data is converted into a numerical value, and the accuracy of the linear model for processing the operation data is improved.

In another embodiment, as shown in fig. 4, a schematic flow chart of a step of normalizing a position feature value is provided, where the step of normalizing the position feature value to obtain a normalized position feature value includes:

and 402, acquiring standard data, calculating a standard characteristic value according to the standard data, and calculating to obtain a maximum difference value corresponding to the standard characteristic value.

Specifically, the standard data refers to historical position data used to normalize position feature values, which are distributed over a large range, to a small range. The standard characteristic value refers to a position characteristic value obtained by standard data according to position statistical logic. The maximum difference corresponding to the standard feature value refers to a difference between the maximum value and the minimum value in the standard feature value, and represents an interval range of the standard feature value distribution. For example, if the standard feature values of 4s stores of the automobile within 3 km from the shop are 0, 1, 2, and 3, the maximum difference value corresponding to the standard feature value is 3.

Specifically, after the server acquires the standard data from the terminal, the standard characteristic value is calculated according to the position statistical logic, the maximum value and the minimum value in the standard characteristic value are extracted, and the difference value between the maximum value and the minimum value obtained through calculation is the maximum difference value corresponding to the standard characteristic value.

For example, after the server acquires the standard data of the automobile 4s store within 3 km from the shop from the terminal, the standard feature values are calculated to be 0, 1, 2 and 3 according to the position statistical logic, the maximum value and the minimum value of the extracted standard feature values are 3 and 0, and the difference between the maximum difference value corresponding to the standard feature value and 3 and 0 is 3, that is, the maximum difference value corresponding to the standard feature value of the automobile 4s store.

And step 404, calculating according to the minimum value in the standard characteristic values and the position characteristic value to obtain a position characteristic value difference value.

Specifically, the position feature value difference refers to a difference of the position feature value from the minimum value among the standard feature values. Specifically, after the server obtains the minimum value of the position characteristic value and the standard characteristic value, the difference between the position characteristic value and the standard characteristic value is obtained and used as the position characteristic value difference.

For example, after the server acquires the standard features of the 4s car stores within 3 km from the store, the minimum value of the extracted standard feature values is 0, and the position feature value is 1.5, and then the difference 1.5 between the position feature value 1.5 and the minimum value 0 of the standard feature values is the position feature value difference.

And 406, calculating through the maximum difference value of the standard characteristic values and the difference value of the position characteristic values to obtain a normalized position characteristic value.

Specifically, the normalized position feature value refers to a value obtained by normalizing the position feature value. May be a number in the range of [0, 1], such as 0.5, 0.8.

Specifically, after obtaining the difference between the position characteristic values and the maximum difference between the standard characteristic values, the server obtains the quotient of the position characteristic values and the maximum difference between the standard characteristic values to obtain the standardized position characteristic value.

For example, the difference of the position characteristic values of the 4s stores of the automobile within 3 km from the shop, which is obtained by the server, is 1.5, and the maximum difference of the standard characteristic values is 3, then the difference is 0.5 obtained by dividing 1.5 by 3, which is the normalized position characteristic value of the 4s stores of the automobile within 3 km from the shop.

In the embodiment, the position characteristic value of the operation position data is standardized, so that the position characteristic value corresponding to the operation position data is input into the linear model, the validity of the operation position data is ensured, and the comprehensiveness of operation data acquisition can be ensured.

In another embodiment, as shown in fig. 5, a flow chart of the operation detection score generating step is provided, and after inputting the data characteristics into the linear model and obtaining the level probability corresponding to the operation data, the method further includes:

and 502, obtaining an operation probability value corresponding to the operation index data according to the level probability.

Specifically, the operation probability refers to the probability of occurrence of a certain piece of operation data obtained by the server according to the level probability. The operation index data may be divided into operation reference index data and operation target index data, wherein the operation reference index data refers to data of a reference type index when analyzing the operation status, for example, the number of visitors to a store, and the operation target index data refers to data of a target type index corresponding to the purpose of analyzing the operation status, for example, the number of sales of vehicles, and the recognition degree. The operation probability value is a probability of occurrence of index data of a certain operation, such as a probability that sales on the first day is 3 or a probability that acceptance on the third day is 90%. It can be understood that, the server obtains the probability value of the occurrence of the operation target index data through a logistic regression function on the operation reference index data and the operation target index data, wherein the logistic regression function may be a Sigmod function.

Specifically, after obtaining the level probability output by the linear model, the server inputs the level probability to the logic function to obtain an index probability value corresponding to the operation index data as an operation probability value.

