CN102567477A - Website value evaluation method and device - Google Patents
Website value evaluation method and device Download PDFInfo
- Publication number
- CN102567477A CN102567477A CN2011104213450A CN201110421345A CN102567477A CN 102567477 A CN102567477 A CN 102567477A CN 2011104213450 A CN2011104213450 A CN 2011104213450A CN 201110421345 A CN201110421345 A CN 201110421345A CN 102567477 A CN102567477 A CN 102567477A
- Authority
- CN
- China
- Prior art keywords
- website
- data
- model
- value
- collection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a website value evaluation method and a website value evaluation device. The method comprises the following steps of: acquiring, namely acquiring user network behavior data and operation information data of a plurality of websites; modeling, namely determining feature factors which influence website values, and making a website value evaluation data model according to the feature factors; and determining parameters, namely performing data training on the website value evaluation data model by using the acquired user network behavior data and operation information data so as to determine the model parameters of the website value evaluation data model and acquire a website value evaluation data determining model which is used for evaluating a commercial value of a website. By the method, the evaluation value of each website is acquired according to the constructed model, user groups which are matched with contents of an intelligent advertisement to be issued by the website can be found according to the evaluation value of the website, and the owner of the advertisement can also decide the website on which the advertisement is issued according to the evaluation value, so that the advertisement issuing effect of the owner of the advertisement is improved.
Description
Technical field
The present invention relates to information network technique, relate in particular to website value assessment method and device.
Background technology
Since Internet technology grows up, freely be regarded as inherent " internet spirit " with sharing always, how when free products & services are provided, keep the operation even the profit of website, be a difficult problem of putting before all website interareas.By up to the present, it the most also is one of profit model the most effectively that Internet advertising is still the website.As far as the Internet advertising master, how from thousands of home Web sites, to select the website that is fit to own specific products and carry out the advertisement displaying, thereby obtain the highest attention rate and be converted into the problem that commercial value is a needs solution.
Under the traditional mode webpage rank (Page Rank is called for short PR) mode is all adopted in the evaluation of website all the time, this mode is estimated the value of website with the flow size.But actual conditions; Have the flow of a lot of websites maybe be very not big, but the unusual specialty, high-end of its content, its user group who faces is very professional high-end also; If estimate these websites from the flow angle merely, can't really embody its due commercial value.The PR mode is utilized the web crawlers technology, grasps webpage, the linking relationship between analyzing web page content and the webpage; Through the linking relationship between the analyzing web page, weigh the significance level of a webpage.
The PR mode is the composite target that Google is used for representing Web page importance, and can not receive the influence of various retrievals.The PR mode just is based on the analysis to " use complicated algorithm and obtain link structure ", thus the characteristic of each webpage that draws itself.The ultimate principle of PR mode is that the website can be made its recommendation degree improve by many page links.Simple saying is " welcome (by many page links) page must be the page of high-quality ".The PR mode is mainly by three decisions of backward chaining index of correlation, that is: backward chaining numbers (the popularity index on the simple meaning); Whether backward chaining is from the high page of recommendation degree (valid welcome index); The link number of the backward chaining source page (selected probability index).
Existing PR mode has only reflected the pouplarity of content of pages; And the advertiser paid close attention to is on this website, to throw in advertisement delivery effect; Like clicking rate and conversion ratio; The same user experience of these indexs, user psychology, user preferences have much relations, and the PR mode can not reflect the input effect of this advertisement.
Summary of the invention
One of technical matters to be solved by this invention is that a kind of website value assessment method and device need be provided.
In order to solve the problems of the technologies described above, the invention provides a kind of website value assessment method, this website value assessment method comprises: obtaining step, obtain the user network behavioral data and the operation information data of a plurality of websites; Modeling procedure, confirming influences the characteristic factor that the website is worth, and formulates website value assessment data model according to said characteristic factor; And parameter is confirmed step; Utilize user network behavioral data and the operation information data obtained to come said website value assessment data model is carried out the data training; To confirm the model parameter of said website value assessment data model, the website value assessment data that obtain being used to estimating the website commercial value are confirmed model.
Website value assessment method according to a further aspect of the invention, said parameter confirms that step specifically comprises:
The said user network behavioral data relevant with said characteristic factor of said operation information data and the said first website collection of first website being concentrated the website is as training dataset; And the said user network behavioral data relevant with said characteristic factor of said operation information data and the said second website collection of second website being concentrated the website be as the verification msg collection, and
Come the said website value assessment of verification data to confirm whether model has caused over-fitting based on the verification msg collection, for being, then confirm again to return said parameter and confirm step in the website that said first website collection and the said second website collection are comprised as if check results, wherein,
Said first website collection and the said second website collection are not identity sets.
Website value assessment method according to a further aspect of the invention, in said modeling procedure,
The website value assessment data model of being formulated be expressed as BR=f (xUi, yFi, zMi); Wherein, said BR is a commercial level, and Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic; F representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter;
The annual income at the operation information packet purse rope station of said website, with the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website.
Website value assessment method according to a further aspect of the invention also comprises:
Evaluation procedure; The user network behavioral data that will carry out the website of commercial value evaluation import resulting website value assessment data confirm in the model in the hope of the BR value, as being used to estimate the said evaluation of estimate that will carry out the website commercial value of the website that commercial value estimates.
Website value assessment method according to a further aspect of the invention, said evaluation procedure also comprises:
Said evaluation of estimate is divided into a plurality of intervals, confirms the commercial value rank of this website according to the residing interval of the evaluation of estimate of website.
Website value assessment method according to a further aspect of the invention; Said characteristic factor comprises user characteristics, traffic characteristic and marketing characteristic; And the characterization factor that constitutes said user characteristics comprises user group, pageview and residence time per capita per capita; The characterization factor that constitutes said traffic characteristic comprises page access amount, IP amount and independent user sessions, and the characterization factor that constitutes said marketing characteristic comprises clicking rate, conversion ratio and rate of return on investment.
Website value assessment method according to a further aspect of the invention, said user network behavioral data are setting-up time section, setting region and/or the crowd's of setting user network behavioral data.
According to a further aspect in the invention, also provide a kind of website commercial value apparatus for evaluating to comprise: acquisition module (61), it obtains the user network behavioral data and the operation information data of a plurality of websites; MBM (62), it confirms to influence the characteristic factor that the website is worth, and formulates website value assessment data model according to said characteristic factor; And parameter determination module (63); It utilizes user network behavioral data and the operation information data obtained to come said website value assessment data model is carried out the data training; To confirm the model parameter of said website value assessment data model, the website value assessment data that obtain being used to estimating the website commercial value are confirmed model.
