CN109002507A - A kind of social user's recommended method that big data platform is combined with brand vehicle - Google Patents
A kind of social user's recommended method that big data platform is combined with brand vehicle Download PDFInfo
- Publication number
- CN109002507A CN109002507A CN201810703587.0A CN201810703587A CN109002507A CN 109002507 A CN109002507 A CN 109002507A CN 201810703587 A CN201810703587 A CN 201810703587A CN 109002507 A CN109002507 A CN 109002507A
- Authority
- CN
- China
- Prior art keywords
- user
- social
- big data
- data platform
- comment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
A kind of social user's recommended method that big data platform is combined with brand vehicle, feature is: after detecting that the first social user completes to the concern of a certain brand vehicle, the user property of the first social user of the social reporting of user of detection first and the first comment to the brand vehicle；The second social user for the good friend for being not belonging to the first social user is inquired from the social user set for having focused on the brand vehicle, the user property of the second social user and the user property of the first social activity user match；The second comment to the brand vehicle of the first comment of comparison and the second social reporting of user, if the two is approximate, it is sent to the first social user and includes at least the corresponding social account of the second social activity user, the user property of the second social user and the recommendation information of the second comment, finding the first social user effectively, there are other social users of approximate view to exchange with the brand vehicle of oneself concern, convenient for the public praise of the deep brand vehicle for understanding oneself concern.
Description
Technical field
The present invention relates to the social use that social technical field more particularly to a kind of big data platform are combined with brand vehicle
Family recommended method.
Background technique
Currently, social user can log in car website (such as vapour that internet provides by social account (such as QQ account)
Vehicle shopping guide website), and social user, after logging in the car website that internet provides by social account, social user can
To pay close attention on car website oneself interested a certain brand vehicle (such as BMW X6).
In practice, it has been found that social user occasionally wants to, there are other societies of approximate view with the brand vehicle of oneself concern
User is handed over to exchange, in order to the public praise of the brand vehicle for understanding oneself concern that can be more deep.However, in existing skill
In art, social user is difficult to effectively find other social users' progress with the brand vehicle of oneself concern there are approximate view
Exchange.
Summary of the invention
Social user's recommended method that the embodiment of the invention discloses a kind of big data platforms to combine with brand vehicle, energy
Enough effectively find exchanges with other social activity users that the brand vehicle of oneself concern has approximate view, in order to can be with
The public praise of the more deep brand vehicle for understanding oneself concern.
Wherein, social user's recommended method that a kind of big data platform is combined with brand vehicle, comprising:
The big data platform prompts institute after detecting that the first social user completes to the concern of a certain brand vehicle
The user property of the first social user described in the first social reporting of user and the first social user are stated to the brand vehicle
First comment of type；
The big data platform detect the user property of the described first social user of the described first social reporting of user with
And first comment of the first social activity user to the brand vehicle；
The big data platform is according to the user property of the described first social user, from the society for having focused on the brand vehicle
Hand over the second social user that the good friend for being not belonging to the described first social user is inquired in user's set；Described second social user
The user property of the social user of described second reported and the user property of the described first social user match；
The big data platform obtains the described second social user of the described second social reporting of user to the brand vehicle
Second comment of type；
The big data platform compares first comment and second comment；
If comparison shows that first comment is approximate with second comment, the big data platform is social to described first
User sends recommendation information, and the recommendation information includes at least the second social activity user corresponding social account, described second
The user property of social user and second comment.
As an alternative embodiment, being commented comparing first comment with described second in the embodiment of the present invention
By after approximation and before the big data platform sends recommendation information to the described first social user, the method is also wrapped
It includes:
The big data platform sends inquiry message to the described second social user, and the inquiry message is described for inquiring
Whether the second social user allows to be recommended to other social users；
If the second social user feedback allows to be recommended to other social users, described in the big data platform executes
The step of sending recommendation information to the described first social user.
As an alternative embodiment, in the embodiment of the present invention, the big data platform by first comment with
Second comment compares, comprising:
The big data platform is disassembled first comment to obtain each candidate sentences；
The big data platform determines the importance scores of each candidate sentences；
The big data platform extracts the target sentence that the importance scores are greater than preset value from each candidate sentences
Key message of the son as first comment；
The big data platform carries out the key message of key message that described first comments on and second comment pair
Than.
As an alternative embodiment, in the embodiment of the present invention, the big data platform by described first comment on into
Row dismantling obtains each candidate sentences, comprising:
The big data platform obtains preset text dismantling rule, the preset text dismantling rule include branch,
Comma, fullstop will be disassembled, and pause mark, colon, quotation marks are without dismantling；
The big data platform disassembles rule according to the preset text, and the sight spot first comment is disassembled,
To obtain each candidate sentences.
