The method, apparatus and server of device-fingerprint are determined according to similarity
Technical field
The present invention relates to field of computer technology, and in particular to a kind of method, dress that device-fingerprint is determined according to similarity
Put, server and computer-readable storage medium.
Background technology
Many websites are intended to carry out user behavior analysis according to the corresponding information of user now, with reach anti-fraud or
The purpose of precision marketing, especially some banks for being related to transaction or electric business etc., it is thus necessary to determine that whether user's logging device becomes
Change.Specifically, can be solved by device-fingerprint technology, wherein, device-fingerprint refers to a variety of attribute informations based on equipment
Not the repeating of generation, unique device identification, are equipment at Virtual Space " identity card ".
And the core of device-fingerprint technology is exactly the judgement of equipment similarity, in the prior art using following two methods come
Determine equipment similarity:
The first, Delphi methods (expert point rating method):This is a kind of qualitative description quantitative method, and it is first according to evaluation
Several assessment items are selected in the specific requirement of object, and evaluation criterion is worked up further according to assessment item, engage some representativenesses
Expert is provided the evaluation score value of projects by the experience of oneself by this evaluation criterion, and then it is concentrated.What it was commonly used
A kind of score method is addition evaluation type.The score value addition evaluated obtained by each index subjet is summed, represents to evaluate by total score
As a result, and it is usually used in the simple person of relation between index.
Although this method is simple, directly perceived, method itself relies on familiar and degree of understanding of the expert for index,
For the service of ground zero device-fingerprint, typically together decided on by business and technical specialist, but for device-fingerprint, it is complete
Quan You expert is not necessarily suitable to score, and takes time and effort.
Second, analytic hierarchy process (AHP):Each index is judged two-by-two, matrix is built, each row is summed, draws spy
It is normalized after sign vector.Ratio is finally adjusted according to consistency check result, if consistency check is by it is determined that power
Weight.But the comparative result two-by-two in the matrix originally built is still artificially to be judged by experience.
The all excessive experience dependent on people of above two method, so as to reduce the accuracy of judgement and preciseness.
The content of the invention
In view of the above problems, it is proposed that the present invention so as to provide one kind overcome above mentioned problem or at least in part solve on
The foundation similarity for stating problem determines the method for device-fingerprint, the device, server and meter of device-fingerprint is determined according to similarity
Calculation machine storage medium.
According to an aspect of the invention, there is provided a kind of method for determining device-fingerprint according to similarity, method include:
Belong to a plurality of device attribute data record of same OS Type, every equipment category in extraction equipment fingerprint base
Property data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form, wherein, each attribute information is by attribute
With attribute corresponding to property value composition;
According to a plurality of device attribute data record, each category is calculated using feature weight computational algorithm or weight model algorithm
First weighted value of property;
Whether the first weighted value of each attribute for judging to be calculated meets the first preparatory condition, if so, then by attribute
Weight assignment be the first weighted value;
The attribute information of equipment to be confirmed is obtained, the category according to corresponding to weight corresponding to the attribute of equipment to be confirmed, attribute
Property value and attributes similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library;
Judge whether each equipment similarity is more than or equal to predetermined threshold value;
If at least one equipment similarity be present is more than or equal to predetermined threshold value, above or equal to predetermined threshold value extremely
The device-fingerprint of equipment similarity highest equipment in a few equipment similarity is assigned to the equipment to be confirmed.
According to another aspect of the present invention, there is provided a kind of device that device-fingerprint is determined according to similarity, device include:
Extraction module, a plurality of device attribute data for belonging to same OS Type in extraction equipment fingerprint base are remembered
Record, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form, wherein, each category
Property information property value corresponding to attribute and attribute forms;
First computing module, for according to a plurality of device attribute data record, utilizing feature weight computational algorithm or weight
Model algorithm calculates the first weighted value of each attribute;
Whether the first judge module, the first weighted value of each attribute for judging to be calculated meet the first default bar
Part;
First assignment module, if the first weighted value for each attribute meets the first preparatory condition, by the power of attribute
Reassignment is the first weighted value;
Acquisition module, for obtaining the attribute information of equipment to be confirmed;
Second computing module, for property value and category corresponding to weight corresponding to the attribute according to equipment to be confirmed, attribute
Property similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library;
Second judge module, for judging whether each equipment similarity is more than or equal to predetermined threshold value;
Second assignment module, if being more than or equal to predetermined threshold value at least one equipment similarity be present, it will be greater than
Or it is assigned to this equal to the device-fingerprint of the equipment similarity highest equipment at least one equipment similarity of predetermined threshold value
Equipment to be confirmed.
According to another aspect of the invention, there is provided a kind of electronic equipment/terminal/server, including:Processor, storage
Device, communication interface and communication bus, the processor, the memory and the communication interface are completed by the communication bus
Mutual communication;
The memory is used to deposit an at least executable instruction, and the executable instruction makes the computing device above-mentioned
Determine to operate corresponding to the method for device-fingerprint according to similarity.
In accordance with a further aspect of the present invention, there is provided a kind of computer-readable storage medium, be stored with the storage medium to
A few executable instruction, the executable instruction make the computing device determine the side of device-fingerprint according to similarity as described above
Operated corresponding to method.
According to scheme provided by the invention, feature weight computational algorithm or weight mould are utilized according to device attribute data record
Type algorithm calculates the first weighted value of each attribute, rather than the weighted value of weight is set according to the experience of people, so as to avoid
Dependence for the experience of people, therefore, is effectively improved the accuracy of equipment Similarity Measure, and then improve judgement and set
It is standby whether be same equipment accuracy;By being verified to the weighted value being calculated, equipment phase can be further improved
The accuracy calculated like degree.In addition, the weighted value of each attribute is calculated using the device attribute information for belonging to same operating system,
Also contribute to reduce amount of calculation, improve the accuracy of equipment Similarity Measure.
Described above is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can
Become apparent, below especially exemplified by the embodiment of the present invention.
Brief description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this area
Technical staff will be clear understanding.Accompanying drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And in whole accompanying drawing, identical part is denoted by the same reference numerals.In the accompanying drawings:
Fig. 1 shows that a kind of according to embodiments of the present invention one foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to;
Fig. 2 shows that a kind of according to embodiments of the present invention two foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to;
Fig. 3 shows that a kind of according to embodiments of the present invention three foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to;
Fig. 4 shows that a kind of according to embodiments of the present invention four foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to;
Fig. 5 shows that a kind of according to embodiments of the present invention five foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to;
Fig. 6 shows that a kind of according to embodiments of the present invention six foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to;
Fig. 7 shows a kind of structural representation of according to embodiments of the present invention eight server.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in accompanying drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
Limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
Embodiment one
Fig. 1 shows that a kind of according to embodiments of the present invention one foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to.As shown in figure 1, this method comprises the following steps:
Step S100, a plurality of device attribute data record of same OS Type is belonged in extraction equipment fingerprint base.
Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form,
Each attribute information property value corresponding to attribute and attribute forms.
In embodiments of the present invention, the similarity of equipment is directed to for the equipment with identical OS Type
, at present, the OS Type of equipment can substantially be divided into:IOS, Android operation system, Windows behaviour
Make system, Windows Phone operating systems, for every kind of OS Type, be stored with device-fingerprint storehouse and largely set
Multiple attribute informations corresponding to standby fingerprint and device-fingerprint, wherein, multiple category corresponding to a device-fingerprint and the device-fingerprint
Property information is identified as a device attribute data record, and each attribute information property value corresponding to attribute and attribute forms,
, can be to extract a plurality of device attribute number of the OS Type in slave unit fingerprint base according to the difference of OS Type
According to record, for example, for IOS, Android operation system, Windows operating system, Windows Phone operations
System extracts 10000 device attribute data records respectively, is merely illustrative of here, without any restriction effect.
Step S101, according to a plurality of device attribute data record, utilize feature weight computational algorithm or weight model algorithm
Calculate the first weighted value of each attribute.
Belong to extracting after a plurality of device attribute data record of same OS Type, it is necessary to be directed to each category
Property information sets weight, wherein, it is relatively important in equipment Similarity Measure that the weight of each attribute embodies the attribute
Degree, specifically, feature weight computational algorithm or weight model algorithm can be utilized to calculate the first weighted value of each attribute,
In the embodiment of the present invention, it is that each attribute sets weight according to device attribute data record, the experience for people can be avoided
Rely on, so as to make the calculating process of similarity more objective, rigorous, further improve the accuracy of similarity.
Whether step S102, the first weighted value of each attribute for judging to be calculated meet the first preparatory condition, if so,
Then perform step S103.
, it is necessary to judge the first power of each attribute being calculated after the first weighted value of each attribute is calculated
Whether weight values meet the first preparatory condition, judge whether the first weighted value meets the first preparatory condition here, primarily to really
Whether the first weighted value being calculated surely is more preferable.
Step S103, it is the first weighted value by the weight assignment of attribute.
