CN110633311B

CN110633311B - Data processing method, device and storage medium

Info

Publication number: CN110633311B
Application number: CN201910864839.2A
Authority: CN
Inventors: 王辰正
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-09-12
Filing date: 2019-09-12
Publication date: 2023-10-20
Anticipated expiration: 2039-09-12
Also published as: CN110633311A

Abstract

The embodiment of the application discloses a data processing method, a data processing device and a storage medium, wherein the embodiment of the application acquires real-time feedback data in a preset time period; calculating a duty ratio interval of a feedback fluctuation frequency and a preset keyword occurrence frequency according to the real-time feedback data; acquiring a feedback reference mapping table generated based on a reference feedback fluctuation frequency and a reference duty ratio interval; and determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table. According to the scheme, the feedback result is obtained through the feedback fluctuation frequency and the duty ratio interval of the preset keyword occurrence frequency and the comprehensive analysis by combining the feedback reference mapping table, the accuracy of the feedback result determination is improved, and the reliability of the feedback result is greatly improved compared with the single analysis performed at the feedback quantity increasing speed.

Description

Data processing method, device and storage medium

Technical Field

The present application relates to the field of communications technologies, and in particular, to a data processing method, apparatus, and storage medium.

Background

At present, a user generally encounters some problems and feeds back in the process of using a browser, after receiving feedback based on the user using the browser, a server analyzes the feedback of the user, and the specific analysis process is to count the number of problem feedback and alarm abnormal feedback according to the number. For example, when the feedback quantity is gradually increased or the feedback quantity is increased at a high speed, the current browser is indicated to have a large problem, and the current browser needs to deal with the problem as soon as possible, and abnormal feedback is alarmed at the moment.

In the research and practice process of the prior art, the inventor of the application finds that single analysis is performed at the rate of increasing the feedback quantity, and the rate of finding abnormal feedback cannot be ensured, so that the analysis result is inaccurate, and the situation of false alarm is easy to occur due to the fact that the user feedback has data fluctuation, so that the reliability is not high.

Disclosure of Invention

The embodiment of the application provides a data processing method, a data processing device and a storage medium, aiming at improving the accuracy and the reliability of a feedback result.

In order to solve the technical problems, the embodiment of the application provides the following technical scheme:

a data processing method, comprising:

acquiring real-time feedback data in a preset time period;

calculating a duty ratio interval of a feedback fluctuation frequency and a preset keyword occurrence frequency according to the real-time feedback data;

acquiring a feedback reference mapping table generated based on a reference feedback fluctuation frequency and a reference duty ratio interval, wherein the feedback reference mapping table comprises a plurality of reference feedback fluctuation frequency ranges and problem feedback probability intervals obtained by dividing the plurality of reference duty ratio intervals;

and determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table.

In some embodiments, the determining, according to the feedback fluctuation frequency, the duty cycle interval, and the feedback reference mapping table, a feedback result corresponding to the real-time feedback data includes:

comparing the feedback fluctuation frequency with a reference feedback fluctuation frequency, and comparing the duty cycle section with a reference duty cycle section;

determining the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table according to the comparison result;

and determining a feedback result corresponding to the real-time feedback data according to the position.

In some embodiments, calculating the feedback fluctuation frequency from the real-time feedback data comprises:

acquiring the quantity of real-time feedback data and the accumulated value of the average value and the standard deviation of the historical feedback data;

calculating a difference between the amount of real-time feedback data and the accumulated value;

acquiring a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency;

and determining the feedback fluctuation frequency corresponding to the real-time feedback data from the historical feedback fluctuation frequency list according to the difference value.

In some embodiments, calculating the duty cycle interval of the occurrence frequency of the preset keyword according to the real-time feedback data includes:

Word segmentation processing is carried out on the real-time feedback data to obtain at least one word;

extracting preset keywords from the at least one word;

acquiring the ratio of the preset keywords to the total quantity of the real-time feedback data;

acquiring a historical duty cycle interval list generated based on a reference duty cycle interval;

and determining a duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio.

In some embodiments, before the obtaining the feedback reference map generated based on the baseline feedback fluctuation frequency and the baseline duty cycle interval, the method further comprises:

acquiring historical feedback data in a preset time period, and screening normal feedback data from the historical feedback data to obtain screened historical feedback data;

dividing the screened historical feedback data according to preset dates to obtain historical feedback data corresponding to a plurality of date intervals;

grouping the historical feedback data in each date interval according to a preset periodic strategy to obtain a plurality of groups of historical feedback data;

calculating a reference feedback fluctuation frequency corresponding to each group of historical feedback data;

Calculating a reference duty ratio interval of the occurrence frequency of the preset key words corresponding to each group of historical feedback data;

and generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval.

In some embodiments, the calculating the reference feedback fluctuation frequency for each set of historical feedback data comprises:

acquiring the number of each group of historical feedback data, and the average value and standard deviation corresponding to each group of historical feedback data;

calculating the accumulated sum among the mean value, standard deviation and preset threshold value of each group of historical feedback data;

and calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data according to the sum of the quantity and the accumulation sum.

In some embodiments, the calculating the reference duty cycle interval of the occurrence frequency of the preset keyword corresponding to each set of historical feedback data includes:

performing word segmentation on each group of historical feedback data to obtain at least one word, and extracting a preset keyword from the at least one word;

acquiring the number of preset keywords in each group of historical feedback data in the current group of historical feedback data and the total number of the current group of historical feedback data;

and calculating a reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each group of historical feedback data according to the ratio and the total number.

In some embodiments, the extracting the preset keyword from the at least one word includes:

extracting verbs and nouns from the at least one word;

calculating the frequency of occurrence of verbs in each set of historical feedback data, and calculating the frequency of occurrence of nouns in each set of historical feedback data;

and screening the verb with the highest frequency and the noun with the highest frequency to obtain the preset keywords.

A data processing apparatus comprising:

the first acquisition unit is used for acquiring real-time feedback data in a preset time period;

the calculating unit is used for calculating a duty ratio interval of the feedback fluctuation frequency and the occurrence frequency of the preset keyword according to the real-time feedback data;

the second acquisition unit is used for acquiring a feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty ratio interval;

and the determining unit is used for determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table.

In some embodiments, the determining unit is specifically configured to:

In some embodiments, the first computing unit is specifically configured to:

extracting preset keywords from the at least one word;

In some embodiments, the data processing apparatus further comprises:

the screening unit is used for acquiring the historical feedback data in a preset time period, and screening normal feedback data from the historical feedback data to obtain screened historical feedback data;

the dividing unit is used for dividing the screened historical feedback data according to preset dates to obtain historical feedback data corresponding to a plurality of date intervals;

the grouping unit is used for grouping the historical feedback data in each date interval according to a preset periodic strategy to obtain a plurality of groups of historical feedback data;

the second calculation unit is used for calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data;

the third calculation unit is used for calculating a reference duty ratio interval of the occurrence frequency of the preset key words corresponding to each group of historical feedback data;

and the generating unit is used for generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval.

In some embodiments, the second computing unit is specifically configured to:

In some embodiments, the third computing unit comprises:

the extraction module is used for carrying out word segmentation on each group of historical feedback data to obtain at least one word, and extracting preset keywords from the at least one word;

the acquisition module is used for acquiring the number of preset keywords in each group of historical feedback data accounting for the current group of historical feedback data and the total number of the current group of historical feedback data;

and the calculation module is used for calculating a reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each group of historical feedback data according to the ratio and the total number.

In some embodiments, the extraction module is specifically configured to:

extracting verbs and nouns from the at least one word;

A storage medium storing a computer program adapted to be loaded by a processor for performing any one of the data processing methods provided by the embodiments of the present application.

The embodiment of the application can acquire real-time feedback data in a preset time period, calculate the feedback fluctuation frequency and the duty ratio interval of the occurrence frequency of the preset keyword according to the real-time feedback data, and then acquire the feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty ratio interval, and at the moment, determine the feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table. According to the scheme, the feedback result is obtained through the feedback fluctuation frequency and the duty ratio interval of the preset keyword occurrence frequency and the comprehensive analysis by combining the feedback reference mapping table, the accuracy of the feedback result determination is improved, and the reliability of the feedback result is greatly improved compared with the single analysis performed at the feedback quantity increasing speed.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic diagram of a scenario of a data processing system provided by an embodiment of the present application;

FIG. 2 is a flow chart of a data processing method according to an embodiment of the present application;

FIG. 3 is a schematic diagram of another flow chart of a data processing method according to an embodiment of the present application;

FIG. 4 is a schematic diagram of a feedback reference mapping table according to an embodiment of the present application;

FIG. 5 is a schematic diagram of generating feedback results according to an embodiment of the present application;

FIG. 6 is another schematic diagram of generating feedback results according to an embodiment of the present application;

FIG. 7 is another schematic diagram of generating feedback results according to an embodiment of the present application;

FIG. 8 is a schematic diagram of a feedback monitoring interface provided by an embodiment of the present application;

FIG. 9 is another schematic diagram of a feedback monitoring interface provided by an embodiment of the present application;

FIG. 10 is a schematic diagram of a data processing apparatus according to an embodiment of the present application;

FIG. 11 is a schematic diagram of another structure of a data processing apparatus according to an embodiment of the present application;

fig. 12 is a schematic structural diagram of a server according to an embodiment of the present application.

Detailed Description

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to fall within the scope of the application.

The embodiment of the application provides a data processing method, a data processing device and a storage medium.

