CN103455938A - Data-processing method and device and server equipment - Google Patents

Abstract

The invention discloses a data-processing method and device and server equipment. The method comprises the steps that numerical values corresponding to feedback data of each version of a product and the number of feedback times of the feedback data are recorded; a weight value corresponding to each version of the product is obtained; according to the numerical values corresponding to the feedback data, the number of feedback times of the feedback data, and the weight value of each version, final feedback data of the product are calculated.

Description

A kind of data processing method, device and server apparatus

Technical field

The present invention relates to the computer science and technology field, relate in particular to a kind of data processing method, device and server apparatus.

Background technology

Along with the fast development of infotech and internet, derive website, forum and the mhkc etc. of various user-generated contents (UGC), and, in these UGC, what user's participation was the highest, threshold is minimum is exactly to comment on points-scoring system.The comment points-scoring system that almost each forum, mhkc, portal website, video website exist the user to participate in.As long as for the user, comment scoring is almost without any difficult point, generally registration of website, provides the true or false essential information of some individuals just can be marked.Exactly because comment scoring is simple, there is no threshold, but user's attention rate, reason that participation is high also allows the information such as some advertisements, pornographic be full of easily therein, affects user's experience.

In addition, the user also marks by other people comment in its comment points-scoring system, determine whether buying or using a certain commodity, and, in these determine whether users download, buy, browse the factor of some commodity, the scoring of these commodity is important factors.Scoring likely directly affects rank and the sales volume of these commodity on whole market.

Current score calculation method is only the most original weight computation method.What the score calculation based on weighting adopted is the average of different specific weight data, is about to raw data and calculates according to rational ratio, if in the n number, x1 occurs f1 time, and x2 occurs f2 time ..., xk occurs fk time, and (x1f1+x2f2+...xkfk)/(f1+f2+...+fk) is called x1, x2 so, ..., the weighted mean of xk, wherein f1, f2 ..., fk can regard x1 as, x2 ..., the weight that xk is corresponding.The shared ratio of raw data is fixed.

But consider same commodity user this kind of factor of quantity of marking because weighting is just simple, the not impact on it with reference to the renewal iteration of commodity fully, the user is difficult to therefrom obtain more real score information.

Summary of the invention

The embodiment of the present invention provides a kind of data processing method, device and server apparatus, for realizing, field feedback is more accurately analyzed.

A kind of data processing method, the method comprises:

The numerical value that record is corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Obtain weighted value corresponding to each version of product;

The final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to described feedback data.

In this programme, consider the iteration of some products, distribute different weighted values by the different spaces of a whole page to product, like this, when calculating the final feedback data of this product, can embody the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Preferably, according to described feedback data, the final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version comprises: according to the final feedback data of following formula counting yield,

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{N}_{\mathrm{ij}}{W}_j},$

Wherein, the final feedback data that S is this product; C _ijfor i the numerical value that feedback data is corresponding to j version of described product; N _ijtimes of Feedback for i the feedback data to j version of described product; W _jweighted value for j version of described product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

In the present embodiment, by the new and old order according to product version, for each version value of assigning weight, like this, when calculating the final feedback data of described product, can embody the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Preferably, obtaining weighted value corresponding to each version of product comprises: by following formula, calculate the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\Π}}}(1-t_{j-1}),$

Wherein, W _jfor the weighted value of j version of described product, j=1,2,3...m; M is described product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

Preferably, described difference in version value is calculated and is obtained by following formula:

Work as z _j<y _jthe time, described difference in version value

Work as z _j>=y _jthe time, described difference in version value

Described z _jthe code update increment obtained for the difference according between j version of described product and j+1 version; y _jsize of code for j version of described product.

In this programme, by weighted value corresponding to each version of code update incremental computations between the size of code according to each version of product and adjacent version, make the weighted value of each version embody the difference between adjacent version, can embody more exactly the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Preferably, the weighted value that each version of described product is corresponding is arithmetic progression or Geometric Sequence, or weighted value corresponding to each version of described product obtains according to the table lookup set in advance.

In this programme, by the new and old order according to product version, for each version distributes the weighted value successively decreased, like this, when calculating the final feedback data of described product, can embody the impact of the feedback data of new and old edition product for final feedback data, the feedback data of latest edition product has the greatest impact to final feedback data, and more the feedback data of legacy version product is less on final feedback data impact, make the analysis of user feedback data more accurate.

Preferably, obtaining weighted value corresponding to each version of product comprises:

Detect whether record of product version corresponding to described feedback data;

When product version corresponding to described feedback data recorded, obtain weighted value corresponding to described product version recorded;

When product version corresponding to described feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

In the present embodiment, by judging that whether feedback data is for the redaction product, when feedback data is for the redaction product, be that product upgrades, need to adjust each weighted value of product, make weighted value maximum corresponding to latest edition, the weighted value that more legacy version is corresponding is less, makes the analysis of user feedback data more accurate.When product does not upgrade, obtain existing weighted value corresponding to version, user feedback data is carried out to analytical calculation.