Website commercial value apparatus for evaluating according to a further aspect of the invention, said parameter determination module (63) comprising:
The training submodule; The said user network behavioral data relevant with said characteristic factor of said operation information data and the said first website collection that is used for the website is concentrated in first website is as training dataset; And the said user network behavioral data relevant with said characteristic factor of said operation information data and the said second website collection of second website being concentrated the website be as the verification msg collection, and
The checking submodule is used for coming the said website value assessment of verification data to confirm whether model has caused over-fitting based on the verification msg collection, if check results is for being; Then confirm the website that said first website collection and the said second website collection are comprised again; Return said parameter determination module (63), wherein
Said first website collection and the said second website collection are not identity sets.
Website commercial value apparatus for evaluating according to a further aspect of the invention; The website value assessment data model of being formulated be expressed as BR=f (xUi, yFi, zMi); Wherein, Said Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic, and f representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter; The annual income at the operation information packet purse rope station that is obtained, with the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website.
Website commercial value apparatus for evaluating according to a further aspect of the invention also comprises:
Evaluation module (64); Its user network behavioral data that will carry out the website of commercial value evaluation is imported resulting website value assessment data and is confirmed in the model in the hope of the BR value, as being used to estimate the said evaluation of estimate that will carry out the website commercial value of the website that commercial value estimates.
According to a further aspect in the invention, also provide a kind of website commercial value appraisal procedure to comprise: the website commercial value to be assessed through following steps:
(1) the user browsing behavior data is collected and stored; Collect the operation and the financial information of website from open approach;
(2) confirm to influence the characteristic factor that the website is worth, assessment models is formulated;
(3) confirm model parameter, the value assessment data model sets up a web site;
(4) website value assessment data model being used to assess the website is worth;
(5) through the verification msg set, model is optimized,, confirms final mask with the robustness and the model accuracy of raising system under real complex situations;
(6) the user behavior parameter is applied mechanically in the final mask, estimate its website commercial value.
Website value assessment method according to a further aspect of the invention, said model parameter are calculated through following formula and are obtained, and (zMi), said Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic to BR=f for xUi, yFi.
According to a further aspect in the invention; A kind of website commercial value evaluating system also is provided; Comprise and be used for business intelligence Information Collection System and network data acquisition system that the user browsing behavior data are collected and stored; Said two systems give the data training system of the value assessment data model that is used to set up a web site with data transmission, and said data training system also receives the model evaluation system that is used for verification msg simultaneously.
Website commercial value evaluating system according to a further aspect of the invention; Said data training system comprises model bank; Said model bank is provided with and is used for the Model Selection module, and said Model Selection module is exported selected model, and said selected model receives the data of business intelligence Information Collection System and network data acquisition system; Said selected model passes through parameter calculating module; Form and confirm module output, said affirmation module gets into the verification module and is optimized, and said verification module receives the model evaluation system data.
Website commercial value evaluating system according to a further aspect of the invention, said model evaluation system comprises business intelligence Information Collection System and network data acquisition system.
Compared with prior art, the present invention has the following advantages at least:
In the present invention; The evaluation of estimate of each website that obtains according to constructed website value assessment data model is in time, region and the crowd's three dimensions; Each website can find according to the evaluation of estimate of himself website with its customer group that is complementary of the content of the intelligent advertisement that will throw in; And the advertiser also can through evaluation of estimate come correct decisions its advertisement that will throw in should on which website, throw in, improve advertiser's advertisement delivery effect.
Other features and advantages of the present invention will be set forth in instructions subsequently, and, partly from instructions, become obvious, perhaps understand through embodiment of the present invention.The object of the invention can be realized through the structure that in instructions, claims and accompanying drawing, is particularly pointed out and obtained with other advantages.
Description of drawings
Accompanying drawing is used to provide further understanding of the present invention, and constitutes the part of instructions, is used to explain the present invention with embodiments of the invention, is not construed as limiting the invention.In the accompanying drawings:
Fig. 1 is the schematic flow sheet according to the website value assessment method of first embodiment of the invention;
Fig. 2 is the structural representation according to the website valve estimating system of second embodiment of the invention;
Fig. 3 is the schematic flow sheet according to the website value assessment method of third embodiment of the invention;
Fig. 4 is the schematic flow sheet according to the creating website value assessment data model of third embodiment of the invention;
Fig. 5 forms topology example figure according to the characteristic that influences evaluation of estimate of third embodiment of the invention;
Fig. 6 is the structural representation according to the website value assessment device of fourth embodiment of the invention.
Embodiment
Below will combine accompanying drawing and embodiment to specify embodiment of the present invention, how the application technology means solve technical matters to the present invention whereby, and the implementation procedure of reaching technique effect can make much of and implement according to this.Need to prove that only otherwise constitute conflict, each embodiment among the present invention and each characteristic among each embodiment can mutually combine, formed technical scheme is all within protection scope of the present invention.
In addition; Can in computer system, carry out in the step shown in the process flow diagram of accompanying drawing such as a set of computer-executable instructions, and, though logical order has been shown in process flow diagram; But in some cases, can carry out step shown or that describe with the order that is different from here.
First embodiment
Fig. 1 is the schematic flow sheet according to the website value assessment method of first embodiment of the invention, specifies each step of this method below with reference to Fig. 1.
Like the described website of Fig. 1 commercial value appraisal procedure, the website commercial value is assessed through following steps:
Step (1) is obtained the user network behavioral data of a plurality of websites and is obtained the operation information data of setting the quantity website through open approach;
Step (2) confirms to influence the characteristic factor that the website is worth, and according to the characteristic factor of confirming website value assessment data model is formulated;
Step (3) through the data training, is confirmed model parameter, and the value assessment data model sets up a web site;
Step (4) is used to assess the website with website value assessment data model and is worth;
Step (5) through the verification msg set, is optimized model, with robustness and the model accuracy of raising system under real complex situations, confirms final mask;
Step (6) is applied mechanically the user behavior parameter in the final mask, estimates the commercial value of other websites.
The website value assessment data model of said formulation be expressed as BR=f (xUi, yFi, zMi); Wherein, BR is a commercial level, and Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic; F representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter.
User characteristics includes but not limited to following characterization factor: user group, the pageview and the residence time per capita per capita.
Traffic characteristic includes but not limited to following characterization factor: page access amount PV, IP and independent user sessions UV.
The marketing characteristic includes but not limited to following characterization factor: clicking rate, conversion ratio and ROI.
Usually, the commercial value of website can be represented with formula:
Commercial value=the f of website (number of users (frequency); The value of unique user (frequency) (advertising income or purchase income); This formula is to calculate the most directly method of website commercial value; But the value of unique user is difficult to quantize, and then this formula is difficult to realize in actual conditions, therefore adopts the BR formula to substitute this formula.