Referring to fig. 1, fig. 1 is a schematic view of a scenario of a data processing system provided by an embodiment of the present application, where the data processing system may include a data processing device, and the data processing device may be specifically integrated in a server, where the server may obtain real-time feedback data within a preset time period, for example, may receive real-time feedback data sent by a terminal, and may then calculate a duty cycle interval of a feedback fluctuation frequency and a preset keyword occurrence frequency according to the real-time feedback data, and obtain a feedback reference mapping table generated based on a reference feedback fluctuation frequency and a reference duty cycle interval. At this time, a feedback result corresponding to the real-time feedback data may be determined according to the feedback fluctuation frequency, the duty cycle interval, and the feedback reference mapping table, for example, the feedback fluctuation frequency is compared with the reference feedback fluctuation frequency, and the duty cycle interval is compared with the reference duty cycle interval, so as to determine a feedback result corresponding to the real-time feedback data according to the comparison result, and the feedback result may be sent to the management background, so that the management personnel may check the feedback result in time, take corresponding measures, and may also send the feedback result to the designated mailbox or the instant messaging account. The feedback result is obtained through the feedback fluctuation frequency and the duty ratio interval of the preset keyword occurrence frequency and the comprehensive analysis by combining the feedback reference mapping table, and the accuracy and the reliability of the feedback result determination are improved.

It should be noted that, the schematic view of the scenario of the data processing system shown in fig. 1 is only an example, and the data processing system and scenario described in the embodiment of the present application are for more clearly describing the technical solution of the embodiment of the present application, and do not constitute a limitation on the technical solution provided by the embodiment of the present application, and those skilled in the art can know that, with the evolution of the data processing system and the appearance of a new service scenario, the technical solution provided by the embodiment of the present application is equally applicable to similar technical problems.

The following will describe in detail.

In the present embodiment, description will be made from the viewpoint of a data processing apparatus which may be integrated in a network device such as a server or gateway.

Referring to fig. 2, fig. 2 is a flow chart of a data processing method according to an embodiment of the application. The data processing method may include:

s101, acquiring real-time feedback data in a preset time period.

For example, the terminal such as a mobile phone or a computer may receive real-time feedback data sent at intervals of a preset time period, or extract real-time feedback data from a database at intervals of a preset time period, or the like. The preset time period can be flexibly set according to actual needs, for example, the preset time period can be 20 minutes or 30 minutes, that is, real-time feedback data can be obtained every 20 minutes or 30 minutes, or feedback data in the first 1 hour can be obtained every 20 minutes, and the like, so that the real-time feedback data can be analyzed, problems can be found timely, and corresponding measures can be taken timely.

The feedback data may be feedback data generated based on user experience or feedback data generated by the application program when running in the process of using the application program such as a browser, instant messaging, mailbox or game, etc. For example, during 2019, 6, 03, 9:00:00 user A uses the browser, feedback is generated: the error code err_connection_time_out refreshes the web page |view solution; for another example, during play of user B using the web page at 2019, 7, 16, 11:30:00, feedback is generated: the XX game cannot be played; for another example, during 2019, 8, 28, 15:20:00, user B uses the browser, feedback is generated: how to download Applications (APP) using a browser, etc.

S102, calculating a duty ratio interval of the feedback fluctuation frequency and the occurrence frequency of the preset keywords according to the real-time feedback data.

S103, a feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty ratio interval is obtained.

In order to accurately analyze data, corresponding measures are convenient to take in time, after real-time feedback data is obtained, feedback fluctuation frequency can be calculated based on the real-time feedback data, and a duty ratio interval of the occurrence frequency of a preset keyword is calculated based on the real-time feedback data, wherein the feedback fluctuation frequency can be a feedback fluctuation frequency interval in a preset time period and is used for representing the condition of problem feedback, for example, more than 80% of the real-time feedback data in one hour are feedback browsers are not opened, or less than 80% of the real-time feedback data in one hour are feedback accounts which cannot be logged in. The preset keywords may be nouns and verbs extracted from the real-time feedback data, or nouns and verbs extracted from the real-time feedback data with highest occurrence frequency, or the like, and the preset keywords may be one or more, for example, keywords extracted from the real-time feedback data, such as "web pages are not opened" and the like. The duty cycle interval of the occurrence frequency of the preset keyword may be a duty cycle interval of the occurrence frequency of the preset keyword in the real-time feedback data within a preset time period, for example, the duty cycle interval of the occurrence frequency of the preset keyword "login" is within 20%.

In some embodiments, before the step S101 of acquiring the real-time feedback data within the preset time period, or before the step S102 of calculating the duty cycle interval of the feedback fluctuation frequency and the preset keyword occurrence frequency according to the real-time feedback data, or before the step S103 of acquiring the feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty cycle interval, the data processing method may further include: acquiring historical feedback data in a preset time period, and screening normal feedback data from the historical feedback data to obtain screened historical feedback data; dividing the screened historical feedback data according to preset dates to obtain historical feedback data corresponding to a plurality of date intervals; grouping the historical feedback data in each date interval according to a preset periodic strategy to obtain a plurality of groups of historical feedback data; calculating a reference feedback fluctuation frequency corresponding to each group of historical feedback data; calculating a reference duty ratio interval of the occurrence frequency of the preset key words corresponding to each group of historical feedback data; and generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval.

Specifically, first, the historical feedback data within a preset period of time, which may be flexibly set according to actual needs, may be acquired, for example, the historical feedback data of the past 1 year or 2 years may be acquired. In order to screen representative data for analysis, the accuracy of subsequent calculation is improved, normal feedback data (normal historical feedback data is used as reference data) can be screened from the historical feedback data, the screened historical feedback data is obtained, and the normal feedback data can be obtained by removing repeated severe feedback data. Since the normal feedback data is the vast majority and the abnormal feedback data is the minimum, the representativeness of the normal feedback data is relatively better, so that the feedback condition is judged through the distribution condition of the normal feedback data.

Because there is a certain difference in the feedback data of different periods (i.e. the user behaviors of different periods are inconsistent), in order to distinguish the feedback data of different periods, the filtered historical feedback data can be divided according to preset dates, so as to obtain the historical feedback data corresponding to a plurality of date intervals, the preset dates can be flexibly set according to actual needs, for example, the filtered historical feedback data can be divided according to the date intervals of the working day (i.e. the day is the working day and the second day is the working day), the rest day, the preparation rest day (i.e. the day is the working day and the second day is the rest day), and the like. Alternatively, it may be divided by working day and non-working day, and so on.

Then, the historical feedback data in each divided date interval can be grouped according to a preset period strategy to obtain multiple groups of historical feedback data, and the preset period strategy can be flexibly set according to actual needs, for example, the historical feedback data in each divided date interval can be grouped according to every 20 minutes, namely, the historical feedback data in every 20 minutes is used as a data group, or the historical feedback data in the previous 1 hour is obtained every 20 minutes as a data group, so that multiple groups of historical feedback data can be obtained. Wherein, each set of historical feedback data may include historical feedback data of the same time node on different days, for example, the historical feedback data set a may include historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 1, historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 2, historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 3, and so on.

At this time, a reference feedback fluctuation frequency corresponding to each set of historical feedback data may be calculated, where the reference feedback fluctuation frequency may be a fluctuation interval of all feedback data corresponding to a certain time node in a certain date period in the screened historical feedback data. In some embodiments, calculating the reference feedback fluctuation frequency for each set of historical feedback data may include: acquiring the number of each group of historical feedback data, and the average value and standard deviation corresponding to each group of historical feedback data; calculating the accumulated sum of the average value, the standard deviation and the preset threshold value of each group of historical feedback data; and calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data according to the sum of the quantity and the accumulation sum.

For example, the number of the same time point in each set of the historical feedback data may be acquired, for example, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 1 in the historical feedback data set a is acquired as a pieces, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 2 is acquired as b pieces, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 3 is acquired as c pieces, and so on; the mean value of each set of historical feedback data is then calculated, and the standard deviation of each set of historical feedback data is calculated.

At this time, the sum of the average value, the standard deviation and the preset threshold value of each set of historical feedback data can be calculated, the preset threshold value can be flexibly set according to actual needs, the preset threshold value can comprise one or more values, for example, the preset threshold value can comprise 1, 2, 3, 4, 5 and the like, when the average value is 7.42857 and the standard deviation is 0.37978, and when the preset threshold value is 1, the accumulated value=7.42857+0.37978+1= 8.80835; when the average value is 7.42857 and the standard deviation is 0.37978, the accumulated value=7.42857+0.37978+2= 9.80835 and so on when the preset threshold value is 2.

According to the number and the sum, calculating the corresponding reference feedback fluctuation frequency of each group of historical feedback data, for example, taking a rest day, 12:00:00 as an example, in the near calendar historical feedback data, acquiring the number of the historical feedback data of the rest day in the time node 12:00:00 (in the previous hour), calculating the average value of the historical feedback data of the time node to be 7.42857, the standard deviation to be 0.37978 and the sum of the average value and the standard deviation to be 7.80835, calculating the sum of the average value and the standard deviation, respectively obtaining accumulated values between the accumulated values and preset thresholds (including +1, +2, +3, +4, +5) to obtain 8.80835, 9.80835, 10.80835, 11.80835, 12.80835 and the like, and calculating the corresponding reference feedback fluctuation frequency of the group of the historical feedback data to be: '12:00:00': [7.42857,0.37978,0.5079,0.619,0.7937,0.8571,0.9206], wherein the data represents: 12:00:00 the mean value at the time node was 7.42857 in the rest day of the last year, the standard deviation was: 0.37978 50.79% of the historical feedback data is less than 8.80835, 61.9% of the historical feedback data is less than 9.80835, 79.37% of the historical feedback data is less than 10.80835, 85.71% of the historical feedback data is less than 11.80835, and 92.06% of the historical feedback data is less than 12.80835. According to the calculation mode, the reference feedback fluctuation frequency of the historical feedback data corresponding to each time node in each date interval can be obtained.