Preferably, described method also comprises:

The user ID of described feedback data submitted in record;

Detect described user according to described user ID and whether the same version of identical product had been submitted to feedback data;

When the feedback data of having submitted to, described user is deleted the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

In this programme, can avoid same user repeatedly described product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of described product, makes the analysis of user feedback data more accurate.

Preferably, described method also comprises:

The user ID of described feedback data submitted in record;

The number of times of the feedback data that detects the number of times of the feedback data that described user ID submits to described product or one of them version of described product is submitted to;

When the number of times of the number of times of the feedback data that described product is submitted to or feedback data that one of them version of described product is submitted to surpasses default first threshold, do not record the feedback data that described user ID is submitted to.

In this programme, can avoid same user repeatedly described product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of described product, makes the analysis of user feedback data more accurate.

Preferably, described method also comprises:

The user ID of described feedback data submitted in record;

Detect described user ID and submit the number of times of feedback data in the Preset Time section;

While submitting to the number of times of feedback data to surpass default Second Threshold when described user ID, do not record the feedback data that described user ID is submitted in the Preset Time section.

In this programme, can avoid same user continually described product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of described product, makes the analysis of user feedback data more accurate.

Preferably, described method also comprises:

Described feedback data is examined, judge in described feedback data and whether comprise information unauthorized;

While in described feedback data, comprising information unauthorized, delete described feedback data.

In this programme, can be filtered the feedback data that contains junk information or harmful content, be avoided the impact of user's malice feedback subsequent analysis, be improved the degree of accuracy to the analysis of user feedback data.

A kind of data processing equipment, described device comprises:

Logging modle, for recording the numerical value corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Acquisition module, for obtaining weighted value corresponding to each version of product;

Final feedback data computing module, for the final feedback data of the weighted value counting yield of the Times of Feedback of the numerical value corresponding according to described feedback data, feedback data and each version.

Preferably, described final feedback data computing module, for calculate the final feedback data of counting yield according to following formula,

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j},$

Wherein, the final feedback data that S is described product; C _ijfor i the numerical value that feedback data is corresponding to j version of described product; N _ijtimes of Feedback for i the feedback data to j version of described product; W _jweighted value for j version of described product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

Preferably, described acquisition module comprises: the weight calculation submodule, for by following formula, calculating the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1}),$

Wherein, W _jfor the weighted value of j version of described product, j=1,2,3...m; M is described product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

Preferably, described weight calculation submodule, for calculate described difference in version value by following formula:

Work as z _j<y _jthe time, described difference in version value

Work as z _j>=y _jthe time, described difference in version value

Described z _jthe code update increment obtained for the difference according between j version of described product and j+1 version; y _jsize of code for j version of described product.

Preferably, described acquisition module also comprises:

Detection sub-module, the corresponding product version record whether for detection of described feedback data;

The Weight Acquisition submodule, for when product version corresponding to described feedback data recorded, obtain weighted value corresponding to described product version recorded;

Described weight calculation submodule, when product version corresponding to described feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

Preferably, the weighted value that each version of described product is corresponding is arithmetic progression or Geometric Sequence, or weighted value corresponding to each version of described product obtains according to the table lookup set in advance.

Preferably, described device also comprises: first detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Whether described first detection module, submitted feedback data to the same version of identical product for detect described user according to described user ID;

Described logging modle, for the feedback data that ought submit to, delete described user to the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

Preferably, described device also comprises: the second detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Described the second detection module, the number of times of the number of times of the feedback data of described product being submitted to for detection of described user ID or feedback data that one of them version of described product is submitted to;

Described logging modle, while for the number of times of the number of times of the feedback data when described product is submitted to or feedback data that one of them version of described product is submitted to, surpassing default first threshold, do not record the feedback data that described user ID is submitted to.

Preferably, described device also comprises: the 3rd detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Described the 3rd detection module is submitted the number of times of feedback data in the Preset Time section for detection of described user ID;

Described logging modle, while for the number of times of submitting feedback data when described user ID in the Preset Time section to, surpassing default Second Threshold, do not record the feedback data that described user ID is submitted to.

Preferably, described device also comprises:

Auditing module, for described feedback data is examined, judge in described feedback data and whether comprise information unauthorized;

Described logging modle, for when described feedback data comprises information unauthorized, delete described feedback data.

A kind of server apparatus, described server apparatus includes storer, and one or more than one program, one of them or an above program are stored in storer, and are configured to carry out described one or above routine package containing for carrying out the instruction of following operation by one or above processor:

The numerical value that record is corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Obtain weighted value corresponding to each version of product;

The final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to described feedback data.

Other features and advantages of the present invention will be set forth in the following description, and, partly from instructions, become apparent, or understand by implementing the present invention.Purpose of the present invention and other advantages can realize and obtain by specifically noted structure in the instructions write, claims and accompanying drawing.

Below by drawings and Examples, technical scheme of the present invention is described in further detail.