As an alternative embodiment, the big data platform determines each candidate sentence in the embodiment of the present invention
The importance scores of son, comprising:
It is described big if the candidate sentences are Chinese sentence for each candidate sentences in each candidate sentences
The Chinese sentence is split as several phrases in the way of semantic analysis by data platform；
The big data platform carries out full text traversal to each phrase that fractionation obtains and calculates, and obtains the appearance of each phrase
Number；
The big data platform carries out all phrases that fractionation obtains according to the sequence of the frequency of occurrence from high to low
Sequence, and each phrase assigns corresponding weight according to the frequency of occurrence, and the frequency of occurrence is higher, and the weight is got over
It is high；
The big data platform calculates the importance scores of each Chinese sentence, and the importance scores are the Chinese
The weights sum for several phrases that sentence is split.
As an alternative embodiment, in the embodiment of the present invention, social activity user's recommended method further include:
If the candidate sentences are webpage link address, the big data platform with opening the web page interlinkage from the background
The corresponding target webpage in location；
The big data platform determines the target network according to the link in the target webpage, being directed toward the target webpage
The importance scores of page, the importance scores of the target webpage are the importance scores of the candidate sentences.
As an alternative embodiment, the big data platform determines the target webpage in the embodiment of the present invention
Importance scores process are as follows:
Wherein, S (Vi) be target webpage importance scores, d is damped coefficient, and In (Vi) is to exist to be directed toward target webpage
Link collections of web pages, out (Vj) it is the collections of web pages that the existing link of link in webpage j is directed toward, out (Vj) take absolutely
Value is the number to indicate element in the collections of web pages, S (Vj) be webpage j importance scores.
As an alternative embodiment, in the embodiment of the present invention, what the big data platform was commented on described first
Key message and the key message of second comment compare, comprising:
The big data platform determines the first sentence in the key message of first comment, and determines described second
The second sentence in the key message of comment；Wherein, the phrase to overlap each other between first sentence and second sentence
Quantity it is most；
The big data platform calculates the cosine similarity of first sentence Yu second sentence；
If the cosine similarity is higher than designated value, the big data platform determines that first comment is commented with described second
By approximation.
As an alternative embodiment, in the embodiment of the present invention, the calculating process of the cosine similarity are as follows:
First sentence is split as several phrases by the big data platform, to obtain one group of phrase；
Second sentence is split as several phrases by the big data platform, to obtain another group of phrase；
The big data platform is compared the phrase in two groups of phrases one by one, and if it exists, 1 is then recorded as, if not depositing
It is being recorded as 0, then to obtain First ray and the second sequence；
The big data platform calculates the cosine similarity between the First ray and second sequence, and as institute
State the cosine similarity between the first sentence and second sentence；
Wherein, the big data platform calculates the cosine similarity between the First ray and second sequence are as follows:
Wherein, ab is integrally added after indicating the element in a sequence and corresponding element multiplication in b sequence, and denominator indicates a sequence
The quadratic sum of all elements is opened and opens radical sign multiplied by the quadratic sum of all elements in b sequence after radical sign in column；Wherein, a sequence indicates
The First ray, b sequence indicate second sequence.
As an alternative embodiment, the big data platform detection described first is social in the embodiment of the present invention
The user property of the social user of described the first of reporting of user and the first social user are to the first of the brand vehicle
After comment and the big data platform is according to the user property of the described first social user, from having focused on the brand vehicle
Before the second social user for inquiring the good friend for being not belonging to the described first social user in the social user set of type, the side
Method further include:
The social reporting of user additional searching information of the big data platform prompt described first；
The big data platform detects the additional searching information of the described first social reporting of user；The additional searching
Information includes at least search time section and region of search；
The big data platform obtains the described second social user of the described second social reporting of user to the brand vehicle
After second comment of type, the method also includes:
The big data platform inquire the second comment described in the described second social reporting of user on call time and whether be located at
In the described search period；
It is described if calling time within the described search period in the second comment described in the described second social reporting of user
Whether the current location that big data platform inquires the described second social user is located in described search region；
If the current location of the described second social user is located in described search region, described in the big data platform execution
By it is described first comment with it is described second comment compare the step of.
In the embodiment of the present invention, big data platform is detecting concern of the first social user's completion to a certain brand vehicle
Later, it can detecte the first social user according to the user property for prompting the report first social user and the first social user
The first comment to the brand vehicle；And big data platform can be from the social user set for having focused on the brand vehicle
The second social user for the good friend for being not belonging to the first social user is inquired, the second social user's of the second social reporting of user
User property and the user property of the first social user match；And big data platform can be by the first comment and the second society
The second social user of reporting of user is handed over to compare the second comment of the brand vehicle, if comparison shows the first comment and the
Two comments are approximate, and big data platform is sent to the first social user includes at least the corresponding social account of the second social user, the
The user property of two social users and the recommendation information of the second comment, allow the first social user effectively find with from
Other social users that the brand vehicle of oneself concern has approximate view exchange, in order to can be more deep be known from
The public praise of the brand vehicle of oneself concern.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to needed in the embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for ability
For the those of ordinary skill of domain, without creative efforts, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is a kind of social user recommendation side that big data platform is combined with brand vehicle disclosed by the embodiments of the present invention
The flow diagram of method；
Fig. 2 is the method flow schematic diagram that the first comment disclosed by the embodiments of the present invention is compared with the second comment；
Fig. 3 is that the social user that another big data platform disclosed by the embodiments of the present invention is combined with brand vehicle recommends
The flow diagram of method.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
It should be noted that the term " includes " of the embodiment of the present invention and " having " and their any deformation, it is intended that
Be to cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units not
Those of be necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for these processes, side
The intrinsic other step or units of method, product or equipment.
