CN110110221A - Government data intelligent recommendation method and system - Google Patents
Government data intelligent recommendation method and system Download PDFInfo
- Publication number
- CN110110221A CN110110221A CN201910221130.0A CN201910221130A CN110110221A CN 110110221 A CN110110221 A CN 110110221A CN 201910221130 A CN201910221130 A CN 201910221130A CN 110110221 A CN110110221 A CN 110110221A
- Authority
- CN
- China
- Prior art keywords
- user
- data
- data item
- feature vector
- vector
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 239000013598 vector Substances 0.000 claims abstract description 141
- 230000003068 static effect Effects 0.000 claims abstract description 70
- 230000007246 mechanism Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 8
- 238000001914 filtration Methods 0.000 abstract description 4
- 238000005457 optimization Methods 0.000 abstract description 3
- 230000006399 behavior Effects 0.000 description 18
- 230000003542 behavioural effect Effects 0.000 description 14
- 230000008569 process Effects 0.000 description 10
- 238000004590 computer program Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000000875 corresponding effect Effects 0.000 description 3
- 238000007405 data analysis Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000002372 labelling Methods 0.000 description 3
- 239000000203 mixture Substances 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 239000013589 supplement Substances 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 241001180404 Warea Species 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000001174 ascending effect Effects 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000006386 memory function Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
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
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Databases & Information Systems (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- General Business, Economics & Management (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Health & Medical Sciences (AREA)
- Operations Research (AREA)
- Quality & Reliability (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 government data intelligent recommendation methods, which comprises obtains user basic information and user's static attribute according to the registration information of user；Recommend pond according to user's static attribute locking data；Acquisition user behavior data and the interest vector for calculating user；Extract the feature vector of each of data recommendation pond data item；According to the feature vector of the interest vector of the user and/or data item to user's recommending data.Using the operation system data of different government department's government affairs staff as basic data supporting, by obtaining user basic information and user's static attribute, the proposed algorithm of the collaborative filtering based on user is respectively adopted, accurately recommend business datum item to user (i.e. government affairs staff) with content-based recommendation algorithm, improve the convenience of data acquisition in user (i.e. government affairs staff) course of work, the effectively quality of optimization acquisition data, assists it more efficiently to handle official business and carry out intelligent decision.
Description
Technical field
The present invention relates to government data analysis management and intelligent data recommended technology fields, specifically, being a kind of political affairs
Data intelligence recommended method of being engaged in and system.
Background technique
With Information Technology Development, government affairs working electronic, informationization have become the need that government efficiently manages social affairs
It wants, however, quick obtaining effective information causes certain journey the magnanimity and randomness of data resource are handled to government affairs again during
The obstacle of degree.In order to improve the network information service efficiency and quality of government affairs network users, pushed away now there are many government website
Personalized recommendation system is gone out.The first kind is government affairs information recommender system towards the public, and the user of such system is numerous people
The people masses formulate customized information according to the personal information of user and personal behavior, provide the service with individual subscriber characteristic.
After user's Website login, personalized recommendation system provides using the user information being collected into similar user for user and feels
The information of interest and care；The web page contents that analysis user accessed, and there is memory function, according to these records of user
The interest and hobby for automatically analyzing out user recommend similar information to it when user logs in again.Second class system is towards political affairs
The big data platform of business department personnel, big data management platform can provide clearly data assets management for data owner,
Convenient and fast data service is provided for data application, service application and applies O&M service, provides the data on basis for data analysis
It relies on and is associated with foundation.It is managed collectively for data standard, and realizes construction and the pipe of java standard library based on data standard
Reason.
Above-mentioned first kind government portals personalization system is to facilitate the people towards the common people
Solve government policy, economy, culture, amusement and educational information.Second class big data platform has the data resource of magnanimity, is government's political affairs
The department personnel working of business provides data supporting, and provides association foundation for data analysis, however is not provided with intelligence and pushes away
Function is recommended, government affairs department personnel can not be enable more easily to obtain accurate data information, assist its more efficient
Ground office and progress intelligent decision.
Summary of the invention
The purpose of the present invention is aiming at the shortcomings in the prior art, provide a kind of government data intelligent recommendation method and be
System, using the operation system data of different department's government affairs staff as basic data supporting, by obtaining user basic information
It whether is that (whether the field feedback recorded in the behavioral data of user by new user according to user with user's static attribute
For sky) different data recommendation methods is respectively adopted: for new user (the user feedback letter recorded in the behavioral data of user
Breath is sky), by constructing user model, using the proposed algorithm of the collaborative filtering based on user, calculate target user and sample
The distance between user acquires neighbour of the similar users as target user, and mesh is predicted using the interest of neighbour as module
Mark the data requirements of user, the data item for recommending it that may be concerned about to target user.For common user (the i.e. behavior number of user
Field feedback according to middle record is not empty), user interest and project data are calculated using content-based recommendation algorithm
The data that distance recommends it to be concerned about to user, in particular by the feature vector for the interest vector and data item for calculating user, and
Further calculate the similarity of the two, the data item for recommending it that may be concerned about to user by sequencing of similarity.Through the invention
Government data intelligent recommendation method the convenience of data acquisition in user (i.e. government affairs staff) course of work can be improved,
The effectively quality of optimization acquisition data, improves office efficiency.
To achieve the above object, the technical solution adopted by the present invention is that: a kind of government data intelligent recommendation method, the side
Method the following steps are included:
User basic information and user's static attribute are obtained according to the registration information of user；
Recommend pond according to user's static attribute locking data；
User behavior data is acquired, the interest vector of user is calculated according to the user behavior data；
Extract the feature vector of each of data recommendation pond data item；
According to the similarity between the interest vector of the user and the feature vector of data item to user's recommending data.
Further,
The user basic information includes: user name, age and gender；User's static attribute includes: the work of user
Make area, operating mechanism, department and the authority of office.
Further,
The interest vector that user is calculated according to the user behavior data, comprising:
Judge whether the field feedback recorded in user behavior data is empty, if field feedback is sky,
Neighbour is found according to user basic information and user's static attribute, by neighbour to user's recommending data；Conversely, then obtaining user
Data item in feedback information, for calculating the interest vector of user.
