CN110222257A - A kind of method, apparatus and data-link node for recommending business information - Google Patents
A kind of method, apparatus and data-link node for recommending business information Download PDFInfo
- Publication number
- CN110222257A CN110222257A CN201910394773.5A CN201910394773A CN110222257A CN 110222257 A CN110222257 A CN 110222257A CN 201910394773 A CN201910394773 A CN 201910394773A CN 110222257 A CN110222257 A CN 110222257A
- Authority
- CN
- China
- Prior art keywords
- user
- business information
- data
- information
- link node
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a kind of method, apparatus and data-link node for recommending business information, and this method includes that data-link node inquires the corresponding user information of user and historical record from the data-link that itself is stored；Determine the user to the preference value of every kind of business information respectively according to the historical record；According to user group belonging to user information acquisition user to the preference value of every kind of business information；The user group according to belonging to user, to the preference value of every kind of business information, obtains business information to be recommended, and business information to be recommended is sent to user to the preference value of every kind of business information and user.Business information is recommended to the preference value of every kind of business information according to preference value and user individual of the entire group of user group belonging to user to every kind of business information in the present invention, it can correctly reflect the hobby of user, the business information for ensuring to recommend is the interested information of user, substantially increases the accuracy of business information recommendation.
Description
Technical field
The present invention relates to Internet communication technology field, in particular to a kind of method, apparatus and number for recommending business information
According to chain node.
Background technique
With the development of generation information technology, Internet application is very universal, and user often browses on network
Different types of business information, such as books, magazine and news information.In order to promote user to browse more business information, phase
In the technology of pass server usually will current newest business information or hot spot service information recommendation to user, but newest business
Information or the interested information of hot spot service information not necessarily user, the industry for causing the seldom browser server of user to be recommended
Be engaged in information, be not able to achieve recommend business information effective purpose, and to the uninterested business information of user recommended user to
Family forms and bothers.
Summary of the invention
The purpose of the present invention is to provide a kind of method, apparatus and data-link node for recommending business information, to overcome
The shortcomings that prior art, the embodiment of the present invention mainly realize above-mentioned purpose by way of following several respects.
In a first aspect, the embodiment of the invention provides a kind of methods for recommending business information, which comprises
Data-link node inquires the corresponding user information of user and historical record from the data-link that itself is stored；
The data-link node determines the user to the preference of every kind of business information respectively according to the historical record
Value；
The data-link node according to the user information obtain the user belonging to user group to every kind of business information
Preference value；
Preference value and the use of the data-link node user group according to belonging to the user to every kind of business information
Family obtains business information to be recommended to the preference value of every kind of business information, and the business information to be recommended is sent to
The user.
With reference to first aspect, the embodiment of the invention provides the first possible implementation of above-mentioned first aspect,
In, the data-link node inquires the corresponding user information of user and historical record from the data-link that itself is stored, comprising:
When data-link nodal test to user logs in, the data-link node obtains the corresponding user's letter of the user
Breath；Alternatively, data-link node obtains the corresponding user information of user every preset duration；
The data-link node orients the data block comprising the user information from the data-link that itself is stored；
The data-link node reads the corresponding historical record of the user from the data block.
With reference to first aspect, the embodiment of the invention provides second of possible implementation of above-mentioned first aspect,
In, the data-link node determines the user to the preference value of every kind of business information, packet respectively according to the historical record
It includes:
The data-link node obtains the operation behavior of every kind of business information of the user's operation from the historical record
Information；
The data-link node is believed according to preset behavior weight information and the corresponding operation behavior of every kind of business information
Breath calculates separately the user to the preference value of every kind of business information by formula (1), and the operation behavior information includes behaviour
Make the type of behavior and the number of operations of every kind of operation behavior；
In the formula (1), i is the number of business information, PiIt is the user to the preference of i-th kind of business information
Value, j are the number of operation behavior information, ωjFor the corresponding behavior weight information of jth kind operation behavior, AjIt operates and goes for jth kind
For corresponding number of operations.
With reference to first aspect, the embodiment of the invention provides the third possible implementation of above-mentioned first aspect,
In, the data-link node determines the user to the preference value of every kind of business information, packet respectively according to the historical record
It includes:
The historical record is divided into the son of multiple chronomeres according to chronological order by the data-link node
Record；
The data-link node obtains the behaviour of every kind of business information of the user's operation from each subrecord respectively
Make behavioural information；
The data-link node is believed according to preset behavior weight information and the corresponding operation behavior of each subrecord
Breath calculates separately in each chronomere the user to the interest-degree of every kind of business information；
The data-link node according to the user in each chronomere to the interest-degree of every kind of business information,
Determine the user to the preference value of every kind of business information.
The third possible implementation with reference to first aspect, the embodiment of the invention provides above-mentioned first aspects
4th kind of possible implementation, wherein the data-link node is according to the user in each chronomere to every kind
The interest-degree of business information determines the user to the preference value of every kind of business information, comprising:
The data-link node determines that each time is single according to each chronomere, by formula (2) respectively
Time difference coefficient between position and current time；
The data-link node is according in the corresponding time difference coefficient of each chronomere, each chronomere
The user calculates separately the user to the inclined of every kind of business information to the interest-degree of every kind of business information, by formula (3)
Good value；
gt=a- (Tt-T)…(2)
In formula (2) and (3), t is the number of chronomere, gtFor the corresponding time difference coefficient of t-th of chronomere, Tt
For t-th of chronomere, T is chronomere nearest apart from current time in each chronomere, and a is default value；
I is the number of business information, PiIt is the user to the preference value of i-th kind of business information, BtIt is corresponding for t-th of chronomere
Interest-degree of the interior user to i-th kind of business information.