The accompanying drawing explanation

Accompanying drawing is used to provide a further understanding of the present invention, and forms the part of instructions, for explaining the present invention, is not construed as limiting the invention together with embodiments of the present invention.In the accompanying drawings:

The schematic flow sheet that Fig. 1 is data processing method in the embodiment of the present invention;

Another schematic flow sheet that Fig. 2 is data processing method in the embodiment of the present invention;

The flow process schematic diagram again that Fig. 3 is data processing method in the embodiment of the present invention;

The structural representation that Fig. 4 is data processing equipment in the embodiment of the present invention;

The structural representation that Fig. 5 is acquisition module in the embodiment of the present invention;

The structural representation that Fig. 6 is server apparatus in the embodiment of the present invention.

Embodiment

Below in conjunction with accompanying drawing, the preferred embodiments of the present invention are described, should be appreciated that preferred embodiment described herein, only for description and interpretation the present invention, is not intended to limit the present invention.

As described in Figure 1, the embodiment of the present invention provides a kind of data processing method, comprises the following steps:

Step 102, the numerical value that the feedback data of each version of record product is corresponding and the Times of Feedback of feedback data;

Step 104, obtain weighted value corresponding to each version of product;

Step 106, the final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to feedback data.

In the present embodiment, consider the iteration of some products, distribute different weighted values by the different spaces of a whole page to product, like this, when the final feedback data of counting yield, can embody the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Preferably, in step 106, according to feedback data, the final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version comprises:

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j}$ Formula (1)

Wherein, the final feedback data that S is product; C _ijfor i the numerical value that feedback data is corresponding to j version of product; N _ijtimes of Feedback for i the feedback data to j version of product; W _jweighted value for j version of product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

Preferably, the weighted value that each version of product is corresponding is arithmetic progression or Geometric Sequence, or weighted value corresponding to each version of product obtains according to the table lookup set in advance.

For example: the weighted value between version is arithmetic progression, as 1,0.8, and 0.6,0.4.Weighted value between version is Geometric Sequence, as 1,0.8, and 0.64,0.512.Perhaps, the weighted value between version is according to the acquisition of tabling look-up of the form that pre-sets.Can also provide according to the renewal of certain emphasis version special weight, as poor two-stage between this version and adjacent version.For example, certain product has just been done once great renewal, and before latest edition and its, the weighted value of version is followed successively by: 1,0.6,0.4.

In the present embodiment, by the new and old order according to product version, for each version distributes the weighted value successively decreased, like this, when the final feedback data of counting yield, can embody the impact of the feedback data of new and old edition product for final feedback data, the feedback data of latest edition product has the greatest impact to final feedback data, and more the feedback data of legacy version product is less on final feedback data impact, make the analysis of user feedback data more accurate.

Preferably, in order more reasonably to show the impact of edition upgrading for each version weight, in step 104, according to the size of code of each version of product and weighted value corresponding to each version of code update incremental computations between adjacent version:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1})$ Formula (2)

Wherein, W _jfor the weighted value of j version of product, j=1,2,3...m; M is the product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.In described formula, according to product version, by the new weighted value to old correspondence, be respectively W ₁, W ₂, W ₃... W _m.It should be pointed out that the corresponding relation between the number order of weighted value and product version new and old, is only a kind of artificial setting, does not affect protection scope of the present invention and limits.

The difference in version value is calculated and is obtained by following formula:

Work as z _j<y _jthe time, the difference in version value

$t_j=\frac{z_j}{y_j};$ Formula (3)

Work as z _j>=y _jthe time, the difference in version value

$t_j=\frac{z_j}{z_j+y_j};$ Formula (4)

Z _jthe code update increment obtained for the difference according between j version of product and j+1 version; y _jsize of code for j version of product.

In reality, can adopt scale-of-two difference (binary diff, bsdiff) difference (diff) file that program pin generates the difference between new and old two versions of product, calculate according to the size of code of diff file and the size of code of each version Android installation kit (AndroidPackage, apk) file the weighted value that each version is corresponding.For example: certain product one has 5 versions, and product is by newly to old edition this shop, being followed successively by 1,2,3,4,5.According to above-mentioned formula (2), the weighted value that calculates each version is followed successively by:

W ₁=1;

$W_2=1-t_1=1-\frac{z_1}{y_1},z_1<y_1;$

$W_3=(1-t_1)(1-t_2)=(1-\frac{z_1}{y_1})(1-\frac{z_2}{y_2}),z_2<y_2;$

$W_4=(1-t_1)(1-t_2)(1-t_3)=-(1-\frac{z_1}{y_1})(1-\frac{z_2}{y_2})(1-\frac{z_3}{y_3}),z_3<y_3;$

$W_5=(1-t_1)(1-t_2)(1-t_3)(1-t_4)=(1-\frac{z_1}{y_1})(1-\frac{z_2}{y_2})(1-\frac{z_3}{y_3})(1-\frac{z_4}{y_4}),z_4<y_4.$

For example " search dog input method " is upgraded to 1.0 from version 2 .0, the apk file 10M of version 2 .0, the apk file 12M of version 1.0, the diff file 5M generated according to algorithm between these 2 versions, after edition upgrading, the weighted value of version 1.0 correspondences is 1, and the weighted value that version 2 .0 is corresponding is 1-5/12=0.58.