, can be by the weight of attribute in the case where the first weighted value for judging each attribute meets the first preparatory condition
It is entered as the first weighted value.
Step S104, the attribute information of equipment to be confirmed is obtained, according to weight, attribute corresponding to the attribute of equipment to be confirmed
Corresponding property value and attributes similarity algorithm calculates the equipment similarity of the equipment to be confirmed and each equipment in equipment library.
When having detected that equipment is logged in or accessed, the attribute information of the equipment to be confirmed is obtained, then basis is got
Equipment to be confirmed attribute corresponding to weight, property value corresponding to attribute and attributes similarity algorithm to be confirmed set to calculate this
The standby equipment similarity with each equipment in equipment library.
Step S105, judge whether that at least one equipment similarity is more than or equal to predetermined threshold value, if so, then performing
Step S106.
Whether each equipment similarity for judging to be calculated is more than or equal to predetermined threshold value, if at least one equipment be present
Similarity is more than or equal to predetermined threshold value, then shows that the equipment to be confirmed is similar at least one equipment in equipment library, if each
Equipment similarity is respectively less than predetermined threshold value, then shows that any one equipment is all dissimilar in equipment to be confirmed and equipment library, wherein,
Predetermined threshold value can be set according to practical experience, not illustrated here.
Step S106, above or equal to the equipment similarity highest at least one equipment similarity of predetermined threshold value
The device-fingerprint of equipment is assigned to the equipment to be confirmed.
If at least one equipment similarity be present is more than or equal to predetermined threshold value, show the equipment to be confirmed and equipment library
In at least one equipment it is similar, in this way, it can be assumed that more than or equal to equipment at least one equipment similarity of predetermined threshold value
Similarity highest equipment and the equipment to be confirmed are same equipment, can be by the device-fingerprint of equipment similarity highest equipment
It is assigned to the equipment to be confirmed.
The method provided according to the above embodiment of the present invention, calculated and calculated using feature weight according to device attribute data record
Method or weight model algorithm calculate the first weighted value of each attribute, rather than the weight of weight is set according to the experience of people
Value, so as to avoid the dependence of the experience for people, therefore, is effectively improved the accuracy of equipment Similarity Measure, and then
Improve judge equipment whether be same equipment accuracy;By being verified to the weighted value being calculated, one can be entered
Step improves the accuracy of equipment Similarity Measure.In addition, calculated using the device attribute information for belonging to same operating system each
The weighted value of attribute, it helps reduce amount of calculation, improve the accuracy of equipment Similarity Measure.
Embodiment two
Fig. 2 shows that a kind of according to embodiments of the present invention two foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to.As shown in Fig. 2 this method comprises the following steps:
Step S200, a plurality of device attribute data record of same OS Type is belonged in extraction equipment fingerprint base.
Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form,
Each attribute information property value corresponding to attribute and attribute forms.
Specifically, attribute information includes:Hardware attributes information, software attributes information and/or behavior property information;Wherein,
Hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence number;Software attributes
Information includes the one or more in following information:OS types, system setting, network settings, agreement fingerprint, browser attribute,
Geographical position;Behavior property information includes the one or more in following information:Visitation frequency, access time, operation trace.
OS Type includes:IOS, Android operation system, Windows operating system and/or
Windows Phone operating systems.
Step S201, a plurality of device attribute data record is pre-processed, obtain pretreated device attribute data
Record.
In order to more effectively calculate the weighted value of weight and equipment similarity, a plurality of device attribute data are being extracted
After record, it is also necessary to a plurality of device attribute data record is pre-processed, specifically, following methods can be used to a plurality of
Device attribute data record is pre-processed:
Method one:The quantity for deleting attribute information is less than the device attribute data record of predetermined threshold value.
Wherein, each device-fingerprint is corresponding with many attribute informations, for the less device-fingerprint of those attribute informations, then
It can be assumed that to be abnormal device-fingerprint, therefore, it is less than the device-fingerprint of predetermined threshold value for the quantity of attribute information, here
Delete processing can be done, that is, the quantity for deleting attribute information is less than the device attribute data record of predetermined threshold value.
Method two:Device-fingerprint corresponding to a plurality of device attribute data record is entered with the device-fingerprint in blacklist respectively
Row matching;If matching, delete the device attribute data record matched with the device-fingerprint in blacklist.
Specifically, some abnormal device-fingerprints may be recorded in device-fingerprint storehouse, for example, entering using the information stolen
Device-fingerprint corresponding to the equipment that row logs in, needs to delete for this kind of device attribute data record, more specifically, Ke Yishe
A blacklist for being used to store abnormal device-fingerprint is put, after a plurality of device attribute data record is extracted, by a plurality of equipment
Device-fingerprint corresponding to attribute data record is matched with the device-fingerprint in blacklist respectively;If matching, delete with it is black
The device attribute data record of device-fingerprint matching in list.
It is, of course, also possible to delete attribute information quantity be less than predetermined threshold value device attribute data record and then
Judge that device-fingerprint corresponding to remaining device attribute data record is matched with the device-fingerprint in blacklist respectively;If
Match somebody with somebody, then delete the device attribute data record matched with the device-fingerprint in blacklist.
Step S202, according to pretreated device attribute data record, each attribute is calculated using equation below (1)
First weighted value:
Wherein, T represents the set of device attribute data record, | T | represent that T gathers included device attribute data record
Bar number, Values (A) represent attribute A all properties values set, TvEquipment category when the property value for being attribute A in T set is v
The subset of property data record, | Tv| when the property value for representing attribute A is v, TvSet includes the bar of device attribute data record
Number, S (T) represent T entropy, S (Tv) represent TvEntropy;
Wherein, c represents the other number of Attribute class in T set, piRepresent the device attribute data of ith attribute classification in T set
Record accounts for the ratio of device attribute data record total in T set;
Wherein, cvWhen the property value for representing attribute A is v, TvThe other number of Attribute class, p in setviRepresent attribute A property value
For v when, TvThe device attribute data record of ith attribute classification accounts for T in setvTotal device attribute data record in set
Ratio.
So that the first weighted value of each attribute can be calculated using the step.In addition, in the embodiment of the present invention
In, it can be assumed that the classification of attribute corresponding to device-fingerprint is identical, and for the attribute that the device-fingerprint does not have, can
To assert property value corresponding to the attribute as sky, for example, device-fingerprint 1, attribute A- property values:1, attribute B- property values:
1, attribute C- property values:Sky, attribute D- property values:1.
After the first weighted value of each attribute is calculated, it is also necessary to the first weighted value is verified, specifically
Verification method may refer to step S203- steps S207:
Step S203, a plurality of device attribute data record is analyzed, obtains first sample and the second sample.
Wherein, first sample is that same equipment is set in device attribute data record at different moments, the second sample for difference
Standby device attribute data record.
Same equipment can be different in device attribute data record at different moments, for example, being visited in behavior property information
Ask that the property value of the attributes such as the frequency, access time, operation trace can be different.The device attribute data record of distinct device is then
Because it is that distinct device can make a big difference.
It is well known that equipment similarity is higher to show that two equipment are more similar, the embodiment of the present invention is by a plurality of device attribute
Data record is divided into same equipment and remembered in device attribute data record at different moments and the device attribute data of distinct device
Record, it is in order to which the equipment similarity between the equipment similarity and distinct device using same equipment verifies the first weighted value phase
It is whether more preferable compared with initial weight value.
Step S204, the between the first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Two equipment similarities.
Specifically, after the first weighted value is calculated, it is also necessary to same equipment is calculated according to the first weighted value
The second equipment similarity between first equipment similarity and distinct device, the attributes similarity of initial setting up can be used here
Algorithm is calculated.
In a preferred embodiment of the invention, first attribute information can also be screened according to the first weighted value, for example, right
The first weighted value being calculated is ranked up according to order from small to large, is filtered out the first weighted value and is less than default weighted value
Attribute information, the attribute information that the first weighted value is more than or equal to default weighted value is obtained, then, according to the attribute after screening
The first weighted value corresponding to attribute is calculated between the first equipment similarity of same equipment and distinct device in information second sets
Standby similarity, by being screened to the first weighted value, it is possible to reduce participate in the attribute information of equipment Similarity Measure process
Number, so as to accelerate the calculating speed of equipment similarity.
Specifically, can first property value and attributes similarity algorithm computation attribute similarity according to corresponding to attribute, then
The second equipment similarity between the first equipment similarity of same equipment and distinct device is calculated respectively using equation below:
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Step S205, judge that the second equipment similarity between the first equipment similarity of same equipment and distinct device is
The first preparatory condition of no satisfaction, if so, then performing step S206;If it is not, then perform step S207.
After the second equipment similarity being calculated between the first equipment similarity of same equipment and distinct device,
Need to judge whether the second equipment similarity between the first equipment similarity of same equipment and distinct device meets that first is pre-
If condition, wherein, the first preparatory condition is specially:First equipment similarity of same equipment is more than the original equipment of same equipment
Similarity, and the second equipment similarity between distinct device is less than the original equipment similarity between distinct device;It is same to set
Original equipment similarity between standby original equipment similarity and distinct device calculates according to the initial weight value of attribute.