With reference to first aspect, the embodiment of the invention provides the 5th kind of possible implementation of above-mentioned first aspect,
In, the data-link node according to the user information obtain the user belonging to user group to the inclined of every kind of business information
Good value, comprising:
The data-link node determines user group belonging to the user according to the user information；
The data-link node obtains each user that the user group includes to the preference of every kind of business information respectively
Value；
The data-link node, to the preference value of every kind of business information, determines the user group according to each user
To the preference value of every kind of business information.
With reference to first aspect, the embodiment of the invention provides the 6th kind of possible implementation of above-mentioned first aspect,
In, preference value and the user pair of the data-link node user group according to belonging to the user to every kind of business information
The preference value of every kind of business information obtains business information to be recommended, comprising:
The data-link node determines the highest preset number kind business information of the corresponding preference value of the user group；
The data-link node obtains the corresponding hot information of the preset number kind business information；
The data-link node, to the preference value of every kind of business information, determines belonging to the user according to the user
Interest group；
The data-link node believes the business that other group members operated in the interest group and the user did not operated
The business information that user described in breath and the hot information did not operated is determined as business information to be recommended.
With reference to first aspect, the embodiment of the invention provides the 7th kind of possible implementation of above-mentioned first aspect,
In, the method also includes:
The data-link node believes the preference value of every kind of business information and the business to be recommended according to the user
Breath generates data block；
The data block is added in the data-link of itself storage by the data-link node, and the data block is wide
It broadcasts and is stored to other data-link nodes in common recognition network.
Second aspect, the embodiment of the invention provides a kind of device for recommending business information, described device includes:
Enquiry module, for inquiring the corresponding user information of user and historical record from the data-link that itself is stored；
Determining module, for determining the user to the preference value of every kind of business information respectively according to the historical record；
Obtain module, for according to the user information obtain the user belonging to user group to every kind of business information
Preference value；
Recommending module, for the user group according to belonging to the user to the preference value and the use of every kind of business information
Family obtains business information to be recommended to the preference value of every kind of business information, and the business information to be recommended is sent to
The user.
The third aspect, the embodiment of the invention provides a kind of data-link nodes, including one or more memories and processing
Device, the memory are stored with executable program, the one or more processor loads of the executable code, to hold
Row following steps:
The corresponding user information of inquiry user and historical record from the data-link that itself is stored；
Determine the user to the preference value of every kind of business information respectively according to the historical record；
According to the user information obtain the user belonging to user group to the preference value of every kind of business information；
The preference value of every kind of business information and the user believe every kind of business according to user group belonging to the user
The preference value of breath obtains business information to be recommended, and the business information to be recommended is sent to the user.
In embodiments of the present invention, the user information of user and the historical record that browse service information generates in the past are to deposit
In data-link, the data in data-link are difficult to be tampered for storage, it can be ensured that in the treatment process for recommending business information,
The historical record that data-link node is based on be it is true and reliable, can correctly reflect the hobby of user, so that finally
The business information of recommendation is the interested information of user, substantially increases the accuracy of business information recommendation.And the present invention is implemented
Example recommends business information to the preference value of every kind of business information according to user group belonging to user and user, it is ensured that recommendation
Business information is the interested information of user.Further, the embodiment of the present invention can use other similar with user preferences
The business information that family operated recommends user, or analysis user as time goes by becomes the hobby of every kind of business information
Change, the business information that user may currently like is recommended into user, it is clear to the business information progress of recommendation to improve user
A possibility that corresponding operatings such as looking at, download or buy.
Detailed description of the invention
Fig. 1 is a kind of method flow schematic diagram for recommending business information provided in an embodiment of the present invention；
Fig. 2 is a kind of structure drawing of device for recommending business information provided in an embodiment of the present invention；
Fig. 3 is a kind of structural schematic diagram of data-link node provided in an embodiment of the present invention.
Specific embodiment
With reference to the accompanying drawing, specific embodiments of the present invention will be described in detail, it is to be understood that guarantor of the invention
Shield range is not limited by the specific implementation.
Unless otherwise explicitly stated, otherwise in entire disclosure and claims, term " includes " or its change
Changing such as "comprising" or " including " etc. will be understood to comprise stated element or component, and not exclude other
Element or other component parts.
The embodiment of the invention provides a kind of methods for recommending business information, and this method is multiple numbers based on data-link
Common recognition network is constituted according to chain node.Being shared by multiple data-link nodes in common recognition network to all users recommends business to believe
The treating capacity of the processing pressure of breath, each data-link node is not too big, and processing speed is fast, can recommend in time to user
User may interested business information.And user goes over the historical record that operation service information generates and is stored in data-link
In, identical data-link is stored in each data-link node, the everyways such as service provider and hacker can not distort number
According to the data stored in chain, Information Security is very high.Data-link node is based on true historical record, can recommend user out
The accuracy of interested business information, recommendation is very high.
And user group belonging to synthetic user of the embodiment of the present invention to the preference and user's individual of business information to business
The preference of information is recommended business information to user, and the business information for substantially increasing recommendation is the interested business letter of user
A possibility that a possibility that breath, the business information that user's browsing, downloading or purchase are recommended, is very high, can be realized recommendation business letter
Effective purpose of breath will not form user and bother, improve user experience.