From the baseline feedback fluctuation frequency of the historical feedback data, a number of conclusions can be drawn, which may include, for example: (1) feedback data is expected; (2) greater than 80% feedback in the feedback data; (3) feedback data exceeds the limit expected feedback, (4) feedback data exceeds the average feedback by more than two times, (5) feedback data exceeds the average feedback by more than three times, and so on.

And calculating a reference duty ratio interval of the occurrence frequency of the preset keyword corresponding to each group of historical feedback data, wherein the reference duty ratio interval can be a duty ratio interval of the occurrence frequency of the keyword with the highest feedback frequency (including nouns, verbs and the like) in the normal historical feedback data, namely the highest frequency of the keyword in a specific feedback data in a certain date interval, and the number of intervals (for example, within 80% or within 20% and the like) in the normal historical feedback data.

In some embodiments, calculating the reference duty cycle interval of the occurrence frequency of the preset keyword corresponding to each set of historical feedback data may include: performing word segmentation on each group of historical feedback data to obtain at least one word, and extracting preset keywords from the at least one word; acquiring the number of preset keywords in each group of historical feedback data in the current group of historical feedback data and the total number of the current group of historical feedback data; and calculating a reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each group of historical feedback data according to the ratio and the total number.

Specifically, word segmentation processing can be performed on each group of historical feedback data according to a preset word segmentation strategy to obtain at least one word, and the preset word segmentation strategy can be flexibly set according to actual needs, for example, words such as "cipher cannot be used" in the historical feedback data are divided into words such as "cipher", "cannot be used" and "use". Then, a preset keyword is extracted from at least one word obtained by word segmentation, for example, a word having the highest frequency of occurrence may be extracted from the at least one word as a preset keyword, or a word of a specific semantic meaning may be extracted from the at least one word according to semantic information as a preset keyword, and so on.

In some embodiments, extracting the preset keyword from the at least one word may include: extracting verbs and nouns from at least one word; calculating the frequency of occurrence of verbs in each set of historical feedback data, and calculating the frequency of occurrence of nouns in each set of historical feedback data; and screening the verb with the highest frequency and the noun with the highest frequency to obtain the preset keywords.

For example, verbs and nouns may be extracted from at least one word, and frequencies of occurrence of verbs in each set of historical feedback data may be calculated, as well as frequencies of occurrence of nouns in each set of historical feedback data; the verb with the highest frequency is selected as a preset keyword, and the noun with the highest frequency is selected as the preset keyword, namely the preset keyword can comprise a plurality of nouns, verbs and the like.

At this time, the number of preset keywords in each set of historical feedback data accounting for the current set of historical feedback data and the total number of the current set of historical feedback data can be obtained, and the reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each set of historical feedback data is calculated according to the ratio and the total number. For example, taking the example that the rest day is at the time node 12:00:00, the number of the history feedback data of the last calendar history feedback data of the rest day time node 12:00:00 (in the previous hour) is 10, the 10 pieces of history feedback data are segmented, for example, the "password cannot be used" to obtain a password (noun), the use (verb) and the failure (non-verb non-noun can be ignored), then, the preset keywords with the largest occurrence number in the 10 pieces of history feedback data are calculated, if the 4 pieces of history feedback data have the "password" (noun) and the 3 pieces of history feedback data have the "account" (noun), the frequency corresponding to the occurrence of the "password" in the 10 pieces of history feedback data is 0.4, and the frequency is the highest and the preset keywords can be used as keywords.

The reference duty cycle interval for calculating the occurrence frequency of the keyword 'password' is 10: {0.8:0.20299,0.85:0,0.9:0,0.95:0.30299,1:0.40299, where 10 represents 10 pieces of historical feedback data, 0.8:0.20299 it represents: there is a 80% frequency of occurrence of keywords less than 0.20299, i.e., in the 80% baseline duty cycle interval when the frequency of occurrence of keyword a is less than 0.20299, and in the 20% baseline duty cycle interval when the frequency of occurrence of keyword a is greater than 0.20299. And so on, 0.85:0 represents: 85% of the keywords are less than 0;1:0.40299 it represents: the frequency of occurrence of 100% keywords is less than 0.40299, etc.

Based on the reference duty cycle interval of the occurrence frequency of the preset keyword, a plurality of conclusions can be drawn, which may include, for example: (1) The frequency of occurrence of the keywords is within an interval of 80% (i.e. the frequency of occurrence of the keywords accounts for 80% of the historical feedback data); (2) The frequency of occurrence of the keywords is within a 20 percent interval (namely, the frequency of occurrence of the keywords accounts for 20 percent in the historical feedback data); (3) the keyword occurrence frequency is within a 15% interval; (4) the occurrence frequency of the keywords is within a 10% interval; (5) the occurrence frequency of the keywords is within a 5% interval; (6) the occurrence frequency of the keywords is within the interval of 0%; etc.

Finally, a feedback reference mapping table may be generated according to the reference feedback fluctuation frequency and the reference duty ratio interval, where the feedback reference mapping table may be a problem interval table, that is, the feedback reference mapping table may include a plurality of reference feedback fluctuation frequency ranges and problem feedback probability intervals obtained by dividing a plurality of reference duty ratio intervals, and a plurality of different intervals may be divided by the reference feedback fluctuation frequency and the reference duty ratio interval respectively into transverse and longitudinal axes, where each interval regular script is distinguished by using different marks such as different colors or numbers, for example, green represents a normal interval, yellow represents a section with a relatively high probability of being problematic, red represents a section with a very high probability of being problematic, and so on. For example, when the reference feedback fluctuation frequency corresponding to the historical feedback data set a is greater than 80% and the reference duty cycle interval of the preset keyword occurrence frequency is within 20%, the interval corresponding to the feedback reference mapping table is a normal interval. When the reference feedback fluctuation frequency corresponding to the historical feedback data set B exceeds the average feedback by 3 times and the reference duty ratio interval of the occurrence frequency of the preset keyword is within 5%, the interval corresponding to the feedback reference mapping table is a problem interval with the maximum probability.

At this time, it is also possible to generate a history feedback fluctuation frequency list based on the reference feedback fluctuation frequency, a history duty section list based on the reference duty section, and so on. The generated feedback reference map, the historical feedback fluctuation frequency list, the historical duty cycle interval list, and the like may be stored for subsequent use.

After the real-time feedback data is acquired, a feedback fluctuation frequency corresponding to the real-time feedback data may be calculated, and in some embodiments, calculating the feedback fluctuation frequency according to the real-time feedback data may include: acquiring the quantity of real-time feedback data and the accumulated value of the average value and the standard deviation of the historical feedback data; calculating a difference between the amount of real-time feedback data and the accumulated value; acquiring a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency; and determining the feedback fluctuation frequency corresponding to the real-time feedback data from the historical feedback fluctuation frequency list according to the difference value.

Specifically, the number of real-time feedback data in the preset time period may be counted, for example, the number of real-time feedback data in the time period from the working day 2019, 8, 28, 9, 00, to the working day 2019, 8, 28, 9, 20, 00 is counted to be 10 pieces. And obtaining the average value and the standard deviation of the historical feedback data, wherein the historical feedback data can be the historical feedback data in a preset time period corresponding to a preset time period, the preset time period can be flexibly set according to actual needs, for example, the historical feedback data in 9:00:00 to 9:20:00 in the working days of the past 1 year, and the average value and the standard deviation of each group of historical feedback data are calculated in the process of generating the feedback reference mapping table, so that the average value and the standard deviation of the historical feedback data corresponding to the preset time period obtained by calculation can be obtained. Then, an accumulated value of the average value and the standard deviation of the historical feedback data is calculated, for example, when the average value is 7.42857 and the standard deviation is 0.37978, the accumulated value=7.42857+0.37978= 7.80835; the difference between the amount of real-time feedback data and the accumulated value may be calculated at this time, for example, when the amount of real-time feedback data in the period of time from 8:00 in 2019, 28:00 in 8, 28:9:00 in 2019, and the difference=10-7.80835 = 2.19165.

Then, a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency can be obtained, and the feedback fluctuation frequency corresponding to the real-time feedback data is determined from the historical feedback fluctuation frequency list according to the difference value, for example, the feedback fluctuation frequency corresponding to the time node 01:00:00 in the historical feedback fluctuation frequency list is used for: '01:00:00': as an example of [2.04762,0.73541,0.6508,0.8095,0.9365,0.9841,1.0], when the number of pieces of real-time feedback data received at 01:00:00 is 10, calculating the difference between the number of pieces of real-time feedback data and the accumulated value between the average value (2.04762) and the standard deviation (0.73541), and accumulating the sum between the average value, the standard deviation and the preset threshold value, wherein 7.21698 obtained by calculation is greater than 6.78303 and less than 7.78303, and the feedback data fluctuation frequency corresponding to the real-time feedback data at 01:00:00 is 0.9841 by querying the historical feedback fluctuation frequency list.

And, the duty cycle interval of the occurrence frequency of the preset keyword may be calculated according to the real-time feedback data, and in some embodiments, the calculating the duty cycle interval of the occurrence frequency of the preset keyword according to the real-time feedback data may include: performing word segmentation processing on the real-time feedback data to obtain at least one word; extracting preset keywords from at least one word; acquiring the ratio of the preset keywords to the total quantity of the real-time feedback data; acquiring a historical duty cycle interval list generated based on a reference duty cycle interval; and determining a duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio.