Further,
It is described that neighbour is found according to user basic information and user's static attribute, comprising:
The essential information and user's static attribute of collecting sample user constructs user model, and the user model includes basic
Information eigenvector and static attribute feature vector；
Weight is assigned to the characteristic value of all feature vectors in the user model；
The essential information and user's static attribute for obtaining target user, obtain the basic letter of target user using user model
Cease feature vector and static attribute feature vector；
The essential information feature vector of the target user and static attribute feature vector are used with each sample respectively
The essential information feature vector and static attribute feature vector at family are compared one by one；
Utilize the similarity between the weight calculation target user and each sample of users；
Similarity is ranked up according to sequence from big to small, 10 sample of users is as mesh before selection similarity ranking
Mark the neighbour of user.
Further,
It is described by neighbour to user's recommending data, comprising:
It obtains sample of users in the feedback information of 10 neighbours respectively to record the access of all data item, the access note
Whether whether record include: access duration, access times, newest access time interval, collect and subscribe to；
The interest score of each data item is calculated according to the access record；
The interest score is ranked up according to descending sequence；
The data item for choosing before the interest score ranking 10, by its data recommendation to target user.
Further,
The interest vector for calculating user, comprising:
It acquires user in field feedback to record the access of all data item, when the access record includes: access
Whether whether length access times, newest access time interval, collected and subscribed to；
The interest score of all data item in field feedback is calculated according to the access record；
The interest score is ranked up according to descending sequence；
It filters out interest score and comes preceding 30 data item and segmented one by one, count the participle and participle of all data item
Frequency；
All participles are ranked up according to the sequence of participle frequency from high to low, choose before participle frequency ranking 10 point
Interest vector of the word as user.
Further,
The feature vector for extracting each of data recommendation pond data item, comprising:
A1. the data item extracted in the data recommendation pond is segmented；
A2. the participle and participle frequency for counting the data item, according to the sequence of participle frequency from high to low to all points
Word is ranked up；
A3. feature vector of 5 participle as the data item before selection ranking；
A4. the feature vector of step A1-A3 to all data item obtained in data recommendation pond is repeated.
Further,
According to the similarity between the interest vector of the user and the feature vector of data item to user's recommending data, wrap
It includes:
Calculate the similarity of both the interest vector of target user and the feature vector of data item；
Similarity is ranked up according to sequence from big to small, before similarity ranking 10 data item is chosen, is counted
According to recommending target user.
On the other hand, the present invention provides a kind of government data intelligent recommendation system, the system comprises:
Module is obtained, accesses record for obtaining user basic information, user's static attribute and user, the user includes
Target user and sample of users；
Data recommendation pond, for determining the scope of business recommended to the user, the data recommendation pond includes different business portion
The data item of door, the data item of each business department by the business department specific transactions item, common data item and outer network chain
Connect composition；
Field feedback library records access behavior of the user to all data item for storing user behavior data；
Computing module, comprising:
Interest vector computing module, for calculating the interest vector of user；Feature vector computing module, for calculating data
The feature vector of item；Similarity calculation module, for calculate similarity between target user and sample of users and interest to
The similarity of amount and both feature vectors；
Recommending module, for according to the feature vector of the interest vector of user and/or data item to user's recommending data.
Further,
The system also includes user models to construct module, essential information and user's static state category for collecting sample user
Property building user model, the user model includes essential information feature vector and static attribute feature vector.
The invention has the advantages that:
Government data intelligent recommendation method and system of the invention are with the business system of different government department's government affairs staff
Data of uniting are respectively adopted by obtaining user basic information and user's static attribute based on user's as basic data supporting
The proposed algorithm of collaborative filtering, and accurately recommend business number to user (i.e. government affairs staff) with content-based recommendation algorithm
According to item, the convenience of data acquisition in user (i.e. government affairs staff) course of work, the effectively matter of optimization acquisition data are improved
Amount, assists it more efficiently to handle official business and carry out intelligent decision.
Detailed description of the invention
For purpose, feature and advantage of the present invention can be clearer to understand, below with reference to attached drawing to preferable reality of the invention
Example is applied to be described in detail, in which:
Fig. 1 is the flow diagram of government data intelligent recommendation method of the present invention in embodiment one；
Fig. 2 is the block schematic illustration of government data intelligent recommendation system of the present invention in embodiment one；
Fig. 3 is the flow diagram of government data intelligent recommendation method of the present invention in embodiment two；
Fig. 4 is the block schematic illustration of government data intelligent recommendation system of the present invention in embodiment two.
Specific embodiment
Many details are explained in the following description in order to fully understand the application.But the application can
Implemented with being much different from other way described herein, those skilled in the art can be in the feelings without prejudice to the application intension
Similar popularization is done under condition, therefore the application is not limited by following public specific implementation.
A kind of government data intelligent recommendation method and system of the present invention described in this specification are based on each portion, government
Door different business platform and its big data shared platform, using the operation system data of different government department's government affairs staff as
Basic data supports, provided government data intelligent recommendation method and system, in government data intelligent recommendation system of the present invention
User refer to government department government affairs staff, obtained in government data intelligent recommendation method of the present invention or the data of acquisition come
Derived from the shared big data of Government departments' different business systems.
Embodiment one
The present embodiment passes through for common user (field feedback recorded in the behavioral data of user is not empty)
Common user is stored in the available user basic information of registration information initial in system database and user's static attribute, leads to
The available user behavior data of all operations carried out in systems after subscriber self-registration is crossed, is remembered in user behavior data
The feedback information for having carried user's regular job provides a field feedback according to the behavioral data of common user accordingly
Library records access behavior of the user to all data item for storing user behavior data.With the increasing of user volume in system
It is more, it, can be with according to the user basic information and user's static attribute and the behavioral data of user recorded in user's registration information
Describe user to draw a portrait and establish communities of users accordingly, the interest vector and data of user can also be calculated separately according to these data
The feature vector of item, and then user interest is calculated using content-based recommendation algorithm and gives user's recommendation at a distance from project data
Its data being concerned about.
The government data intelligent recommendation method and system that the present embodiment is embodied below in conjunction with attached drawing 1 and attached drawing 2 into
Row detailed description.