In the commercial value formula of above-mentioned website, number of users reflection commercial size, for any business model, scale and benefit are very important.Such as the Microsoft of software industry, the Google of search engine and the Tengxun of IM, all be to rely on scale and benefit, obtained the income scale that traditional business model is difficult to reach.Scale and benefit are particularly important at information service field.For the advertiser, number of users has comprised multiple parameter, like IP, and PV, UV etc.To different advertisers' different demands, its importance is also different.
It is consumption figure and the advertising income that each user can bring that unique user is worth; And customer consumption amount and advertisement acceptance level have substantial connection with user's multidimensional property; Like sex, age, occupation, income, schooling, family role, hobby, Regional Distribution etc., therefore when considering that unique user is worth, to consider above-mentioned factor.
Present embodiment is based on time, region, crowd's three dimensions the commercial value of website is estimated dynamically.
The commercial value of website is by the decision of the factor of a plurality of dimensions, it often one in time, the difference of region, industry and the value that constantly changes.In order to safeguard the efficient of value assessment algorithm, need between the complicacy of dimension granularity of division and algorithm, keep balance.If but these three dimensions of undue segmentation can cause algorithm too complicated again, have lost meaning on the contrary.As time granularity can be per year, month, day divide, the region can be with country, administrative area, city division, the crowd then can wait with demographic dimension, hobby and divide.
The BR value can instruct the website to find with its content, the advertisement that customer group etc. are complementary, and the advertiser also can come its advertisement of correct decisions to represent in which website through the BR value.
Among the three big effect characteristics factor Ui that influence the BR value in the present embodiment, Fi, the Mi, the website can drive the lifting of Ui and Mi through continuing to optimize Fi, and the direct lifting of Ui and Mi then is to decide through the performance of this website in actual commercialization.The promotional period that the advertiser can be in through himself product is comprehensively weighed Ui, Fi, and Mi carries out the decision-making of advertisement putting.
Website commercial value evaluation system can adopt staging hierarchy, like the 1-10 level, will divide in one-level in all interval websites of a certain BR value.In order to make different web sites that unified evaluation criteria arranged and to calculate the factor of influence coefficient, present embodiment adopts the object of reference of the annual income of website as the website commercial value.At first user capture is monitored, obtain Ui through network monitoring or website, Fi, the parameter in the Mi characteristic is obtained the annual income of this website then, obtains following equation:
BR
1＝f(A
1U
1...AnU
n，B
1F
1...B
nF
n，C
1M
1...C
nM
n)
BR
2＝f(A
1U
1...A
nU
n，B
1F
1...B
nF
n，C
1M
1...C
nM
n)
BR
n＝f(A
1U
1...A
nU
n，B
1F
1...B
nF
n，C
1M
1...C
nM
n)
Through repeatedly curve fitting algorithm or neural network algorithm can derived parameter A
1... A
n, B
1... B
n, C
1... C
n, obtain the BR computing formula thus, when the BR of a certain website of needs value, only need influence factor Ui with BR, Fi, Mi substitution formula can obtain the BR value of this website.
Second embodiment
Fig. 2 illustrates the structural representation according to the website commercial value evaluating system of second embodiment of the invention.The each several part composition of present embodiment is described with reference to figure 2 below.
A kind of website as shown in Figure 2 commercial value evaluating system; Comprise and be used for business intelligence Information Collection System and network data acquisition system that the user browsing behavior data are collected and stored; Business intelligence Information Collection System and network data acquisition system give the data training system of the value assessment data model that is used to set up a web site with data transmission, and the data training system also receives the data that are used to verify from the model evaluation system simultaneously.
Wherein, The data training system comprises model bank; Model bank comprises the Model Selection module, and the Model Selection module is exported selected model, and selected model receives the data that are stored in business intelligence Information Collection System and network data acquisition system; Selected model passes through parameter calculating module; Form and confirm model output, the verification module is optimized the model of confirming through the verification msg that receives the model evaluation system, and wherein the model evaluation system also comprises business intelligence Information Collection System and network data acquisition system.
The 3rd embodiment
Fig. 3 is the schematic flow sheet according to the website commercial value appraisal procedure of third embodiment of the invention, specifies each step of this method below with reference to Fig. 3.
Step 310 is obtained the user network behavioral data and the operation information data of a plurality of websites, and this step is an obtaining step.
Particularly; Obtain technology through network data and obtain user in a plurality of websites and carry out all data of network behavior and the data that obtained are carried out pre-service obtaining embodying the user network behavioral data of user network behavior, and obtain the operation information data of a plurality of websites through open approach.Wherein, user's network behavior can be user's the behavior of browsing, and the operation information of website mainly comprises the annual income of website.
More specifically, owing to be used for characterizing the journal file collection of the general data of user network behavior from the webserver of website.These journal files have comprised the executive logging about each visitor's HTTP (being called for short HTTP) affairs of visiting this website, therefore can be through utilizing network data capture techniques such as network packet sniff technology or web crawlers technology, concentrating from journal file and obtain the data that are used to characterize the user network behavior.But; Because the actual data that obtain have incomplete correlativity, redundancy and notional ambiguity and possibly there are a large amount of problems such as meaningless information in the mass data that is used for auditing; Cause the data of being obtained to carry out pre-service and have the information that can embody the user network behavior to make up and to recognize; For example, can carry out pre-service to these data through introducing Packet Filtering mechanism.
Need to prove; The user network behavioral data that this step 310 is obtained is the setting-up time section, sets the region and/or the crowd's of setting user network behavioral data, and these three factors of time period, region and crowd are estimated the value of website as three dimensions dynamically.
Commercial value when particularly, these three factors being estimated the crowd of region and/or different set of time period that a website is in different set, different set as a three dimensional space coordinate.When the partition dimension granularity, need the efficient of maintaining web value assessment method simultaneously, with the balance between the complicacy that keeps dimension granularity of division and website value assessment method.For example; With the granularity division of time dimension is year, month, day; With the region dimension be divided into country, administrative area, city, crowd's dimension is divided into the size of population and hobby, obtain the user network behavioral data based on the dividing mode of above dimension granularity.
To be used to calculate commercial level (Business Rank in the present embodiment; Being called for short BR) formula of value is as formulation website value assessment data model, with the evaluation of estimate of BR value as the website commercial value that is used to estimate the said website that will carry out the commercial value evaluation.Therefore, with the characteristic factor that influences evaluation of estimate as influencing the characteristic factor that the website is worth.