For example, the real-time feedback data may be subjected to word segmentation according to a preset word segmentation policy, so as to obtain at least one word, where the preset word segmentation policy may be flexibly set according to actual needs, for example, the real-time feedback data "web page super card" may be divided into words such as "web page", "super" and "card". Then extracting preset keywords from at least one word, for example, extracting verbs and nouns from at least one word; and calculating the frequency of verbs in the real-time feedback data, calculating the frequency of nouns in the real-time feedback data, screening out the verbs with the highest frequency and the nouns with the highest frequency, and obtaining preset keywords.

And acquiring the ratio of the preset keywords to the total amount of the real-time feedback data, and when the preset keywords comprise a plurality of preset keywords, acquiring the ratio of each preset keyword to the total amount of the real-time feedback data. For example, the ratio=the number of preset keywords in real-time feedback data (e.g., 4)/the total number of real-time feedback data (e.g., 10) =0.4,

and then, acquiring a historical duty ratio interval list generated based on the reference duty ratio interval, and determining the duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio. For example, the total number of real-time feedback data is 10, and the corresponding historical duty cycle interval list is 10: {0.8:0,0.85:0,0.9:0,0.95:0.31556,1:0.41556, since the ratio is 0.4,0.4 is greater than 0.31 and less than 0.41, the history duty cycle interval list can be queried to 0.95:0.31556, namely the duty ratio interval corresponding to the occurrence frequency of the preset keyword is 1-0.95=0.05 (i.e. 5%).

When the feedback reference map is required to be used, a feedback reference map generated based on the reference feedback fluctuation frequency and the reference duty cycle section, which is stored in advance, may be acquired.

S104, determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table.

After the feedback fluctuation frequency of the real-time feedback data and the duty ratio interval of the occurrence frequency of the preset keyword are obtained, and the feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty ratio interval, the feedback result corresponding to the real-time feedback data can be determined by inquiring the feedback reference mapping table, so that problems can be found in time according to the feedback result, and corresponding measures can be taken. The feedback result may include normal feedback, a problem with a high probability and a type of the problem, and so on.

In some embodiments, determining the feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty cycle interval, and the feedback reference mapping table may include: comparing the feedback fluctuation frequency with a reference feedback fluctuation frequency, and comparing the duty cycle section with a reference duty cycle section; determining the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table according to the comparison result; and determining a feedback result corresponding to the real-time feedback data according to the position.

Specifically, the feedback fluctuation frequency of the real-time feedback data can be compared with the reference feedback fluctuation frequency, the reference feedback fluctuation frequency matched with the feedback fluctuation frequency is determined, the duty ratio interval of the preset keyword occurrence frequency in the real-time feedback data is compared with the reference duty ratio interval, the reference duty ratio interval matched with the duty ratio interval is determined, then the positions of the reference feedback fluctuation frequency and the reference duty ratio interval obtained by matching in the feedback reference mapping table are determined, and the positions (namely the intervals) of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table can be obtained, and at the moment, the feedback result corresponding to the real-time feedback data can be determined according to the positions. For example, when the feedback fluctuation frequency of the real-time feedback data exceeds twice the average feedback and the duty ratio interval of the occurrence frequency of the preset keyword is within 15%, the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table is obtained to be the R interval, and at this time, it is possible to determine that the feedback result corresponding to the real-time feedback data is a larger probability according to the R interval.

The method described in the above embodiments is described in further detail below by way of example.

In this embodiment, the data processing device is taken as a server, and the server can accurately analyze feedback data generated in the process of using the browser by the terminal in multiple dimensions and alarm against abnormal feedback, so that problems can be quickly found (i.e., abnormal feedback), the problems can be timely processed, the data processing efficiency and reliability are improved, and negative effects caused by online problem feedback of the browser are reduced.

Referring to fig. 3, fig. 3 is a flowchart illustrating a data processing method according to an embodiment of the application. The method flow may include:

s201, the server receiving terminal is based on historical feedback data sent when the browser is used in a preset time period.

The preset time period can be flexibly set according to actual needs, for example, the server can acquire historical feedback data sent by the terminal when the terminal uses the browser in the past 1 year or 2 years, the historical feedback data can comprise normal feedback data, abnormal feedback data and the like, and the normal feedback data can be data fed back when the browser uses normally, for example, "the browser is good, all uses normally", "how to download the APP through the browser", and the like. The abnormal feedback data may be data fed back when the browser is abnormal in use, for example, "the new tab is frequently opened at a slow speed", "the browser has a problem, an error code occurs, and the browser is annoying", etc.

S202, the server screens normal feedback data from the historical feedback data to obtain screened historical feedback data.

In order to screen representative data for analysis, the accuracy of subsequent calculation is improved, the server can screen normal feedback data from the historical feedback data to obtain screened historical feedback data, and the normal feedback data can be data fed back when the browser is in normal use. Since the normal feedback data is the vast majority and the abnormal feedback data is the minimum, the representativeness of the normal feedback data is relatively better, so that the feedback condition is judged through the distribution condition of the normal feedback data.

And S203, the server divides the screened historical feedback data according to preset dates to obtain historical feedback data corresponding to a plurality of date intervals.

Because there is a certain difference in the feedback data of different periods, that is, the conditions of the user behaviors of different periods are inconsistent, for example, the durations of using the browser on the working day and the rest day may be different, so in order to distinguish the feedback data of different periods, the server may divide the filtered historical feedback data according to preset dates to obtain the historical feedback data corresponding to a plurality of date intervals, where the preset dates may be flexibly set according to actual needs, for example, the filtered historical feedback data may be divided into date intervals according to the working day WorkDay (i.e., the day is the working day and the second day is the working day), the rest day, the ready rest day ReadyRest (i.e., the day is the working day and the second day is the rest day), and the like.

S204, grouping the historical feedback data in each date interval by the server according to a preset periodic strategy to obtain a plurality of groups of historical feedback data.

The server may group the historical feedback data in each divided date interval according to a preset period policy to obtain multiple groups of historical feedback data, where the preset period policy may be flexibly set according to actual needs, for example, the historical feedback data in each divided date interval may be obtained as a data group according to the historical feedback data in the first 1 hour every 20 minutes, so as to obtain multiple groups of historical feedback data. Wherein, each set of historical feedback data may include historical feedback data of the same time node on different days, for example, the historical feedback data set a may include historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 1, historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 2, historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 3, and so on. For another example, historical feedback data for 9:20:00 to 9:40:00 for holiday 1, historical feedback data for 9:20:00 to 9:40:00 for holiday 2, and so forth may be included in the historical feedback data set B.

S205, the server calculates the reference feedback fluctuation frequency corresponding to each group of historical feedback data and the reference duty ratio interval of the preset keyword occurrence frequency.

The server may calculate a reference feedback fluctuation frequency corresponding to each set of historical feedback data, where the reference feedback fluctuation frequency may be a fluctuation interval of all feedback data corresponding to a certain time node in a certain date period in the filtered historical feedback data. Specifically, the server may acquire the number of each set of historical feedback data, and the average value and the standard deviation corresponding to each set of historical feedback data; and then calculating the accumulated sum of the mean value and the standard deviation of each group of historical feedback data and a preset threshold value, and calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data according to the accumulated sum of the number.

For example, the number of the same time point in each set of the historical feedback data may be acquired, for example, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 1 in the historical feedback data set a is acquired as a pieces, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 2 is acquired as b pieces, the historical feedback data of 12:00:00 to 12:20:00 corresponding to the working day 3 is acquired as c pieces, and so on; the average of each set of historical feedback data is calculated, e.g., n data of historical feedback data a, b, and c are included in the historical feedback data set a, and the average of the set of historical feedback data may be x= (a+b+c+)/n. And calculating a standard deviation of each set of historical feedback data, e.g., standard deviation s=sqrt (((a-x)/(2+ (b-x)/(2+) (xn-x)/(2)/n) for historical feedback data set a).

Then, calculating the sum of the average value, the standard deviation and a preset threshold value of each set of historical feedback data, wherein the preset threshold value can be flexibly set according to actual needs, and the preset threshold value can comprise one or more values, for example, 1, 2, 3, 4, 5 and the like, and when the average value is 7.42857 and the standard deviation is 0.37978, the preset threshold value is 1, the accumulated value=7.42857+0.37978+1= 8.80835; when the average value is 7.42857 and the standard deviation is 0.37978, the accumulated value=7.42857+0.37978+2= 9.80835 and so on when the preset threshold value is 2.

The server may calculate the reference feedback fluctuation frequency corresponding to each set of historical feedback data according to the sum of the number and the accumulation, for example, taking a rest day, 12:00:00 as an example, in the near calendar history feedback data, obtain the number of the historical feedback data of the rest day in the time node 12:00:00 (in the previous hour), calculate the average value corresponding to the historical feedback data of the time node as 7.42857, the standard deviation as 0.37978, and the sum of the average value and the standard deviation as 7.80835, calculate the sum of the average value and the standard deviation, and obtain the accumulated values between the average value and the preset thresholds (including +1, +2, +3, +4, +5) as 7.42857+0.37978+1= 8.80835, 7.42857+0.37978+2= 9.80835, 7.42857+0.37978+3= 10.80835, 7.37978+4= 11.80835, and 7.42857+5= 12.80835, and so on, and calculate the reference fluctuation frequency corresponding to the set of the historical feedback data as the reference fluctuation frequency. '12:00:00': [7.42857,0.37978,0.5079,0.619,0.7937,0.8571,0.9206], wherein the data represents: 12:00:00 the mean value at the time node was 7.42857 in the rest day of the last year, the standard deviation was: 0.37978 50.79% of the historical feedback data is less than 7.42857+0.37978+1, 61.9% of the historical feedback data is less than 7.42857+0.37978+2, 79.37% of the historical feedback data is less than 7.42857+0.37978+3, 85.71% of the historical feedback data is less than 7.42857+0.37978+4, and 92.06% of the historical feedback data is less than 7.42857+0.37978+5. According to the calculation mode, the reference feedback fluctuation frequency of the historical feedback data corresponding to each time node in each date interval can be obtained.