Referring to Fig. 1, this method specifically includes the following steps:
Step 101: data-link node inquires the corresponding user information of user and history note from the data-link that itself is stored
Record.
In embodiments of the present invention, for user recommend business information opportunity can there are two types of.One is when user is logical
Client logon data chain node is crossed, when data-link nodal test to user logs in, data-link node is according to the embodiment of the present invention
The mode of offer recommends business information to user.Another kind is that data-link node is every preset duration, automatically according to the present invention
The mode that embodiment provides recommends business information to user.
When data-link nodal test to user log in, alternatively, detect last time recommend between be spaced preset duration,
The period of business information is recommended to reach, data-link node obtains the corresponding user information of user at this time.User information may include
The information such as user account, gender, age, occupation, geographical location, user account can be the use for the unique identification user
Customized character string of user etc. when family contact method, identification card number or registration.When data-link nodal test is stepped on to user
When record, data-link node can be directly obtained user information from the log-on message of user.It is arrived when the period of recommendation business information
Up to when, data-link node can obtain stored user information from local.
After data-link node gets user information, from the data-link that itself is stored, orient comprising the user information
Data block；The corresponding historical record of the user is read from the data block oriented.It wherein, include user in historical record
Process operates the operation behavior information of every kind of business information, and operation behavior information includes the type and every kind of operation of operation behavior
The number of operations of behavior.User includes the types such as browsing, downloading, purchase to the operation behavior of business information.
Step 102: data-link node determines user to the preference value of every kind of business information respectively according to historical record.
The embodiment of the invention provides the modes that following the first and second two kinds determine preference value, specifically include:
First, data-link node determines user to the preference value of every kind of business information only in accordance with historical record.
In embodiments of the present invention, it has been pre-configured in each data-link node in common recognition network each type of
The corresponding behavior weight information of operation behavior, the corresponding weight of operation behavior that can more embody user preferences business information is higher,
Weight such as purchase is higher, and the weight of browsing is lower.
Data-link node obtains the operation behavior information of every kind of business information of user's operation from historical record, according to pre-
If the corresponding operation behavior information of behavior weight information and every kind of business information, user is calculated separately to every by formula (1)
The preference value of kind business information, operation behavior information includes the type of operation behavior and the number of operations of every kind of operation behavior.
In formula (1), i is the number of business information, PiIt is user to the preference value of i-th kind of business information, j is behaviour
Make the number of behavioural information, ωjFor the corresponding behavior weight information of jth kind operation behavior, AjIt is corresponding for jth kind operation behavior
Number of operations.
I.e. when specifically calculating preference value, for certain business information, the number of operations and row of every kind of operation behavior are calculated
Product between weight obtains the corresponding behavior coefficient of every kind of operation behavior.By the corresponding behavior coefficient of every kind of operation behavior
It sums up, obtains user to the preference value of this kind of business information.
For example, it is assumed that including that user browsed swordsman's class figure for the book information of swordsman's class, in the historical record of user
Letter ceases 10 times, downloaded swordsman's class book information 6 times, bought swordsman's class book information 1 time.It is preset on data-link node
Behavior weight information include the corresponding weight of browsing behavior be 1, the corresponding weight of downloading behavior is 5, and buying behavior is corresponding
Weight is 10.It is 10 that then data-link node, which calculates the corresponding behavior coefficient of browsing behavior, the corresponding behavior coefficient of downloading behavior
It is 30, the corresponding behavior coefficient of buying behavior is 10, and then calculating user is 50 to the preference value of swordsman's class book information.
For other every kind business information being related in historical record, user is calculated separately out all in accordance with aforesaid way
The preference value of corresponding other every kind business information.
Second, data-link node is according to user in historical record as time goes by the operation row of every kind of business information
Determine the user to the preference value of every kind of business information to change.
Historical record is divided into the subrecord of multiple chronomeres according to chronological order by data-link node.Time
Unit can be one day or one week etc..Data-link node obtains every kind of business information of user's operation from each subrecord respectively
Operation behavior information calculated separately according to preset behavior weight information and the corresponding operation behavior information of each subrecord
Interest-degree of the user to every kind of business information in each chronomere.Preference in the calculating of interest-degree and above-mentioned first way
The calculation of value is identical, and details are not described herein.
Data-link node, to the interest-degree of every kind of business information, determines user to every according to user in each chronomere
The preference value of kind business information.Specifically, data-link node is determined each respectively according to each chronomere by formula (2)
Time difference coefficient between chronomere and current time, according to the corresponding time difference coefficient of each chronomere, each chronomere
Interior user calculates user to the preference value of every kind of business information by formula (3) to the interest-degree of every kind of business information.
gt=a- (Tt-T)…(2)
In formula (2), t is the number of chronomere, gtFor the corresponding time difference coefficient of t-th of chronomere, TtFor t
A chronomere, T are chronomere nearest apart from current time in above-mentioned each chronomere, and a is default value.
It is i.e. when determining the corresponding time difference coefficient of each chronomere, distance in all chronomeres of division is current
Time nearest chronomere is known as current time unit, calculate separately other each chronomeres and current time unit it
Between chronomere's number for differing.For some chronomere, by the time of default value difference corresponding with the chronomere
Unit number is subtracted each other, and the corresponding time difference coefficient of the chronomere is obtained.For other each chronomeres, all in accordance with above-mentioned side
Formula determines the corresponding time difference coefficient of other each chronomeres respectively.Above-mentioned default value can be 10,20 or 30 etc..It is right
In current time unit, the chronomere's number differed between itself is 0, then the corresponding time difference system of current time unit
Number is above-mentioned default value.