Government data intelligent recommendation method as shown in Figure 1, the first registration information according to user in system are available
User basic information and user's static attribute, wherein user basic information includes: user name, age and gender etc., and user is static
Attribute includes: the static attributes such as operational area, operating mechanism, department and the authority of office of user.
After the user basic information and the user's static attribute that obtain user, it can be pushed away according to user's static attribute locking data
Pond is recommended, specifically user can be carried out by community's division by user's static attribute, it can be according to user location, affiliated
Relative or its data item that may be concerned about is recommended to user by operating mechanism or department, user's functional authority etc..
The data recommendation pond of user can be linked by the specific transactions item, common data item and outer net of user affiliated function and be formed.
Behavioral data after logging in system by user can reflect the preferences and needs of user, what user was carried out in systems
All operations are recorded by user behavior data, and feedback information acquired in user accesses data item will be used to calculate the emerging of user
Inclination amount.
At the same time, user interest can be calculated at a distance from project data by the behavioral data of user, specifically
The feature vector of data item can be established according to the user basic information and its static attribute for accessing each data item.
Finally by the similarity between the feature vector of the interest vector and data item that calculate user, recommend number to user
According to.
The government data intelligent recommendation method of the present embodiment can be realized simultaneously outer net Link Recommendation, specific transactions recommend and
Three kinds of recommended requirements forms of personalized recommendation.Specific implementation process is described as follows:
One, outer net Link Recommendation
Outer net Link Recommendation is the link for recommending its other possible interested government website data to user, the recommendation shape
The main thought of formula is: by the behavioral data of the government website link of record user's browsing, calculating the interest vector of user, together
When the feature vector of the website is calculated by the access situation of users all in system according to corresponding website, calculate interest vector and special
The similarity for levying vector recommends the higher outer net of similarity to link user, and this recommendation form makes user have other political affairs of browsing
When the demand of mansion department website, it is precisely recommended, the working efficiency of user can be improved, optimizes the use of recommender system
Experience sense.During using outer net Link Recommendation, the feature vector of interest vector and website to user calculates simultaneously, data
Recommend to include all government website link data item in pond.
Calculate interest vector such as step 101~step 105 of user.
User is to all government websites in step 101. acquisition user's access government website link feedback information obtained
The access record for linking data item, the access record in this example include: user in section in those years (can be one day,
One week or one month) to the access duration of each website, access times, newest access time interval, whether collect and whether order
Read, wherein access duration is obtained by the accumulative total residence time for accessing the website each time, it should be noted that for
Family single reference stops the website no more than 2 seconds, is considered as ineffective access, is not involved in calculating.Newest access time interval refers to: being
System current time accesses the interval between the data time last time apart from user, and the interval the short, illustrates that interest value is higher；
Step 102. according to access record, to access duration, access times, newest access time interval, whether collect and
Whether subscription sets preference weight, calculates the user in field feedback to the interest of each government website link data item
Score, for example:
Qi=0.6*freqi-0.2*sclicki+0.1*stimei+0.3*collecti+0.2*subi,
Wherein, QiIt is user to the interest score of web site url data item i, freqiIt is right within the past period for user
Total access times (data are carried out through normalized during obtaining total access times) of web site url data item i,
sclickiThe time interval that web site url data item i is accessed apart from this, which is accessed, for user's last time (is obtaining total access times
During data are carried out through normalized), stimeiFor user to web site url data item i within the past period
Total access duration (data are carried out through normalized during obtaining total access times), collectiFor user couple
(collection is 1 to the collecting state of web site url data item i, is not collected as 0), subiIt is user to belonging to web site url data item i
(subscribe to is 1 to the subscription status of theme, is not subscribed to as 0), 0.6 is the weight of access times, and -0.2 is the power at access time interval
Weight (the interval the short, illustrates that interest value is higher, i.e. time interval and interest value is negatively correlated), 0.1 is the power of total access duration
Weight, 0.3 is the weight of collecting state, and 0.2 is the weight of subscription status.
Step 103. is ranked up the interest score according to descending sequence；
Step 104. filter out interest score come preceding 30 government website link data item segmented one by one, statistics institute
There are the participle and participle frequency of data item；
Step 105. is ranked up all participles according to the sequence of participle frequency from high to low, chooses participle frequency ranking
Interest vector of preceding 10 participle as user.
Calculate feature vector such as StepA1~StepA4 of website.
A1. all government website link data item are extracted, each data item is segmented；
A2. the participle of statistical data item and participle frequency, according to participle frequency sequence from high to low to it is all segment into
Row sequence；
A3. feature vector of 5 participle as data item before selection ranking；
A4. the feature vector of step A1-A3 to all data item obtained in data recommendation pond is repeated.
The above-mentioned user interest vector calculated separately and data item feature vector are subjected to similarity skill between the two
Art；Similarity is ranked up according to sequence from big to small, before similarity ranking 10 data item is chosen, by its corresponding political affairs
Mansion web site url recommends user.
Two, specific transactions are recommended
Specific transactions recommendation is to recommend relative specific transactions to target user, according to the place of working of target user
The specific informations such as area, operating mechanism, department, the authority of office, the particular traffic data for recommending it to be concerned about.The recommendation form it is main
Thought is: recommending pond, institute in data recommendation pond according to the user basic information of target user and user's static attribute locking data
Include is the data item according to specific transactions after the labelling classification of user's static attribute.According to operational area, operating mechanism, work
It is labelled to all business datums to make department, authority of office etc., is classified according to label to business, according to the static state of target user
Attribute removes matching business datum label, so that locking data recommends pond, then calculates the interest vector sum number of target user simultaneously
According to the feature vector for recommending particular traffic data in pond.Calculate the similarity of interest vector and feature vector to recommend similarity compared with
High particular traffic data.
The labelling classification of all business datums is given first.For example:
Stick first layer label to particular traffic data according to operational area: regional label (area), provincial data are just
Paste province, the data post province of city-level and city, the counties and districts, data post provinces and cities of counties and districts' grade.Label example: XX province, the city XX, XX province, XX province
The city YY, the area XX, the city XX, XX province, area YY, the city YY, XX province etc..For distinguishing area belonging to user, division can check particular traffic data
Range；
Second layer label: operating mechanism's label (agency) is sticked to particular traffic data according to operating mechanism.Label shows
Example: certain city's civil administration is then recommended when target user is the staff of the Bureau of Finance, city to it by the Bureau of Finance, city, city's public security, the tax bureau, city
The particular traffic data of office.