For example, the operation objective index is sales volume per day, and within 30 days, the operation probability value of the sales volume can be calculated by using formula (1), where formula (1) is:

wherein P is the operation probability value of sales volume per day, g_iIs an index level probability, r_iAs the coefficient corresponding to the index, the coefficient corresponding to the level having the highest level probability is 1, and the coefficients corresponding to the remaining levels are 0.

And step 504, calculating to obtain an operation detection score according to the operation probability value and the operation index data.

Specifically, the operation detection score refers to a weighted sum of the operation probability value and the actual value in the operation index data, and may be a weighted sum of the operation probability value and the actual value in the operation target index data, for example, the operation detection score of the sales amount is a weighted sum of the sales amount of a certain day and the probability of the sales amount occurring in the certain day.

Specifically, the server obtains the operation probability value and the operation target index data, extracts the index true value in the operation target index data, and obtains the operation detection score by weighting and summing the operation probability value and the index true value.

For example, the operation objective index is sales volume per day, and within 30 days, the operation detection score of the sales volume can be calculated by using formula (2), where formula (2) is:

wherein S is the sales operation detection score, x_jSales on day j, y_jAccording to the index grade probabilitySales by day j was x_jThe probability of (c).

Step 506, the operation detection score is sent to the terminal.

Specifically, the server obtains an operation detection score according to the operation probability value and the index true value of the operation target index data, and sends the operation detection score to the terminal.

In this embodiment, the server obtains the operation detection score by weighting and summing the operation probability value and the true value of the operation data, and analyzes the operation status by obtaining the operation detection score in combination with the operation probability, thereby avoiding the influence of special conditions on the operation status and effectively improving the general applicability of the operation data processing method.

In another embodiment, as shown in fig. 6, a schematic flow chart of the operation detection result generating step is provided, where the operation detection result generating step, that is, the step of comparing the first detection result with the second detection result to obtain an operation detection result and sending the operation detection result to the terminal, includes:

step 602, extracting a target grade corresponding to the first detection result, and generating the target grade according to the grade probability.

Specifically, the target level is a level determined according to the level probability, and is also a level that the index item is expected to reach, and may be a level of the index item in a period determined according to the level probability in the period.

Specifically, the server extracts a target level corresponding to the first detection result according to the level probability corresponding to the operation index data. The level probabilities corresponding to the index items may be sorted, and the level with the higher probability value may be used as the target level of the index item.

For example, the server outputs, through the linear model, that when the level corresponding to the number of visiting persons in the store within 30 days is a high level, the probability is 40%, when the level is a medium level, the probability is 25%, and when the level is a low level, the probability is 35%, and the probability values are sorted to obtain a higher probability value of 40%, where when the level corresponding to 40% is a high level, the level corresponding to the number of visiting persons in the store within 30 days is a high level, that is, the target level of the index item is a high level.

Step 604, obtaining a to-be-compared grade corresponding to the second detection result, wherein the to-be-compared grade is calculated according to the operation data.

Specifically, the to-be-compared grade refers to an actual grade obtained by the server according to the operation index data after the operation index data is accumulated. In the preset period, the server may obtain the actual level of the accumulated operation index data in the preset period according to the conversion logic.

Specifically, the server obtains the grade to be compared corresponding to the second detection result according to the operation data. The server may obtain the operation data in the preset period, obtain the corresponding operation index data according to the data type information, respectively obtain the actual index values corresponding to the operation index data of each day according to the conversion logic, sum the actual index values to obtain the accumulated actual index value, and further obtain the grade corresponding to the operation index data, that is, the actual grade corresponding to the operation index data in the preset period, that is, the grade to be compared corresponding to the second detection result.

For example, one of the operation data is the number of visiting persons to the store, the preset period is 30 days, the server can obtain one number of visiting persons to the store every day, the actual values corresponding to the number of visiting persons to the store in each number of visiting persons to the store are respectively extracted, if the number of visiting persons to the store in the first day is 20, the actual value of the number of visiting persons to the store in the first day is 20, and the number of visiting persons to the store in the second day is 25, the actual value of the number of visiting persons to the store in the second day is 25, and 30 numerical values are counted up, the numerical values are weighted and summed to obtain 220, so that the accumulated number of visiting persons to the store in 30 days is 220, and the actual level corresponding to the number of visiting persons to the store in 30 days. The grade corresponding to the accumulated data is the actual grade corresponding to the operation index data within 30 days, and is also the grade to be compared corresponding to the second detection result.