After calculating the corresponding time difference coefficient of each chronomere in the manner described above, user is calculated by formula (3)
To the preference value of every kind of business information.For certain business information, the corresponding time difference coefficient of each chronomere and interest are calculated
The sum of products between degree obtains user to the preference value of this kind of business information.
In formula (3), t is the number of chronomere, gtFor the corresponding time difference coefficient of t-th of chronomere, i is industry
The number for information of being engaged in, PiIt is the user to the preference value of i-th kind of business information, BtFor corresponding interior institute, t-th of chronomere
User is stated to the interest-degree of i-th kind of business information.
For example, it is assumed that user one month historical record was divided into three son notes according to 10 days for a chronomere
Record, subrecord 3, subrecord 2 and subrecord 1 are denoted as according to the time respectively from front to back.Wherein, subrecord 1 is apart from current time
It recently, is current time unit.1 chronomere is differed between subrecord 2 and subrecord 1, between subrecord 3 and subrecord 1
Differ 2 chronomeres.Assuming that default value is 5, then the corresponding time difference coefficient of subrecord 1 is 5, the corresponding time difference of subrecord 2
Coefficient is 4, and the corresponding time difference coefficient of subrecord 3 is 3.Assuming that for swordsman's class book information, the interest of user in subrecord 1
Degree is 1, and the corresponding interest-degree of subrecord 2 is 1.5, and the corresponding interest-degree of subrecord 3 is 2, then the user is to swordsman's class books
The preference value of information is 17.
For other every kind business information, user is calculated separately out in the same way to other every kind business information
Preference value.The second way changes with time due to considering user preferences, and calculated preference value can more embody
The actual preferences of user, accuracy are higher.
Each user is calculated separately out to the preference of every kind of business information for each operation with per family by this step
Value, and the user information for storing each user respectively and user are to the corresponding relationship of the preference value of every kind of business information.
Step 103: data-link node according to above-mentioned user information obtain the user belonging to user group every kind of business is believed
The preference value of breath.
Specifically, data-link node determines user group belonging to the user according to above-mentioned user information, and obtaining respectively should
Each user that user group includes is to the preference value of every kind of business information, then according to each user to every kind of business information
Preference value determines the user group to the preference value of every kind of business information.
In embodiments of the present invention, gender, age, occupation and the geographical position that data-link node includes according to user information
It the information such as sets to cluster all users, all users is divided into multiple user groups, the user for including in same user group
With similar user property, as the gender of the user in same user group be located at women, age 20 years old to 30 years old it
Between, occupation be company's white collar, geographical location is in Beijing etc..
The user property that data-link node includes according to the user information of user is determined for compliance with the user property of the user
User group, which is user group belonging to the user.According to the user's letter for each user that the user group includes
Breath obtains each user that the user group includes to the preference value of every kind of business information respectively.Then every kind of business is believed
Breath, data-link node, to the preference value of this kind of business information, calculate the user group to this kind of business information according to each user
Preference value.Specifically, data-link node calculates each user to the average value of the preference value of this kind of business information, this is averaged
Value is as the user group to the preference value of this kind of business information.
By the above-mentioned means, data-link node can calculate user group belonging to the user to every kind of business information
Preference value.
User's individual is calculated through the above way to user group belonging to the preference value of every kind of business information and user
As a group to the preference value of every kind of business information after, 104 operation recommends business to believe to user as follows
Breath.
Step 104: preference value and user pair of the data-link node user group according to belonging to user to every kind of business information
The preference value of every kind of business information obtains business information to be recommended, and business information to be recommended is sent to user.
Data-link node determines the highest preset number kind business letter of the corresponding preference value of user group belonging to the user
Breath, obtains the corresponding hot information of above-mentioned preset number kind business information, the user in these hot informations was not operated
Business information is determined as business information to be recommended.
Other users in user group as belonging to the user and its have similar user property, these use
For being also likely on the Preference of business information there are very high similitude, user group belonging to the user is corresponding partially at family
It is worth highest preset number kind business information well, it is likely that and the user compares the business information type of preference.These types
The corresponding hot information of business information be likely to the very interested business information of user, therefore by these hot informations
In the business information that did not operated of the user recommend user, a possibility that these information of user's operation, is very high.
Data-link node, which is also referred to the historical record of other users similar in user interest, obtains business to be recommended
Information.Specifically, data-link node calculates the preference value of user every time in the past, and after recommending business information to user, all will
The preference value of user and the business information of recommendation are stored into data-link.Current data chain node through the above steps 102 behaviour
After the user is calculated to the preference value of every kind of business information, other users are obtained from currently stored data-link to every
The preference value of kind of business information, according to the user to the preference value and other users of every kind of business information to every kind of business information
Preference value, determine the preference similarity between user and other users, and then according to inclined between the user and other users
Good similarity, determines interest group belonging to user.In embodiments of the present invention, it can choose the preference phase between the user
Like highest predetermined number other users are spent, the predetermined number other users and the user are formed into an interest group,
Member in the interest group has similar interest preference.Data-link node operated other group members in interest group and the use
The business information that family did not operated is determined as business information to be recommended.
It will be operated with the approximate other users of user interest and business information that the user did not operated recommends use
Family, then the user is very high to the interested possibility of the business information of recommendation, improves the accuracy of business information recommendation and has
Effect property.