Social user's recommended method that the embodiment of the invention discloses a kind of big data platforms to combine with brand vehicle, energy
Enough effectively find exchanges with other social activity users that the brand vehicle of oneself concern has approximate view, in order to can be with
The public praise of the more deep brand vehicle for understanding oneself concern.Attached drawing is combined below to be described in detail.
Referring to Fig. 1, Fig. 1 is a kind of social activity that big data platform is combined with brand vehicle disclosed by the embodiments of the present invention
The flow diagram of user's recommended method.As shown in Figure 1, the social user that the big data platform is combined with brand vehicle recommends
Method may comprise steps of:
101, big data platform is after detecting that the first social user completes to the concern of a certain brand vehicle, prompts the
First comment of the user property and the first social user of the one social user of social reporting of user first to the brand vehicle.
In the embodiment of the present invention, the first social user can log in internet by its social account (such as QQ account) and provide
Car website (such as automobile shopping guide website), and the first social user is in the vapour for logging in internet offer by its social account
After vehicle website, the first social user can on car website to oneself interested a certain brand vehicle (such as BMW X6) into
Row concern；Correspondingly, big data platform is detecting that the first social user completes on car website to a certain brand vehicle
The user property that can prompt the first social user of social reporting of user first after concern and the first social user are to the product
First comment of board vehicle.For example, big data platform, which can control the car website, pops up the first interactive interface, first interaction
Interface includes user property for prompting the first social user of social reporting of user first and the first social user to the product
The prompt information of first comment of board vehicle and for the user property for the first social user of social reporting of user first with
And first social user to the reporting port of the first comment of the brand vehicle.
Wherein, brand vehicle is the combination of brand and vehicle, such as when brand is BMW, vehicle is X6, brand vehicle are as follows:
BMW X6.
102, the user property of the first social user of the social reporting of user of big data platform detection first and the first society
Hand over first comment of the user to the brand vehicle.
In the embodiment of the present invention, what the first social user can include according to first interactive interface is used to prompt the first society
Hand over the prompt of the user property and the first social user of the social user of reporting of user first to the first comment of the brand vehicle
The prompt of information, the reporting port for including by first interactive interface report the user of the first social user to big data platform
The first comment of attribute and the first social user to the brand vehicle.
In the embodiment of the present invention, user property may include age of user and gender.
103, big data platform is according to the user property of the first social user, from the social user for having focused on the brand vehicle
The second social user for the good friend for being not belonging to the first social user is inquired in set；Wherein, the of the second social reporting of user
The user property of two social users and the user property of the first social user match.
As an alternative embodiment, in the embodiment of the present invention, after big data platform executes step 102, and
Before executing step 103, following steps can also be performed:
Big data platform judges whether the first social user has opened that there are approximate views with the brand vehicle of oneself concern
Social user recommend business；
If the first social user has opened, there are the social users of approximate view to recommend industry with the brand vehicle of oneself concern
Business, big data platform can execute step 103；Conversely, being deposited if the first social user does not open with the brand vehicle of oneself concern
Recommend business in the social user of approximate view, big data platform can terminate this process.
Wherein, implement above embodiment, can prevent to not yet open with oneself pay close attention to brand vehicle there are approximate
When the social user of view recommends the first social user of business to recommend other social users, the first social user is caused to recommend
Harassing and wrecking.
104, the second social user that big data platform obtains the second social reporting of user comments the second of the brand vehicle
By.
105, big data platform compares the first comment and the second comment.
If 106, comparison shows that the first comment is approximate with the second comment, big data platform is sent to the first social user recommends
Information, recommendation information include at least the corresponding social account of the second social user, the user property of the second social user and the
Two comments.
Wherein, the first social user can comment on according to the user property of the second social user and second, decide whether
It needs to be made friends according to the corresponding social account of the second social activity user with the second social user.
As an alternative embodiment, big data platform comments on the first comment with second in above-mentioned step 105
It compares, may comprise steps of:
201, big data platform is disassembled the first comment to obtain each candidate sentences；
202, big data platform determines the importance scores of each candidate sentences；
203, big data platform extracts target sentences of the importance scores greater than preset value as the from each candidate sentences
The key message of one comment；
204, big data platform compares first key message commented on and the key message of second comment.
As an alternative embodiment, first comment is disassembled to obtain by big data platform in above-mentioned steps 201
Each candidate sentences may include:
Big data platform obtains preset text dismantling rule, which includes branch, comma, sentence
It number to be disassembled, and pause mark, colon, quotation marks are without dismantling；
And big data platform disassembles rule according to the preset text, the first comment is disassembled, to obtain each time
Select sentence.