Third layer label: department's label (department) is sticked to particular traffic data according to department.Mark
Sign example: certain Bureau of Civil Affairs Information Center, certain Bureau of Civil Affairs office.
The 4th layer of label is sticked to particular traffic data according to the authority of office: the label of position power where user
(authority), belong to Hold sticker, according to actual recommendation need to decide whether use the label.
The static attributes such as operational area, operating mechanism, department, the authority of office of target user are obtained, according to target user
Static attribute matches the particular traffic data of above-mentioned labelling, and locking particular traffic data recommends pond.It should be noted that for
The data item of newest publication, due to apart from user's login time it is short in the behavioral data of target user there are no feedback information,
In order to guarantee to recommend to be not missed, using the data recommendation of recent renewal as override recommending data item: calculating data first push away
The issuing time of all data item in pond is recommended away from the time difference of target user's login time, according to ascending sequence to the time difference
It is ranked up, 3 data item is placed on the feature for not needing to calculate these data item 3 before recommendation list before retention time difference ranking
Vector recommends target user directly as newest particular traffic data.
Calculate interest vector such as step 111~step 115 of user.
Step 111. obtains user and accesses feedback information obtained to the data item in data recommendation pond, acquires feedback letter
User records the access of all data item in breath, and the access record in this example includes: that user (can in section in those years
To be one day, one week or one month) to the access duration of each data item, access times, (user works as at newest access time interval
The last interval accessed between the data time of preceding access time distance, is spaced more short, illustrates that interest value is higher), whether receive
(collection is 1, is not collected 0) and whether to subscribe to that (subscription be 1, is not subscribed to as 0), wherein it is a certain by adding up for accessing duration for hiding
Data item browsing is obtained total time in period, it should be noted that user's single browsing time is not more than 2 seconds
Data item, it will rejected from data recommendation pond, be not involved in calculating；
Whether step 112. pair access record to access duration, access times, newest access time interval, is collected and is
No subscription carries out accumulation calculating, obtains target user to the interest score of all data item in data recommendation pond；
Step 113. is ranked up the interest score according to descending sequence；
Step 114. filters out interest score and comes preceding 30 data item to be segmented one by one, counts preceding 30 all data item
Participle and participle frequency；
Step 115. is ranked up all participles according to the sequence of participle frequency from high to low, chooses participle frequency ranking
Interest vector of preceding 10 participle as user.
Calculate feature vector such as StepA1~StepA4 of 3 data item before recommendation list in data recommendation pond.
A1. extract recommendation list before 3 data item, each data item is segmented；
A2. the participle of statistical data item and participle frequency, according to participle frequency sequence from high to low to it is all segment into
Row sequence；
A3. feature vector of 5 participle as data item before selection ranking；
A4. step A1-A3 is repeated to obtaining the feature vector of 3 all data item before recommendation list in data recommendation pond.
The above-mentioned target user's interest vector calculated separately is carried out to data item feature vector between the two similar
Degree technology；Similarity is ranked up according to sequence from big to small, before similarity ranking 7 data item is chosen, is corresponded to
Government website Link Recommendation to user.
By specific transactions recommend user in total include 10 data item, by 3 newest specific transactions and 7 targets
The interested specific transactions composition of user.
Three, personalized recommendation
Personalized recommendation is to recommend relative business to target user.The master that personalized recommendation and specific transactions are recommended
It is similar for wanting thought: recommending pond according to the user basic information of target user and user's static attribute locking data, data push away
Data item all in system included in pond is recommended, i.e., by the specific transactions item of user affiliated function, common data item and outer net
Link composition.
The specific recommendation step that personalized recommendation and specific transactions are recommended is also, difference is: personalized recommendation
It is suitable for the user that new user is login system for the first time simultaneously, the field feedback recorded in behavioral data is sky.For
The government data intelligent recommendation method of new user will be described in detail by embodiment two.
As shown in Fig. 2, it is based on design identical with embodiment of the present invention method, the embodiment of the invention provides a kind of political affairs
Business data intelligence recommender system.Since system embodiment is substantially similar to embodiment of the method, so describe fairly simple, it is related
Place illustrates with reference to the part of embodiment of the method.System embodiment described below is only schematical.
The government data intelligent recommendation system of the present embodiment, comprising: obtain module 101, for obtain user basic information,
User's static attribute and user access record, and the user includes target user and sample of users；Data recommendation pond 102, is used for
Determine the scope of business recommended to the user, the data recommendation pond includes the data item of different business department, each business department
The data item of door is made of specific transactions item, common data item and the outer net link of the business department；Field feedback library
103, for storing user behavior data, record access behavior of the user to all data item；Computing module 104, comprising: interest
Vector calculation module 141, for calculating the interest vector of user；Feature vector computing module 142, for calculating the spy of data item
Levy vector；Similarity calculation module 143, for calculate similarity between target user and sample of users and interest vector with
The similarity of both feature vectors；Recommending module 105, for according to the interest vector of user and/or the feature vector of data item
To user's recommending data.
Embodiment two
The present embodiment passes through building for new user (field feedback recorded in the behavioral data of user is sky)
User model is calculated the distance between target user and sample of users, is asked using the proposed algorithm of the collaborative filtering based on user
Neighbour of the similar users as target user is obtained, the data requirements of target user is predicted using the interest of neighbour as module,
The data item for recommending it that may be concerned about to target user.
The government data intelligent recommendation method and system that the present embodiment is embodied below in conjunction with attached drawing 3 and attached drawing 4 into
Row detailed description.
Government data intelligent recommendation method as shown in Figure 3, the first registration information according to user in system are available
User basic information and user's static attribute, wherein user basic information includes: user name, age and gender etc., and user is static
Attribute includes: the static attributes such as operational area, operating mechanism, department and the authority of office of user.Further, continue to obtain and use
Family behavioral data judges whether the field feedback recorded in user behavior data is empty, if field feedback is sky,
That is the user user that is new registration, it is there are no generating corresponding behavioral data, then quiet according to the user basic information and user
State attribute finds neighbour, by neighbour to user's recommending data；Conversely, i.e. the user is common user, then user feedback is obtained
Data item in information uses and method pair described in embodiment one in combination with user basic information and user's static attribute
Common user is recommended.
For the government data intelligence of common user (field feedback recorded in the behavioral data of user is not empty)
Can process described in recommended method and embodiment one be it is the same, repeat no more in the present embodiment.In this example mainly in combination with
The government data intelligent recommendation method and system of the new user of attached drawing 3 and 4 pair is described in detail.
It is to its central idea for carrying out data item recommendation for new user: constructs user model, meter using sample of users
The similarity for calculating new user (i.e. target user) and user model, finds the neighbour of new user (i.e. target user), neighbour be with
The very high user of the similarities such as essential information, the static attribute of target user, while neighbour is also the common user in system,
Interest score by calculating neighbour obtains the interested data item of neighbour, and being estimated as new user (i.e. target user) may feel emerging
The data item of interest is recommended.As shown in figure 3, finding neighbour according to target user's essential information and user's static attribute, pass through
Following steps are realized:
The essential information and user's static attribute of step 201. collecting sample user constructs user model, and user model includes
Essential information feature vector and static attribute feature vector；
Step 202. assigns weight to the characteristic value of all feature vectors in the user model；
Step 203. obtains the essential information and user's static attribute of target user, obtains target user using user model
Essential information feature vector and static attribute feature vector；
Step 204. by the essential information feature vector of the target user and static attribute feature vector respectively with it is each
The essential information feature vector and static attribute feature vector of a sample of users are compared one by one；
Step 205. utilizes the similarity between the weight calculation target user and each sample of users；
Step 206. is ranked up similarity according to sequence from big to small, chooses before similarity ranking 10 sample use
Neighbour of the family as target user.
The specific implementation process of above-mentioned steps can be described as follows:
Sample of users includes all registered users in system, can be through user according to feature by building user model
Similarity is classified, and different communities of users is divided.User model provides calculating for the division of communities of users and accurate recommendation
Basis, communities of users partitioning standards user's self attributes feature, user model include that user basic information (Base) and user are quiet
Effective attribute combination of eigenvectors of state attribute (Static), wherein Base={ uid, age, gender }, indicates the base of user
This information eigenvector；Static={ area, agency, department, authority }, expression contain the work of user
Make the static attribute feature vector of the static attributes such as area, operating mechanism, department, the authority of office.Similarity judgement for example:
User model can be expressed as 5 dimensional vector U=(area, agency, department, age, gender), by target user with
Sample of users in user model is compared two-by-two, and the value range of dimension is 0 or 1, when value is 1, indicates feature phase
Together, indicate that feature is different when value is 0.Such as: there are sample of users A, sample of users B, sample of users C, compare vector
Uab=(1,0,1,1,1), the operational area of expression A and B, department, age are identical with gender, and operating mechanism is different；
Uac=(1,0,1,0,0) indicates that the operational area of A and C is identical with department, and operating mechanism, age and gender not phase
Together.
Using above-mentioned judgment method, by the essential information feature vector of target user and static attribute feature vector respectively with
The essential information feature vector of each sample of users and static attribute feature vector are compared one by one in user model, can be with
Multiple groups comparison vector is obtained, for convenient for for example, being only listed below four groups of comparison vectors in this example: U1=(1,1,1,1,1),
U2=(1,1,1,1,0), U3=(1,1,1,0,1), U4=(1,1,1,0,0), according to operational area, operating mechanism, work department
Door, the specific gravity of age and gender during similarity comparison, respectively to U=(area, agency, department, age,
Gender each characteristic value in) assigns weight, such as: Warea=0.3；Wagency=0.25；Wdepartment=0.2；
Wage=0.15Wgender=0.1 then utilizes the mistake of the similarity between weight calculation target user and each sample of users
Journey is expressed as follows:
Compare the similarity of vector U1=(1,1,1,1,1), S1=1*0.3+1*0.25+1*0.2+1*0.15+1*0.1=
1；
Compare the similarity of vector U2=(1,1,1,1,0), S2=1*0.3+1*0.25+1*0.2+1*0.15+0*0.1=
0.9；
Compare the similarity of vector U3=(1,1,1,0,1), S3=1*0.3+1*0.25+1*0.2+0*0.15+1*0.1=
0.85；
Compare the similarity of vector U4=(1,1,1,0,0), S4=1*0.3+1*0.25+1*0.2+1*0.15+1*0.1=
0.75……
By the above method, obtain the similarity in target user and user model between all sample of users, according to from
It arrives small sequence greatly to be ranked up similarity, neighbour of 10 sample of users as target user before selection similarity ranking.
After the neighbour for determining target user, by neighbour, to user's recommending data, the specific implementation steps are as follows:
Step 211. obtains sample of users in the feedback information of 10 neighbours respectively and records to the access of all data item, visits
Ask record include: user to the access duration of data item, access times, whether collect and whether subscribe to, it should be noted that it is right
It is not more than 2 seconds data item in user's single browsing time, will be removed and be not involved in calculating；
Step 212. is accumulated as an access note according to total access recording integrating that 10 neighbours access the same data item
Record, calculates the interest score of each data item；
Step 213. is ranked up interest score according to descending sequence；
Step 214. chooses before interest score ranking 10 data item, by its data recommendation to target user.
As shown in Fig. 4, it is based on design identical with embodiment of the present invention method, the embodiment of the invention provides a kind of political affairs
Business data intelligence recommender system.Since system embodiment is substantially similar to embodiment of the method, so describe fairly simple, it is related
Place illustrates with reference to the part of embodiment of the method.System embodiment described below is only schematical.
The government data intelligent recommendation system of the present embodiment, comprising: obtain module 201, for obtain user basic information,
User's static attribute and user access record, and the user includes target user and sample of users；Data recommendation pond 202, is used for
Determine the scope of business recommended to the user, the data recommendation pond includes the data item of different business department, each business department
The data item of door is made of specific transactions item, common data item and the outer net link of the business department；Field feedback library
203, for storing user behavior data, record access behavior of the user to all data item；User model constructs module 204,
Essential information and user's static attribute for collecting sample user construct user model, and wherein user model includes essential information
Feature vector and static attribute feature vector；Computing module 205, comprising: interest vector computing module 251, for calculating user
Interest vector；Feature vector computing module 252, for calculating the feature vector of data item；Similarity calculation module 253 is used
The similarity of similarity and interest vector and both feature vectors between calculating target user and sample of users；Recommend mould
Block 206, for according to the feature vector of the interest vector of user and/or data item to user's recommending data.
In the possible embodiment of other, government data intelligent recommendation method and system of the invention is additionally provided most
New recommendation form and most hot recommendation form, both recommendation forms both can be adapted for new user or common user can be used.
Wherein, newest recommendation is to recommend the data that are newly joined in advertisement data pond to user, and this recommendation form can be with
Solve the problems, such as that project is cold-started.The operation system that one completely new nobody accessed, in no user's history data and other
In the case where all data, data item is added into data recommendation pond, in order to avoid occur user may it is interested in it but because
The situation missed not know the presence of the data item carries out newest recommendation to the data item being newly added.Implement step
Are as follows:
Step 301. reads the issuing time that common data recommends all data item in pond；
Step 302. to the issuing time of data item according to by the short time to being sequentially ranked up for a long time；
Step 303. is chosen common data and is recommended in pond, issuing time ranking most preceding new 10 item data items (and when publication
Between nearest first 10) Xiang Suoyou user recommends.
Most hot recommendation form is according in a period of time, by the most concerned hot spot datas of user all in system, i.e. user
Access time, longest data item recommended user.What such recommendation form solved is that user is cold-started problem, mainly includes three
Kind of situation: 1: user registers logins for the first time, it is not known that oneself wants to see user's cold start-up problem when what data.2: as new user
User when distant is cold-started problem with old user's similarity distance.3: user does not know when oneself wanting what data to using
Most concerned data in other users nearest a period of time in the recommender system of family.Illustrate most hot recommendation form by taking week age as an example
Specific implementation step are as follows:
Step 401. obtained in this Monday to current time, when common data recommends all data item single references in pond
It is long；
Step 402. judges whether the single reference duration of each data item is greater than 2 seconds, is then considered as effectively if more than 2 seconds
Access, the data item will participate in temperature and calculate；It is considered as ineffective access if less than 2 seconds, which is not involved in temperature calculating；
Step 403. obtains total access times of the same data item within this Monday to current time, and add up all access time
Several single reference durations calculates total access duration within this Monday to current time of the data item；
Step 404. assigns weight to access times and total access duration respectively, passes through the heat of each data item of weight calculation
It spends (H), citing: H_i=0.8*freq_i+0.2*stime_i
Wherein, H_iFor the hot value of data item i, freq_i is total access of the data item i within this Monday to current time
Number, stime_i be total access duration of the data item i within this Monday to current time, 0.8 be access times weight, 0.2
For the weight for always accessing duration；
Step 405. presses the sequence of the temperature (H) of data item from big to small and carries out ranking；
10 data item is recommended to user before step 406. selection temperature ranking.
System, device or the module that above-described embodiment illustrates can specifically be realized, Huo Zheyou by computer chip or entity
Product with certain function is realized.It is a kind of typically to realize that equipment is computer.Specifically, computer for example can be cloud
In server, personal computer, laptop computer, smart phone, tablet computer, wearable device or these equipment
The combination of any equipment.
For convenience of description, it is divided into various modules when description apparatus above with function to describe respectively.Certainly, implementing this
The function of each module can be realized in the same or multiple software and or hardware when invention.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
The present invention be referring to according to the method for the embodiment of the present invention, the flow chart of system and computer program product and/or
Block diagram describes.It should be understood that each process that can be realized by computer program instructions in flowchart and/or the block diagram and/or
The combination of process and/or box in box and flowchart and/or the block diagram.It can provide these computer program instructions to arrive
General purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor to generate one
Machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for realizing flowing
The device for the function of being specified in journey figure one process or multiple processes and/or block diagrams one box or multiple boxes.
It should also be noted that, the terms "include", "comprise" or its any other variant are intended to nonexcludability
It include so that the process, method, commodity or the equipment that include a series of elements not only include those elements, but also to wrap
Include other elements that are not explicitly listed, or further include for this process, method, commodity or equipment intrinsic want
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including described want
There is also other identical elements in the process, method of element, commodity or equipment.
It will be understood by those skilled in the art that the embodiment of the present invention can provide as method, system or computer program product.
Therefore, complete hardware embodiment, complete software embodiment or embodiment combining software and hardware aspects can be used in the present invention
Form.It is deposited moreover, the present invention can be used to can be used in the computer that one or more wherein includes computer usable program code
The shape for the computer program product implemented on storage media (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
Formula.
The above is only a preferred embodiment of the present invention, it is noted that for the ordinary skill people of the art
Member, under the premise of not departing from the method for the present invention, can also make several improvement and supplement, these are improved and supplement also should be regarded as
Protection scope of the present invention.
Claims (10)
1. a kind of government data intelligent recommendation method, which is characterized in that the described method comprises the following steps:
User basic information and user's static attribute are obtained according to the registration information of user；
Recommend pond according to user's static attribute locking data；
User behavior data is acquired, the interest vector of user is calculated according to the user behavior data；
Extract the feature vector of each of data recommendation pond data item；
According to the similarity between the interest vector of the user and the feature vector of data item to user's recommending data.
2. government data intelligent recommendation method according to claim 1, which is characterized in that the user basic information packet
It includes: user name, age and gender；User's static attribute include: the operational area of user, operating mechanism, department and
The authority of office.
3. government data intelligent recommendation method according to claim 2, which is characterized in that described according to the user behavior
Data calculate the interest vector of user, comprising:
Judge whether the field feedback recorded in user behavior data is empty, if field feedback is sky, basis
User basic information and user's static attribute find neighbour, by neighbour to user's recommending data；Conversely, then obtaining user feedback
Data item in information, for calculating the interest vector of user.
4. government data intelligent recommendation method according to claim 3, which is characterized in that described according to user basic information
Neighbour is found with user's static attribute, comprising:
The essential information and user's static attribute of collecting sample user constructs user model, and the user model includes essential information
Feature vector and static attribute feature vector；
Weight is assigned to the characteristic value of all feature vectors in the user model；
The essential information and user's static attribute for obtaining target user, the essential information for obtaining target user using user model are special
Levy vector sum static attribute feature vector；
By the essential information feature vector of the target user and static attribute feature vector respectively with each sample of users
Essential information feature vector and static attribute feature vector are compared one by one；
Utilize the similarity between the weight calculation target user and each sample of users；
Similarity is ranked up according to sequence from big to small, 10 sample of users is used as target before selection similarity ranking
The neighbour at family.
5. government data intelligent recommendation method according to claim 3, which is characterized in that described to be pushed away by neighbour to user
Recommend data, comprising:
It obtains sample of users in the feedback information of 10 neighbours respectively to record the access of all data item, the access record packet
Include: whether access duration access times, newest access time interval, collected and subscribed to；
The interest score of each data item is calculated according to the access record；
The interest score is ranked up according to descending sequence；
The data item for choosing before the interest score ranking 10, by its data recommendation to target user.
6. government data intelligent recommendation method according to claim 3, which is characterized in that it is described calculate user interest to
Amount, comprising:
It acquires user in field feedback to record the access of all data item, the access record includes: access duration, visits
It asks number, newest access time interval, whether collect and whether subscribe to；
The interest score of all data item in field feedback is calculated according to the access record；
The interest score is ranked up according to descending sequence；
It filters out interest score and comes preceding 30 data item and segmented one by one, count the participle and participle frequency of all data item
Rate；
All participles are ranked up according to the sequence of participle frequency from high to low, choose before participle frequency ranking 10 participle work
For the interest vector of user.
7. government data intelligent recommendation method according to claim 1, which is characterized in that described to extract the data recommendation
The feature vector of each of pond data item, comprising:
A1. the data item extracted in the data recommendation pond is segmented；
A2. the participle and participle frequency for counting the data item, according to participle frequency sequence from high to low to it is all segment into
Row sequence；
A3. feature vector of 5 participle as the data item before selection ranking；
A4. the feature vector of step A1-A3 to all data item obtained in data recommendation pond is repeated.
8. government data intelligent recommendation method according to claim 6 or 7, which is characterized in that described according to the user
Interest vector and data item feature vector between similarity to user's recommending data, comprising:
Calculate the similarity of both the interest vector of target user and the feature vector of data item；
Similarity is ranked up according to sequence from big to small, before similarity ranking 10 data item is chosen, its data is pushed away
It recommends to target user.
9. a kind of government data intelligent recommendation system, which is characterized in that the system comprises:
Module is obtained, accesses record for obtaining user basic information, user's static attribute and user, the user includes target
User and sample of users；
Data recommendation pond, for determining the scope of business recommended to the user, the data recommendation pond includes different business department
Data item, the data item of each business department by the business department specific transactions item, common data item and outer net link-group
At；
Field feedback library records access behavior of the user to all data item for storing user behavior data；
Computing module, comprising:
Interest vector computing module, for calculating the interest vector of user；Feature vector computing module, for calculating data item
Feature vector；Similarity calculation module, for calculate similarity between target user and sample of users and interest vector with
The similarity of both feature vectors；
Recommending module, for according to the feature vector of the interest vector of user and/or data item to user's recommending data.
10. government data intelligent recommendation system according to claim 9, which is characterized in that the system also includes users
Model construction module, essential information and user's static attribute for collecting sample user construct user model, user's mould
Type includes essential information feature vector and static attribute feature vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910221130.0A CN110110221A (en) | 2019-03-22 | 2019-03-22 | Government data intelligent recommendation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910221130.0A CN110110221A (en) | 2019-03-22 | 2019-03-22 | Government data intelligent recommendation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110110221A true CN110110221A (en) | 2019-08-09 |
Family
ID=67484509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910221130.0A Pending CN110110221A (en) | 2019-03-22 | 2019-03-22 | Government data intelligent recommendation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110110221A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737694A (en) * | 2019-09-17 | 2020-01-31 | 中电万维信息技术有限责任公司 | Intelligent subscription method for government affair data resource directories |
| CN110852852A (en) * | 2019-11-15 | 2020-02-28 | 成都航天科工大数据研究院有限公司 | Industrial Internet product recommendation system and method |
| CN111027926A (en) * | 2019-11-28 | 2020-04-17 | 北明软件有限公司 | Multi-event package generation method, system and storage medium |
| CN111079009A (en) * | 2019-12-11 | 2020-04-28 | 中国地质大学（武汉） | User interest detection method and system for government map service |
| CN111191117A (en) * | 2019-12-11 | 2020-05-22 | 中国地质大学（武汉） | Accurate user interest detection method and system for government map service |
| CN111400474A (en) * | 2020-03-20 | 2020-07-10 | 腾讯云计算（北京）有限责任公司 | Intelligent customer service question and answer method and device for government affair service |
| CN111832945A (en) * | 2020-07-15 | 2020-10-27 | 中国科学技术大学智慧城市研究院(芜湖) | Government affair big data application maturity evaluation method and system |
| CN112015962A (en) * | 2020-07-24 | 2020-12-01 | 北京艾巴斯智能科技发展有限公司 | Government affair intelligent big data center system architecture |
| CN112053078A (en) * | 2020-09-15 | 2020-12-08 | 山东爱城市网信息技术有限公司 | System and implementation method for government affair service PGC community |
| CN112148983A (en) * | 2020-09-30 | 2020-12-29 | 神思电子技术股份有限公司 | Content updating and recommending method for tax industry |
| CN116975396A (en) * | 2023-09-25 | 2023-10-31 | 北京市大数据中心 | Intelligent recommendation method, system, equipment and storage medium for government service |
| CN117149859A (en) * | 2023-10-27 | 2023-12-01 | 中国市政工程华北设计研究总院有限公司 | Urban waterlogging point information recommendation method based on government user portrait |