In embodiments of the present invention, data-link node can also come only with reference to preference value of the user to every kind of business information
Determine business information to be recommended.Specifically, for data-link node according to user to the preference value of every kind of business information, selection is full
The information category of sufficient preset condition obtains business letter to be recommended from the corresponding business information set of information category of selection
Breath.
Above-mentioned preset condition can be the highest preset number kind information category of preference value.In data-link node in advance
It is configured with the corresponding collection of services of every kind of information category, includes multiple information categories in the corresponding collection of services of information category
Business information.Preset number kind information category highest for preference value, data-link node can be corresponding from information category
The business information that the user did not operated is determined in collection of services, is chosen at random from the business information that the user did not operated
Select a certain number of business information as business information to be recommended.In addition, in the embodiment of the present invention, it is every in collection of services
A business information can have corresponding hot value, which is user-operably with the business information in the past period
Operation behavior and number of operations it is related.Data-link node can be from the business information that the user determined did not operated
The highest business information of a certain amount of hot value is selected as business information to be recommended.
Business information in the highest information category of user preference value is recommended into user, then business of the user to recommendation
Interested probability is very high, and user is likely to the operations row such as the business information of recommendation be browsed, downloads or bought
For.Further, the highest business information of temperature in the highest information category of user preference value can also be recommended user,
Then the interested possibility of user is bigger, promotes the accuracy of business information higher.
Business information to be recommended is sent to the corresponding client of the user by data-link node.Client, which receives, to be pushed away
These business information are shown after the business information recommended, so that user operates these business information.
In embodiments of the present invention, while data-link node determines business information to be recommended or later, data-link
Node, to the preference value and business information to be recommended of every kind of business information, generates data block also according to user.Specifically, number
According to chain node using the head cryptographic Hash of the last one data block in itself currently stored data-link as data block to be generated
Father's cryptographic Hash obtains to be generated to user to the preference value and business information progress Hash operation to be recommended of every kind of business information
At data block head cryptographic Hash.Data-link node is according to the father's cryptographic Hash, head cryptographic Hash and user couple of data block to be generated
The preference value and business information generation data block to be recommended of every kind of business information.Data-link node adds the data block of generation
It is added in the data-link of itself storage, and other data-link nodes that data block is broadcast in common recognition network is stored.
After other data-link nodes in common recognition network receive the data block, demonstrate,proved by proof of work mechanism, equity
Mechanism etc. authenticates the data block, which is added to the data-link respectively itself stored most after passing through by certification
Afterwards.
In embodiments of the present invention, the user information of user and the historical record that browse service information generates in the past are to deposit
In data-link, the data in data-link are difficult to be tampered for storage, it can be ensured that in the treatment process for recommending business information,
The historical record that data-link node is based on be it is true and reliable, can correctly reflect the hobby of user, so that finally
The business information of recommendation is the interested information of user, substantially increases the accuracy of business information recommendation.And the present invention is implemented
Example recommends business information to the preference value of every kind of business information according to user group belonging to user and user, it is ensured that recommendation
Business information is the interested information of user.Further, the embodiment of the present invention can use other similar with user preferences
The business information that family operated recommends user, or analysis user as time goes by becomes the hobby of every kind of business information
Change, the business information that user may currently like is recommended into user, it is clear to the business information progress of recommendation to improve user
A possibility that corresponding operatings such as looking at, download or buy.
Referring to fig. 2, the embodiment of the invention provides a kind of device for recommending business information, the device is above-mentioned for executing
Recommend the method for business information provided by embodiment, which includes:
Enquiry module 20, for inquiring the corresponding user information of user and historical record from the data-link that itself is stored；
Determining module 21, for being determined user to the preference value of every kind of business information respectively according to historical record；
Obtain module 22, for according to the user information obtain the user belonging to user group every kind of business is believed
The preference value of breath；
Recommending module 23, preference value and user for the user group according to belonging to the user to every kind of business information
To the preference value of every kind of business information, business information to be recommended is obtained, and business information to be recommended is sent to user.
Above-mentioned enquiry module 20, for when detecting that user logs in, obtaining the corresponding user information of user；Alternatively, every
Every preset duration, the corresponding user information of user is obtained；From the data-link that itself is stored, orient comprising user information
Data block；The corresponding historical record of user is read from data block.
Above-mentioned determining module 21, comprising:
First acquisition unit, for obtaining the operation behavior information of every kind of business information of user's operation from historical record；
First computing unit, for according to preset behavior weight information and the corresponding operation behavior of every kind of business information
Information calculates separately user to the preference value of every kind of business information by formula (1), and the operation behavior information includes operation
The number of operations of the type of behavior and every kind of operation behavior；
In the formula (1), i is the number of business information, PiIt is the user to the preference of i-th kind of business information
Value, j are the number of operation behavior information, ωjFor the corresponding behavior weight information of jth kind operation behavior, AjIt operates and goes for jth kind
For corresponding number of operations.
In another embodiment, above-mentioned determining module 21 includes:
Division unit, for according to chronological order, historical record to be divided into the subrecord of multiple chronomeres；
Second acquisition unit, for obtaining the operation row of every kind of business information of user's operation from each subrecord respectively
For information；
Second computing unit, for being believed according to preset behavior weight information and the corresponding operation behavior of each subrecord
Breath calculates separately in each chronomere user to the interest-degree of every kind of business information；
First determination unit is determined and is used for the interest-degree according to user in each chronomere to every kind of business information
Preference value of the family to every kind of business information.