In the present embodiment, by weighted value corresponding to each version of code update incremental computations between the size of code according to each version of product and adjacent version, make the weighted value of each version embody the difference between adjacent version, can embody more exactly the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Preferably, in the present embodiment, step 104 comprises:

Detect whether record of product version corresponding to feedback data;

When product version corresponding to feedback data recorded, obtain weighted value corresponding to product version recorded;

When product version corresponding to feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

In the present embodiment, by judging that whether feedback data is for the redaction product, when feedback data is for the redaction product, be that product upgrades, need to adjust each weighted value of product, make weighted value maximum corresponding to latest edition, the weighted value that more legacy version is corresponding is less, makes the analysis of user feedback data more accurate.When product does not upgrade, obtain existing weighted value corresponding to version, user feedback data is carried out to analytical calculation.

Preferably, the method also comprises: also record the user ID of submitting feedback data in step 102; Before step 104, detect the user according to user ID and whether the same version of identical product had been submitted to feedback data; When the feedback data of having submitted to, the user is deleted the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.Like this, can avoid same user repeatedly product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of product, makes the analysis of user feedback data more accurate.

Preferably, the method also comprises: also record the user ID of submitting feedback data in step 102; Before step 104, detect the number of times of the feedback data that user ID submits to product or the number of times of feedback data that one of them version of product is submitted to; When the number of times of the number of times of the feedback data that product is submitted to or feedback data that one of them version of product is submitted to surpasses default first threshold, the feedback data that recording user ID does not submit to.Like this, can avoid same user repeatedly product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of product, makes the analysis of user feedback data more accurate.

Preferably, the method also comprises: also record the user ID of submitting feedback data in step 102; Before step 104, detect user ID and submit the number of times of feedback data in the Preset Time section; While submitting to the number of times of feedback data to surpass default Second Threshold in the Preset Time section when user ID, the feedback data that recording user ID does not submit to.Like this, can avoid same user continually product to be carried out to malice feedback, the user that also can upgrade in time, to the up-to-date feedback data of product, makes the analysis of user feedback data more accurate.

Preferably, the method also comprises: before step 104, feedback data is examined, judge in feedback data and whether comprise information unauthorized; While in feedback data, comprising information unauthorized, delete feedback data.Like this, can be filtered the feedback data that contains junk information or harmful content, be avoided the impact of user's malice feedback subsequent analysis, be improved the degree of accuracy to the analysis of user feedback data.

The user of below take is elaborated to the different editions software method embodiment of the present invention provided as example of being marked.

As shown in Figure 2, when having the user to submit new scoring to, the method that the embodiment of the present invention provides comprises the following steps:

Step 202, the score information that this user is submitted to is recorded in database; Wherein in score information, at least comprise: fractional value and for software version information;

Step 204, the scoring number of times to this fractional value while in database, recording score information is upgraded;

Step 206, detect this new submission scoring institute for version information whether be to having the scoring of software version in database, if so, perform step 208, if not, performing step 210;

Step 208, directly utilize the final scoring of the data software for calculation recorded in database;

Step 210, after obtaining the weighted value of software version, the final scoring of software for calculation.

In the present embodiment, more reasonable to the calculating of software scoring.The developer of software can see the hobby of user for the software of newly reaching the standard grade or submitting to faster.When the user comments when more software is poor, can be timely from the rapid reacting condition of scoring out, be convenient to the developer and submit in time redaction to.Calculating to the software scoring is more accurate, and the recommendation of the information such as the seniority among brothers and sisters that scoring affects for software, elaboration is more reasonable, and in seniority among brothers and sisters, in a series of personalized recommendations such as elaboration, the poor software of scoring is not recommended.Can contrast same developer's software, recommend the higher software of this developer's scoring of user to the user.

As shown in Figure 3, the method that the embodiment of the present invention provides comprises the following steps:

Step 302, the score information that this user is submitted to is recorded in database, wherein in score information, at least comprises: fractional value, for software version information and user ID;

Step 304, the scoring number of times to this fractional value when the data-base recording score information is upgraded;

Step 306, detect this user according to user ID and whether the same version of this software submitted to feedback data, if not, performs step 308, if so, performs step 310;

Step 308, directly utilize the final scoring of the data software for calculation recorded in database;

Step 310 is deleted the original scoring record of this user in database, adds new scoring record;

Step 312, according to the final scoring of the database software for calculation after upgrading.

Like this, can avoid same user repeatedly this software to be carried out to the malice scoring, the up-to-date evaluation of user to this software also can upgrade in time.

In order to embody the fancy grade of user to software, adopt the at present popular star mode of commenting, the 1-5 star has represented respectively the fancy grade of user to software, and 1 star is least liked, and 5 stars like best.