And step 606, when the target grade is lower than or equal to the grade to be compared, taking the target grade as an operation detection result.

Specifically, the server compares the target level with the level to be compared, sends the target level as an operation detection result to the terminal when the target level is the same as the level to be compared, represents that the operation index data is stable when the target level is lower than the level to be compared, and sends the target level as the operation detection result to the terminal.

For example, when the target level and the comparison level of the number of visiting persons in the store are high within 30 days, the high level is used as an operation detection result of the number of visiting persons in the store and is sent to the terminal; and when the target grade is a medium grade and the to-be-compared grade is a high grade, sending the high grade serving as an operation detection result to the terminal.

Step 608, when the target level is higher than the to-be-compared level, querying a target value corresponding to the to-be-compared level, and taking the target value as an operation detection result.

Specifically, the target value is a level threshold corresponding to a level one level higher than the level to be compared. After the server preprocesses the operation index data, the minimum index true value in the grade range can be obtained, and the minimum index true value is used as the grade threshold value of the grade.

Specifically, the target grade is compared with the grade to be compared, when the target grade is higher than the grade to be compared, the grade to be compared is inquired, a grade threshold value corresponding to the grade higher than the grade to be compared by one grade is inquired to serve as a target value, and the target value is sent to the terminal as an operation detection result.

For example, if the target level of the number of store visits within 30 days is a high level, the actual level of the number of store visits is a medium level, that is, if the comparison level is a medium level, the level higher than the medium level is a high level, and a level threshold corresponding to the high level is extracted as a target value, if the number of store visits of a store is 220, and the high level threshold of the number of store visits is 250, the medium level threshold is 100, and the low level threshold is 0, the high level threshold of 250 is the target value, and 250 is output to the terminal as an operation detection result of the number of store visits. If the number of the accumulated number of the visiting persons of the shop is 80 and the grade threshold value is not changed, if the comparison grade of the number of the visiting persons of the shop is a low grade, and if the grade higher than the low grade is a medium grade, 100 persons of the grade threshold value of the medium grade is the target value, and 100 persons are output to the terminal as the operation detection result of the number of the visiting persons of the shop.

In the embodiment, the target grade is compared with the grade to be compared, and the operation index data is specifically analyzed according to the comparison result to give the operation detection result, so that the operation detection result can clearly represent the operation state and the target, and the referential property of the operation detection result is improved.

In another embodiment, a flow diagram of a linear model generation step is provided, the step comprising:

and acquiring sample characteristics corresponding to the sample data, inputting the sample characteristics into the initial model to obtain a sample result corresponding to the sample characteristics, and acquiring a target result corresponding to the sample data.

Specifically, the sample data refers to historical data for building a model and optimizing the model, which contains all of the historical index data and the historical position data. The initial model refers to a model obtained by training according to sample data and sample results, and the linear model can be obtained by optimizing the initial model. And the sample result is a detection result obtained by inputting the sample characteristics to the initial model after the sample characteristics are obtained according to the sample data. The target result refers to an actual result corresponding to the sample data.

Specifically, after receiving sample data sent by the terminal, the server processes the sample data to obtain corresponding sample characteristics, inputs the sample characteristics into the initial model to obtain corresponding sample results, and queries actual results corresponding to the sample data as target results.

And when the sample result is the same as the target result, taking the initial model as a linear model.

Specifically, when the sample result is the same as the target result, the initial model is matched with the detection data, and the initial model is suitable for detection of the detection data and is stored in the server as the linear model. In this embodiment, the linear model is obtained after the initial model is optimized, so that the adaptability of the linear model to the operation parameters is ensured.

It should be understood that although the various steps in the flow charts of fig. 2-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps in fig. 2-6 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performance of the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternating with other steps or at least some of the sub-steps or stages of other steps.

In one embodiment, as shown in fig. 7, there is provided an operation data detecting apparatus including: receiving module, preprocessing module, first detection module, second detection module and output module, wherein:

an obtaining module 702, configured to obtain operation data of a user, and analyze the operation data to obtain data classification information.

And a calculating module 704, configured to calculate data characteristics according to the data classification information.

The first detection module 706 is configured to input the data characteristics to the linear model, obtain a level probability corresponding to the operation data, and obtain a first detection result according to the level probability;

a second detection module 708, configured to calculate, according to the data classification information, a second detection result corresponding to the operation data by using a preset conversion logic;

the comparing module 710 is configured to compare the first detection result with the second detection result to obtain an operation detection result.