If the second equipment similarity between the first equipment similarity and distinct device of same equipment meets that first is default
Condition, then the equipment similarity for the same equipment that explanation is calculated using the first weighted value of each attribute is higher, shows this
First weighted value is more preferable compared to initial weight value;If second between the first equipment similarity and distinct device of same equipment
Equipment similarity is unsatisfactory for the first preparatory condition, then illustrates the same equipment being calculated using the initial weight value of each attribute
The equipment similarity of same equipment that is calculated higher than the first weighted value of equipment similarity, show initial weight value compared to
First weighted value is more preferable.
Step S206, it is the first weighted value by the weight assignment of attribute.
If the second equipment similarity between the first equipment similarity and distinct device of same equipment meets that first is default
Condition, then can be the first weighted value by the weight assignment of attribute.
Step S207, it is initial weight value by the weight assignment of attribute.
It is pre- that if the second equipment similarity between the first equipment similarity and distinct device of same equipment is unsatisfactory for first
Then can be initial weight value by the weight assignment of attribute if condition.
Step S208, the attribute information of equipment to be confirmed is obtained, determine that equipment to be confirmed participates in each equipment Similarity Measure
Attribute information, and weight corresponding to attribute.
When having detected that equipment is logged in or accessed, the attribute information of the equipment to be confirmed is obtained, for the weight of weight
The attribute that value is less than default weighted value should filter out, and be not involved in equipment Similarity Measure process, in addition, two equipment of calculating
Equipment similarity is that the attribute according to common to two equipment is calculated, accordingly, it is determined that equipment to be confirmed participates in each equipment
The attribute information of Similarity Measure, and weight corresponding to attribute.
Step S209, attribute pair in the attribute information of each equipment Similarity Measure is participated according to identified equipment to be confirmed
Property value and attributes similarity algorithm corresponding to the weight answered, attribute calculate the equipment to be confirmed and each equipment in equipment library
Equipment similarity.
After being determined that equipment to be confirmed participates in weight corresponding to the attribute information and attribute of each equipment Similarity Measure,
It is corresponding according to weight, attribute corresponding to attribute in the attribute information of each equipment Similarity Measure of identified equipment participation to be confirmed
Property value and attributes similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library, wherein,
Attributes similarity algorithm can be initial attributes similarity algorithm.
Specifically, the equipment similarity of equipment to be confirmed and each equipment in equipment library can be calculated with the following method:
Property value and attributes similarity algorithm computation attribute similarity according to corresponding to attribute, then being calculated using equation below (2) should
The equipment similarity of equipment to be confirmed and each equipment in equipment library:
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Step S210, judge whether that at least one equipment similarity is more than or equal to predetermined threshold value, if so, then performing
Step S211;If it is not, then perform step S212.
Whether each equipment similarity for judging to be calculated is more than or equal to predetermined threshold value, if at least one equipment be present
Similarity is more than or equal to predetermined threshold value, then shows that the equipment to be confirmed is similar at least one equipment in equipment library, if each
Equipment similarity is respectively less than predetermined threshold value, then shows that any one equipment is all dissimilar in equipment to be confirmed and equipment library, wherein,
Predetermined threshold value can be set according to practical experience, not illustrated here.
Step S211, above or equal to the equipment similarity highest at least one equipment similarity of predetermined threshold value
The device-fingerprint of equipment is assigned to the equipment to be confirmed.
If at least one equipment similarity be present is more than or equal to predetermined threshold value, show the equipment to be confirmed and equipment library
In at least one equipment it is similar, in this way, it can be assumed that more than or equal to equipment at least one equipment similarity of predetermined threshold value
Similarity highest equipment and the equipment to be confirmed are same equipment, can be by the device-fingerprint of equipment similarity highest equipment
It is assigned to the equipment to be confirmed.
Step S212, according to the attribute information computing device fingerprint of new equipment, and it is assigned to the new equipment.
If each similarity is respectively less than predetermined threshold value, show new equipment and any one equipment in equipment library not phases
Seemingly, in this way, can be according to the attribute information computing device fingerprint of new equipment, and the new equipment is assigned to, wherein, equipment refers to
Line specifically can be made up of following information:Attribute coding (8), timestamp coding (14), check code (2).
The method provided according to the above embodiment of the present invention, calculated and calculated using feature weight according to device attribute data record
Method calculates the weighted value of weight, rather than the weighted value of weight is set according to the experience of people, so as to avoiding for people's
The dependence of experience, therefore, the accuracy of equipment Similarity Measure is effectively improved, and then improves and judge whether equipment is same
The accuracy of one equipment;By being verified to the weighted value being calculated, equipment Similarity Measure can be further improved
Accuracy.In addition, calculate the weighted value of each attribute using the device attribute information for belonging to same operating system, it helps subtract
Few amount of calculation, improve the accuracy of equipment Similarity Measure.
Embodiment three
Fig. 3 shows that a kind of according to embodiments of the present invention three foundation similarity determines that the flow of the method for device-fingerprint is shown
It is intended to.As shown in figure 3, this method comprises the following steps:
Step S300, a plurality of device attribute data record of same OS Type is belonged in extraction equipment fingerprint base.
Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form,
Each attribute information property value corresponding to attribute and attribute forms.
Specifically, attribute information includes:Hardware attributes information, software attributes information and/or behavior property information;Wherein,
Hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence number;Software attributes
Information includes the one or more in following information:OS types, system setting, network settings, agreement fingerprint, browser attribute,
Geographical position;Behavior property information includes the one or more in following information:Visitation frequency, access time, operation trace.
OS Type includes:IOS, Android operation system, Windows operating system and/or
Windows Phone operating systems.
Step S301, a plurality of device attribute data record is pre-processed, obtain pretreated device attribute data
Record.
In order to more effectively calculate the weighted value of weight and equipment similarity, a plurality of device attribute data are being extracted
After record, it is also necessary to a plurality of device attribute data record is pre-processed, specifically, following methods can be used to a plurality of
Device attribute data record is pre-processed:
Method one:The quantity for deleting attribute information is less than the device attribute data record of predetermined threshold value.
Wherein, each device-fingerprint is corresponding with many attribute informations, for the less device-fingerprint of those attribute informations, then
It can be assumed that to be abnormal device-fingerprint, therefore, it is less than the device-fingerprint of predetermined threshold value for the quantity of attribute information, here
Delete processing can be done, that is, the quantity for deleting attribute information is less than the device attribute data record of predetermined threshold value.
Method two:Device-fingerprint corresponding to a plurality of device attribute data record is entered with the device-fingerprint in blacklist respectively
Row matching;If matching, delete the device attribute data record matched with the device-fingerprint in blacklist.
Specifically, some abnormal device-fingerprints may be recorded in device-fingerprint storehouse, for example, entering using the information stolen
Device-fingerprint corresponding to the equipment that row logs in, needs to delete for this kind of device attribute data record, more specifically, Ke Yishe
A blacklist for being used to store abnormal device-fingerprint is put, after a plurality of device attribute data record is extracted, by a plurality of equipment
Device-fingerprint corresponding to attribute data record is matched with the device-fingerprint in blacklist respectively;If matching, delete with it is black
The device attribute data record of device-fingerprint matching in list.
It is, of course, also possible to delete attribute information quantity be less than predetermined threshold value device attribute data record and then
Judge that device-fingerprint corresponding to remaining device attribute data record is matched with the device-fingerprint in blacklist respectively;If
Match somebody with somebody, then delete the device attribute data record matched with the device-fingerprint in blacklist.
It is to calculate the first weighted value using weight model algorithm in the present embodiment, specifically, step can be used
Method described in S302- steps S305:
Step S302, pretreated device attribute data record is analyzed, obtains first sample and the second sample.
Wherein, first sample is that same equipment is set in device attribute data record at different moments, the second sample for difference
Standby device attribute data record.
Same equipment can be different in device attribute data record at different moments, for example, being visited in behavior property information
Ask that the property value of the attributes such as the frequency, access time, operation trace can be different.The device attribute data record of distinct device is then
Because it is that distinct device can make a big difference.
Step S303, using initial attributes similarity algorithm calculate same equipment attributes similarity and distinct device it
Between attributes similarity.
Initial attributes similarity algorithm is rule of thumb set, and attributes similarity algorithm includes:Equal judgement is similar
Spend algorithm, cosine similarity algorithm, most short editing distance similarity algorithm and Longest Common Substring similarity algorithm, this area skill
Art personnel can set initial attributes similarity algorithm as needed, for example, setting initial attributes similarity algorithm as most
Short editing distance similarity algorithm.
Step S304, the average value for calculating the attributes similarity between the attributes similarity and distinct device of same equipment are made
For the attributes similarity of each attribute.