Same version scoring to same software, adopt the final scoring of weighting scheme software for calculation.When the user marks for the first time, corresponding star value adds 1, and the scoring sum adds 1.For the second time and later scoring, corresponding star value scoring number of times adds 1 to the user, and star value scoring number of times originally subtracts 1.

For example, software version 1 scoring 1,2,3,4 at present that appId is 10, the number of 5 stars is respectively 1,2,3,2,1, comments the 1*1+2*2+3*3+2*4+1*5=27 that adds up to of star, and the scoring total number of persons is 1+2+3+2+1=9 people, and the software scoring is 27/9=3.When some users comment 3 star for the first time, comment the number of 3 stars to add 1, comment the sum of star to add 3, the software scoring is the 30/10=3 star.It is 5 minutes that this user marks for the second time, and the number that 3 original star numbers subtract 1,5 star adds 1.Now, comment 1,2,3,4, the number of 5 stars is respectively 1,2,3,2,2, comments star sum 30-3+5=32, and the software scoring is 32/10=3.2.

But after software release upgrade, comment accordingly the star number also to change, the software that appId is 10 is that example (marks 1 thereupon, 2,3,4, the number of 5 stars is respectively 1,2, and 3,2,1), when version 2 is arrived in application upgrade, comment accordingly the star number all to be multiplied by the value between a coefficient 0.8(or other 0-1), comment 1,2,3, the number of 4,5 stars becomes 0.8,1.6,2.4,1.6,0.8, the total number of persons of scoring becomes 0.8+1.6+2.4+1.6+0.8=7.2, commenting the star sum is 27*0.8=21.6, and the software scoring is 21.6/7.2=3.Visible, after edition upgrading, if when also nobody marks, the software scoring remains unchanged.

When a user (no matter the user who never marked before being or the user who marked) is arranged, to the scoring of software, be 1 timesharing, this is to comment 1,2,3, the number of 4,5 stars is 1.8,1.6,2.4,1.6,0.8, the scoring total number of persons becomes 8.2, the scoring sum becomes 22.6, and the software scoring is 22.6/8.2=2.756.A visible user comments (1 star) directly to cause software scoring to drop to 2.75 from 3 for the difference of redaction, and has also retained the impact of the scoring of early version for the software scoring simultaneously.

In order to improve the favorable comment number, by program or artificial mode, issue some advertisements as some users maliciously, during yellow information, the calculating of software scoring also can be greatly affected, and can have following 2 kinds of modes to solve:

1) first by audit, rear deletion comment scoring.

The software that the appId of front of take is 10 (mark 1,2,3,4, the number of 5 stars is respectively 1,2,3,2,1) is example, and some users have delivered a yellow advertising commentary, and scoring is 5 minutes, and this news commentary star number becomes 1,2,3,2,2, and the software scoring is 3.2.But this scoring should not be accumulated in the software score calculation, when deleting, whether first detecting this comment scoring, to record corresponding version be the latest edition of app, if, show that version is not also upgraded, at this moment can be simply by for star number subtract 1, overall score subtracts 1, the software scoring is (1*1+2*2+3*3+4*2+5*(2-1))/(1+2+3+2+2-1)=3.If not the latest edition of app, need to detect the app how many times of having upgraded, suppose to have upgraded 1 time, the software scoring should be just (1*1+2*2+3*3+4*2+5*(2-1*0.8))/(1+2+3+2+2-1*0.8)=3.04.

2) directly by program, be judged as the rubbish comment, the calculating not impact of its scoring on average mark

By some machine learning and the comment of the rubbish based on regularity detection mode, can find timely waste advertisements, its state is set to only own visible, and the comment scoring that he sends like this can be seen at his mobile phone terminal.

For example, the recognition method of rubbish comment is as follows: A) can limit the number of times that same user recommends; B) for the app found out, search its scoring in Commentary Systems, scoring is less than 2, thinks that it is the application of a rubbish, shows in various lists; C) each account (such as a day) within a period of time can not surpass the comment of 10 repetitions, and each No. ime within a period of time, (such as one day) can not the identical comment over 20.

Based on same inventive concept, the embodiment of the present invention also provides a kind of data processing equipment, and as shown in Figure 4, this device comprises:

Logging modle 41, for recording the numerical value corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Acquisition module 42, for obtaining weighted value corresponding to each version of product;

Final feedback data computing module 43, for the final feedback data of the weighted value counting yield of the Times of Feedback of the numerical value corresponding according to feedback data, feedback data and each version.

Preferably, final feedback data computing module 43 calculates the final feedback data of counting yield according to following formula,

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j}$ Formula (1)

Wherein, the final feedback data that S is this product; C _ijfor i the numerical value that feedback data is corresponding to j version of this product; N _ijtimes of Feedback for i the feedback data to j version of this product; W _jweighted value for j version of this product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

Preferably, as shown in Figure 5, acquisition module 42 comprises: weight calculation submodule 421, for by following formula, calculating the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1})$ Formula (2)

Wherein, W _jfor the weighted value of j version of product, j=1,2,3...m; M is the product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

Preferably, the weight calculation submodule, for by following formula calculated version difference value:

Work as z _j<y _jthe time, the difference in version value

$t_j=\frac{z_j}{y_j};$ Formula (3)

Work as z _j>=y _jthe time, the difference in version value

$t_j=\frac{z_j}{z_j+y_j};$ Formula (4)

Z _jthe code update increment obtained for the difference according between j version of product and j+1 version; y _jsize of code for j version of product.