As an alternative embodiment, big data platform determines the importance of each candidate sentences in above-mentioned steps 202
Score may include:
For each candidate sentences in each candidate sentences, if the candidate sentences are Chinese sentence, big data platform is pressed
The Chinese sentence is split as several phrases according to the mode of semantic analysis；
And big data platform carries out full text traversal to each phrase that fractionation obtains and calculates, and obtains going out for each phrase
Occurrence number；
And big data platform is arranged according to all phrases that the sequence of frequency of occurrence from high to low obtains fractionation
Sequence, and each phrase assigns corresponding weight according to frequency of occurrence, and frequency of occurrence is higher, and the weight is higher；
And big data platform calculates the importance scores of each Chinese sentence, importance scores are the Chinese sentence
Split the weights sum of several obtained phrases.
For example, the first comment are as follows:
The oil consumption of XX brand vehicle is low, space is big, power is sufficient, comfort is good；The cost performance of XX brand vehicle is high；XX brand
The appearance of vehicle is beautiful, interior trim is luxurious.
Correspondingly, big data platform disassembles rule according to the preset text, the first comment is disassembled, what is obtained is each
Candidate sentences include:
Candidate sentences include:
A, the oil consumption of XX brand vehicle is low, space is big, power is sufficient, comfort is good；
B, the cost performance of XX brand vehicle is high；
C, the appearance of XX brand vehicle is beautiful, interior trim is luxurious.
Correspondingly, the phrase that big data platform is disassembled includes:
XX brand vehicle；Occur 3 times, weight 1；
Oil consumption is low；Occur 1 time, weight 1
Space is big；Occur 1 time, weight 1
Power foot；Occur 1 time, weight 1
Comfort is good；Occur 1 time, weight 1
Cost performance is high；Occur 1 time, weight 1
Appearance: occur 1 time, weight 1
It is beautiful: to occur 1 time, weight 1
Interior trim；Occur 1 time, weight 1
It is luxurious；Occur 1 time, weight 1.
Correspondingly, the importance scores of candidate sentences above are respectively as follows: No. A 7 points, No. B 4 points, No. C 7 points.
Further, it is assumed that preset value is 6 points, then target sentences are No. A and No. C；
Correspondingly, the key message of the key message of first comment are as follows: the oil consumption of XX brand vehicle is low, space is big, dynamic
Power foot, comfort are good；The appearance of XX brand vehicle is beautiful, interior trim is luxurious.
As one kind optionally implement it is above-mentioned, if the not Chinese sentence of the candidate sentences, webpage link address, that
Big data platform can open from the background the corresponding target webpage of the webpage link address；And big data platform can root
The importance scores of target webpage, the importance point of the target webpage are determined according to the link in target webpage, being directed toward target webpage
Number is the importance scores of the candidate sentences.
Wherein it is determined that the process of the importance scores of target webpage are as follows:
Wherein, S (Vi) be target webpage importance scores, d is damped coefficient, and In (Vi) is to exist to be directed toward target webpage
Link collections of web pages, out (Vj) it is the collections of web pages that the existing link of link in webpage j is directed toward, out (Vj) take absolutely
Value is the number to indicate element in collections of web pages, S (Vj) be webpage j importance scores.
As an alternative embodiment, the key message of first comment is actually the set of sentence, to pass
The comparison of key information is actually to compare to sentence.Therefore, in above-mentioned steps 204, big data platform first comments this
The key message of opinion and the key message of second comment compare, and may include:
Big data platform determines the first sentence in the key message of first comment, and determines the key of the second comment
The second sentence in information；Wherein, the quantity of the phrase to overlap each other between the first sentence and the second sentence is most；
And big data platform calculates the cosine similarity of the first sentence and the second sentence；
If cosine similarity is higher than designated value, big data platform determines that the first comment is approximate with the second comment.
In the embodiment of the present invention, the calculating process of above-mentioned cosine similarity are as follows:
First sentence is split as several phrases by big data platform, to obtain one group of phrase；
And the second sentence is split as several phrases by big data platform, to obtain another group of phrase；
And big data platform is compared the phrase in two groups of phrases one by one, and if it exists, 1 is then recorded as, if not depositing
It is being recorded as 0, then to obtain First ray and the second sequence；
And big data platform calculates the cosine similarity between First ray and the second sequence, and as the first sentence
And the second cosine similarity between sentence.
As an example it is assumed that:
First sentence are as follows: the oil consumption of XX brand vehicle is low, space is big, power is sufficient, comfort is good；
Second sentence are as follows: the comfort of XX brand vehicle is good, space is big, power is sufficient；
So, the comparison for two groups of phrases that the first sentence and the second sentence are split into is as shown in table 1, it may be assumed that
Table 1
| XX brand vehicle | Oil consumption is low | Space is big | Power foot | Comfort is good | |
| First sentence | 1 | 1 | 1 | 1 | 1 |
| Second sentence | 1 | 0 | 1 | 1 | 1 |
As shown in table 1, First ray a is (1,1,1,1), and the second sequence b is (1,1,0,1).