| CN117473168A (en) * | 2023-12-27 | 2024-01-30 | 中电数据产业有限公司 | Intelligent feature recommendation method, device, equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102473183A (en) * | 2009-07-14 | 2012-05-23 | 索尼公司 | Content recommendation system, content recommendation method, content recommendation device, and information recording medium |
| CN102768685A (en) * | 2012-07-24 | 2012-11-07 | 杭州东方网升科技有限公司 | Content recommendation method based on keyword matching |
| CN107193829A (en) * | 2016-03-14 | 2017-09-22 | 百度在线网络技术（北京）有限公司 | Application program recommends method and apparatus |
| CN107391687A (en) * | 2017-07-24 | 2017-11-24 | 华中师范大学 | A kind of mixing commending system towards local chronicle website |
-
2019
- 2019-03-22 CN CN201910221130.0A patent/CN110110221A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102473183A (en) * | 2009-07-14 | 2012-05-23 | 索尼公司 | Content recommendation system, content recommendation method, content recommendation device, and information recording medium |
| CN102768685A (en) * | 2012-07-24 | 2012-11-07 | 杭州东方网升科技有限公司 | Content recommendation method based on keyword matching |
| CN107193829A (en) * | 2016-03-14 | 2017-09-22 | 百度在线网络技术（北京）有限公司 | Application program recommends method and apparatus |
| CN107391687A (en) * | 2017-07-24 | 2017-11-24 | 华中师范大学 | A kind of mixing commending system towards local chronicle website |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110737694A (en) * | 2019-09-17 | 2020-01-31 | 中电万维信息技术有限责任公司 | Intelligent subscription method for government affair data resource directories |
| CN110852852A (en) * | 2019-11-15 | 2020-02-28 | 成都航天科工大数据研究院有限公司 | Industrial Internet product recommendation system and method |
| CN111027926A (en) * | 2019-11-28 | 2020-04-17 | 北明软件有限公司 | Multi-event package generation method, system and storage medium |
| CN111027926B (en) * | 2019-11-28 | 2023-04-18 | 北明软件有限公司 | Multi-event package generation method, system and storage medium |
| CN111191117A (en) * | 2019-12-11 | 2020-05-22 | 中国地质大学（武汉） | Accurate user interest detection method and system for government map service |
| CN111079009A (en) * | 2019-12-11 | 2020-04-28 | 中国地质大学（武汉） | User interest detection method and system for government map service |
| CN111079009B (en) * | 2019-12-11 | 2023-05-26 | 中国地质大学（武汉） | User interest detection method and system for government map service |
| CN111400474A (en) * | 2020-03-20 | 2020-07-10 | 腾讯云计算（北京）有限责任公司 | Intelligent customer service question and answer method and device for government affair service |
| CN111400474B (en) * | 2020-03-20 | 2023-02-03 | 腾讯云计算（北京）有限责任公司 | Intelligent customer service question and answer method and device for government affair service |
| CN111832945A (en) * | 2020-07-15 | 2020-10-27 | 中国科学技术大学智慧城市研究院(芜湖) | Government affair big data application maturity evaluation method and system |
| CN111832945B (en) * | 2020-07-15 | 2023-05-23 | 长三角信息智能创新研究院 | Government affair big data application maturity evaluation method |
| CN112015962A (en) * | 2020-07-24 | 2020-12-01 | 北京艾巴斯智能科技发展有限公司 | Government affair intelligent big data center system architecture |
| CN112053078A (en) * | 2020-09-15 | 2020-12-08 | 山东爱城市网信息技术有限公司 | System and implementation method for government affair service PGC community |
| CN112148983B (en) * | 2020-09-30 | 2023-02-03 | 神思电子技术股份有限公司 | Content updating and recommending method for tax industry |
| CN112148983A (en) * | 2020-09-30 | 2020-12-29 | 神思电子技术股份有限公司 | Content updating and recommending method for tax industry |
| CN116975396A (en) * | 2023-09-25 | 2023-10-31 | 北京市大数据中心 | Intelligent recommendation method, system, equipment and storage medium for government service |
| CN117149859A (en) * | 2023-10-27 | 2023-12-01 | 中国市政工程华北设计研究总院有限公司 | Urban waterlogging point information recommendation method based on government user portrait |
| CN117149859B (en) * | 2023-10-27 | 2024-02-23 | 中国市政工程华北设计研究总院有限公司 | Urban waterlogging point information recommendation method based on government user portrait |
| CN117473168A (en) * | 2023-12-27 | 2024-01-30 | 中电数据产业有限公司 | Intelligent feature recommendation method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110221A (en) | Government data intelligent recommendation method and system | |
| Quijano-Sánchez et al. | Recommender systems for smart cities | |
| Majid et al. | A system for mining interesting tourist locations and travel sequences from public geo-tagged photos | |
| Zhao et al. | Activities, ringmaps and geovisualization of large human movement fields | |
| Tsai et al. | A location-item-time sequential pattern mining algorithm for route recommendation | |
| CN105210061B (en) | Tagged search result maintenance | |
| CN102667840A (en) | Context information utilizing systems, apparatus and methods | |
| WO2014056408A1 (en) | Information recommending method, device and server | |
| Cui et al. | Travel behavior classification: an approach with social network and deep learning | |
| CN108595461A (en) | Interest heuristic approach, storage medium, electronic equipment and system | |
| Lu et al. | Travel attractions recommendation with knowledge graphs | |
| CN116738066B (en) | Rural travel service recommendation method and device, electronic equipment and storage medium | |
| Wang et al. | Knowledge graph-based spatial-aware user community preference query algorithm for lbsns | |
| Husin et al. | News recommendation based on web usage and web content mining | |
| Ma et al. | Characterizing urban lifestyle signatures using motif properties in network of places | |
| Wang et al. | Time-aware user profiling from personal service ecosystem | |
| Chu et al. | TRSO: A tourism recommender system based on ontology | |
| Suresh Kumar et al. | Multi-ontology based points of interests (MO-POIS) and parallel fuzzy clustering (PFC) algorithm for travel sequence recommendation with mobile communication on big social media | |
| Vavpotič et al. | Using a market basket analysis in tourism studies | |
| Alghieth et al. | A map-based job recommender model | |
| Medury et al. | Pedestrian count expansion methods: bridging the gap between land use groups and empirical clusters | |
| Zheng et al. | TGTM: Temporal-geographical topic model for point-of-interest recommendation | |
| Xiong et al. | A situation information integrated personalized travel package recommendation approach based on TD-LDA model | |
| CN111460300B (en) | Network content pushing method, device and storage medium | |
| Sun et al. | Build smart campus using human behavioral data |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190809 |
|
| RJ01 | Rejection of invention patent application after publication |