Wherein, the attributes similarity between the attributes similarity and distinct device of the same equipment that step S303 is calculated
It is the average value for seeking multiple attributes similarities here, using the average value of multiple attributes similarities as each attribute to be multiple
Attributes similarity.
Step S305, the attributes similarity of each attribute is input in weight model, obtains the first power of each attribute
Weight values.
Wherein, weight model is that the weighted value according to corresponding to the attributes similarity and attribute of great amount of samples equipment is trained
Obtain, be the model on attributes similarity and weighted value, therefore, after the attributes similarity of each attribute is calculated,
, can be to obtain the first weighted value of each attribute by the way that the attributes similarity of each attribute is input in weight model.
After the first weighted value of each attribute is calculated, it is also necessary to the first weighted value is verified, with checking
Whether the equipment similarity being calculated using the first weighted value is higher, and specific verification method may refer to step S306- steps
Rapid S309:
Step S306, the between the first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Two equipment similarities.
Specifically, after the first weighted value is calculated, it is also necessary to same equipment is calculated according to the first weighted value
The second equipment similarity between first equipment similarity and distinct device, the attributes similarity of initial setting up can be used here
Algorithm is calculated.
In a preferred embodiment of the invention, first attribute information can also be screened according to the first weighted value, for example, right
The first weighted value being calculated is ranked up according to order from small to large, is filtered out the first weighted value and is less than default weighted value
Attribute information, the attribute information that the first weighted value is more than or equal to default weighted value is obtained, then, according to the attribute after screening
The first weighted value corresponding to attribute is calculated between the first equipment similarity of same equipment and distinct device in information second sets
Standby similarity, by being screened to the first weighted value, it is possible to reduce participate in the attribute information of equipment Similarity Measure process
Number, so as to accelerate the calculating speed of equipment similarity.
Specifically, the attributes similarity of computation attribute can be carried out first with attributes similarity algorithm, then using following public
Formula calculates the second equipment similarity between the first equipment similarity of same equipment and distinct device respectively:
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Step S307, judge that the second equipment similarity between the first equipment similarity of same equipment and distinct device is
The first preparatory condition of no satisfaction, if so, then performing step S308;If it is not, then perform step S309.
After the second equipment similarity being calculated between the first equipment similarity of same equipment and distinct device,
Need to judge whether the second equipment similarity between the first equipment similarity of same equipment and distinct device meets that first is pre-
If condition, wherein, the first preparatory condition is specially:First equipment similarity of same equipment is more than the original equipment of same equipment
Similarity, and the second equipment similarity between distinct device is less than the original equipment similarity between distinct device;It is same to set
Original equipment similarity between standby original equipment similarity and distinct device calculates according to the initial weight value of attribute.
If the second equipment similarity between the first equipment similarity and distinct device of same equipment meets that first is default
Condition, then the equipment similarity for the same equipment that explanation is calculated using the first weighted value of each attribute is higher, shows this
First weighted value is more preferable compared to initial weight value;If second between the first equipment similarity and distinct device of same equipment
Equipment similarity is unsatisfactory for the first preparatory condition, then illustrates the same equipment being calculated using the initial weight value of each attribute
The equipment similarity of same equipment that is calculated higher than the first weighted value of equipment similarity, show initial weight value compared to
First weighted value is more preferable.
Step S308, it is the first weighted value by the weight assignment of attribute.
If the second equipment similarity between the first equipment similarity and distinct device of same equipment meets that first is default
Condition, then can be the first weighted value by the weight assignment of attribute.
Step S309, it is initial weight value by the weight assignment of attribute.
It is pre- that if the second equipment similarity between the first equipment similarity and distinct device of same equipment is unsatisfactory for first
Then can be initial weight value by the weight assignment of attribute if condition.
In order to obtain optimal equipment similarity, the embodiment of the present invention can pass through Alternative Attribute similarity algorithm
Mode, the attributes similarity of attribute is recalculated, weighted value is calculated according to rear attributes similarity is recalculated, specifically, can be with
Referring to step S310- steps S312:
Step S310, Alternative Attribute similarity algorithm, same set is recalculated using the attributes similarity algorithm after change
Attributes similarity between standby attributes similarity and distinct device.
It is assumed that initial attributes similarity algorithm is most short editing distance similarity algorithm, can be more for different attributes
Change attributes similarity algorithm, for example, being directed to attribute:CPU, attributes similarity algorithm can be changed to:Equal judgement phase
Like degree algorithm, for attribute:Hostname, attributes similarity algorithm can be changed to:Cosine similarity algorithm, only it is here
For example, do not have any restriction effect.
Step S311, calculate the attributes similarity between the attributes similarity and distinct device that recalculate rear same equipment
Attributes similarity of the average value as each attribute.
Step S312, input into weight model, obtain each using the attributes similarity of each attribute as matching input item
Second weighted value of individual attribute.
Wherein, step S310- steps S312 is similar with step S303- steps S305, repeats no more here.
In order to determine whether the second weighted value is better than the first weighted value, it is also necessary to verify have to the second weighted value
The verification method of body may refer to step S313- steps S315:
Step S313, the 3rd equipment similarity of same equipment is calculated according to the second weighted value of each attribute and difference is set
The 4th equipment similarity between standby.
Step S314, judge that the 4th equipment similarity between the 3rd equipment similarity of same equipment and distinct device is
The second preparatory condition of no satisfaction, if so, then performing step S315;If it is not, then perform step S316.
Second preparatory condition is specially:3rd equipment similarity of same equipment is more than the first equipment of same equipment
Similarity, and the 4th equipment similarity between distinct device is less than the second equipment similarity between distinct device.
Wherein, step S313- steps S314 is similar with step S306- steps S307, repeats no more here.
Step S315, it is the second weighted value by the weight assignment of attribute, and determines the attributes similarity algorithm of each attribute.
The 4th equipment similarity between the 3rd equipment similarity and distinct device of same equipment meets second default article
Part, it is the second weighted value by the weight assignment of attribute, and the attributes similarity algorithm after change is determined to the attribute of each attribute
Similarity algorithm.
Wherein, step S310- steps 315 are optional step.
Step S316, the attribute information of equipment to be confirmed is obtained, determine that equipment to be confirmed participates in each equipment Similarity Measure
Attribute information, and weight corresponding to attribute.
When having detected that equipment is logged in or accessed, the attribute information of the equipment to be confirmed is obtained, for the weight of weight
The attribute that value is less than default weighted value should filter out, and not participate in equipment Similarity Measure process, in addition, calculating two equipment
Equipment similarity be that attribute according to common to two equipment is calculated, accordingly, it is determined that equipment to be confirmed participates in respectively setting
The attribute information of standby Similarity Measure, and weight corresponding to attribute, so as to accelerate calculating process, improve computational efficiency.
Step S317, attribute pair in the attribute information of each equipment Similarity Measure is participated according to identified equipment to be confirmed
Property value and attributes similarity algorithm corresponding to the weight answered, attribute calculate the equipment to be confirmed and each equipment in equipment library
Equipment similarity.
After being determined that equipment to be confirmed participates in weight corresponding to the attribute information and attribute of each equipment Similarity Measure,
It is corresponding according to weight, attribute corresponding to attribute in the attribute information of each equipment Similarity Measure of identified equipment participation to be confirmed
Property value and attributes similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library, wherein,
Attributes similarity algorithm can be the attributes similarity algorithm after initial attributes similarity algorithm or change.
Specifically, the equipment similarity of equipment to be confirmed and each equipment in equipment library can be calculated with the following method:
Property value and attributes similarity algorithm computation attribute similarity according to corresponding to attribute, then being calculated using equation below (2) should
The equipment similarity of equipment to be confirmed and each equipment in equipment library:
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Step S318, judge whether that at least one equipment similarity is more than or equal to predetermined threshold value, if so, then performing
Step S319;If it is not, then perform step S320.
Whether each equipment similarity for judging to be calculated is more than or equal to predetermined threshold value, if at least one equipment be present
Similarity is more than or equal to predetermined threshold value, then shows that the equipment to be confirmed is similar at least one equipment in equipment library, if each
Equipment similarity is respectively less than predetermined threshold value, then shows that any one equipment is all dissimilar in equipment to be confirmed and equipment library, wherein,
Predetermined threshold value can be set according to practical experience, not illustrated here.
Step S319, above or equal to the equipment similarity highest at least one equipment similarity of predetermined threshold value
The device-fingerprint of equipment is assigned to the equipment to be confirmed.
If at least one equipment similarity be present is more than or equal to predetermined threshold value, show the equipment to be confirmed and equipment library
In at least one equipment it is similar, in this way, it can be assumed that more than or equal to equipment at least one equipment similarity of predetermined threshold value
Similarity highest equipment and the equipment to be confirmed are same equipment, can be by the device-fingerprint of equipment similarity highest equipment
It is assigned to the equipment to be confirmed.
Step S320, according to the attribute information computing device fingerprint of new equipment, and it is assigned to the new equipment.