Wherein, big data platform calculates the cosine similarity between First ray and the second sequence are as follows:
Wherein, ab is integrally added after indicating the element in a sequence and corresponding element multiplication in b sequence, and denominator indicates a sequence
The quadratic sum of all elements is opened and opens radical sign multiplied by the quadratic sum of all elements in b sequence after radical sign in column；Wherein, a sequence indicates
First ray, b sequence indicate the second sequence.
For example, the result that the first above-mentioned sentence and the second sentence calculate be cos (a, b)=0.89 (i.e. First ray and
Cosine similarity between second sequence), be higher than designated value when 0.89, big data platform can determine first comment with this
Two comments are approximate.
In the embodiment of the present invention, the above-mentioned implementation to step 105 is described in detail, and implements above-mentioned embodiment party
Formula can promote the accuracy that the first comment is compared with the second comment, so as to accurately judge to report first to comment
There are approximate views between the social user of the first of opinion and the second social user for reporting the second comment.
Wherein, implement method described in Fig. 1, find the first social user effectively, accurately and closed with oneself
Other social users that the brand vehicle of note has approximate view exchange, in order to understanding oneself pass that can be more deep
The public praise of the brand vehicle of note.
Referring to Fig. 3, Fig. 3 is the society that another big data platform disclosed by the embodiments of the present invention is combined with brand vehicle
Hand over the flow diagram of user's recommended method.As shown in figure 3, the social user that the big data platform is combined with brand vehicle pushes away
The method of recommending may comprise steps of:
301, big data platform is after detecting that the first social user completes to the concern of a certain brand vehicle, prompts the
First comment of the user property and the first social user of the one social user of social reporting of user first to the brand vehicle.
302, the user property of the first social user of the social reporting of user of big data platform detection first and the first society
Hand over first comment of the user to the brand vehicle.
As an alternative embodiment, in the embodiment of the present invention, after big data platform executes step 302, and
Before executing step 303, following steps can also be performed:
Big data platform judges whether the first social user has opened that there are approximate views with the brand vehicle of oneself concern
Social user recommend business；
If the first social user has opened, there are the social users of approximate view to recommend industry with the brand vehicle of oneself concern
Business, big data platform can execute step 303；Conversely, being deposited if the first social user does not open with the brand vehicle of oneself concern
Recommend business in the social user of approximate view, big data platform can terminate this process.
Wherein, implement above embodiment, can prevent to not yet open with oneself pay close attention to brand vehicle there are approximate
When the social user of view recommends the first social user of business to recommend other social users, the first social user is caused to recommend
Harassing and wrecking.
303, the social reporting of user additional searching information of big data platform prompt first.
In the embodiment of the present invention, big data platform can control the car website and pop up the second interactive interface, second friendship
Mutual interface includes prompt information for prompt the first social reporting of user additional searching information and for social for first
Family reports the reporting port of additional searching information.
304, the additional searching information of the social reporting of user of big data platform detection first；Additional searching information includes at least
Search time section and region of search.
In the embodiment of the present invention, what the first social user can include according to second interactive interface is used to prompt the first society
The prompt for handing over the prompt information of reporting of user additional searching information, the reporting port for including by second interactive interface are counted to big
Additional searching information is reported according to platform, wherein additional searching information includes at least search time section and region of search.
305, big data platform is according to the user property of the first social user, from the social user for having focused on the brand vehicle
The second social user for the good friend for being not belonging to the first social user is inquired in set；Wherein, the of the second social reporting of user
The user property of two social users and the user property of the first social user match.
306, the second social user that big data platform obtains the second social reporting of user comments the second of the brand vehicle
By.
307, the social reporting of user second of big data platform inquiry second comment on call time whether be located at search time
In section；If calling time in search time section in the second social comment of reporting of user second, step 308 is executed；If second
It calls time and is not located in search time section in the social comment of reporting of user second, terminate this process.
308, whether the current location of the social user of big data platform inquiry second is located in region of search；If second is social
The current location of user is located in region of search, executes step 309- step 311；If the current location of the second social user not position
In in region of search, terminate this process.
In the embodiment of the present invention, the search time section that the first social user reports to big data platform is by the first social user
Self-setting, the such first social reporting of user can be relatively close apart from current time to the search time section of big data platform, from
And other social users and approximate view existing for the brand vehicle of first social user's concern can be prevented no longer approximate, thus
The first social user can be improved and find the reliable of other social activity users that there is approximate view with the brand vehicle of oneself concern
Property.
In the embodiment of the present invention, region of search that the first social user reports to big data platform by the first social user from
Row setting, such first social reporting of user may include the present bit of the first social user to the region of search of big data platform
It sets, so as to be convenient for and other social users and first of approximate view existing for the brand vehicle of first social user's concern
Aspectant exchange is carried out between social user, in region of search in order to which the first social user can more deep, face
Opposite understands oneself concern to other social users of approximate view existing for the brand vehicle paid close attention to the first social user
Brand vehicle public praise.