From the baseline feedback fluctuation frequency of the historical feedback data, a number of conclusions can be drawn, which may include, for example: (1) feedback data is in line with expectations (normal feedback); (2) greater than 80% feedback in the feedback data; (3) the feedback data exceeds the limit expected feedback (e.g., exceeds the mean + standard deviation + 5), (4) the feedback data exceeds the average feedback by more than two times, (5) the feedback data exceeds the average feedback by more than three times, etc.

The server can perform word segmentation processing on each group of historical feedback data according to a preset word segmentation strategy to obtain at least one word, and the preset word segmentation strategy can be flexibly set according to actual needs, for example, words such as "cipher" can not be used by a password "of the historical feedback data are divided into words such as" cipher "," unable "and" use ". Then, a preset keyword is extracted from at least one word obtained by word segmentation, for example, a word having the highest frequency of occurrence may be extracted from the at least one word as a preset keyword, or a word of a specific semantic meaning may be extracted from the at least one word according to semantic information as a preset keyword, and so on.

For example, the server may extract verbs and nouns from at least one word, calculate the frequency of occurrence of the verbs in each set of historical feedback data, and calculate the frequency of occurrence of the nouns in each set of historical feedback data; the verb with the highest frequency is selected as a preset keyword, and the noun with the highest frequency is selected as the preset keyword, namely the preset keyword can comprise a plurality of nouns, verbs and the like.

At this time, the server may obtain the number of preset keywords in each set of historical feedback data in the current set of historical feedback data, and the total number of the current set of historical feedback data, and calculate the reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each set of historical feedback data according to the ratio and the total number. For example, taking the example that the rest day is at the time node 12:00:00, the number of the history feedback data of the last calendar history feedback data of the rest day time node 12:00:00 (in the previous hour) is 10, the 10 pieces of history feedback data are segmented, for example, the "password cannot be used" to obtain a password (noun), the use (verb) and the failure (non-verb non-noun can be ignored), then, the preset keywords with the highest occurrence number (only 1 of the repeated occurrence of the keywords in each piece of history feedback data) in the 10 pieces of history feedback data are calculated, if the 4 pieces of history feedback data have the "password" (noun) and the 3 pieces of account (noun), the frequency corresponding to the occurrence of the "password" in the 10 pieces of history feedback data is 0.4, and the frequency is the highest and the preset keywords can be used as the keywords.

The server may calculate a reference duty cycle interval of the occurrence frequency of the keyword "password" as 10: {0.8:0.20299,0.85:0,0.9:0,0.95:0.30299,1:0.40299, where 10 represents 10 pieces of historical feedback data, 0.8:0.20299 it represents: there is a 80% frequency of occurrence of keywords less than 0.20299, i.e., in the 80% baseline duty cycle interval when the frequency of occurrence of keyword a is less than 0.20299, and in the 20% baseline duty cycle interval when the frequency of occurrence of keyword a is greater than 0.20299. That is, when the history feedback data is 10, the reference duty cycle section indicating the frequency of occurrence of the keyword is within 20% (100% -80%) when the frequency of occurrence of the keyword is 0.203 (> 0.20299), and the reference duty cycle section indicating the frequency of occurrence of the keyword is within 80% when the frequency of occurrence of the keyword is 0.2 (<= 0.20299). And so on, 0.85:0 represents: the frequency of occurrence of 85% of the keywords is less than 0 (0 indicates that the interval is consistent with 80%); 0.9:0 represents: the occurrence frequency of 90% of keywords is less than 0;0.95:0.30299 it represents: the frequency of occurrence of 95% of keywords is less than 0.30299;1:0.40299 it represents: the frequency of occurrence of 100% of the keywords is less than 0.40299. According to the calculation mode, the reference duty ratio interval of the keyword occurrence frequency of the historical feedback data corresponding to each time node in each date interval can be obtained.

S206, the server generates a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio section, a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency, and a historical duty ratio section list generated based on the reference duty ratio section.

The server may generate a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval, where the feedback reference mapping table may be a problem interval table, that is, the feedback reference mapping table may include a plurality of reference feedback fluctuation frequency ranges and problem feedback probability intervals obtained by dividing a plurality of reference duty ratio intervals, and may divide a plurality of different intervals with the reference feedback fluctuation frequency and the reference duty ratio interval respectively as transverse and longitudinal axes, and each interval regular script is distinguished by different color or number identifiers, for example, green represents a normal interval, yellow represents a section with a relatively high probability of being problematic, red represents a section with a very high probability of being problematic, and so on. For example, when the reference feedback fluctuation frequency corresponding to the historical feedback data set a is greater than 80% and the reference duty cycle interval of the preset keyword occurrence frequency is within 20%, the interval corresponding to the feedback reference mapping table is a normal interval. When the reference feedback fluctuation frequency corresponding to the historical feedback data set B exceeds the average feedback by 3 times and the reference duty ratio interval of the occurrence frequency of the preset keyword is within 5%, the interval corresponding to the feedback reference mapping table is a problem interval with the maximum probability.

At this time, the server may generate a history feedback fluctuation frequency list based on the reference feedback fluctuation frequency, for example, the history feedback fluctuation frequency list may be as follows:

{'ReadyRest'：{'01:00:00'：[2.04762，0.73541，0.6508，0.8095，0.9365，0.9841，1.0]，'01:20:00'：[1.4127，0.83313，0.7778，0.9683，1.0，1.0，1.0]，'01:40:00'：[1.1746，1.01234，0.873，0.9365，0.9841，1.0，1.0]，'02:00:00'：[0.87302，1.19862，0.9206，0.9841，0.9841，1.0，1.0]，'02:20:00'：[0.71429，1.20738，0.873，0.9841，0.9841，0.9841，1.0]，'02:40:00'：[0.55556，1.19044，0.9048，1.0，1.0，1.0，1.0]，'03:00:00'：[0.50794，1.44089，0.8889，0.9841，1.0，1.0，1.0]，'03:20:00'：[0.52381，1.51576，0.9841，1.0，1.0，1.0，1.0]，'03:40:00'：[0.50794，1.5239，0.9683，1.0，1.0，1.0，1.0]，......，'23:40:00'：[3.77778，0.50894，0.6825，0.8413，0.9365，0.9683，0.9683]，'00:00:00'：[3.33333，0.61412，0.5714，0.7143，0.8889，0.9365，0.9524]}；

'RestDay'：{'01:00:00'：[1.92254，0.7537，0.6972，0.8732，0.9577，0.9789，0.993]，'01:20:00'：[1.62676，0.7878，0.7746，0.9366，0.9789，0.9859，0.993]，'01:40:00'：[1.39437，0.79824，0.8732，0.9507，0.9859，1.0，1.0]，'02:00:00'：[1.21127，0.9691，0.8592，0.9366，0.993，1.0，1.0]，'02:20:00'：[1.02817，1.09469，0.8944，0.9648，0.9859，1.0，1.0]，'02:40:00'：[0.79577，1.16992，0.8099，0.9296，0.993，1.0，1.0]，'03:00:00'：[0.64789，1.33603，0.8592，0.9507，0.993，1.0，1.0]，......，'23:40:00'：[3.4507，0.64304，0.6972，0.838，0.9014，0.9507，0.9648]，'00:00:00'：[3.09859，0.73803，0.6479，0.7958，0.8662，0.8944，0.9366]}；

'WorkDay'：{'01:00:00'：[1.89069，0.74651，0.7085，0.8947，0.9433，0.9798，0.996]，'01:20:00'：[1.49798，0.82556，0.8219，0.9312，0.9676，1.0，1.0]，'01:40:00'：[1.21862，0.94302，0.8704，0.9636，0.9838，0.996，1.0]，'02:00:00'：[0.97166，1.04529，0.9069，0.9838，0.996，1.0，1.0]，'02:20:00'：[0.82996，1.16484，0.8178，0.9231，0.9919，0.996，0.996]，'02:40:00'：[0.65587，1.30788，0.8704，0.9757，0.9798，0.996，1.0]，'03:00:00'：[0.56275，1.32233，0.8947，0.9757，1.0，1.0，1.0]，......，'23:40:00'：[3.76923，0.5823，0.6599，0.8178，0.9109，0.9555，0.9798]，'00:00:00'：[3.21862，0.6581，0.5587，0.7449，0.8785，0.9555，0.9757]}}。

wherein ReadyRest represents the feedback fluctuation frequency corresponding to the preparation rest day, restDay represents the feedback fluctuation frequency corresponding to the rest day, workDay represents the feedback fluctuation frequency corresponding to the working day, and in one record, for example { '01:00:00': [1.89069,0.74651,0.7085,0.8947,0.9433,0.9798,0.996], the first number 01:00:00 represents a time node, the second number 1.89069 represents a mean value corresponding to feedback data, the third number 0.74651 represents a standard deviation corresponding to feedback data, the fourth number 0.7085 represents that 70.85% of feedback data is smaller than the mean value+the standard deviation+a first preset threshold (e.g. 1), the fifth number 0.8947 represents that 89.47% of feedback data is smaller than the mean value+the standard deviation+a second preset threshold (e.g. 2), the sixth number 0.9433 represents that 94.33% of feedback data is smaller than the mean value+the standard deviation+a third preset threshold (e.g. 3), the seventh number 0.9798 represents that 97.98% of feedback data is smaller than the mean value+the standard deviation+a fourth preset threshold (e.g. 4), and the 8 th number 0.996 represents that 99.6% of feedback data is smaller than the mean value+the standard deviation+a fifth preset threshold (e.g. 5). The understanding of other records and so on.