Above-mentioned determination unit, for determining that each time is single respectively by formula (2) according to each chronomere
Time difference coefficient between position and current time；According to user in the corresponding time difference coefficient of each chronomere, each chronomere
To the interest-degree of every kind of business information, user is calculated separately to the preference value of every kind of business information by formula (3)；
gt=a- (Tt-T)…(2)
In formula (2) and (3), t is the number of chronomere, gtFor the corresponding time difference coefficient of t-th of chronomere, Tt
For t-th of chronomere, T is chronomere nearest apart from current time in each chronomere, and a is default value；
I is the number of business information, PiIt is the user to the preference value of i-th kind of business information, BtIt is corresponding for t-th of chronomere
Interest-degree of the interior user to i-th kind of business information.
Above-mentioned acquisition module 22, for determining user group belonging to the user according to the user information；It obtains respectively
Preference value of each user for taking the user group to include to every kind of business information；According to each user to every kind of business
The preference value of information determines the user group to the preference value of every kind of business information.
Above-mentioned recommending module 23, for determining the highest preset number kind business letter of the corresponding preference value of the user group
Breath；Obtain the corresponding hot information of the preset number kind business information；According to user to the preference value of every kind of business information,
Determine interest group belonging to user；By other group members operated in interest group and user did not operated business information and described
The business information that user described in hot information did not operated is determined as business information to be recommended.
In embodiments of the present invention, which further includes memory module, for according to user to the inclined of every kind of business information
Good value and business information to be recommended generate data block；Data block is added in the data-link of itself storage, and by data
Other data-link nodes that block is broadcast in common recognition network are stored.
In embodiments of the present invention, the user information of user and the historical record that browse service information generates in the past are to deposit
In data-link, the data in data-link are difficult to be tampered for storage, it can be ensured that in the treatment process for recommending business information,
The historical record that data-link node is based on be it is true and reliable, can correctly reflect the hobby of user, so that finally
The business information of recommendation is the interested information of user, substantially increases the accuracy of business information recommendation.And the present invention is implemented
Example recommends business information to the preference value of every kind of business information according to user group belonging to user and user, it is ensured that recommendation
Business information is the interested information of user.Further, the embodiment of the present invention can use other similar with user preferences
The business information that family operated recommends user, or analysis user as time goes by becomes the hobby of every kind of business information
Change, the business information that user may currently like is recommended into user, it is clear to the business information progress of recommendation to improve user
A possibility that corresponding operatings such as looking at, download or buy.
Referring to Fig. 3, the embodiment of the invention provides a kind of data-link nodes, and network of knowing together is by multiple data-link node structures
At each data-link node corresponds to multiple nets about vehicle, and intersection, institute is not present between the corresponding net of different data chain node about vehicle
It states memory and is stored with executable program, connected between the memory and the processor by bus communication, it is described to hold
Line program is loaded by one or more processors, to execute following steps:
The corresponding user information of inquiry user and historical record from the data-link that itself is stored；
Determine the user to the preference value of every kind of business information respectively according to the historical record；
According to the user information obtain the user belonging to user group to the preference value of every kind of business information；
The preference value of every kind of business information and the user believe every kind of business according to user group belonging to the user
The preference value of breath obtains business information to be recommended, and the business information to be recommended is sent to the user.
The user information of user and the historical record that browse service information generates in the past are stored in data-link, number
It is difficult to be tampered according to the data in chain, it can be ensured that in the treatment process for recommending business information, data-link node is based on
Historical record be it is true and reliable, can correctly reflect the hobby of user so that consequently recommended business information be use
The interested information in family substantially increases the accuracy of business information recommendation.And use according to belonging to user of the embodiment of the present invention
Family group and user recommend business information to the preference value of every kind of business information, it is ensured that the business information of recommendation is that user feels emerging
The information of interest.Further, the business information that the embodiment of the present invention can operate other users similar with user preferences
User is recommended, or analysis user as time goes by changes the hobby of every kind of business information, it currently may happiness by user
Good business information recommends user, the business information of recommendation is browsed, downloads or buy etc. accordingly with improving user
A possibility that operation.
The aforementioned description to specific exemplary embodiment of the invention is in order to illustrate and illustration purpose.These are retouched
It states and is not wishing to limit the invention to disclosed precise forms, and it will be apparent that according to the above instruction, can carry out very much
Change and changes.The purpose of selecting and describing the exemplary embodiment is that explaining specific principle of the invention and in fact
Border application, so that those skilled in the art can be realized and utilize a variety of different exemplary embodiment party of the invention
Case and various chooses and changes.The scope of the present invention is intended to be limited by claims and its equivalents.