309, big data platform compares the first comment and the second comment.
Wherein, the specific implementation of above-mentioned step 309 is described in detail in embodiment in front, the present invention
Embodiment does not repeat herein.
If 310, comparison shows that the first comment is approximate with the second comment, big data platform sends inquiry to the second social user
Information, inquiry message is for inquiring whether the second social user allows to be recommended to other social users.
If 311, the second social user feedback allows to be recommended to other social users, big data platform is used to the first social activity
Family sends recommendation information, and recommendation information includes at least the user of the corresponding social account of the second social activity user, the second social user
Attribute and the second comment.
Wherein, implementation steps 310- step 311, can be in the case where the second social user knows, just to the first social activity
User, which sends, includes at least the corresponding social account of the second social activity user, the user property of the second social user and the second comment
Recommendation information, improve the informed degree of the second social user.
As an alternative embodiment, above-mentioned inquiry message further includes the first social use in the embodiment of the present invention
First social user of the corresponding social account in family, the user property of the first social user and the first social reporting of user is to this
First comment of brand vehicle.Correspondingly, the second social user can be according to the user property and first of the first social user
Comment decides whether to allow to be recommended to the first social user.
Wherein, method described in implementing Fig. 3 finds the first social user effectively, accurately and closes with oneself
Other social users that the brand vehicle of note has approximate view exchange, in order to understanding oneself pass that can be more deep
The public praise of the brand vehicle of note.
As an alternative embodiment, the first social user is receiving big data platform hair in the embodiment of the present invention
That send includes at least the corresponding social account of the second social activity user, the second user property of social user and pushing away for the second comment
After recommending information, the corresponding social account of the second social activity user can be added to the social account of good friend by the first social user,
And after the good friend that the second social user becomes the first social user, the first social user can with the second social user into
Row exchange, in order to the public praise of the brand vehicle for understanding oneself concern that can be more deep.In addition, the first social user may be used also
To score the second of the brand vehicle the second social user, and the first social activity user can will be to the second society
It hands over user to carry out scoring acquisition score value to the second comment of the brand vehicle and is reported to big data platform；Correspondingly, big data
Platform can add up the score value and current general comment score value of the second social user on car website, to generate second
Newest general comment score value of the social user on car website；Correspondingly, big data platform can decide whether there are it is described most
New general comment score value is corresponding and the still none obtained user right of the second social user is (for example, the purchase vehicle of the brand vehicle of concern
Preferential permission), if there is have newest general comment score value corresponding and the second social user still none obtained user right (for example,
The preferential permission of the brand vehicle of concern), big data platform can newest general comment score value is corresponding and the second social user
Still none obtained user right (for example, preferential permission of purchase vehicle of the brand vehicle of concern) distributes to the second social user.Its
In, implement this embodiment, the second social user can be encouraged more to report the brand vehicle to concern to big data platform
The comment of type, to not only contribute to other, social user's (the such as first social use) is effective, accurately finds and pays close attention to oneself
Brand vehicle the second social user for there is approximate view exchange, also help the second social user and obtain more use
Family permission (for example, preferential permission of purchase vehicle of the brand vehicle of concern).
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium include read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory,
RAM), programmable read only memory (Programmable Read-only Memory, PROM), erasable programmable is read-only deposits
Reservoir (Erasable Programmable Read Only Memory, EPROM), disposable programmable read-only memory (One-
Time Programmable Read-Only Memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory
(Electrically-Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact
Disc Read-Only Memory, CD-ROM) or other disc memories, magnetic disk storage, magnetic tape storage or can
For carrying or any other computer-readable medium of storing data.
Above to a kind of social user that big data platform is combined with brand vehicle recommendation disclosed by the embodiments of the present invention
Method is described in detail, and used herein a specific example illustrates the principle and implementation of the invention, with
The explanation of upper embodiment is merely used to help understand method and its core concept of the invention；Meanwhile for the general of this field
Technical staff, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion
The contents of this specification are not to be construed as limiting the invention.
Claims (10)
1. social user's recommended method that a kind of big data platform is combined with brand vehicle, which is characterized in that the method packet
It includes:
The big data platform prompts described the after detecting that the first social user completes to the concern of a certain brand vehicle
The user property of first social user described in one social reporting of user and the first social user are to the brand vehicle
First comment；
The big data platform detects user property and the institute of the described first social user of the described first social reporting of user
State first comment of the first social user to the brand vehicle；
The big data platform is used according to the user property of the described first social user from the social of the brand vehicle is had focused on
The second social user for the good friend for being not belonging to the described first social user is inquired in the set of family；Described second social reporting of user
The user property of the described second social user and the user property of the first social activity user match；
The big data platform obtains the described second social user of the described second social reporting of user to the brand vehicle
Second comment；
The big data platform compares first comment and second comment；
If comparison show it is described first comment it is approximate with second comment, the big data platform to it is described first social activity user
Recommendation information is sent, the recommendation information includes at least the corresponding social account of the second social activity user, second social activity
The user property of user and second comment.