And, the server may generate a history duty cycle section list based on the reference duty cycle section, for example, the history duty cycle section list may be as follows:

{"ReadyRest_n"：{......，4：{0.8:0.29863，0.85:0，0.9:0，0.95:0.59863，1:0.79863}，5：{0.8:0，0.85:0，0.9:0，0.95:0.45961，1:0.85961}，6：{0.8:0，0.85:0，0.9:0，0.95:0.34467，1:0.74467}，......}；

"ReadyRest_v"：{......，3：{0.8:0.3859，0.85:0，0.9:0，0.95:0.6859，1:1.0859}，4：{0.8:0，0.85:0，0.9:0，0.95:0.52875，1:0.82875}，5：{0.8:0，0.85:0，0.9:0，0.95:0.48621，1:0.68621}，6：{0.8:0，0.85:0，0.9:0，0.95:0.36648，1:0.56648}，......，}；

"RestDay_n"：{......，4：{0.8:0.29511，0.85:0，0.9:0，0.95:0.59511，1:0.79511}，5：{0.8:0，0.85:0，0.9:0，0.95:0.45846，1:1.05846}，6：{0.8:0，0.85:0，0.9:0，0.95:0.34113，1:0.74113}，......，}；

"RestDay_v"：{......，4：{0.8:0，0.85:0，0.9:0，0.95:0.51695，1:0.81695}，5：{0.8:0，0.85:0，0.9:0，0.95:0.48493，1:0.88493}，6：{0.8:0，0.85:0，0.9:0.37037，0.95:0.57037，1:0.67037}，......，}；

"WorkDay_n"：{......，4：{0.8:0.30208，0.85:0，0.9:0，0.95:0.50208，1:0.80208}，5：{0.8:0，0.85:0，0.9:0，0.95:0.46307，1:0.86307}，6：{0.8:0，0.85:0，0.9:0，0.95:0.3392，1:0.8392}，......，}；

"WorkDay_v"：{......，4：{0.8:0，0.85:0，0.9:0，0.95:0.52418，1:1.02418}，5：{0.8:0，0.85:0，0.9:0，0.95:0.48927，1:0.68927}，6：{0.8:0，0.85:0，0.9:0.36976，0.95:0.56976，1:0.66976}，......，}}。

where n represents a noun, v represents a verb, readyrest_n represents a duty cycle section in which a holiday noun is prepared as a frequency of occurrence of a keyword, readyrest_v represents a duty cycle section in which a holiday verb is prepared as a frequency of occurrence of a keyword, restday_n represents a duty cycle section in which a holiday noun is a frequency of occurrence of a keyword, restday_v represents a duty cycle section in which a holiday verb is a frequency of occurrence of a keyword, workday_n represents a duty cycle section in which a WorkDay noun is a frequency of occurrence of a keyword, and workday_v represents a duty cycle section in which a WorkDay verb is a frequency of occurrence of a keyword. In one record, for example, 6: {0.8:0,0.85:0,0.9:0.36976,0.95:0.56976,1:0.66976}, the first number 6 represents the number of pieces of feedback data occupied by the keywords, and the second number 0.8:0 represents that the occurrence frequency of 80% of the keywords is less than 0; the third number 0.85:0 indicates that the frequency of occurrence of 85% of the keywords is less than 0; the fourth number 0.9:0.36976 indicates that 90% of the keywords occur less frequently than 0.36976; the fifth number 0.95:0.56976 indicates that 95% of the keywords occur less frequently than 0.56976; the sixth number, 1:0.66976, indicates that 100% of the keywords occur less frequently than 0.66976. The understanding of other records and so on.

The server may also generate a feedback reference map (which may also be referred to as a problem interval table) from the baseline feedback fluctuation frequency and the baseline duty cycle interval, for example, as shown in fig. 4. In the feedback reference mapping table, a plurality of different sections may be divided by taking the reference feedback fluctuation frequency and the reference duty ratio section as the horizontal axis and the vertical axis, and each section is distinguished by using marks such as different colors or numbers, for example, in fig. 4, when the reference feedback fluctuation frequency is greater than 80%, and the reference duty ratio section of the preset keyword occurrence frequency is within 20%, the section corresponding to the section in the feedback reference mapping table is a normal section.

The server may store the generated feedback reference map, the historical feedback fluctuation frequency list, the historical duty cycle interval list, and the like for subsequent use.

S207, the server receives real-time feedback data sent by the terminal when the browser is used in a preset time period.

For example, due to the optical fiber being cut off or the network, the feedback that the browser is used to have problems is continuous, and the server may receive real-time feedback data sent when the browser is used in a preset time period, where the preset time period may be flexibly set according to actual needs, for example, the real-time feedback data received in the first 1 hour is acquired every 20 minutes. Specific real-time feedback data may include: (1) XX space is not opened; (2) the XX game cannot be played; (3) Error code err_connection_time_out refresh web page |view solution|; (4) new tab pages often do not open at a very slow rate; (5) how the browser does not open today; (6) browser has problems and annoyance; etc.

For another example, since the flash chinese proxy pulls the browser into the blacklist, the terminal uses the browser to watch the video and other scenes through the flash, and pops up the flash unauthorized window to cause feedback, and at this time, the server can receive real-time feedback data sent by the terminal when using the browser. Specific real-time feedback data may include: (1) unauthorized use of flashlayer by the browser; (2); (3) a live video using window popup function is provided with a BUG; (4) Playing the web game prompts that the software is not authorized to run the adobe fash player; (5) the webpage is not opened; (6) the flash plug-in is unauthorized; etc.

S208, the server acquires the quantity of the real-time feedback data, and an accumulated value of the average value and the standard deviation of the historical feedback data, and calculates a difference value between the quantity of the real-time feedback data and the accumulated value.

The server may count the number of real-time feedback data in a preset time period, for example, the number of real-time feedback data in a period from 8/28/9/00/5/00 in 2019/8/28/9/20/00 is counted as 10 pieces. And obtaining the average value and the standard deviation of the historical feedback data, wherein the historical feedback data can be the historical feedback data in a preset time period corresponding to a preset time period, the preset time period can be flexibly set according to actual needs, for example, the historical feedback data in 9:00:00 to 9:20:00 in the working days of the past 1 year, and the average value and the standard deviation of each group of historical feedback data are calculated in the process of generating the feedback reference mapping table, so that the average value and the standard deviation of the historical feedback data corresponding to the preset time period obtained by calculation can be obtained. Then, an accumulated value of the average value and the standard deviation of the historical feedback data is calculated, for example, when the average value is 7.42857 and the standard deviation is 0.37978, the accumulated value=7.42857+0.37978= 7.80835; the difference between the amount of real-time feedback data and the accumulated value may be calculated at this time, for example, when the amount of real-time feedback data in the period of time from 8:00 in 2019, 28:00 in 8, 28:9:00 in 2019, and the difference=10-7.80835 = 2.19165.

S209, the server determines real-time feedback data to calculate the feedback fluctuation frequency from the historical feedback fluctuation frequency list according to the difference value.

The server may obtain a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency, and determine, according to the difference, a feedback fluctuation frequency corresponding to the real-time feedback data from the historical feedback fluctuation frequency list, for example, with a feedback fluctuation frequency corresponding to a time node 01:00:00 in the historical feedback fluctuation frequency list: '01:00:00': for example, [2.04762,0.73541,0.6508,0.8095,0.9365,0.9841,1.0] when the real-time feedback data received at 01:00:00 is 10 pieces, the difference between the number of real-time feedback data and the accumulated value between the mean value (2.04762) and the standard deviation (0.73541) is calculated: 10-2.04762-0.73541 = 7.21698, and the sum of the mean, standard deviation and preset threshold is: 2.04762+0.73541+1=3.78303, 2.04762+0.73541+2=4.78303, 2.04762+0.73541+3=5.78303, 2.04762+0.73541+4=6.78303, 2.04762+0.73541+5= 7.78303, where 7.21698 is calculated to be greater than 6.78303 and less than 7.78303, and by querying the historical feedback fluctuation frequency list, the feedback data fluctuation frequency corresponding to the real-time feedback data of 01:00:00 is 0.9841.

S210, the server performs word segmentation processing on the real-time feedback data to obtain at least one word, and extracts a preset keyword from the at least one word.

For example, the server may perform word segmentation processing on the real-time feedback data according to a preset word segmentation policy to obtain at least one word, where the preset word segmentation policy may be flexibly set according to actual needs, for example, the real-time feedback data "web page super card" may be divided into words such as "web page", "super" and "card". Then extracting preset keywords from at least one word, for example, extracting verbs and nouns from at least one word; and calculating the frequency of verbs in the real-time feedback data, calculating the frequency of nouns in the real-time feedback data, screening out the verbs with the highest frequency and the nouns with the highest frequency, and obtaining preset keywords.

S211, the server acquires the ratio of the preset keywords to the total quantity of the real-time feedback data and the total quantity of the real-time feedback data.

The server may obtain a ratio between the number of preset keywords and the total number of the real-time feedback data, and when the preset keywords include a plurality of preset keywords, obtain a ratio between the number of each preset keyword and the total number of the real-time feedback data, respectively. For example, the ratio=the number of preset keywords (e.g., 4)/the total number of real-time feedback data (e.g., 10) =0.4,

S212, the server determines a duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio.

The server can acquire a historical duty ratio interval list generated based on the reference duty ratio interval, and determine the duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio. For example, the total number of real-time feedback data is 10, and the corresponding historical duty cycle interval list is 10: {0.8:0,0.85:0,0.9:0,0.95:0.31556,1:0.41556, since the ratio is 0.4,0.4 is greater than 0.31 and less than 0.41, the history duty cycle interval list can be queried to 0.95:0.31556, namely the duty ratio interval corresponding to the occurrence frequency of the preset keyword is 1-0.95=0.05 (i.e. 5%).

S213, the server compares the feedback fluctuation frequency with the reference feedback fluctuation frequency, compares the duty ratio section with the reference duty ratio section, and determines the positions of the feedback fluctuation frequency and the duty ratio section in the feedback reference mapping table according to the comparison result.

The server may compare the feedback fluctuation frequency of the real-time feedback data with the reference feedback fluctuation frequency, determine the reference feedback fluctuation frequency matched with the feedback fluctuation frequency, compare the duty ratio interval of the preset keyword occurrence frequency in the real-time feedback data with the reference duty ratio interval, determine the reference duty ratio interval matched with the duty ratio interval, and then determine the positions of the reference feedback fluctuation frequency and the reference duty ratio interval obtained by matching in the feedback reference mapping table, so as to obtain the positions (i.e. the intervals) of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table.

S214, the server determines a feedback result corresponding to the real-time feedback data according to the position, and sends the feedback result to a management background corresponding to the browser.

The server can determine a feedback result corresponding to the real-time feedback data according to the position. For example, when the feedback fluctuation frequency of the real-time feedback data exceeds twice the average feedback and the duty ratio interval of the occurrence frequency of the preset keyword is within 15%, the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table is obtained to be the R interval, and at this time, it is possible to determine that the feedback result corresponding to the real-time feedback data is a larger probability according to the R interval.

For example, as shown in fig. 5, in the process of analyzing real-time feedback data of workday WordDay (on the date of 2019, 06, 03), 15:40:00, the analysis results are: there is a problem with a small probability for the noun keyword total_c_n and a problem with a small probability for the verb keyword total_c_v, wherein 7 pieces of feedback data are judged by the feedback number, and there is no problem, wherein n_info represents an analysis result of a duty ratio section of the occurrence frequency of the noun keyword, which is lower than 80% in terms of similar fluctuation, is 0.1492, an account represents the noun keyword, and an analysis result of a duty ratio section of the occurrence frequency of the v_info verb keyword, and a registration represents the verb keyword.

As another example, as shown in fig. 6, analysis results of real-time feedback data of workday WordDay (2019, 06, 03) 16:00:00; for another example, as shown in FIG. 7, the result of analysis of real-time feedback data for the weekday WordDay (day 03 of 2019) is 16:20:00.

The problem feedback existing in the present embodiment may include: (1) browser new tab reprinting grayscale: the user is unaccustomed to use and cannot find the website collected before, so that problem feedback is caused; the keywords corresponding to the feedback data may include labels. (2) Browser not open (e.g., XX space or XX farm not open): the optical fiber is shoveled by a construction party, so that a plurality of products are affected; the keywords corresponding to the feedback data can comprise a browser. (3) new tab page reprint gray scale expansion: the user is not used to and cannot find the website collected before, so that a large amount of feedback is started; the keywords corresponding to the feedback data can comprise the unauthorized problem of the tag (4) flash: the flash Chinese proxy pulls the browser into a blacklist, so that a user can watch scenes such as video and the like through the terminal by using the browser, and a flash unauthorized window is popped up, thereby causing problem feedback; the keywords corresponding to the feedback data can comprise flash, authorization and the like. (5) browser starting to automatically open XX live broadcast problem: aiming at upgrading plug-ins, the XX live broadcast is automatically opened when the terminal is started, the problem of follow-up strategies is solved, release is enlarged, and a large amount of feedback is caused; the keywords corresponding to the feedback data may include live broadcast. Of course, other types of problem feedback may also be included.

In order to timely inform related personnel to process feedback with problems, the server can send feedback results to a management background corresponding to the browser so that the management background can monitor feedback of the user through the terminal, the management background can be the management background corresponding to the browser, the management background can also be a background such as a mailbox or instant messaging, and specific types are not limited here. For example, about 16 points in 3 months and 23 days in 2019 are fed back and increased rapidly due to the reasons that the optical fiber is shoveled off, and the problem can be found in time within 1 hour, and when the feedback number fluctuation (namely the feedback fluctuation frequency) or the keyword fluctuation (namely the duty ratio interval of the occurrence frequency of the keyword) exceeds the historical expectation (namely the expectation of the feedback reference mapping table), an alarm is given and the management background is notified. For another example, the flash chinese proxy pulls the browser into the blacklist, which causes the terminal to watch the video and other scenes through the browser using flash, pops up the flash unauthorized window, causes a lot of feedback, and the fluctuation of the feedback number or the fluctuation of the keyword exceeds the history expectation, alarms and notifies the management background.

The interface and information for monitoring the feedback of the user through the terminal can be flexibly set according to actual needs, for example, as shown in fig. 8, feedback monitoring can be performed on browser use based on network problems, and monitoring can be performed on analysis time, noun analysis, verb analysis, feedback number analysis, noun details, verb details, specific feedback and the like.

For another example, as shown in fig. 9, feedback monitoring of browser usage based on flash authorization issues may include monitoring of analyzed time, noun analysis, verb analysis, feedback number analysis, noun details, verb details, specific feedback, and so on.

According to the embodiment of the application, through the combination of the feedback fluctuation frequency of the real-time feedback data and the duty ratio interval of the keyword occurrence frequency and the feedback reference mapping table generated based on the reference feedback fluctuation frequency of the historical feedback data and the reference duty ratio interval of the keyword occurrence frequency, the server can quickly find out the abnormality and process in time in large-scale data feedback, can extract more data information in small-scale data feedback (less feedback data from several to tens of data per hour), perform multidimensional analysis, quickly find out the abnormality feedback, and perform quick alarm at an hour level, even if the user feedback has data fluctuation, the situation of false alarm cannot occur, the problem level is small, the accuracy and the alarm timeliness can be ensured, and the accuracy and the efficiency of data processing are greatly improved.

In order to facilitate better implementation of the data processing method provided by the embodiment of the application, the embodiment of the application also provides a device based on the data processing method. Where the meaning of a noun is the same as in the data processing method described above, specific implementation details may be referred to in the description of the method embodiments.

Referring to fig. 10, fig. 10 is a schematic structural diagram of a data processing apparatus according to an embodiment of the present application, where the data processing apparatus may include a first obtaining unit 301, a first calculating unit 302, a second obtaining unit 303, a determining unit 304, and the like.

The first acquiring unit 301 is configured to acquire real-time feedback data in a preset time period;

a calculating unit 302, configured to calculate a duty ratio interval between a feedback fluctuation frequency and a preset keyword occurrence frequency according to real-time feedback data;

a second obtaining unit 303, configured to obtain a feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty cycle interval;

the determining unit 304 is configured to determine a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty cycle interval and the feedback reference mapping table.

In some embodiments, the determining unit 304 is specifically configured to: comparing the feedback fluctuation frequency with a reference feedback fluctuation frequency, and comparing the duty cycle section with a reference duty cycle section; determining the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table according to the comparison result; and determining a feedback result corresponding to the real-time feedback data according to the position.

In some embodiments, the first computing unit 301 is specifically configured to: acquiring the quantity of real-time feedback data and the accumulated value of the average value and the standard deviation of the historical feedback data; calculating a difference between the amount of real-time feedback data and the accumulated value; acquiring a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency; and determining the feedback fluctuation frequency corresponding to the real-time feedback data from the historical feedback fluctuation frequency list according to the difference value.

In some embodiments, the first computing unit 301 is specifically configured to: performing word segmentation processing on the real-time feedback data to obtain at least one word; extracting preset keywords from at least one word; acquiring the ratio of the preset keywords to the total quantity of the real-time feedback data; acquiring a historical duty cycle interval list generated based on a reference duty cycle interval; and determining a duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio.

In some embodiments, as shown in fig. 11, the data processing apparatus may further include a filtering unit 305, a dividing unit 306, a grouping unit 307, a second calculating unit 308, a third calculating unit 309, and a generating unit 310, which may specifically be as follows:

The screening unit 305 is configured to obtain historical feedback data in a preset time period, and screen normal feedback data from the historical feedback data to obtain screened historical feedback data;

the dividing unit 306 is configured to divide the screened historical feedback data according to a preset date, so as to obtain historical feedback data corresponding to a plurality of date intervals;

a grouping unit 307, configured to group the historical feedback data in each date interval according to a preset periodic policy, so as to obtain multiple groups of historical feedback data;

a second calculation unit 308, configured to calculate a reference feedback fluctuation frequency corresponding to each set of historical feedback data;

a third calculation unit 309, configured to calculate a reference duty interval of the occurrence frequency of the preset keyword corresponding to each set of historical feedback data;

and a generating unit 310, configured to generate a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty cycle interval.

In some embodiments, the second computing unit 308 is specifically configured to: acquiring the number of each group of historical feedback data, and the average value and standard deviation corresponding to each group of historical feedback data; calculating the accumulated sum of the average value, the standard deviation and the preset threshold value of each group of historical feedback data; and calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data according to the sum of the quantity and the accumulation sum.

In some embodiments, the third computing unit 309 may include an extracting module, an obtaining module, a computing module, and the like, and may specifically be as follows:

the acquisition module is used for acquiring the number of preset keywords in each group of historical feedback data in the current group of historical feedback data and the total number of the current group of historical feedback data;

In some embodiments, the extraction module is specifically configured to: extracting verbs and nouns from at least one word; calculating the frequency of occurrence of verbs in each set of historical feedback data, and calculating the frequency of occurrence of nouns in each set of historical feedback data; and screening the verb with the highest frequency and the noun with the highest frequency to obtain the preset keywords.