In one embodiment, the calculation module 704 includes: and the grade evaluation logic query unit is used for querying the grade evaluation logic corresponding to the operation index data when the data classification information represents that the operation data is the operation index data. And the target code query unit is used for calculating the index grade corresponding to the operation index data according to the grade evaluation logic and querying the target code corresponding to the index grade. And the data characteristic generating unit is used for generating the index characteristic as the data characteristic according to the target code.

In one embodiment, the calculation module 704 includes: and the position data coefficient acquisition unit is used for acquiring a position data coefficient corresponding to the operation position data when the data classification information represents that the operation data is the operation position data. And a location data amount extraction unit configured to extract a location data amount corresponding to the operation location data. And the characteristic value calculating unit is used for calculating according to the position data coefficient and the position data quantity to obtain a position characteristic value, and standardizing the position characteristic value to obtain a standardized position characteristic value. And a data feature generation unit configured to generate a position feature as a data feature from the normalized position feature value.

In one embodiment, the feature value calculation unit includes: and the maximum difference value calculating subunit is used for acquiring the standard data, calculating the standard characteristic value according to the standard data, and calculating to obtain the maximum difference value corresponding to the standard characteristic value. And the characteristic value difference operator unit is used for calculating according to the minimum value in the standard characteristic values and the position characteristic values to obtain the position characteristic value difference. And the characteristic value standardization subunit is used for calculating through the maximum difference value of the standard characteristic values and the difference value of the position characteristic values to obtain the position characteristic values after standardization.

In one embodiment, the first detection module 706 includes: and the operation probability value calculation unit is used for obtaining the operation probability value corresponding to the operation index data through the level probability. And the score calculating unit is used for calculating to obtain an operation detection score according to the operation probability value and the operation index data. And the score sending unit is used for sending the operation detection score to the terminal.

In one embodiment, the alignment module 710 includes: and the target grade extracting unit is used for extracting a target grade corresponding to the first detection result, and the target grade is generated according to the grade probability. And the to-be-compared grade acquisition unit is used for acquiring the to-be-compared grade corresponding to the second detection result, and the to-be-compared grade is calculated according to the operation data. And the comparison unit is used for taking the target grade as an operation detection result when the target grade is lower than or equal to the grade to be compared. And the sending unit is used for inquiring a target value corresponding to the grade to be compared when the target grade is higher than the grade to be compared, and taking the target value as an operation detection result.

In one embodiment, the operation data detection apparatus 700 further includes: and the model calculation module is used for acquiring sample characteristics corresponding to the sample data, inputting the sample characteristics into the initial model, acquiring a sample result corresponding to the sample characteristics, and acquiring a target result corresponding to the sample data. And the result comparison module is used for taking the initial model as a linear model when the sample result is the same as the target result.

For specific limitations of the operation data detection apparatus, reference may be made to the above limitations of the operation data detection method, which is not described herein again. The modules in the operation data detection device can be wholly or partially realized by software, hardware and a combination thereof. The modules can be embedded in a hardware form or independent from a processor in the computer device, and can also be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 8. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer device is used for storing operational data. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of operational data detection.

Those skilled in the art will appreciate that the architecture shown in fig. 8 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.

In one embodiment, there is provided a computer device comprising a memory storing a computer program and a processor implementing the following steps when the processor executes the computer program: and acquiring operation data, and analyzing the operation data to obtain data classification information. And calculating to obtain data characteristics according to the data classification information. And inputting the data characteristics into the linear model to obtain the grade probability corresponding to the operation data, and obtaining a first detection result according to the grade probability. And calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information. And comparing the first detection result with the second detection result to obtain an operation detection result.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and when the data classification information represents that the operation data is the operation index data, inquiring the grade evaluation logic corresponding to the operation index data. And according to the grade evaluation logic, calculating the index grade corresponding to the operation index data, and inquiring the target code corresponding to the index grade. And generating index characteristics according to the target codes as data characteristics.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and when the data classification information represents that the operation data is operation position data, acquiring a position data coefficient corresponding to the operation position data. A location data amount corresponding to the operation location data is extracted. And calculating according to the position data coefficient and the position data quantity to obtain a position characteristic value, and standardizing the position characteristic value to obtain a standardized position characteristic value. And generating a position feature as a data feature from the normalized position feature value.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and acquiring standard data, calculating a standard characteristic value according to the standard data, and calculating to obtain a maximum difference value corresponding to the standard characteristic value. And calculating according to the minimum value in the standard characteristic values and the position characteristic value to obtain a position characteristic value difference value. And calculating through the maximum difference value of the standard characteristic values and the difference value of the position characteristic values to obtain the position characteristic values after standardization.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and obtaining an operation probability value corresponding to the operation index data according to the level probability. And calculating to obtain an operation detection score according to the operation probability value and the operation index data. And sending the operation detection score to the terminal.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and extracting a target grade corresponding to the first detection result, wherein the target grade is generated according to the grade probability. And acquiring a grade to be compared corresponding to the second detection result, and calculating the grade to be compared according to the operation data. And when the target grade is lower than or equal to the grade to be compared, taking the target grade as an operation detection result. And when the target grade is higher than the grade to be compared, inquiring a target value corresponding to the grade to be compared, and taking the target value as an operation detection result.