2. social activity user recommended method according to claim 1, which is characterized in that compare first comment with it is described
After second comment is approximate and before the big data platform sends recommendation information to the described first social user, the side
Method further include:
The big data platform sends inquiry message to the described second social user, and the inquiry message is for inquiring described second
Whether social user allows to be recommended to other social users；
If the second social user feedback allows to be recommended to other social users, the big data platform execute it is described to
The step of described first social user sends recommendation information.
3. social activity user's recommended method according to claim 1 or 2, which is characterized in that the big data platform will be described
First comment is compared with second comment, comprising:
The big data platform is disassembled first comment to obtain each candidate sentences；
The big data platform determines the importance scores of each candidate sentences；
The big data platform is extracted the importance scores from each candidate sentences and is made greater than the target sentences of preset value
For the key message of first comment；
The big data platform compares the key message of key message and second comment that described first comments on.
4. social activity user recommended method according to claim 3, which is characterized in that the big data platform is by described first
Comment is disassembled to obtain each candidate sentences, comprising:
The big data platform obtains preset text dismantling rule, the preset text dismantling rule include branch, comma,
Fullstop will be disassembled, and pause mark, colon, quotation marks are without dismantling；
The big data platform disassembles rule according to the preset text, and the sight spot first comment is disassembled, with
To each candidate sentences.
5. social activity user's recommended method according to claim 4, which is characterized in that the big data platform determines described each
The importance scores of candidate sentences, comprising:
For each candidate sentences in each candidate sentences, if the candidate sentences are Chinese sentence, the big data
The Chinese sentence is split as several phrases in the way of semantic analysis by platform；
The big data platform carries out full text traversal to each phrase that fractionation obtains and calculates, and obtain each phrase goes out occurrence
Number；
The big data platform is ranked up according to all phrases that the sequence of the frequency of occurrence from high to low obtains fractionation,
And each phrase assigns corresponding weight according to the frequency of occurrence, and the frequency of occurrence is higher, and the weight is higher；
The big data platform calculates the importance scores of each Chinese sentence, and the importance scores are the Chinese sentence
Split the weights sum of several obtained phrases.
6. social activity user's recommended method according to claim 5, which is characterized in that further include:
If the candidate sentences are webpage link address, the big data platform is opening the webpage link address pair from the background
The target webpage answered；
The big data platform determines the target webpage according to the link in the target webpage, being directed toward the target webpage
Importance scores, the importance scores of the target webpage are the importance scores of the candidate sentences.
7. social activity user's recommended method according to claim 6, which is characterized in that the big data platform determines the mesh
Mark the process of the importance scores of webpage are as follows:
Wherein, S (Vi) be target webpage importance scores, d is damped coefficient, and In (Vi) is in the presence of the chain for being directed toward target webpage
The collections of web pages connect, out (Vj) it is the collections of web pages that the existing link of link in webpage j is directed toward, out (Vj) taking absolute value is
To indicate the number of element in the collections of web pages, S (Vj) be webpage j importance scores.
8. according to the described in any item social user's recommended methods of claim 3-7, which is characterized in that the big data platform will
The key message of first comment and the key message of second comment compare, comprising:
The big data platform determines the first sentence in the key message of first comment, and determines second comment
Key message in the second sentence；Wherein, the number of the phrase to overlap each other between first sentence and second sentence
Amount is most；
The big data platform calculates the cosine similarity of first sentence Yu second sentence；
If the cosine similarity is higher than designated value, the big data platform determines that first comment is close with second comment
Seemingly.
9. social activity user's recommended method according to claim 8, which is characterized in that the calculating process of the cosine similarity
Are as follows:
First sentence is split as several phrases by the big data platform, to obtain one group of phrase；
Second sentence is split as several phrases by the big data platform, to obtain another group of phrase；
The big data platform is compared the phrase in two groups of phrases one by one, and if it exists, is then recorded as 1, if it does not exist, then
It is recorded as 0, to obtain First ray and the second sequence；
The big data platform calculates the cosine similarity between the First ray and second sequence, and as described
Cosine similarity between one sentence and second sentence；
Wherein, the big data platform calculates the cosine similarity between the First ray and second sequence are as follows:
Wherein, ab is integrally added after indicating the element in a sequence and corresponding element multiplication in b sequence, and denominator indicates in a sequence
The quadratic sum of all elements is opened and opens radical sign multiplied by the quadratic sum of all elements in b sequence after radical sign；Wherein, described in the expression of a sequence
First ray, b sequence indicate second sequence.