If each similarity is respectively less than predetermined threshold value, show new equipment and any one equipment in equipment library not phases
Seemingly, in this way, can be according to the attribute information computing device fingerprint of new equipment, and the new equipment is assigned to, wherein, equipment refers to
Line specifically can be made up of following information:Attribute coding (8), timestamp coding (14), check code (2).
The method provided according to the above embodiment of the present invention, power is calculated using weight model according to record according to device attribute
The weighted value of weight, rather than the weighted value of weight is set according to the experience of people, so as to avoid the dependence of the experience for people,
Therefore, be effectively improved the accuracy of equipment Similarity Measure, so improve judge equipment whether be same equipment standard
True property;By being verified to the weighted value being calculated, and Alternative Attribute similarity algorithm, it can further improve equipment
The accuracy of Similarity Measure.In addition, calculate the weight of each attribute using the device attribute information for belonging to same operating system
Value, it helps reduce amount of calculation, improve the accuracy of equipment Similarity Measure.
Example IV
Fig. 4 shows that a kind of according to embodiments of the present invention four foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to.As shown in figure 4, the device includes:Extraction module 400, the first computing module 401, the first judge module 402, first are assigned
It is worth module 403, acquisition module 404, the second computing module 405, the second judge module 406 and the second assignment module 407.
Extraction module 400, for belonging to a plurality of device attribute number of same OS Type in extraction equipment fingerprint base
According to record.Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form, often
Individual attribute information property value corresponding to attribute and attribute forms.
First computing module 401, for according to a plurality of device attribute data record, utilizing feature weight computational algorithm or power
Weight model algorithm calculates the first weighted value of each attribute.
First judge module 402, it is pre- whether the first weighted value of each attribute for judging to be calculated meets first
If condition.
First assignment module 403, if the first weighted value for each attribute meets the first preparatory condition, by attribute
Weight assignment is the first weighted value.
Acquisition module 404, for obtaining the attribute information of equipment to be confirmed.
Second computing module 405, for property value corresponding to weight corresponding to the attribute according to equipment to be confirmed, attribute and
Attributes similarity algorithm calculates the equipment similarity of the equipment to be confirmed and each equipment in equipment library.
Second judge module 406, for judging whether each equipment similarity is more than or equal to predetermined threshold value.
Second assignment module 407, will be big if being more than or equal to predetermined threshold value at least one equipment similarity be present
The equipment to be confirmed is assigned in or equal to the device-fingerprint of equipment corresponding to highest equipment similarity in predetermined threshold value.
The device provided according to the above embodiment of the present invention, calculated and calculated using feature weight according to device attribute data record
Method or weight model algorithm calculate the first weighted value of each attribute, rather than the weight of weight is set according to the experience of people
Value, so as to avoid the dependence of the experience for people, therefore, is effectively improved the accuracy of equipment Similarity Measure, and then
Improve judge equipment whether be same equipment accuracy;By being verified to the weighted value being calculated, one can be entered
Step improves the accuracy of equipment Similarity Measure.In addition, calculated using the device attribute information for belonging to same operating system each
The weighted value of attribute, it helps reduce amount of calculation, improve the accuracy of equipment Similarity Measure.
Embodiment five
Fig. 5 shows that a kind of according to embodiments of the present invention five foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to.As shown in figure 5, the device includes:Extraction module 500, pretreatment module 501, the first computing module 502, first judge
Module 503, the first assignment module 504, acquisition module 505, the second computing module 506, the second judge module 507, the second assignment
The computing module 509 of module 508 and the 3rd.
Extraction module 500, for belonging to a plurality of device attribute number of same OS Type in extraction equipment fingerprint base
According to record.
Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form,
Each attribute information property value corresponding to attribute and attribute forms.
Specifically, attribute information includes:Hardware attributes information, software attributes information and/or behavior property information;Wherein,
Hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence number;Software attributes
Information includes the one or more in following information:OS types, system setting, network settings, agreement fingerprint, browser attribute,
Geographical position;Behavior property information includes the one or more in following information:Visitation frequency, access time, operation trace.
OS Type includes:IOS, Android operation system, Windows operating system and/or
Windows Phone operating systems.
Pretreatment module 501, for being pre-processed to a plurality of device attribute data record, obtain pretreated equipment
Attribute data records.
In a preferred embodiment of the invention, pretreatment module 501 is further used for:The quantity for deleting attribute information is less than in advance
If the device attribute data record of threshold value.
In addition, pretreatment module 501 is further used for:By device-fingerprint corresponding to a plurality of device attribute data record point
Do not matched with the device-fingerprint in blacklist;If matching, delete the equipment category matched with the device-fingerprint in blacklist
Property data record.
First computing module 502, for according to pretreated device attribute data record, being counted using equation below (1)
Calculate the first weighted value of each attribute:
Wherein, T represents the set of device attribute data record, | T | represent that T gathers included device attribute data record
Bar number, Values (A) represent attribute A all properties values set, TvEquipment category when the property value for being attribute A in T set is v
The subset of property data record, | Tv| when the property value for representing attribute A is v, TvSet includes the bar of device attribute data record
Number, S (T) represent T entropy, S (Tv) represent TvEntropy;
Wherein, c represents the other number of Attribute class in T set, piRepresent the device attribute data of ith attribute classification in T set
Record accounts for the ratio of device attribute data record total in T set;
Wherein, cvWhen the property value for representing attribute A is v, TvThe other number of Attribute class, p in setviRepresent attribute A property value
For v when, TvThe device attribute data record of ith attribute classification accounts for T in setvTotal device attribute data record in set
Ratio.
First judge module 503, for analyzing pretreated device attribute data record, obtain first sample
With the second sample, wherein, first sample be same equipment in device attribute data record at different moments, the second sample is difference
The device attribute data record of equipment;
The second equipment phase between first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Like degree;
In a preferred embodiment of the invention, the first judge module 503 can also enter according to the first weighted value to attribute information
Row screening, obtain the attribute information that the first weighted value is more than or equal to default weighted value;Then, according to the attribute information after screening
First weighted value corresponding to middle attribute calculates the second equipment phase between the first equipment similarity of same equipment and distinct device
Like degree;
Judge whether the second equipment similarity between the first equipment similarity of same equipment and distinct device meets
One preparatory condition.
Wherein, the first preparatory condition is specially:First equipment similarity of same equipment is more than the initial of same equipment and set
Standby similarity, and the second equipment similarity between distinct device is less than the original equipment similarity between distinct device, it is same
Original equipment similarity between the original equipment similarity and distinct device of equipment calculates according to the initial weight value of attribute.
First assignment module 504, if being set for second between the first equipment similarity and distinct device of same equipment
Standby similarity meets the first preparatory condition, then is the first weighted value by the weight assignment of attribute.
Acquisition module 505, for obtaining the attribute information of equipment to be confirmed.
Second computing module 506, for determining that equipment to be confirmed participates in the attribute information of each equipment Similarity Measure, and
Weight corresponding to attribute;
Participated according to identified equipment to be confirmed in the attribute information of each equipment Similarity Measure weight corresponding to attribute,
It is similar to the equipment of each equipment in equipment library that property value corresponding to attribute and attributes similarity algorithm calculate the equipment to be confirmed
Degree.
In a preferred embodiment of the invention, the second computing module 506 is further used for:According to corresponding to attribute property value and
Attributes similarity algorithm computation attribute similarity;
The equipment similarity of the equipment to be confirmed and each equipment in equipment library is calculated using equation below (2):
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Second judge module 507, for judging whether each equipment similarity is more than or equal to predetermined threshold value;
Second assignment module 508, will be big if being more than or equal to predetermined threshold value at least one equipment similarity be present
The equipment to be confirmed is assigned in or equal to the device-fingerprint of equipment corresponding to highest equipment similarity in predetermined threshold value.
3rd computing module 509, if being respectively less than predetermined threshold value for each similarity, the attribute according to equipment to be confirmed
Information calculating device fingerprint, and it is assigned to the equipment to be confirmed.
The device provided according to the above embodiment of the present invention, the weight of weight is calculated using feature weight computational algorithm
It is worth, rather than the weighted value of weight is set according to the experience of people, so as to avoid the dependence of the experience for people, therefore, has
Effect improve the accuracy of equipment Similarity Measure, and then improve judge equipment whether be same equipment accuracy;It is logical
Cross and the weighted value being calculated is verified, can further improve the accuracy of equipment Similarity Measure.In addition, utilize category
The weighted value of each attribute is calculated in the device attribute information of same operating system, it helps reduce amount of calculation, improve equipment
The accuracy of Similarity Measure.
Embodiment six
Fig. 6 shows that a kind of according to embodiments of the present invention six foundation similarity determines that the structure of the device of device-fingerprint is shown
It is intended to.As shown in fig. 6, the device includes:Extraction module 600, pretreatment module 601, the first computing module 602, first judge
Module 603, the first assignment module 604, change module 605, the 4th computing module 606, the 5th computing module 607, input module
608th, the 6th computing module 609, the 3rd judge module 610, determining module 611, acquisition module 612, the second computing module 613,
Second judge module 614, the second assignment module 615 and the 3rd computing module 616.