In one embodiment, the processor, when executing the computer program, further performs the steps of: and acquiring sample characteristics corresponding to the sample data, inputting the sample characteristics into the initial model to obtain a sample result corresponding to the sample characteristics, and acquiring a target result corresponding to the sample data. And when the sample result is the same as the target result, taking the initial model as a linear model.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of: and acquiring operation data, and analyzing the operation data to obtain data classification information. And calculating to obtain data characteristics according to the data classification information. And inputting the data characteristics into the linear model to obtain the grade probability corresponding to the operation data, and obtaining a first detection result according to the grade probability. And calculating a second detection result corresponding to the operation data by adopting a preset conversion logic according to the data classification information. And comparing the first detection result with the second detection result to obtain an operation detection result.

In one embodiment, the computer program when executed by the processor further performs the steps of: and when the data classification information represents that the operation data is the operation index data, inquiring the grade evaluation logic corresponding to the operation index data. And according to the grade evaluation logic, calculating the index grade corresponding to the operation index data, and inquiring the target code corresponding to the index grade. And generating index characteristics according to the target codes as data characteristics.

In one embodiment, the computer program when executed by the processor further performs the steps of: and when the data classification information represents that the operation data is operation position data, acquiring a position data coefficient corresponding to the operation position data. A location data amount corresponding to the operation location data is extracted. And calculating according to the position data coefficient and the position data quantity to obtain a position characteristic value, and standardizing the position characteristic value to obtain a standardized position characteristic value. And generating a position feature as a data feature from the normalized position feature value.

In one embodiment, the computer program when executed by the processor further performs the steps of: and acquiring standard data, calculating a standard characteristic value according to the standard data, and calculating to obtain a maximum difference value corresponding to the standard characteristic value. And calculating according to the minimum value in the standard characteristic values and the position characteristic value to obtain a position characteristic value difference value. And calculating through the maximum difference value of the standard characteristic values and the difference value of the position characteristic values to obtain the position characteristic values after standardization.

In one embodiment, the computer program when executed by the processor further performs the steps of: and obtaining an operation probability value corresponding to the operation index data through the grade probability. And calculating to obtain an operation detection score according to the operation probability value and the operation index data. And sending the operation detection score to the terminal.

In one embodiment, the computer program when executed by the processor further performs the steps of: and extracting a target grade corresponding to the first detection result, wherein the target grade is generated according to the grade probability. And acquiring a grade to be compared corresponding to the second detection result, and calculating the grade to be compared according to the operation data. And when the target grade is lower than or equal to the grade to be compared, taking the target grade as an operation detection result. And when the target grade is higher than the grade to be compared, inquiring a target value corresponding to the grade to be compared, and taking the target value as an operation detection result.

In one embodiment, the computer program when executed by the processor further performs the steps of: and acquiring sample characteristics corresponding to the sample data, inputting the sample characteristics into the initial model to obtain a sample result corresponding to the sample characteristics, and acquiring a target result corresponding to the sample data. And when the sample result is the same as the target result, taking the initial model as a linear model.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above-mentioned embodiments only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the concept of the present application, which falls within the scope of protection of the present application. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A method of operational data detection, the method comprising:

2. The method of claim 1, wherein the calculating data features from the data classification information comprises:

3. The method of claim 1, wherein the calculating data features from the data classification information comprises:

extracting a location data amount corresponding to the operation location data;

4. The method of claim 3, wherein the normalizing the position feature value to obtain a normalized position feature value comprises:

5. The method of claim 2, wherein after inputting the data features into a linear model and obtaining a class probability corresponding to the operational data, the method further comprises:

and sending the operation detection score to a terminal.

6. The method of claim 1, wherein the comparing the first and second detection results to obtain an operation detection result comprises:

7. The method of claim 1, wherein the linear model is generated in a manner that includes:

8. An operational data detection apparatus, the apparatus comprising:

9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor implements the steps of the method of any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 7.