10. -9 described in any item social user's recommended methods according to claim 1, which is characterized in that the big data platform
The user property for detecting the described first social user of the described first social reporting of user and the first social user are to institute
It states after the first comment of brand vehicle and the big data platform is according to the user property of the described first social user, from
Have focused on the second society for inquiring in the social user set of the brand vehicle and being not belonging to the good friend of the described first social user
Before handing over user, the method also includes:
The social reporting of user additional searching information of the big data platform prompt described first；
The big data platform detects the additional searching information of the described first social reporting of user；The additional searching information
Including at least search time section and region of search；
The big data platform obtains the described second social user of the described second social reporting of user to the brand vehicle
After second comment, the method also includes:
The big data platform inquire the second comment described in the described second social reporting of user on call time whether be located at it is described
In search time section；
If calling time within the described search period in the second comment described in the described second social reporting of user, the big number
Whether it is located in described search region according to the current location that platform inquires the described second social user；
If the current location of the second social user is located in described search region, the big data platform execute described in general
The step of first comment is compared with second comment.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810703587.0A CN109002507A (en) | 2018-07-01 | 2018-07-01 | A kind of social user's recommended method that big data platform is combined with brand vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810703587.0A CN109002507A (en) | 2018-07-01 | 2018-07-01 | A kind of social user's recommended method that big data platform is combined with brand vehicle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109002507A true CN109002507A (en) | 2018-12-14 |
Family
ID=64600457
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810703587.0A Pending CN109002507A (en) | 2018-07-01 | 2018-07-01 | A kind of social user's recommended method that big data platform is combined with brand vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109002507A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230046696A1 (en) * | 2021-08-10 | 2023-02-16 | Shusheng Fang | Method and system for provisioning realtime social community based on location service |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315622A (en) * | 2007-05-30 | 2008-12-03 | 香港中文大学 | System and method for detecting file similarity |
| CN106980999A (en) * | 2016-01-19 | 2017-07-25 | 阿里巴巴集团控股有限公司 | The method and apparatus that a kind of user recommends |
| CN107526800A (en) * | 2017-08-20 | 2017-12-29 | 平安科技（深圳）有限公司 | Device, method and the computer-readable recording medium of information recommendation |
| CN107967357A (en) * | 2017-12-21 | 2018-04-27 | 广东小天才科技有限公司 | A kind of method for pushing of good friend, system and terminal device |
-
2018
- 2018-07-01 CN CN201810703587.0A patent/CN109002507A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101315622A (en) * | 2007-05-30 | 2008-12-03 | 香港中文大学 | System and method for detecting file similarity |
| CN106980999A (en) * | 2016-01-19 | 2017-07-25 | 阿里巴巴集团控股有限公司 | The method and apparatus that a kind of user recommends |
| CN107526800A (en) * | 2017-08-20 | 2017-12-29 | 平安科技（深圳）有限公司 | Device, method and the computer-readable recording medium of information recommendation |
| CN107967357A (en) * | 2017-12-21 | 2018-04-27 | 广东小天才科技有限公司 | A kind of method for pushing of good friend, system and terminal device |
Non-Patent Citations (1)
| Title |
|---|
| 杨茂: "基于句子相似度的文本比对算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20230046696A1 (en) * | 2021-08-10 | 2023-02-16 | Shusheng Fang | Method and system for provisioning realtime social community based on location service |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bohnet et al. | Attributed question answering: Evaluation and modeling for attributed large language models | |
| Carmel et al. | Personalized social search based on the user's social network | |
| US9916366B1 (en) | Query augmentation | |
| US8935192B1 (en) | Social search engine | |
| US8321278B2 (en) | Targeted advertisements based on user profiles and page profile | |
| Guha | An early environmental debate: The making of the 1878 forest act | |
| US8515975B1 (en) | Search entity transition matrix and applications of the transition matrix | |
| CN108694647B (en) | Method and device for mining merchant recommendation reason and electronic equipment | |
| US20110106746A1 (en) | Affiliate linking | |
| US8515986B2 (en) | Query pattern generation for answers coverage expansion | |
| AU2014204091A1 (en) | Determining product categories by mining chat transcripts | |
| US20110307432A1 (en) | Relevance for name segment searches | |
| Sung et al. | Booming up the long tails: Discovering potentially contributive users in community-based question answering services | |
| Rasler et al. | Borders, rivalry, democracy, and conflict in the European region, 1816-1994 | |
| KR20160089177A (en) | Polarity-based user opinion ranking algorithm and system | |
| CN108572971A (en) | It is a kind of to be used to excavate and the method and apparatus of the relevant keyword of term | |
| Stojanovic | On Value‐Attributions: Semantics and Beyond | |
| WO2012047214A2 (en) | Visual display of semantic information | |
| Amiri et al. | Target-dependent churn classification in microblogs | |
| Blanco et al. | Overview of NTCIR-13 Actionable Knowledge Graph (AKG) Task. | |
| Li et al. | Understanding an enriched multidimensional user relevance model by analyzing query logs | |
| Olson | Hyaline arteriolosclerosis: new meaning for an old lesion | |
| CN109002507A (en) | A kind of social user's recommended method that big data platform is combined with brand vehicle | |
| US20130117301A1 (en) | Triggering social pages | |
| CN109002506A (en) | Information recommendation method caused by one kind is paid close attention to by brand vehicle |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181214 |
|
| WD01 | Invention patent application deemed withdrawn after publication |