In this embodiment of the present application, the first acquiring unit 301 may acquire real-time feedback data within a preset time period, the first calculating unit 302 calculates, according to the real-time feedback data, a duty cycle interval between the feedback fluctuation frequency and the occurrence frequency of the preset keyword, and then the second acquiring unit 303 acquires, according to the feedback fluctuation frequency and the reference duty cycle interval, a feedback reference mapping table generated based on the reference feedback fluctuation frequency and the reference duty cycle interval, and at this time, the determining unit 304 may determine a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty cycle interval and the feedback reference mapping table. According to the scheme, the feedback result is obtained through the feedback fluctuation frequency and the duty ratio interval of the preset keyword occurrence frequency and the comprehensive analysis by combining the feedback reference mapping table, the accuracy of the feedback result determination is improved, and the reliability of the feedback result is greatly improved compared with the single analysis performed at the feedback quantity increasing speed.

The embodiment of the application also provides a server, as shown in fig. 12, which shows a schematic structural diagram of the server according to the embodiment of the application, specifically:

the server may include one or more processors 401 of a processing core, memory 402 of one or more computer readable storage media, a power supply 403, and an input unit 404, among other components. Those skilled in the art will appreciate that the server architecture shown in fig. 12 is not limiting of the server and may include more or fewer components than shown, or may combine certain components, or a different arrangement of components.

Wherein:

the processor 401 is a control center of the server, connects respective portions of the entire server using various interfaces and lines, and performs various functions of the server and processes data by running or executing software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby performing overall monitoring of the server. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor and a modem processor, wherein the application processor mainly processes an operating system, a user interface, an application program, etc., and the modem processor mainly processes wireless communication. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by executing the software programs and modules stored in the memory 402. The memory 402 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like; the storage data area may store data created according to the use of the server, etc. In addition, memory 402 may include high-speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 with access to the memory 402.

The server also includes a power supply 403 for powering the various components, and preferably, the power supply 403 may be logically connected to the processor 401 by a power management system so as to implement functions such as charge, discharge, and power consumption management by the power management system. The power supply 403 may also include one or more of any of a direct current or alternating current power supply, a recharging system, a power failure detection circuit, a power converter or inverter, a power status indicator, and the like.

The server may also include an input unit 404, which input unit 404 may be used to receive entered numeric or character information and to generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

Although not shown, the server may further include a display unit or the like, which is not described herein. In particular, in this embodiment, the processor 401 in the server loads executable files corresponding to the processes of one or more application programs into the memory 402 according to the following instructions, and the processor 401 executes the application programs stored in the memory 402, so as to perform the following functions, as follows:

acquiring real-time feedback data in a preset time period; calculating a duty ratio interval of a feedback fluctuation frequency and a preset keyword occurrence frequency according to the real-time feedback data; acquiring a feedback reference mapping table generated based on a reference feedback fluctuation frequency and a reference duty ratio interval, wherein the feedback reference mapping table comprises a plurality of reference feedback fluctuation frequency ranges and problem feedback probability intervals obtained by dividing the plurality of reference duty ratio intervals; and determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table.

In some embodiments, when determining the feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty cycle interval, and the feedback reference map, the processor 401 is further configured to perform: comparing the feedback fluctuation frequency with a reference feedback fluctuation frequency, and comparing the duty cycle section with a reference duty cycle section; determining the position of the feedback fluctuation frequency and the duty ratio interval in the feedback reference mapping table according to the comparison result; and determining a feedback result corresponding to the real-time feedback data according to the position.

In some embodiments, in calculating the feedback fluctuation frequency from the real-time feedback data, the processor 401 is further configured to perform: acquiring the quantity of real-time feedback data and the accumulated value of the average value and the standard deviation of the historical feedback data; calculating a difference between the amount of real-time feedback data and the accumulated value; acquiring a historical feedback fluctuation frequency list generated based on the reference feedback fluctuation frequency; and determining the feedback fluctuation frequency corresponding to the real-time feedback data from the historical feedback fluctuation frequency list according to the difference value.

In some embodiments, in calculating the duty cycle interval of the occurrence frequency of the preset keyword according to the real-time feedback data, the processor 401 is further configured to perform: performing word segmentation processing on the real-time feedback data to obtain at least one word; extracting preset keywords from at least one word; acquiring the ratio of the preset keywords to the total quantity of the real-time feedback data; acquiring a historical duty cycle interval list generated based on a reference duty cycle interval; and determining a duty ratio interval corresponding to the occurrence frequency of the preset keyword from the historical duty ratio interval list according to the ratio.

In some embodiments, prior to obtaining the feedback reference map generated based on the baseline feedback fluctuation frequency and the baseline duty cycle interval, the processor 401 is further configured to perform: acquiring historical feedback data in a preset time period, and screening normal feedback data from the historical feedback data to obtain screened historical feedback data; dividing the screened historical feedback data according to preset dates to obtain historical feedback data corresponding to a plurality of date intervals; grouping the historical feedback data in each date interval according to a preset periodic strategy to obtain a plurality of groups of historical feedback data; calculating a reference feedback fluctuation frequency corresponding to each group of historical feedback data; calculating a reference duty ratio interval of the occurrence frequency of the preset key words corresponding to each group of historical feedback data; and generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval.

In some embodiments, in calculating the reference feedback fluctuation frequency for each set of historical feedback data, the processor 401 is further configured to perform: acquiring the number of each group of historical feedback data, and the average value and standard deviation corresponding to each group of historical feedback data; calculating the accumulated sum of the average value, the standard deviation and the preset threshold value of each group of historical feedback data; and calculating the reference feedback fluctuation frequency corresponding to each group of historical feedback data according to the sum of the quantity and the accumulation sum.

In some embodiments, during the calculation of the reference duty cycle interval of the occurrence frequency of the preset keyword corresponding to each set of historical feedback data, the processor 401 is further configured to: performing word segmentation on each group of historical feedback data to obtain at least one word, and extracting preset keywords from the at least one word; acquiring the number of preset keywords in each group of historical feedback data in the current group of historical feedback data and the total number of the current group of historical feedback data; and calculating a reference duty ratio interval of the occurrence frequency of the preset keywords corresponding to each group of historical feedback data according to the ratio and the total number.

In some embodiments, in extracting the preset keyword from the at least one word, the processor 401 is further configured to perform: extracting verbs and nouns from at least one word; calculating the frequency of occurrence of verbs in each set of historical feedback data, and calculating the frequency of occurrence of nouns in each set of historical feedback data; and screening the verb with the highest frequency and the noun with the highest frequency to obtain the preset keywords.

In the foregoing embodiments, the descriptions of the embodiments are focused on, and the portions of a certain embodiment that are not described in detail may be referred to the above detailed description of the data processing method, which is not repeated herein.

It will be appreciated by those of ordinary skill in the art that all or part of the steps of the various methods of the above embodiments may be performed by a computer program, or by computer program control related hardware, which may be stored in a computer readable storage medium (i.e., a storage medium) and loaded and executed by a processor.

To this end, an embodiment of the present application provides a storage medium in which a computer program is stored, the computer program being capable of being loaded by a processor to perform a data processing method provided by the embodiment of the present application. For example, the computer program may perform the steps of:

The specific implementation of each operation above may be referred to the previous embodiments, and will not be described herein.

Wherein the storage medium may include: read-only memory (ROM, readOnlyMemory), random access memory (RAM, randomAccessMemory), magnetic or optical disk, and the like.

The steps of any data processing method provided by the embodiment of the present application can be executed by the computer program stored in the storage medium, so that the beneficial effects of any data processing method provided by the embodiment of the present application can be achieved, and detailed descriptions of the foregoing embodiments are omitted.

The foregoing has described in detail a data processing method, apparatus and storage medium according to embodiments of the present application, and specific examples have been applied to illustrate the principles and embodiments of the present application, where the foregoing examples are only for aiding in the understanding of the method and core idea of the present application; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in light of the ideas of the present application, the present description should not be construed as limiting the present application.

Claims

1. A method of data processing, comprising:

Acquiring real-time feedback data in a preset time period;

generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval;

acquiring a feedback reference mapping table generated based on a reference feedback fluctuation frequency and a reference duty ratio interval;

2. The data processing method of claim 1, wherein calculating a feedback fluctuation frequency from the real-time feedback data comprises:

3. The data processing method according to claim 1, wherein calculating a duty cycle interval of occurrence frequency of a preset keyword according to the real-time feedback data comprises:

extracting preset keywords from the at least one word;

4. The data processing method according to claim 1, wherein calculating the reference feedback fluctuation frequency corresponding to each set of historical feedback data includes:

5. The data processing method according to claim 1, wherein the calculating the reference duty interval of the occurrence frequency of the preset keyword corresponding to each set of the historical feedback data includes:

6. The data processing method according to claim 5, wherein the extracting the preset keyword from the at least one word includes:

extracting verbs and nouns from the at least one word;

7. A data processing apparatus, comprising:

the first calculation unit is used for calculating a duty ratio interval of the feedback fluctuation frequency and the occurrence frequency of the preset keywords according to the real-time feedback data;

the generation unit is used for generating a feedback reference mapping table according to the reference feedback fluctuation frequency and the reference duty ratio interval;

the determining unit is used for determining a feedback result corresponding to the real-time feedback data according to the feedback fluctuation frequency, the duty ratio interval and the feedback reference mapping table;

wherein, the determining unit is specifically configured to:

8. A storage medium storing a computer program adapted to be loaded by a processor to perform the data processing method of any one of claims 1 to 6.

9. A server comprising a memory storing an application program and a processor for running the application program in the memory to perform the data processing method of any one of claims 1 to 6.