Preferably, as shown in Figure 5, acquisition module 42 also comprises:

Detection sub-module 422, the corresponding product version record whether for detection of feedback data;

Weight Acquisition submodule 423, for when product version corresponding to feedback data recorded, obtain weighted value corresponding to product version recorded;

Weight calculation submodule 421, when product version corresponding to feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

As shown in Figure 4, preferably, this device also comprises: first detection module 44,

Logging modle 41, for recording the user ID of submitting feedback data to;

Whether first detection module 44, submitted feedback data to the same version of identical product for detect this user according to user ID;

Logging modle 41, for the feedback data that ought submit to, delete this user to the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

Preferably, this device also comprises: the second detection module 45,

Logging modle 41, for recording the user ID of submitting feedback data to;

The second detection module 45, the number of times of the number of times of the feedback data of product being submitted to for detection of user ID or feedback data that one of them version of product is submitted to;

Logging modle 41, while for the number of times of the number of times of the feedback data when product is submitted to or feedback data that one of them version of product is submitted to, surpassing default first threshold, the feedback data that recording user ID does not submit to.。

Preferably, this device also comprises: the 3rd detection module 46,

Logging modle 41, for recording the user ID of submitting feedback data to;

The 3rd detection module 46 is submitted the number of times of feedback data in the Preset Time section for detection of user ID;

Logging modle 41, while for the number of times of submitting feedback data when user ID in the Preset Time section to, surpassing default Second Threshold, the feedback data that recording user ID does not submit to.

Preferably, this device also comprises:

Auditing module 47, for feedback data is examined, judge in feedback data and whether comprise information unauthorized;

Logging modle 41, for when feedback data comprises information unauthorized, delete feedback data.

Fig. 6 is a kind of server device topology schematic diagram that the embodiment of the present invention provides.As shown in Figure 6, the data processing method that this server apparatus can provide for implementing above-described embodiment.Wherein, this server apparatus can be the high-performance computer in network environment or computer system etc., intercepts the services request that other computing machines (client computer) on network are submitted to, and corresponding service is provided.Preferential:

This server apparatus 600 includes but not limited to following structure or kinetic energy.Preferentially, this server apparatus 600 at least comprise one or more central processing units (CPU) 610, one or more internal memory 630, for one or more media 620(of storage operation system 621, application program 622 or data one or more mass storages for example).

These one or more internal memories 630 and media 620 can be set to interim or non-interim.The program be stored in one or more media 620 can comprise one or more modules.Each module can comprise the operational order collection of this server apparatus 600.Closer, CPU610 can be configured to communicate, carry out instruction set and carry out the operation on server apparatus 600 with media 620.

This server apparatus can also comprise one or more power supplys 660, one or more wired or wireless network interface 640, the one or more input and output of one or more keyboard (I/O) interface 650 and/or one or more operating system 621, as Windows Server ^tM, Mac OS X ^tM, Unix ^tM, Linux ^tM, FreeBSD ^tMetc. operating system.

Specifically in the present embodiment, server apparatus includes storer, and one or more than one program, one of them or an above program are stored in storer, and are configured to carry out one or above routine package containing for carrying out the instruction of following operation by one or above processor:

The numerical value that the feedback data of each version of record product is corresponding and the Times of Feedback of feedback data;

Obtain weighted value corresponding to each version of product;

The final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to feedback data.

Preferably, also comprise for carrying out the instruction of following operation: according to feedback data, the final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version comprises:

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j}$ Formula (1)

Wherein, the final feedback data that S is this product; C _ijfor i the numerical value that feedback data is corresponding to j version of this product; N _ijtimes of Feedback for i the feedback data to j version of this product; W _jweighted value for j version of this product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

Preferably, also comprise for carrying out the instruction of following operation: by following formula, calculate the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1}),$

Wherein, W _jfor the weighted value of j version of product, j=1,2,3 ... m; M is the product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

Preferably, also comprise for carrying out the instruction of following operation: the difference in version value is calculated and is obtained by following formula:

Work as z _j<y _jthe time, the difference in version value

$t_j=\frac{z_j}{y_j};$ Formula (3)

Work as z _j>=y _jthe time, the difference in version value

$t_j=\frac{z_j}{z_j+y_j};$ Formula (4)

Z _jthe code update increment obtained for the difference according between j version of product and j+1 version; y _jsize of code for j version of product.

Preferably, also comprise for carrying out the instruction of following operation: detect whether record of product version corresponding to feedback data; When product version corresponding to feedback data recorded, obtain weighted value corresponding to product version recorded; When product version corresponding to feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

Preferably, also comprise for carrying out the instruction of following operation: the user ID of feedback data submitted in record; Detect this user according to user ID and whether the same version of identical product had been submitted to feedback data; When the feedback data of having submitted to, this user is deleted the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

Preferably, also comprise for carrying out the instruction of following operation: the user ID of feedback data submitted in record; The number of times of the feedback data that detects the number of times of the feedback data that user ID submits to product or one of them version of product is submitted to; When the number of times of the number of times of the feedback data that product is submitted to or feedback data that one of them version of product is submitted to surpasses default first threshold, the feedback data that recording user ID does not submit to.

Preferably, also comprise for carrying out the instruction of following operation: the user ID of feedback data submitted in record; Detect user ID and submit the number of times of feedback data in the Preset Time section; While submitting to the number of times of feedback data to surpass default Second Threshold in the Preset Time section when user ID, the feedback data that recording user ID does not submit to.

Preferably, also comprise for carrying out the instruction of following operation: feedback data is examined, judge in feedback data and whether comprise information unauthorized; While in feedback data, comprising information unauthorized, delete feedback data.

The data processing method of present device embodiment, device and server apparatus, consider the iteration of some products, distribute different weighted values by the different spaces of a whole page to product, like this, when calculating the final feedback data of this product, can embody the impact of the feedback data of different editions product for final feedback data, make the analysis of user feedback data more accurate.

Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt complete hardware implementation example, implement software example or in conjunction with the form of the embodiment of software and hardware aspect fully.And the present invention can adopt the form that wherein includes the upper computer program of implementing of computer-usable storage medium (including but not limited to magnetic disk memory and optical memory etc.) of computer usable program code one or more.

The present invention describes with reference to process flow diagram and/or the block scheme of method, equipment (system) and computer program according to the embodiment of the present invention.Should understand can be in computer program instructions realization flow figure and/or block scheme each flow process and/or the flow process in square frame and process flow diagram and/or block scheme and/or the combination of square frame.Can provide these computer program instructions to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, make the instruction of carrying out by the processor of computing machine or other programmable data processing device produce for realizing the device in the function of flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame appointments.

These computer program instructions also can be stored in energy vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work, make the instruction be stored in this computer-readable memory produce the manufacture that comprises command device, this command device is realized the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame.

These computer program instructions also can be loaded on computing machine or other programmable data processing device, make and carry out the sequence of operations step to produce computer implemented processing on computing machine or other programmable devices, thereby the instruction of carrying out on computing machine or other programmable devices is provided for realizing the step of the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame.

Obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention also is intended to comprise these changes and modification interior.

Claims (21)

1. a data processing method, is characterized in that, described method comprises:

The numerical value that record is corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Obtain weighted value corresponding to each version of product;

The final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to described feedback data.

2. the method for claim 1, it is characterized in that, according to described feedback data, the final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version, comprising: according to the final feedback data of following formula counting yield

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j},$

Wherein, the final feedback data that S is this product; C _ijfor i the numerical value that feedback data is corresponding to j version of described product; N _ijtimes of Feedback for i the feedback data to j version of described product; W _jweighted value for j version of described product; I=1,2,3......n, j=1,2,3......m, the number that n is numerical value corresponding to feedback data, the version number that m is described product.

3. method as claimed in claim 1 or 2, is characterized in that, obtains weighted value corresponding to each version of product and comprise: by following formula, calculate the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1}),$

Wherein, W _jfor the weighted value of j version of described product, j=1,2,3...m; M is described product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

4. method as claimed in claim 3, is characterized in that, described difference in version value is calculated and obtained by following formula:

Work as z _j<y _jthe time, described difference in version value

Work as z _j>=y _jthe time, described difference in version value

Described z _jthe code update increment obtained for the difference according between j version of described product and j+1 version; y _jsize of code for j version of described product.

5. the method for claim 1, is characterized in that, the weighted value that each version of described product is corresponding is arithmetic progression or Geometric Sequence, or weighted value corresponding to each version of described product obtains according to the table lookup set in advance.

6. the method for claim 1, is characterized in that, obtains weighted value corresponding to each version of product and comprise:

Detect whether record of product version corresponding to described feedback data;

When product version corresponding to described feedback data recorded, obtain weighted value corresponding to described product version recorded;

When product version corresponding to described feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

7. the method for claim 1, is characterized in that, described method also comprises:

The user ID of described feedback data submitted in record;

Detect described user according to described user ID and whether the same version of identical product had been submitted to feedback data;

When the feedback data of having submitted to, described user is deleted the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

8. the method for claim 1, is characterized in that, described method also comprises:

The user ID of described feedback data submitted in record;

The number of times of the feedback data that detects the number of times of the feedback data that described user ID submits to described product or one of them version of described product is submitted to;

When the number of times of the number of times of the feedback data that described product is submitted to or feedback data that one of them version of described product is submitted to surpasses default first threshold, do not record the feedback data that described user ID is submitted to.