Fig. 5 forms topology example figure according to the characteristic factor that influences evaluation of estimate of third embodiment of the invention; Please refer to Fig. 5; Preferably, the present embodiment characteristic factor that will influence the BR value is set at user characteristics (being called for short Ui), the traffic characteristic (being called for short Fi) and marketing characteristic (abbreviation Mi) that objectively responds the website.The website value assessment data model of being formulated is expressed as:
BR＝f(xUi，yFi，zMi) (1)
In formula (1), BR is a commercial level, and Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic, and f representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter.
Need to prove; Each characteristic that influences the BR value is made up of the characterization factor of set amount; As shown in Figure 5; Preferably, the characterization factor of the formation user characteristics in the present embodiment is user group, pageview and residence time per capita per capita, and the characterization factor that constitutes traffic characteristic is page access amount (being called for short PV), IP amount and independent user sessions (being called for short the UV amount); The characterization factor that constitutes the marketing characteristic is clicking rate, conversion ratio and rate of return on investment (being called for short ROI), and then the computing formula of BR value can be as follows:
BR＝f(A
1U
1，A
2U
2，...，A
nU
n，B
1F
1，B
2F
2，...，B
kF
k，C
1M
1，C
2M
2，...，C
jM
j) (2)
In formula (2), A
1... A
nThe representation feature factor is n the characteristic parameter of user characteristics Ui, is equivalent to the model parameter x in the formula (1);
B
1... B
kThe representation feature factor is k the characteristic parameter of traffic characteristic Fi, is equivalent to the model parameter y in the formula (1);
C
1... C
jThe representation feature factor is equivalent to the model parameter z in the formula (1) for j the characteristic parameter of marketing characteristic Mi;
U
1... U
nThe representation feature factor is n the characterization factor of user characteristics Ui, is equivalent to the user characteristics Ui in the formula (1);
F
1... F
kThe representation feature factor is k the characterization factor of traffic characteristic Fi, is equivalent to the user characteristics Fi in the formula (1);
M
1... M
jThe representation feature factor is equivalent to the user characteristics Mi in the formula (1) for j the characterization factor of marketing characteristic Mi.
In the present embodiment; The characteristic factor that influences the BR value is set at the characteristic factor of website value assessment data model; Characteristic parameter in the BR value computing formula is corresponding to the model parameter of website value assessment data model, and wherein, the quantitative value of characteristic parameter equates with the quantitative value of characteristic factor.
Particularly; Make up training dataset and verification msg collection based on the operation information data of being obtained with the user network behavioral data relevant with aspect of model factor; Utilize training dataset to train website value assessment data then and confirm model; That is train the model parameter of this model and the mapping relations between the concentrated data of training data, because A
1... A
nBe equivalent to the model parameter x in the formula (1), B
1... B
kBe equivalent to the model parameter y in the formula (1), C
1... C
jBe equivalent to the model parameter z in the formula (1), obtain A in the above-mentioned website value assessment data model formula (2)
1... A
n, B
1... B
k, C
1... C
jConcrete numerical value, promptly can obtain model parameter x, model parameter y and model parameter z, be that website value assessment data are confirmed model thereby obtain final mask.
Then; Website value assessment data based on the verification msg collection comes verification to be trained by training dataset confirm whether model has caused over-fitting; If check results is for being; Then rebuild with before nonrecoverable training dataset training dataset and verification msg collection data different with verification msg collection data, come value assessment data model in website is carried out the data training, confirm that again value assessment data in website confirm the model parameter of model.
For example, Fig. 4 is the schematic flow sheet according to the creating website value assessment data model of third embodiment of the invention, specifies each step of this model below with reference to Fig. 4.
Particularly; In a plurality of websites, set the first website collection and the second website collection; The user network behavioral data relevant with characteristic factor of operation information data and the first website collection of first website being concentrated the website is as training dataset; And the user network behavioral data relevant with characteristic factor of operation information data and the second website collection of second website being concentrated the website be as the verification msg collection, and wherein, the first website collection and the second website collection are not identity sets.
Need to prove; The data of training dataset and checking training set can be merely the part in operation information data and the user network behavioral data relevant with characteristic factor, and the data that training data is concentrated can the identical or complete difference of part with the data of verifying training set.
For example; In the user network behavioral data of a plurality of websites that obtain, choose the user's of N website network behavior data and corresponding operation information data; Aspect of model factor based on having set obtains training dataset, and training dataset can be represented as follows
w
1＝(U
11，U
12...U
1n，F
11，F
12...F
1k，M
11，M
12...M
1j；BR
1)
w
2＝(U
21，U
22...U
2n，F
21，F
22...F
2k，M
21，M
22...M
2j；BR
2)
......
w
N＝(U
N1，U
N2...U
Nn，F
N1，F
N2...F
Nk，M
N1，M
N2...M
Nj；BR
N)
Wherein, U
NnN characterization factor that influences user characteristics representing N website, F
NkRepresent k the characterization factor that influences traffic characteristic in N website, M
NjThe characterization factor of representing j the influence marketing in N website characteristic, BR
NRepresent the annual income in the operation information data of N website.
Then, network behavior data and corresponding operation information data with the user of M website in a plurality of websites that obtain obtain the verification msg collection based on the aspect of model factor of having set, and the verification msg collection can represent as follows,
w
1＝(U
11，U
12...U
1n，F
11，F
12...F
1k，M
11，M
12...M
1j；BR
1)
w
2＝(U
21，U
22...U
2n，F
21，F
22...F
2k，M
21，M
22...M
2j；BR
2)
......
w
M＝(U
M1，U
M2...U
Mn，F
M1，F
M2...F
Mk，M
M1，M
M2...M
Mj；BR
M)
Wherein, U
MnN characterization factor that influences user characteristics representing M website, F
MkRepresent k the characterization factor that influences traffic characteristic in M website, M
MjThe characterization factor of representing j the influence marketing in M website characteristic, BR
MRepresent the annual income in the operation information data of M website, the N that wherein a chooses website and M website can part identical.
Need to prove; The above; Be merely simple example of the present invention, the characteristic factor of website value assessment data model also can comprise other factors, is not limited to above related content; In order to make different websites that unified evaluation criteria arranged, the annual income in the operation information data of the setting quantity website that present embodiment will openly obtain is as the assessment reference of assessment website commercial value.
Particularly, for example, will be based on the user network behavioral data of N website and the training dataset substitution formula (2) of operation information data gained, i.e. BR=f (A
1U
1... A
nU
n, B
1F
1... B
kF
k, C
1M
1... C
jM
j) in, can get as follows,
BR
1＝f(A
1U
11...A
nU
1n，B
1F
11...B
kF
1k，C
1M
1...C
jM
1j)
BR
2＝f(A
1U
21...A
nU
2n，B
1F
21...B
kF
2k，C
1M
21...C
jM
2j)
......