Extraction module 600, for belonging to a plurality of device attribute number of same OS Type in extraction equipment fingerprint base
According to record.
Wherein, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form,
Each attribute information property value corresponding to attribute and attribute forms.
Specifically, attribute information includes:Hardware attributes information, software attributes information and/or behavior property information;
Wherein, hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence
Row number;Software attributes information includes the one or more in following information:OS types, system setting, network settings, agreement refer to
Line, browser attribute, geographical position;Behavior property information includes the one or more in following information:When visitation frequency, access
Between, operation trace.
OS Type includes:IOS, Android operation system, Windows operating system and/or
Windows Phone operating systems.
Pretreatment module 601, for being pre-processed to a plurality of device attribute data record, obtain pretreated equipment
Attribute data records.
In a preferred embodiment of the invention, pretreatment module 601 is further used for:The quantity for deleting attribute information is less than in advance
If the device attribute data record of threshold value.
In addition, pretreatment module 601 is further used for:By device-fingerprint corresponding to a plurality of device attribute data record point
Do not matched with the device-fingerprint in blacklist;If matching, delete the equipment category matched with the device-fingerprint in blacklist
Property data record.
First computing module 602, for analyzing pretreated device attribute data record, obtain first sample
With the second sample, wherein, first sample be same equipment in device attribute data record at different moments, the second sample is difference
The device attribute data record of equipment;
The attribute between the attributes similarity and distinct device of same equipment is calculated using initial attributes similarity algorithm
Similarity;
The average value of the attributes similarity between the attributes similarity and distinct device of same equipment is calculated as each category
The attributes similarity of property;
The attributes similarity of each attribute is input in weight model, obtains the first weighted value of each attribute, wherein,
Weight model is that the weighted value according to corresponding to the attributes similarity and attribute of great amount of samples equipment is trained what is obtained.
First judge module 603, for analyzing pretreated device attribute data record, obtain first sample
With the second sample, wherein, first sample be same equipment in device attribute data record at different moments, the second sample is difference
The device attribute data record of equipment;
The second equipment phase between first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Like degree;
In a preferred embodiment of the invention, the first judge module 603 can also enter according to the first weighted value to attribute information
Row screening, obtain the attribute information that the first weighted value is more than or equal to default weighted value;Then, according to the attribute information after screening
First weighted value corresponding to middle attribute calculates the second equipment phase between the first equipment similarity of same equipment and distinct device
Like degree;
Judge whether the second equipment similarity between the first equipment similarity of same equipment and distinct device meets
One preparatory condition.
Wherein, the first preparatory condition is specially:First equipment similarity of same equipment is more than the initial of same equipment and set
Standby similarity, and the second equipment similarity between distinct device is less than the original equipment similarity between distinct device, it is same
Original equipment similarity between the original equipment similarity and distinct device of equipment calculates according to the initial weight value of attribute.
First assignment module 604, if being set for second between the first equipment similarity and distinct device of same equipment
Standby similarity meets the first preparatory condition, then is the first weighted value by the weight assignment of attribute.
Module 605 is changed, for Alternative Attribute similarity algorithm.
4th computing module 606, for recalculating the attribute of same equipment using the attributes similarity algorithm after change
Attributes similarity between similarity and distinct device.
5th computing module 607, between attributes similarity and distinct device that rear same equipment is recalculated for calculating
Attributes similarity attributes similarity of the average value as each attribute.
Input module 608, for being inputted the attributes similarity of each attribute as matching input item into weight model,
Obtain the second weighted value of each attribute.
6th computing module 609, for calculating the 3rd equipment phase of same equipment according to the second weighted value of each attribute
Like the 4th equipment similarity between degree and distinct device.
3rd judge module 610, for judging the 4th between the 3rd equipment similarity of same equipment and distinct device
Whether equipment similarity meets the second preparatory condition.
First assignment module 604 is further used for:If between the 3rd equipment similarity and distinct device of same equipment
4th equipment similarity meets the second preparatory condition, then is the second weighted value by the weight assignment of attribute.
Determining module 611, if for the 4th equipment phase between the 3rd equipment similarity and distinct device of same equipment
Meet the second preparatory condition like degree, it is determined that the attributes similarity algorithm of each attribute.
Acquisition module 612, for obtaining the attribute information of equipment to be confirmed.
Second computing module 613, for determining that equipment to be confirmed participates in the attribute information of each equipment Similarity Measure, and
Weight corresponding to attribute.
Participated according to identified equipment to be confirmed in the attribute information of each equipment Similarity Measure weight corresponding to attribute,
It is similar to the equipment of each equipment in equipment library that property value corresponding to attribute and attributes similarity algorithm calculate the equipment to be confirmed
Degree.
In a preferred embodiment of the invention, the second computing module 613 is further used for:According to corresponding to attribute property value and
Attributes similarity algorithm computation attribute similarity;
The equipment similarity of the equipment to be confirmed and each equipment in equipment library is calculated using equation below (2):
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
Second judge module 614, for judging whether each equipment similarity is more than or equal to predetermined threshold value.
Second assignment module 615, will be big if being more than or equal to predetermined threshold value at least one equipment similarity be present
The equipment to be confirmed is assigned in or equal to the device-fingerprint of equipment corresponding to highest equipment similarity in predetermined threshold value.
3rd computing module 616, if being respectively less than predetermined threshold value for each similarity, the attribute according to equipment to be confirmed
Information calculating device fingerprint, and it is assigned to the equipment to be confirmed.
The device provided according to the above embodiment of the present invention, the weighted value of weight is calculated using weight model, rather than
The weighted value of weight is set according to the experience of people, so as to avoid the dependence of the experience for people, therefore, is effectively improved
The accuracy of equipment Similarity Measure, so improve judge equipment whether be same equipment accuracy;By to calculating
To weighted value verified that and Alternative Attribute similarity algorithm can further improve the accurate of equipment Similarity Measure
Property.In addition, calculate the weighted value of each attribute using the device attribute information for belonging to same operating system, it helps reduce meter
Calculation amount, improve the accuracy of equipment Similarity Measure.
Embodiment seven
It provides a kind of nonvolatile computer storage media to the embodiment of the present application, computer-readable storage medium be stored with to
A few executable instruction, the foundation similarity that the computer executable instructions can perform in above-mentioned any means embodiment determine to set
The method of standby fingerprint.
Embodiment eight
Fig. 7 shows a kind of structural representation of according to embodiments of the present invention eight server, the specific embodiment of the invention
The specific implementation to server does not limit.
As shown in fig. 7, the server can include:Processor (processor) 702, communication interface
(Communications Interface) 704, memory (memory) 706 and communication bus 708.
Wherein:
Processor 702, communication interface 704 and memory 706 complete mutual communication by communication bus 708.
Communication interface 704, for being communicated with the network element of miscellaneous equipment such as client or other servers etc..
Processor 702, for configuration processor 710, it can specifically perform the above-mentioned side that device-fingerprint is determined according to similarity
Correlation step in method embodiment.
Specifically, program 710 can include program code, and the program code includes computer-managed instruction.
Processor 702 is probably central processor CPU, or specific integrated circuit ASIC (Application
Specific Integrated Circuit), or it is arranged to implement the integrated electricity of one or more of the embodiment of the present invention
Road.The one or more processors that server includes, can be same type of processor, such as one or more CPU;Can also
It is different types of processor, such as one or more CPU and one or more ASIC.
Memory 706, for depositing the first data acquisition system, the second data acquisition system and program 710.Memory 706 may
Include high-speed RAM memory, it is also possible to also including nonvolatile memory (non-volatile memory), for example, at least one
Individual magnetic disk storage.
Program 710 specifically can be used for causing processor 702 to perform the method in embodiment one to embodiment three.
Algorithm and display be not inherently related to any certain computer, virtual system or miscellaneous equipment provided herein.
Various general-purpose systems can also be used together with teaching based on this.As described above, required by constructing this kind of system
Structure be obvious.In addition, the present invention is not also directed to any certain programmed language.It should be understood that it can utilize various
Programming language realizes the content of invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the specification that this place provides, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice in the case of these no details.In some instances, known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it will be appreciated that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description to the exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention:I.e. required guarantor
The application claims of shield features more more than the feature being expressly recited in each claim.It is more precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following embodiment are expressly incorporated in the embodiment, wherein each claim is in itself
Separate embodiments all as the present invention.
Those skilled in the art, which are appreciated that, to be carried out adaptively to the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.Can be the module or list in embodiment
Member or component be combined into a module or unit or component, and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it can use any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, summary and accompanying drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification (including adjoint power
Profit requires, summary and accompanying drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
Replace.