Claims (7)
1. a kind of method for recommending business information, which is characterized in that the described method includes:
Data-link node inquires the corresponding user information of user and historical record from the data-link that itself is stored；
The data-link node determines the user to the preference value of every kind of business information respectively according to the historical record；
The data-link node according to the user information obtain the user belonging to user group to the inclined of every kind of business information
Good value；
Preference value and the user pair of the data-link node user group according to belonging to the user to every kind of business information
The preference value of every kind of business information, obtains business information to be recommended, and the business information to be recommended is sent to described
User, wherein
The data-link node inquires the corresponding user information of user and historical record from the data-link that itself is stored, comprising:
When data-link nodal test to user logs in, the data-link node obtains the corresponding user information of the user；Or
Person, every preset duration, data-link node obtains the corresponding user information of user；
The data-link node orients the data block comprising the user information from the data-link that itself is stored；
The data-link node reads the corresponding historical record of the user from the data block；
The data-link node determines the user to the preference value of every kind of business information, packet respectively according to the historical record
It includes:
The data-link node obtains the operation behavior information of every kind of business information of the user's operation from the historical record；
The data-link node passes through according to preset behavior weight information and the corresponding operation behavior information of every kind of business information
Formula (1) calculates separately the user to the preference value of every kind of business information, and the operation behavior information includes operation behavior
The number of operations of type and every kind of operation behavior；
In the formula (1), i is the number of business information, PiIt is the user to the preference value of i-th kind of business information, j is
The number of operation behavior information, ωjFor the corresponding behavior weight information of jth kind operation behavior, AjIt is corresponding for jth kind operation behavior
Number of operations；
The data-link node determines the user to the preference value of every kind of business information, packet respectively according to the historical record
It includes:
The historical record is divided into the subrecord of multiple chronomeres according to chronological order by the data-link node；
The data-link node obtains the operation row of every kind of business information of the user's operation from each subrecord respectively
For information；
The data-link node divides according to preset behavior weight information and the corresponding operation behavior information of each subrecord
The user is not calculated in each chronomere to the interest-degree of every kind of business information；
The data-link node, to the interest-degree of every kind of business information, determines institute according to the user in each chronomere
User is stated to the preference value of every kind of business information.
2. recommending the method for business information according to claim 1, which is characterized in that the data-link node is according to described every
The user determines the user to the preference of every kind of business information the interest-degree of every kind of business information in a chronomere
Value, comprising:
The data-link node according to each chronomere, by formula (2) determine respectively each chronomere with
Time difference coefficient between current time；
The data-link node is according to the use in the corresponding time difference coefficient of each chronomere, each chronomere
Family calculates separately the user to the preference value of every kind of business information by formula (3) to the interest-degree of every kind of business information；
gt=a- (Tt-T)…(2)
In formula (2) and (3), t is the number of chronomere, gtFor the corresponding time difference coefficient of t-th of chronomere, TtFor t
A chronomere, T are chronomere nearest apart from current time in each chronomere, and a is default value；I is industry
The number for information of being engaged in, PiIt is the user to the preference value of i-th kind of business information, BtIt is corresponding interior described for t-th of chronomere
Interest-degree of the user to i-th kind of business information.
3. recommending the method for business information according to claim 1, which is characterized in that the data-link node is according to the use
Preference value of the user group belonging to user described in the acquisition of information of family to every kind of business information, comprising:
The data-link node determines user group belonging to the user according to the user information；
The data-link node obtains each user that the user group includes to the preference value of every kind of business information respectively；
The data-link node, to the preference value of every kind of business information, determines the user group to every kind according to each user
The preference value of business information.
4. recommending the method for business information according to claim 1, which is characterized in that the data-link node is according to the use
User group belonging to family to the preference value of every kind of business information, obtains wait push away the preference value of every kind of business information and the user
The business information recommended, comprising:
The data-link node determines the highest preset number kind business information of the corresponding preference value of the user group；
The data-link node obtains the corresponding hot information of the preset number kind business information；
The data-link node, to the preference value of every kind of business information, determines interest belonging to the user according to the user
Group；
The data-link node by the business information that other group members operated in the interest group and the user did not operated and
The business information that user described in the hot information did not operated is determined as business information to be recommended.
5. recommending the method for business information according to claim 1, which is characterized in that the method also includes:
The data-link node is according to the user to the preference value and the business information to be recommended of every kind of business information, life
At data block；
The data block is added in the data-link of itself storage, and the data block is broadcast to altogether by the data-link node
Other data-link nodes known in network are stored.
6. a kind of device for recommending business information, for realizing method of any of claims 1-5, feature exists
In described device includes:
Enquiry module, for inquiring the corresponding user information of user and historical record from the data-link that itself is stored；
Determining module, for determining the user to the preference value of every kind of business information respectively according to the historical record；
Obtain module, for according to the user information obtain the user belonging to user group to the preference of every kind of business information
Value；
Recommending module, for the user group according to belonging to the user to the preference value of every kind of business information and the user to every
The preference value of kind business information, obtains business information to be recommended, and the business information to be recommended is sent to the use
Family.