BR
N＝f(A
1U
N1...A
nU
Nn，B
1F
21...B
kF
Nk，C
1M
N1...C
jM
Nj)
Then, obtain characteristic parameter based on the annual income in user network behavioral data and the said operation information data, that is the model parameter of website value assessment data model.
More specifically, for example, the user network behavioral data of N the website that training data is concentrated is as the input variable in the following formula, i.e. U in the following formula
N1... U
Nn, F
21... F
NkAnd M
N1... M
Nj, the annual income in N the website operation information data that training data is concentrated is as the output variable in the following formula, i.e. BR in the following formula
1, BR
2... BR
N, then through repeatedly curve fitting algorithm or neural network algorithm can obtain characteristic parameter A
1... A
n, B
1... B
k, C
1... C
jAnd the mapping relations of input variable and output variable, the computing formula that can obtain being used to calculate the BR value based on the characteristic parameter and the mapping relations of gained.
Need to prove; The annual income at the operation information packet purse rope station of website; With the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website; The operation information data of website are obtained through open approach in this step, and the website can be some well-known larger websites, are prone to obtain the website that the operation information data are little compared to some flow, content is professional or the user group is high-end of these websites.
Need to prove; The definite website value assessment data of parameter based on model in step 3302 are confirmed model; Can be used to estimate the commercial value of the website that will carry out the commercial value evaluation; But definite website value assessment data confirm that model is not optimum website value assessment data model, the precision of its robustness and model is not very high under real complex situations, therefore further the execution in step 3303 website value assessment data that obtain optimum are confirmed model.
Usually; The characteristic that makes up through training dataset concerns; Be resulting BR value computing formula, have right excessively (also claiming over-fitting) problem to a certain degree, that is to say; The website value assessment data model that training dataset trains out can only embody training data and concentrate obvious relation between the characteristic, but may not necessarily represent all user network behavioral datas.
The verification msg collection is the one group of data that is independent of training dataset, but the verification msg collection is obeyed and the same probability distribution of training dataset.Come the initial website value assessment data model of gained is carried out verification through the verification msg collection; If this assessment models well match training dataset also can well match verification msg collection; Then the over-fitting phenomenon is just not obvious; The website value assessment data of gained confirm that model can be used as the website value assessment data that are used to assess the website commercial value and confirms model, and website value assessment data confirm that model has strong robustness and the high model accuracy under real complex situations.
If these website value assessment data confirm that model can only the match training dataset and can not well match verification msg collection; Then the over-fitting phenomenon produces; The website value assessment data of gained confirm that the robustness of model and model accuracy are lower; Then need rebuild training dataset and verification msg collection, then return step 3301 and operate, through carrying out the operation of step 3301, step 3302 and step 3303 repeatedly; Assess the website value assessment data of website commercial value and confirm model to obtain best being used to; If there is not an over-fitting, then the definite website value assessment data of institute in the step 3302 are confirmed that model confirms model as the website value assessment data of the value that is used to estimate the website, in the entering step 340.
Need to prove; The example of above-mentioned network commercial appraisal Model is merely a simple example; Described content only is the embodiment that adopts for the ease of understanding the present invention; Be not that the characteristic that influences evaluation of estimate is not limited to user characteristics, traffic characteristic and marketing characteristic, can also comprise further feature in order to qualification the present invention; And each characteristic factor that influences each characteristic is not limited to the above-mentioned characteristic factor of mentioning, and can also comprise the occurrence rate, website speed, surfing of search engine or by characteristic factors such as link.
Further, this website value assessment method can also comprise:
Particularly; The user network behavioral data that will carry out the website of commercial value evaluation inputs to the website value assessment data of step 330 gained and confirms in the model; To obtain the BR value of this website; Wherein, user's the network behavior data that carry out the website that commercial value estimates are relevant with the characteristic factor of website value assessment data model, and user's network behavior data are based on, and the factors such as time, region and crowd of network behavior generation obtain.
In the present embodiment; The commercial value of website is a value that changes along with the variation of factors such as time, region and crowd, and for example, the commercial value of the different time sections of same website in a week is different; The commercial value of Monday possibly be minimum; And weekend, especially evening Saturday this website commercial value possibly reach maximum, apparently higher than the commercial value of section At All Other Times; Therefore, the advertiser can select evening Saturday carry out the input of advertisement on the website that will throw in.
In sum; Because the evaluation of estimate of resulting each website is in time, region and the crowd's three dimensions; Each website can find according to the evaluation of estimate of himself website with its customer group that is complementary of the content of the intelligent advertisement that will throw in; And the advertiser also can through evaluation of estimate come correct decisions its advertisement that will throw in should on which website, throw in, improve advertiser's advertisement delivery effect.
In addition, in the step 340 of present embodiment, can also further carry out following content.
Said evaluation of estimate is divided into a plurality of intervals, confirms the commercial value rank of this website according to the residing interval of the evaluation of estimate of website.
Particularly; Evaluation of estimate based on each website is carried out classification to confirm the commercial value rank of each website to each website; For example, can the classification of 1-10 level be carried out in the website, will be in all interval website branches of a certain evaluation of estimate in same rank through delimiting the evaluation of estimate interval; Present embodiment can be made as minimum commercial value rank with 1 grade, and 10 grades are made as the highest commercial value rank.
Based on the commercial value rank of resulting website, the advertiser can carry out advertisement putting according to the higher website of rank in the commercial value rank of time, region and crowd's different choice website.
In the commercial value appraisal procedure of the website of present embodiment, through the evaluation of estimate based on the website is carried out classification to each website, make the advertiser can select to be applicable to the higher website of rank of advertisement delivery more intuitively, improve advertiser's advertisement delivery effect.
The 4th embodiment
Fig. 6 illustrates the structural representation according to the website commercial value apparatus for evaluating of fourth embodiment of the invention.The each several part composition of present embodiment is described with reference to figure 6 below.
As shown in Figure 6, this device comprises: acquisition module 61, MBM 62, parameter determination module 63 and evaluation module 64.Wherein, acquisition module 61 is connected with MBM 62, and MBM 62 is connected with parameter determination module 63, and parameter determination module 63 is connected with evaluation module 64, below the function of each module of explanation.
Further, in MBM 62, the website value assessment data model of being formulated is expressed as BR=f (xUi; YFi; ZMi), wherein, said Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic; F representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter;
The annual income at the operation information packet purse rope station that is obtained, with the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website.
In addition; Parameter determination module 63; It utilizes user network behavioral data and the operation information data obtained to come value assessment data model in website is carried out the data training; To confirm the model parameter of website value assessment data model, the website value assessment data that obtain being used to estimating the website commercial value are confirmed model.
Wherein, parameter determination module 63 can comprise: training submodule 631 and checking submodule 632.
Checking submodule 632; Be used for coming value assessment data in verification website to confirm whether model has caused over-fitting based on the verification msg collection; If check results is for being the website that the then definite again first website collection and the second website collection are comprised, return parameters determination module 63; Wherein, the first website collection and the second website collection are not identity sets.
Again further; This system can also comprise: evaluation module 64; Its user network behavioral data that will carry out the website of commercial value evaluation is imported resulting website value assessment data and is confirmed in the model in the hope of the BR value, as the evaluation of estimate of the website commercial value that is used to estimate the website that will carry out the commercial value evaluation.
At embodiment of the invention device; The evaluation of estimate that is in different time, different region and different crowd of each website that obtains according to constructed website value assessment data model; Each website can find according to the evaluation of estimate of himself website with its customer group that is complementary of the content of the intelligent advertisement that will throw in; And the advertiser also can through evaluation of estimate come correct decisions its advertisement that will throw in should on which website, throw in, improve advertiser's advertisement delivery effect.
Those skilled in the art should be understood that; Above-mentioned each module of the present invention or each step can realize that they can concentrate on the single calculation element with the general calculation device, perhaps are distributed on the network that a plurality of calculation element forms; Alternatively; They can realize with the executable program code of calculation element, thereby, can they be stored in the memory storage and carry out by calculation element; Perhaps they are made into each integrated circuit modules respectively, perhaps a plurality of modules in them or step are made into the single integrated circuit module and realize.Like this, the present invention is not restricted to any specific hardware and software combination.
Though the embodiment that the present invention disclosed as above, the embodiment that described content just adopts for the ease of understanding the present invention is not in order to limit the present invention.Technician under any the present invention in the technical field; Under the prerequisite of spirit that does not break away from the present invention and disclosed and scope; Can do any modification and variation what implement in form and on the details; But scope of patent protection of the present invention still must be as the criterion with the scope that appending claims was defined.
Claims (16)
1. a website value assessment method is characterized in that, comprising:
Obtaining step obtains the user network behavioral data and the operation information data of a plurality of websites;
Modeling procedure, confirming influences the characteristic factor that the website is worth, and formulates website value assessment data model according to said characteristic factor; And
Parameter is confirmed step; Utilize user network behavioral data and the operation information data obtained to come said website value assessment data model is carried out the data training; To confirm the model parameter of said website value assessment data model, the website value assessment data that obtain being used to estimating the website commercial value are confirmed model.
2. method according to claim 1 is characterized in that, said parameter confirms that step specifically comprises:
The said user network behavioral data relevant with said characteristic factor of said operation information data and the said first website collection of first website being concentrated the website is as training dataset; And the said user network behavioral data relevant with said characteristic factor of said operation information data and the said second website collection of second website being concentrated the website be as the verification msg collection, and
Come the said website value assessment of verification data to confirm whether model has caused over-fitting based on the verification msg collection, for being, then confirm again to return said parameter and confirm step in the website that said first website collection and the said second website collection are comprised as if check results, wherein,
Said first website collection and the said second website collection are not identity sets.
3. method according to claim 1 and 2 is characterized in that, in said modeling procedure,
The website value assessment data model of being formulated be expressed as BR=f (xUi, yFi, zMi); Wherein, said BR is a commercial level, and Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic; F representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter;
The annual income at the operation information packet purse rope station of said website, with the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website.
4. method according to claim 3 is characterized in that, also comprises:
Evaluation procedure; The user network behavioral data that will carry out the website of commercial value evaluation import resulting website value assessment data confirm in the model in the hope of the BR value, as being used to estimate the said evaluation of estimate that will carry out the website commercial value of the website that commercial value estimates.
5. method according to claim 4 wherein is characterised in that, said evaluation procedure also comprises:
Said evaluation of estimate is divided into a plurality of intervals, confirms the commercial value rank of this website according to the residing interval of the evaluation of estimate of website.
6. method according to claim 1 and 2 is characterized in that,
Said characteristic factor comprises user characteristics, traffic characteristic and marketing characteristic, and
The characterization factor that constitutes said user characteristics comprises user group, pageview and residence time per capita per capita; The characterization factor that constitutes said traffic characteristic comprises page access amount, IP amount and independent user sessions, and the characterization factor that constitutes said marketing characteristic comprises clicking rate, conversion ratio and rate of return on investment.
7. method according to claim 1 and 2 is characterized in that,
Said user network behavioral data is setting-up time section, setting region and/or the crowd's of setting a user network behavioral data.
8. a website commercial value apparatus for evaluating is characterized in that, comprising:
Acquisition module (61), it obtains the user network behavioral data and the operation information data of a plurality of websites;
MBM (62), it confirms to influence the characteristic factor that the website is worth, and formulates website value assessment data model according to said characteristic factor; And
Parameter determination module (63); It utilizes user network behavioral data and the operation information data obtained to come said website value assessment data model is carried out the data training; To confirm the model parameter of said website value assessment data model, the website value assessment data that obtain being used to estimating the website commercial value are confirmed model.
9. device according to claim 8 is characterized in that, said parameter determination module (63) comprising:
The training submodule; The said user network behavioral data relevant with said characteristic factor of said operation information data and the said first website collection that is used for the website is concentrated in first website is as training dataset; And the said user network behavioral data relevant with said characteristic factor of said operation information data and the said second website collection of second website being concentrated the website be as the verification msg collection, and
The checking submodule is used for coming the said website value assessment of verification data to confirm whether model has caused over-fitting based on the verification msg collection, if check results is for being; Then confirm the website that said first website collection and the said second website collection are comprised again; Return said parameter determination module (63), wherein
Said first website collection and the said second website collection are not identity sets.
10. according to Claim 8 or 9 described devices, it is characterized in that,
The website value assessment data model of being formulated be expressed as BR=f (xUi, yFi, zMi); Wherein, Said Ui is that user characteristics, the Fi of website is that traffic characteristic, Mi are the marketing characteristic, and f representes repeatedly curve fitting algorithm or neural network algorithm, and x, y and z are model parameter;
The annual income at the operation information packet purse rope station that is obtained, with the annual income of said website as BR value corresponding in the website value assessment data model of being formulated with this website.
11. device according to claim 10 is characterized in that, also comprises:
Evaluation module (64); Its user network behavioral data that will carry out the website of commercial value evaluation is imported resulting website value assessment data and is confirmed in the model in the hope of the BR value, as being used to estimate the said evaluation of estimate that will carry out the website commercial value of the website that commercial value estimates.
12. a website commercial value appraisal procedure is characterized in that: through following steps the website commercial value is assessed:
(1) the user browsing behavior data is collected and stored; Collect the operation and the financial information of website from open approach;
(2) confirm to influence the characteristic factor that the website is worth, assessment models is formulated;
(3) confirm model parameter, the value assessment data model sets up a web site;
(4) website value assessment data model being used to assess the website is worth;
(5) through the verification msg set, model is optimized,, confirms final mask with the robustness and the model accuracy of raising system under real complex situations;
(6) the user behavior parameter is applied mechanically in the final mask, estimate its website commercial value.
13. website according to claim 12 commercial value appraisal procedure; It is characterized in that: said model parameter is calculated through following formula and is obtained BR=f (xUi, yFi; ZMi), said Ui is that user characteristics, the Fi of website are that traffic characteristic, Mi are the marketing characteristic.
14. website commercial value evaluating system; It is characterized in that: comprise being used for business intelligence Information Collection System and network data acquisition system that the user browsing behavior data are collected and stored; Said two systems give the data training system of the value assessment data model that is used to set up a web site with data transmission, and said data training system also receives the model evaluation system that is used for verification msg simultaneously.
15. website according to claim 14 commercial value evaluating system; It is characterized in that: said data training system comprises model bank; Said model bank is provided with and is used for the Model Selection module, and said Model Selection module is exported selected model, and said selected model receives the data of business intelligence Information Collection System and network data acquisition system; Said selected model passes through parameter calculating module; Form and confirm module output, said affirmation module gets into the verification module and is optimized, and said verification module receives the model evaluation system data.
16. website according to claim 14 commercial value evaluating system is characterized in that: said model evaluation system comprises business intelligence Information Collection System and network data acquisition system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011104213450A CN102567477A (en) | 2011-06-16 | 2011-12-15 | Website value evaluation method and device |
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110162088 | 2011-06-16 | ||
| CN201110162088.3 | 2011-06-16 | ||
| CN2011104213450A CN102567477A (en) | 2011-06-16 | 2011-12-15 | Website value evaluation method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102567477A true CN102567477A (en) | 2012-07-11 |
Family
ID=46412886
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011104213450A Pending CN102567477A (en) | 2011-06-16 | 2011-12-15 | Website value evaluation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102567477A (en) |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103906027A (en) * | 2014-03-28 | 2014-07-02 | 中国联合网络通信集团有限公司 | User value evaluation method and system based on mobile user internet surfing behaviors |
| CN104050595A (en) * | 2014-06-27 | 2014-09-17 | 北京奇虎科技有限公司 | Method and device for selecting flow supply according to quality of flow suppliers |
| CN104133837A (en) * | 2014-06-24 | 2014-11-05 | 上海交通大学 | Internet information putting channel optimizing system based on distributed computing |
| CN104794154A (en) * | 2015-03-11 | 2015-07-22 | 南通天呈医流互联网技术有限公司 | O2O service quality evaluation model for medical apparatus based on text mining |
| CN105574741A (en) * | 2015-12-22 | 2016-05-11 | 北京奇虎科技有限公司 | Flow value assessment method, device, application method and system |
| CN105630957A (en) * | 2015-12-24 | 2016-06-01 | 北京大学 | User management application behavior-based application quality distinguishing method and system |
| CN106170815A (en) * | 2015-08-21 | 2016-11-30 | 广州博鳌纵横网络科技有限公司 | A kind of brand value appraisal procedure and system |
| CN106296388A (en) * | 2015-06-12 | 2017-01-04 | 阿里巴巴集团控股有限公司 | A kind of data analysing method for banking system and banking system |
| CN106355432A (en) * | 2016-08-19 | 2017-01-25 | 焦点科技股份有限公司 | Method for monitoring effectiveness of television advertisements |
| CN106713929A (en) * | 2017-02-16 | 2017-05-24 | 清华大学深圳研究生院 | Video interframe prediction enhancement method based on deep neural network |
| CN103647665B (en) * | 2013-12-13 | 2017-07-14 | 北京启明星辰信息技术股份有限公司 | Network traffics tracing analysis method and apparatus |
| CN107038377A (en) * | 2016-02-03 | 2017-08-11 | 阿里巴巴集团控股有限公司 | A kind of site certificate method and device, website credit method and device |
| CN107145508A (en) * | 2017-03-23 | 2017-09-08 | 广州要啦网络有限公司 | Website data processing method, device and system |
| CN107808333A (en) * | 2016-09-08 | 2018-03-16 | 阿里巴巴集团控股有限公司 | A kind of commodity launch decision system, method and device |
| CN107886357A (en) * | 2017-11-06 | 2018-04-06 | 北京希格斯科技发展有限公司 | The method and system of content value is judged based on user behavior data |
| CN108205770A (en) * | 2016-12-20 | 2018-06-26 | 北京神州泰岳智能数据技术有限公司 | The acquisition methods and device of a kind of target medium |
| CN108768774A (en) * | 2018-05-29 | 2018-11-06 | 北京广成同泰科技有限公司 | A kind of network safety evaluation method and assessment system of quantification |
| CN109039737A (en) * | 2018-07-31 | 2018-12-18 | 深圳市诺信连接科技有限责任公司 | A kind of Website operation management system |
| CN109670976A (en) * | 2018-12-18 | 2019-04-23 | 泰康保险集团股份有限公司 | Characterization factor determines method and apparatus |
| CN110598076A (en) * | 2019-08-27 | 2019-12-20 | 江苏启承知识产权代理有限公司 | Big data grabbing type intellectual property online retrieval and transaction system |
| CN111027833A (en) * | 2019-11-29 | 2020-04-17 | 珠海随变科技有限公司 | Commodity conversion index calculation method, commodity conversion index calculation device, commodity conversion index calculation equipment and storage medium |
| CN111192602A (en) * | 2019-12-03 | 2020-05-22 | 广州荔支网络技术有限公司 | White noise audio content value evaluation method based on audio content portrait system |
-
2011
- 2011-12-15 CN CN2011104213450A patent/CN102567477A/en active Pending
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103647665B (en) * | 2013-12-13 | 2017-07-14 | 北京启明星辰信息技术股份有限公司 | Network traffics tracing analysis method and apparatus |
| CN103906027A (en) * | 2014-03-28 | 2014-07-02 | 中国联合网络通信集团有限公司 | User value evaluation method and system based on mobile user internet surfing behaviors |