In addition, it will be appreciated by those of skill in the art that although some embodiments described herein include other embodiments
In included some features rather than further feature, but the combination of the feature of different embodiments means in of the invention
Within the scope of and form different embodiments.For example, in the following claims, embodiment claimed is appointed
One of meaning mode can use in any combination.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of some different elements and being come by means of properly programmed computer real
It is existing.In if the unit claim of equipment for drying is listed, several in these devices can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
The invention discloses:
A1. a kind of method for determining device-fingerprint according to similarity, it is characterised in that methods described includes:
Belong to a plurality of device attribute data record of same OS Type, every equipment category in extraction equipment fingerprint base
Property data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form, wherein, each attribute information is by attribute
With attribute corresponding to property value composition;
According to a plurality of device attribute data record, each category is calculated using feature weight computational algorithm or weight model algorithm
First weighted value of property;
Whether the first weighted value of each attribute for judging to be calculated meets the first preparatory condition, if so, then by attribute
Weight assignment be the first weighted value;
The attribute information of equipment to be confirmed is obtained, the category according to corresponding to weight corresponding to the attribute of equipment to be confirmed, attribute
Property value and attributes similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library;
Judge whether each equipment similarity is more than or equal to predetermined threshold value;
If at least one equipment similarity be present is more than or equal to predetermined threshold value, above or equal to predetermined threshold value extremely
The device-fingerprint of equipment similarity highest equipment in a few equipment similarity is assigned to the equipment to be confirmed.
A2. the method according to A1, it is characterised in that methods described also includes:If each similarity is respectively less than default
Threshold value, then the attribute information computing device fingerprint according to equipment to be confirmed, and be assigned to the equipment to be confirmed.
A3. the method according to A1 or A2, it is characterised in that the first power of each attribute for judging to be calculated
Whether weight values, which meet the first preparatory condition, further comprises:
A plurality of device attribute data record is analyzed, obtains first sample and the second sample, wherein, first sample
This is same equipment in device attribute data record at different moments, and second sample is the device attribute data of distinct device
Record;
The second equipment phase between first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Like degree;
Judge whether the second equipment similarity between the first equipment similarity of the same equipment and distinct device is full
The first preparatory condition of foot;
If so, it is then the first weighted value by the weight assignment of attribute.
A4. the method according to A3, it is characterised in that described to calculate the first of same equipment according to the first weighted value and set
Standby the second equipment similarity between similarity and distinct device further comprises:
Attribute information is screened according to the first weighted value, the first weighted value is obtained and is more than or equal to default weighted value
Attribute information;
The first equipment that the first weighted value according to corresponding to attribute in the attribute information after screening calculates same equipment is similar
The second equipment similarity between degree and distinct device.
A5. the method according to A3 or A4, it is characterised in that first preparatory condition is specially:Same equipment
First equipment similarity is more than the original equipment similarity of same equipment, and the second equipment similarity between distinct device is less than
Original equipment similarity between distinct device;
Wherein, the original equipment similarity between the original equipment similarity of same equipment and distinct device is according to attribute
Initial weight value calculates.
A6. the method according to any one of A3-A5, it is characterised in that it is described according to more datas, utilize feature weight
The first weighted value that computational algorithm calculates each attribute further comprises:
According to a plurality of device attribute data record, the first weighted value of each attribute of equation below (1) calculating is utilized:
Wherein, T represents the set of device attribute data record, | T | represent that T gathers included device attribute data record
Bar number, Values (A) represent attribute A all properties values set, TvEquipment category when the property value for being attribute A in T set is v
The subset of property data record, | Tv| when the property value for representing attribute A is v, TvSet includes the bar of device attribute data record
Number, S (T) represent T entropy, S (Tv) represent TvEntropy;
Wherein, c represents the other number of Attribute class in T set, piRepresent the device attribute data of ith attribute classification in T set
Record accounts for the ratio of device attribute data record total in T set;
Wherein, cvWhen the property value for representing attribute A is v, TvThe other number of Attribute class, p in setviRepresent attribute A property value
For v when, TvThe device attribute data record of ith attribute classification accounts for T in setvTotal device attribute data record in set
Ratio.
A7. the method according to any one of A3-A5, it is characterised in that it is described according to a plurality of device attribute data record,
The first weighted value that each attribute is calculated using weight model algorithm is further comprised:
A plurality of device attribute data record is analyzed, obtains first sample and the second sample, wherein, first sample
This is same equipment in device attribute data record at different moments, and second sample is the device attribute data of distinct device
Record;
The attribute between the attributes similarity and distinct device of same equipment is calculated using initial attributes similarity algorithm
Similarity;
The average value of the attributes similarity between the attributes similarity and distinct device of same equipment is calculated as each category
The attributes similarity of property;
The attributes similarity of each attribute is input in weight model, obtains the first weighted value of each attribute, wherein,
The weight model is that the weighted value according to corresponding to the attributes similarity and attribute of great amount of samples equipment is trained what is obtained.
A8. the method according to A7, it is characterised in that before the attribute information of new equipment is obtained, methods described
Also include:
Alternative Attribute similarity algorithm, the attribute phase of same equipment is recalculated using the attributes similarity algorithm after change
Like the attributes similarity between degree and distinct device;
Calculate the average value of the attributes similarity between the attributes similarity and distinct device that recalculate rear same equipment
Attributes similarity as each attribute;
Inputted the attributes similarity of each attribute as matching input item into weight model, obtain the of each attribute
Two weighted values;
According to the second weighted value of each attribute calculate same equipment the 3rd equipment similarity and distinct device it
Between the 4th equipment similarity;
Judge whether the 4th equipment similarity between the 3rd equipment similarity of the same equipment and distinct device is full
The second preparatory condition of foot;
If so, it is then the second weighted value by the weight assignment of attribute, and determine the attributes similarity algorithm of each attribute.
A9. the method according to A8, it is characterised in that second preparatory condition is specially:The 3rd of same equipment
Equipment similarity is more than the first equipment similarity of same equipment, and the 4th equipment similarity between distinct device is less than difference
The second equipment similarity between equipment.
A10. the method according to any one of A1-A9, it is characterised in that described corresponding according to the attribute of equipment to be confirmed
Weight, property value corresponding to attribute and attributes similarity algorithm calculate the equipment to be confirmed and set with each equipment in equipment library
Standby similarity further comprises:
Determine that equipment to be confirmed participates in the attribute information of each equipment Similarity Measure, and weight corresponding to attribute;
Participated according to identified equipment to be confirmed in the attribute information of each equipment Similarity Measure weight corresponding to attribute,
It is similar to the equipment of each equipment in equipment library that property value corresponding to attribute and attributes similarity algorithm calculate the equipment to be confirmed
Degree.
A11. the method according to A10, it is characterised in that equipment to be confirmed participates in each equipment determined by the basis
Weight corresponding to attribute, property value corresponding to attribute and attributes similarity algorithm calculate this and treated in the attribute information of Similarity Measure
Confirm that the equipment similarity of equipment and each equipment in equipment library further comprises:
Property value and attributes similarity algorithm computation attribute similarity according to corresponding to attribute;
The equipment similarity of the equipment to be confirmed and each equipment in equipment library is calculated using equation below (2):
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
A12. the method according to any one of A1-A11, it is characterised in that according to a plurality of device attribute data record,
Before calculating the first weighted value of each attribute using feature weight computational algorithm or weight model algorithm, methods described is also wrapped
Include:
A plurality of device attribute data record is pre-processed, obtains pretreated device attribute data record;
It is described according to a plurality of device attribute data record, calculated using feature weight computational algorithm or weight model algorithm each
First weighted value of individual attribute further comprises:
According to pretreated device attribute data record, calculated using feature weight computational algorithm or weight model algorithm
First weighted value of each attribute.
A13. the method according to A12, it is characterised in that described to be pre-processed to a plurality of device attribute data record
Further comprise:The quantity for deleting attribute information is less than the device attribute data record of predetermined threshold value.
A14. the method according to A12 or A13, it is characterised in that described to be carried out to a plurality of device attribute data record
Pretreatment further comprises:
Device-fingerprint corresponding to a plurality of device attribute data record is matched with the device-fingerprint in blacklist respectively;
If matching, delete the device attribute data record matched with the device-fingerprint in blacklist.
A15. the method according to any one of A1-A14, it is characterised in that the attribute information includes:Hardware attributes are believed
Breath, software attributes information and/or behavior property information;
Wherein, hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence
Row number;
Software attributes information includes the one or more in following information:OS types, system setting, network settings, agreement
Fingerprint, browser attribute, geographical position;
Behavior property information includes the one or more in following information:Visitation frequency, access time, operation trace.
A16. the method according to any one of A1-A15, it is characterised in that the OS Type includes:IOS is grasped
Make system, Android operation system, Windows operating system and/or Windows Phone operating systems.
B17. a kind of device that device-fingerprint is determined according to similarity, it is characterised in that described device includes:
Extraction module, a plurality of device attribute data for belonging to same OS Type in extraction equipment fingerprint base are remembered
Record, every device attribute data record multiple attribute informations corresponding to device-fingerprint and device-fingerprint form, wherein, each category
Property information property value corresponding to attribute and attribute forms;
First computing module, for according to a plurality of device attribute data record, utilizing feature weight computational algorithm or weight
Model algorithm calculates the first weighted value of each attribute;
Whether the first judge module, the first weighted value of each attribute for judging to be calculated meet the first default bar
Part;
First assignment module, if the first weighted value for each attribute meets the first preparatory condition, by the power of attribute
Reassignment is the first weighted value;
Acquisition module, for obtaining the attribute information of equipment to be confirmed;
Second computing module, for property value and category corresponding to weight corresponding to the attribute according to equipment to be confirmed, attribute
Property similarity algorithm calculate the equipment similarity of each equipment in the equipment to be confirmed and equipment library;
Second judge module, for judging whether each equipment similarity is more than or equal to predetermined threshold value;
Second assignment module, if being more than or equal to predetermined threshold value at least one equipment similarity be present, it will be greater than
Or it is assigned to this equal to the device-fingerprint of the equipment similarity highest equipment at least one equipment similarity of predetermined threshold value
Equipment to be confirmed.
B18. the device according to B17, it is characterised in that described device also includes:3rd computing module, if for each
Individual similarity is respectively less than predetermined threshold value, then the attribute information computing device fingerprint according to equipment to be confirmed, and be assigned to this and treat really
Recognize equipment.
B19. the device according to B17 or B18, it is characterised in that first judge module is further used for:To more
Bar device attribute data record is analyzed, and obtains first sample and the second sample, wherein, the first sample is same equipment
In device attribute data record at different moments, second sample is the device attribute data record of distinct device;
The second equipment phase between first equipment similarity of same equipment and distinct device is calculated according to the first weighted value
Like degree;
Judge whether the second equipment similarity between the first equipment similarity of the same equipment and distinct device is full
The first preparatory condition of foot;
First assignment module is further used for:If between the first equipment similarity and distinct device of the same equipment
Second equipment similarity meets the first preparatory condition, then is the first weighted value by the weight assignment of attribute.
B20. the device according to B19, it is characterised in that first judge module is further used for:According to first
Weighted value screens to attribute information, obtains the attribute information that the first weighted value is more than or equal to default weighted value;
The first equipment that the first weighted value according to corresponding to attribute in the attribute information after screening calculates same equipment is similar
The second equipment similarity between degree and distinct device.
B21. the device according to B19 or B20, it is characterised in that first preparatory condition is specially:Same equipment
The first equipment similarity be more than the original equipment similarity of same equipment, and the second equipment similarity between distinct device is small
Original equipment similarity between distinct device;
Wherein, the original equipment similarity between the original equipment similarity of same equipment and distinct device is according to attribute
Initial weight value calculates.
B22. the device according to any one of B19-B21, it is characterised in that first computing module is further used
In:
According to a plurality of device attribute data record, the first weighted value of each attribute of equation below (1) calculating is utilized:
Wherein, T represents the set of device attribute data record, | T | represent that T gathers included device attribute data record
Bar number, Values (A) represent attribute A all properties values set, TvEquipment category when the property value for being attribute A in T set is v
The subset of property data record, | Tv| when the property value for representing attribute A is v, TvSet includes the bar of device attribute data record
Number, S (T) represent T entropy, S (Tv) represent TvEntropy;
Wherein, c represents the other number of Attribute class in T set, piRepresent the device attribute data of ith attribute classification in T set
Record accounts for the ratio of device attribute data record total in T set;
Wherein, cvWhen the property value for representing attribute A is v, TvThe other number of Attribute class, p in setviRepresent attribute A property value
For v when, TvThe device attribute data record of ith attribute classification accounts for T in setvTotal device attribute data record in set
Ratio.
B23. the device according to any one of B19-B21, it is characterised in that first computing module is further used
In:A plurality of device attribute data record is analyzed, obtains first sample and the second sample, wherein, the first sample is
Same equipment is remembered in device attribute data record at different moments, second sample for the device attribute data of distinct device
Record;
The attribute between the attributes similarity and distinct device of same equipment is calculated using initial attributes similarity algorithm
Similarity;
The average value of the attributes similarity between the attributes similarity and distinct device of same equipment is calculated as each category
The attributes similarity of property;
The attributes similarity of each attribute is input in weight model, obtains the first weighted value of each attribute, wherein,
The weight model is that the weighted value according to corresponding to the attributes similarity and attribute of great amount of samples equipment is trained what is obtained.
B24. the device according to B23, it is characterised in that described device also includes:
Module is changed, for Alternative Attribute similarity algorithm;
4th computing module, the attribute for recalculating same equipment using the attributes similarity algorithm after change are similar
Attributes similarity between degree and distinct device;
5th computing module, for calculating the category between the attributes similarity and distinct device that recalculate rear same equipment
Attributes similarity of the average value of property similarity as each attribute;
Input module, for being inputted the attributes similarity of each attribute as matching input item into weight model, obtain
To the second weighted value of each attribute;
6th computing module, for calculating the 3rd equipment phase of same equipment according to the second weighted value of each attribute
Like the 4th equipment similarity between degree and distinct device;
3rd judge module, for judging the 4th between the 3rd equipment similarity of the same equipment and distinct device
Whether equipment similarity meets the second preparatory condition;
First assignment module is further used for:If between the 3rd equipment similarity and distinct device of the same equipment
4th equipment similarity meets the second preparatory condition, then is the second weighted value by the weight assignment of attribute;
Determining module, if for the 4th equipment phase between the 3rd equipment similarity and distinct device of the same equipment
Meet the second preparatory condition like degree, it is determined that the attributes similarity algorithm of each attribute.
B25. the device according to B24, it is characterised in that second preparatory condition is specially:The of same equipment
Three equipment similarities are more than the first equipment similarity of same equipment, and the 4th equipment similarity between distinct device is less than not
With the second equipment similarity between equipment.
B26. the device according to any one of B17-B25, it is characterised in that second computing module is further used
In:Determine that equipment to be confirmed participates in the attribute information of each equipment Similarity Measure, and weight corresponding to attribute;
Participated according to identified equipment to be confirmed in the attribute information of each equipment Similarity Measure weight corresponding to attribute,
It is similar to the equipment of each equipment in equipment library that property value corresponding to attribute and attributes similarity algorithm calculate the equipment to be confirmed
Degree.
B27. the device according to B26, it is characterised in that second computing module is further used for:According to attribute
Corresponding property value and attributes similarity algorithm computation attribute similarity;
The equipment similarity of the equipment to be confirmed and each equipment in equipment library is calculated using equation below (2):
Wherein, SdEquipment similarity is represented,Represent the attributes similarity of ith attribute classification, WiRepresent ith attribute
The weight of attribute corresponding to classification, N represent to participate in the other number of Attribute class of equipment Similarity Measure.
B28. the device according to any one of B17-B27, it is characterised in that described device also includes:Pretreatment module,
For being pre-processed to a plurality of device attribute data record, pretreated device attribute data record is obtained;
First computing module is further used for:According to pretreated device attribute data record, weighed using feature
Re-computation algorithm or weight model algorithm calculate the first weighted value of each attribute.
B29. the device according to B28, it is characterised in that the pretreatment module is further used for:Delete attribute letter
The quantity of breath is less than the device attribute data record of predetermined threshold value.
B30. the device according to B28 or B29, it is characterised in that the pretreatment module is further used for:Will be a plurality of
Device-fingerprint is matched with the device-fingerprint in blacklist respectively corresponding to device attribute data record;
If matching, delete the device attribute data record matched with the device-fingerprint in blacklist.
B31. the device according to any one of B17-B30, it is characterised in that the attribute information includes:Hardware attributes
Information, software attributes information and/or behavior property information;
Wherein, hardware attributes information includes the one or more in following information:MAC Address, brand, model, IMEI, sequence
Row number;
Software attributes information includes the one or more in following information:OS types, system setting, network settings, agreement
Fingerprint, browser attribute, geographical position;
Behavior property information includes the one or more in following information:Visitation frequency, access time, operation trace.
B32. the device according to any one of B17-B31, it is characterised in that the OS Type includes:IOS is grasped
Make system, Android operation system, Windows operating system and/or Windows Phone operating systems.
C33. a kind of server, it is characterised in that the server includes:Processor, memory, communication interface and communication
Bus, the processor, the memory and the communication interface complete mutual communication by the communication bus;
The memory is used to deposit an at least executable instruction, and the executable instruction makes the computing device such as
Foundation similarity any one of A1-A16 determines to operate corresponding to the method for device-fingerprint.
D34. a kind of computer-readable storage medium, it is characterised in that at least one executable finger is stored with the storage medium
Order, the executable instruction make foundation similarity of the computing device as any one of A1-A16 determine device-fingerprint
Method corresponding to operate.