9. the method for claim 1, is characterized in that, described method also comprises:

The user ID of described feedback data submitted in record;

Detect described user ID and submit the number of times of feedback data in the Preset Time section;

While submitting to the number of times of feedback data to surpass default Second Threshold when described user ID, do not record the feedback data that described user ID is submitted in the Preset Time section.

10. the method for claim 1, is characterized in that, described method also comprises:

Described feedback data is examined, judge in described feedback data and whether comprise information unauthorized;

While in described feedback data, comprising information unauthorized, delete described feedback data.

11. a data processing equipment, is characterized in that, described device comprises:

Logging modle, for recording the numerical value corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Acquisition module, for obtaining weighted value corresponding to each version of product;

Final feedback data computing module, for the final feedback data of the weighted value counting yield of the Times of Feedback of the numerical value corresponding according to described feedback data, feedback data and each version.

12. device as claimed in claim 11, is characterized in that, described final feedback data computing module, and for calculate the final feedback data of counting yield according to following formula,

$S=\frac{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{C}_{\mathrm{ij}}{N}_{\mathrm{ij}}{W}_j}{\underset{j=1}{\overset{m}{\mathrm{\&Sigma;}}}\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{N}_{\mathrm{ij}}{W}_j},$

Wherein, the final feedback data that S is described product; C _ijfor i the numerical value that feedback data is corresponding to j version of described product; N _ijtimes of Feedback for i the feedback data to j version of described product; W _jweighted value for j version of described product; I=1,2,3......n, j=1,2,3......m, the number that n is feedback data, the number that m is version.

13. device as described as claim 11 or 12, is characterized in that, described acquisition module comprises: the weight calculation submodule, for by following formula, calculating the weighted value that each version is corresponding:

$W_j=\underset{j=1}{\overset{j}{\mathrm{\&Pi;}}}(1-t_{j-1}),$

Wherein, W _jfor the weighted value of j version of described product, j=1,2,3...m; M is described product version number; t _jbe the difference in version value between j version and j+1 version, wherein t ₀=0.

14. device as claimed in claim 13, is characterized in that, described weight calculation submodule, for calculate described difference in version value by following formula:

Work as z _j<y _jthe time, described difference in version value

Work as z _j>=y _jthe time, described difference in version value

Described z _jthe code update increment obtained for the difference according between j version of described product and j+1 version; y _jsize of code for j version of described product.

15. device as claimed in claim 11, is characterized in that, the weighted value that each version of described product is corresponding is arithmetic progression or Geometric Sequence, or weighted value corresponding to each version of described product obtains according to the table lookup set in advance.

16. device as claimed in claim 11, is characterized in that, described acquisition module also comprises:

Detection sub-module, the corresponding product version record whether for detection of described feedback data;

The Weight Acquisition submodule, for when product version corresponding to described feedback data recorded, obtain weighted value corresponding to described product version recorded;

Described weight calculation submodule, when product version corresponding to described feedback data do not record, recalculate the product version and weighted value corresponding to Unrecorded product version that have recorded.

17. device as claimed in claim 11, is characterized in that, described device also comprises: first detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Whether described first detection module, submitted feedback data to the same version of identical product for detect described user according to described user ID;

Described logging modle, for the feedback data that ought submit to, delete described user to the feedback data of formerly submitting to of the same version of identical product, be retained in the feedback data of rear submission.

18. device as claimed in claim 11, is characterized in that, described device also comprises: the second detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Described the second detection module, the number of times of the number of times of the feedback data of described product being submitted to for detection of described user ID or feedback data that one of them version of described product is submitted to; Described logging modle, while for the number of times of the number of times of the feedback data when described product is submitted to or feedback data that one of them version of described product is submitted to, surpassing default first threshold, do not record the feedback data that described user ID is submitted to.

19. device as claimed in claim 11, is characterized in that, described device also comprises: the 3rd detection module,

Described logging modle, for recording the user ID of submitting described feedback data to;

Described the 3rd detection module is submitted the number of times of feedback data in the Preset Time section for detection of described user ID;

Described logging modle, while for the number of times of submitting feedback data when described user ID in the Preset Time section to, surpassing default Second Threshold, do not record the feedback data that described user ID is submitted to.

20. device as claimed in claim 11, is characterized in that, described device also comprises:

Auditing module, for described feedback data is examined, judge in described feedback data and whether comprise information unauthorized;

Described logging modle, for when described feedback data comprises information unauthorized, delete described feedback data.

A 21. server apparatus, it is characterized in that, described server apparatus includes storer, and one or more than one program, one of them or an above program are stored in storer, and are configured to carry out described one or above routine package containing for carrying out the instruction of following operation by one or above processor:

The numerical value that record is corresponding to the feedback data of each version of product and the Times of Feedback of feedback data;

Obtain weighted value corresponding to each version of product;

The final feedback data of the weighted value counting yield of the Times of Feedback of corresponding numerical value, feedback data and each version according to described feedback data.

Images