| CN104133837A (en) * | 2014-06-24 | 2014-11-05 | 上海交通大学 | Internet information putting channel optimizing system based on distributed computing |
| CN104133837B (en) * | 2014-06-24 | 2017-10-31 | 上海交通大学 | A kind of internet information based on Distributed Calculation delivers channel optimization systems |
| CN104050595B (en) * | 2014-06-27 | 2017-12-15 | 北京奇虎科技有限公司 | A kind of apparatus and method for selecting flow to supply according to the quality of flow supplier |
| CN104050595A (en) * | 2014-06-27 | 2014-09-17 | 北京奇虎科技有限公司 | Method and device for selecting flow supply according to quality of flow suppliers |
| CN104794154A (en) * | 2015-03-11 | 2015-07-22 | 南通天呈医流互联网技术有限公司 | O2O service quality evaluation model for medical apparatus based on text mining |
| CN106296388A (en) * | 2015-06-12 | 2017-01-04 | 阿里巴巴集团控股有限公司 | A kind of data analysing method for banking system and banking system |
| CN106170815A (en) * | 2015-08-21 | 2016-11-30 | 广州博鳌纵横网络科技有限公司 | A kind of brand value appraisal procedure and system |
| CN105574741A (en) * | 2015-12-22 | 2016-05-11 | 北京奇虎科技有限公司 | Flow value assessment method, device, application method and system |
| CN105630957A (en) * | 2015-12-24 | 2016-06-01 | 北京大学 | User management application behavior-based application quality distinguishing method and system |
| WO2017107571A1 (en) * | 2015-12-24 | 2017-06-29 | 北京大学 | Method and system for determining quality of application on basis of user behaviors of application management |
| CN105630957B (en) * | 2015-12-24 | 2019-05-21 | 北京大学 | A kind of application quality method of discrimination and system based on subscriber management application behavior |
| CN107038377A (en) * | 2016-02-03 | 2017-08-11 | 阿里巴巴集团控股有限公司 | A kind of site certificate method and device, website credit method and device |
| CN107038377B (en) * | 2016-02-03 | 2021-04-27 | 创新先进技术有限公司 | Website authentication method and device and website credit granting method and device |
| CN106355432A (en) * | 2016-08-19 | 2017-01-25 | 焦点科技股份有限公司 | Method for monitoring effectiveness of television advertisements |
| CN107808333A (en) * | 2016-09-08 | 2018-03-16 | 阿里巴巴集团控股有限公司 | A kind of commodity launch decision system, method and device |
| CN108205770A (en) * | 2016-12-20 | 2018-06-26 | 北京神州泰岳智能数据技术有限公司 | The acquisition methods and device of a kind of target medium |
| CN106713929A (en) * | 2017-02-16 | 2017-05-24 | 清华大学深圳研究生院 | Video interframe prediction enhancement method based on deep neural network |
| CN106713929B (en) * | 2017-02-16 | 2019-06-28 | 清华大学深圳研究生院 | A kind of video inter-prediction Enhancement Method based on deep neural network |
| CN107145508B (en) * | 2017-03-23 | 2020-06-05 | 广州要啦网络有限公司 | Website data processing method, device and system |
| CN107145508A (en) * | 2017-03-23 | 2017-09-08 | 广州要啦网络有限公司 | Website data processing method, device and system |
| CN107886357A (en) * | 2017-11-06 | 2018-04-06 | 北京希格斯科技发展有限公司 | The method and system of content value is judged based on user behavior data |
| CN108768774A (en) * | 2018-05-29 | 2018-11-06 | 北京广成同泰科技有限公司 | A kind of network safety evaluation method and assessment system of quantification |
| CN108768774B (en) * | 2018-05-29 | 2020-09-29 | 北京广成同泰科技有限公司 | Quantitative network security evaluation method and system |
| CN109039737A (en) * | 2018-07-31 | 2018-12-18 | 深圳市诺信连接科技有限责任公司 | A kind of Website operation management system |
| CN109670976B (en) * | 2018-12-18 | 2021-02-26 | 泰康保险集团股份有限公司 | Feature factor determination method and device |
| CN109670976A (en) * | 2018-12-18 | 2019-04-23 | 泰康保险集团股份有限公司 | Characterization factor determines method and apparatus |
| CN110598076A (en) * | 2019-08-27 | 2019-12-20 | 江苏启承知识产权代理有限公司 | Big data grabbing type intellectual property online retrieval and transaction system |
| CN111027833A (en) * | 2019-11-29 | 2020-04-17 | 珠海随变科技有限公司 | Commodity conversion index calculation method, commodity conversion index calculation device, commodity conversion index calculation equipment and storage medium |
| CN111027833B (en) * | 2019-11-29 | 2020-11-10 | 珠海随变科技有限公司 | Commodity conversion index calculation method, commodity conversion index calculation device, commodity conversion index calculation equipment and storage medium |
| CN111192602A (en) * | 2019-12-03 | 2020-05-22 | 广州荔支网络技术有限公司 | White noise audio content value evaluation method based on audio content portrait system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102567477A (en) | Website value evaluation method and device | |
| Bekavac et al. | Web analytics tools and web metrics tools: An overview and comparative analysis | |
| Müller et al. | The adoption of photovoltaic systems in Wiesbaden, Germany | |
| Lu et al. | BizSeeker: a hybrid semantic recommendation system for personalized government‐to‐business e‐services | |
| US8478746B2 (en) | Operationalizing search engine optimization | |
| JP5596152B2 (en) | Information matching method and system on electronic commerce website | |
| Shay et al. | Household travel decision chains: Residential environment, automobile ownership, trips and mode choice | |
| Grilli et al. | A multi-criteria framework to assess the sustainability of renewable energy development in the Alps | |
| Pant et al. | Visibility of corporate websites: The role of information prosociality | |
| US9020922B2 (en) | Search engine optimization at scale | |
| CN104252679A (en) | Construction method and system of brand advertisement evaluation system | |
| CN102110265A (en) | Network advertisement effect estimating method and network advertisement effect estimating system | |
| Posel et al. | The relationship between sex ratios and marriage rates in South Africa | |
| CN105654361A (en) | Method and system for assessing credit based on community O2O | |
| Liu et al. | A meta-analysis of contingent valuation studies in coastal and near-shore marine ecosystems | |
| Gradus et al. | The effects of market concentration on costs of local public services: empirical evidence from Dutch waste collection | |
| Bilbao et al. | Evaluation of the profitability of quality labels in rural tourism accommodation: a hedonic approach using propensity score matching | |
| Weitzel et al. | Evaluating quality of health information sources | |
| Ekanayake et al. | Trade and economic growth: Does the sophistication of traded goods matter? | |
| US20160342699A1 (en) | Systems, methods, and devices for profiling audience populations of websites | |
| Ma et al. | Does better Internet access lead to more adoption? A new empirical study using household relocation | |
| Gould et al. | Multidimensional connectivity: Why the interplay of international connections matters for knowledge transfers | |
| Guo et al. | Global network centrality of university rankings | |
| Ashley et al. | On the Granger causality between median inflation and price dispersion | |
| Nepal et al. | Public values on offshore wind farm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120711 |