7. a kind of data-link node, which is characterized in that including one or more memories and processor, the memory is stored with
Executable program, the one or more processor loads of the executable code, requires any one of 1-5 with perform claim
The method, comprising the following steps:
The corresponding user information of inquiry user and historical record from the data-link that itself is stored；
Determine the user to the preference value of every kind of business information respectively according to the historical record；
According to the user information obtain the user belonging to user group to the preference value of every kind of business information；
According to user group belonging to the user to the preference value of every kind of business information and the user to every kind of business information
Preference value obtains business information to be recommended, and the business information to be recommended is sent to the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910394773.5A CN110222257B (en) | 2019-05-13 | 2019-05-13 | method and device for recommending service information and data link node |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910394773.5A CN110222257B (en) | 2019-05-13 | 2019-05-13 | method and device for recommending service information and data link node |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110222257A true CN110222257A (en) | 2019-09-10 |
| CN110222257B CN110222257B (en) | 2020-01-31 |
Family
ID=67820951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910394773.5A Active CN110222257B (en) | 2019-05-13 | 2019-05-13 | method and device for recommending service information and data link node |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110222257B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751219A (en) * | 2019-10-23 | 2020-02-04 | 郑州阿帕斯科技有限公司 | Content sending method and device |
| CN113536108A (en) * | 2021-05-31 | 2021-10-22 | 珠海大横琴科技发展有限公司 | Information recommendation method and device |
| CN114036374A (en) * | 2021-09-29 | 2022-02-11 | 国家电网有限公司 | Novel service business intelligent matching method and device for different clients |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103603A (en) * | 2009-12-18 | 2011-06-22 | 百度在线网络技术（北京）有限公司 | User behavior data analysis method and device |
| CN103164804A (en) * | 2011-12-16 | 2013-06-19 | 阿里巴巴集团控股有限公司 | Personalized method and personalized device of information push |
| CN103927673A (en) * | 2014-03-24 | 2014-07-16 | 腾讯科技（深圳）有限公司 | Method and device for recommending service information |
| CN104298679A (en) * | 2013-07-18 | 2015-01-21 | 腾讯科技（深圳）有限公司 | Application service recommendation method and device |
| CN105095267A (en) * | 2014-05-09 | 2015-11-25 | 阿里巴巴集团控股有限公司 | User involving project recommendation method and apparatus |
| CN106503269A (en) * | 2016-12-08 | 2017-03-15 | 广州优视网络科技有限公司 | Method, device and server that application is recommended |
| CN108198019A (en) * | 2017-12-27 | 2018-06-22 | 网易无尾熊（杭州）科技有限公司 | Item recommendation method and device, storage medium, electronic equipment |
| WO2018112696A1 (en) * | 2016-12-19 | 2018-06-28 | 深圳大学 | Content pushing method and content pushing system |
-
2019
- 2019-05-13 CN CN201910394773.5A patent/CN110222257B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102103603A (en) * | 2009-12-18 | 2011-06-22 | 百度在线网络技术（北京）有限公司 | User behavior data analysis method and device |
| CN103164804A (en) * | 2011-12-16 | 2013-06-19 | 阿里巴巴集团控股有限公司 | Personalized method and personalized device of information push |
| CN104298679A (en) * | 2013-07-18 | 2015-01-21 | 腾讯科技（深圳）有限公司 | Application service recommendation method and device |
| CN103927673A (en) * | 2014-03-24 | 2014-07-16 | 腾讯科技（深圳）有限公司 | Method and device for recommending service information |
| CN105095267A (en) * | 2014-05-09 | 2015-11-25 | 阿里巴巴集团控股有限公司 | User involving project recommendation method and apparatus |
| CN106503269A (en) * | 2016-12-08 | 2017-03-15 | 广州优视网络科技有限公司 | Method, device and server that application is recommended |
| WO2018112696A1 (en) * | 2016-12-19 | 2018-06-28 | 深圳大学 | Content pushing method and content pushing system |
| CN108198019A (en) * | 2017-12-27 | 2018-06-22 | 网易无尾熊（杭州）科技有限公司 | Item recommendation method and device, storage medium, electronic equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110751219A (en) * | 2019-10-23 | 2020-02-04 | 郑州阿帕斯科技有限公司 | Content sending method and device |
| CN113536108A (en) * | 2021-05-31 | 2021-10-22 | 珠海大横琴科技发展有限公司 | Information recommendation method and device |
| CN114036374A (en) * | 2021-09-29 | 2022-02-11 | 国家电网有限公司 | Novel service business intelligent matching method and device for different clients |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110222257B (en) | 2020-01-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109559208B (en) | Information recommendation method, server and computer readable medium | |
| CN109741146B (en) | Product recommendation method, device, equipment and storage medium based on user behaviors | |
| US8112308B1 (en) | Targeting using generated bundles of content sources | |
| JP5671633B2 (en) | Determining preference categories based on user access attribute values | |
| AU2010210726B2 (en) | Determining conversion probability using session metrics | |
| US10423999B1 (en) | Performing personalized category-based product sorting | |
| US9607273B2 (en) | Optimal time to post for maximum social engagement | |
| US20100211464A1 (en) | Targeted Online Advertising | |
| US20060129463A1 (en) | Method and system for automatic product searching, and use thereof | |
| US20120078725A1 (en) | Method and system for contextual advertisement recommendation across multiple devices of content delivery | |
| US20100262692A1 (en) | Recommendation of network object information to user | |
| US20110066507A1 (en) | Context Enhanced Marketing of Content and Targeted Advertising to Mobile Device Users | |
| US20140108096A1 (en) | Clickstream analysis methods and systems related to modifying an offline promotion for a consumer good | |
| US20090187486A1 (en) | Method and apparatus for delivering targeted content | |
| US20090144133A1 (en) | Context related advertisement/information exposure method and recommendation service system using the same | |
| US8712985B1 (en) | Social network user data advertising | |
| CN101589360A (en) | Generic online ranking system that is suitable for uniting and method | |
| CA3103311C (en) | Multi-source data analytics system, data manager and related methods | |
| WO2005019994A2 (en) | Personalized selection and display of user-supplied content to enhance browsing of electronic catalogs | |
| US9965526B1 (en) | Data mining for multiple item comparisons | |
| CN107784504B (en) | Method for generating return visit event of client and terminal equipment | |
| CN110674391B (en) | Product data pushing method and system based on big data and computer equipment | |
| US20120215664A1 (en) | Epurchase model | |
| JP2010113542A (en) | Information provision system, information processing apparatus and program for the information processing apparatus | |
| CN110222257A (en) | A kind of method, apparatus and data-link node for recommending business information |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |