CN110147498A - A kind of knowledge method for pushing, device and storage equipment, program product - Google Patents
A kind of knowledge method for pushing, device and storage equipment, program product Download PDFInfo
- Publication number
- CN110147498A CN110147498A CN201910419506.9A CN201910419506A CN110147498A CN 110147498 A CN110147498 A CN 110147498A CN 201910419506 A CN201910419506 A CN 201910419506A CN 110147498 A CN110147498 A CN 110147498A
- Authority
- CN
- China
- Prior art keywords
- knowledge
- user
- label
- tag set
- user tag
- 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
- 238000004590 computer program Methods 0.000 claims description 6
- 238000002716 delivery method Methods 0.000 claims description 2
- 230000036541 health Effects 0.000 description 49
- 206010020772 Hypertension Diseases 0.000 description 13
- 238000004364 calculation method Methods 0.000 description 10
- 238000010586 diagram Methods 0.000 description 6
- 201000010099 disease Diseases 0.000 description 6
- 208000037265 diseases, disorders, signs and symptoms Diseases 0.000 description 6
- 206010012601 diabetes mellitus Diseases 0.000 description 5
- 238000001356 surgical procedure Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000001631 hypertensive effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 206010019280 Heart failures Diseases 0.000 description 2
- 230000009471 action Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 208000019622 heart disease Diseases 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012827 research and development Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 108010001267 Protein Subunits Proteins 0.000 description 1
- 238000001467 acupuncture Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 208000029078 coronary artery disease Diseases 0.000 description 1
- 230000009193 crawling Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004400 mucous membrane Anatomy 0.000 description 1
- 208000010125 myocardial infarction Diseases 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 210000000664 rectum Anatomy 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 208000001072 type 2 diabetes mellitus Diseases 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
- 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)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- General Business, Economics & Management (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present application discloses a kind of knowledge method for pushing and device, obtains the user tag set of the user first, and each knowledge tag set of knowledge to be pushed.Wherein, user tag set can react the personal feature of the user, and knowledge tag set is used to reflect the classification of the knowledge to be pushed.Then, it according to the knowledge tag set of the user tag set of user and knowledge to be pushed, calculates user and is each pushed to user wait push the degree of correlation between knowledge, then by the knowledge to be pushed that the degree of correlation meets preset condition.Namely, before pushing knowledge to user, user is calculated with each wait push the correlation between knowledge, the biggish knowledge to be pushed of correlation is pushed to user, to realize accurate to user's push and there is personalized knowledge, the accuracy of knowledge push is improved.
Description
Technical field
The application Internet technology is related to field, and in particular to a kind of knowledge method for pushing, device and storage equipment, program
Product.
Background technique
In order to reinforce population health informatization, various regions combine actual conditions, have carried out related Huimin service item
Construction.As a content of Huimin service and intelligent Service, Health information service is provided to region resident, for realizing full people
Group, the health service management etc. of Life cycle are significant.
In the prior art, can the simple rule based on some fixations to user push relevant knowledge, for example, to whole
User push a certain knowledge of the age at 60 years old or more.But this knowledge push mode can not be directed to the need of different user
It asks and pushes more accurate, personalized knowledge.
Summary of the invention
In view of this, the embodiment of the present application provides a kind of knowledge method for pushing, device and storage equipment, program product, with
Realize the accurate and targeted knowledge to user's push.
To solve the above problems, technical solution provided by the embodiments of the present application is as follows:
A kind of knowledge method for pushing, which comprises
Obtain the user tag set of user and the knowledge tag set of knowledge to be pushed, the user tag set packet
At least one user tag is included, the knowledge tag set includes at least one knowledge label；
According to the knowledge tag set of the user tag set of the user and the knowledge to be pushed, the use is calculated
The degree of correlation at family and the knowledge to be pushed；
The knowledge to be pushed that the degree of correlation meets preset condition is pushed to the user.
In one possible implementation, the user tag set according to the user and described know wait push
The knowledge tag set of knowledge calculates the degree of correlation of the user Yu the knowledge to be pushed, comprising:
Calculate i-th of user tag and j-th of knowledge label in the knowledge tag set in the user tag set
Semantic similarity and semantic association degree；
According to the semanteme of each knowledge label in each user tag in the user tag set and the knowledge tag set
Similarity and semantic association degree determine the degree of correlation of the user Yu the knowledge to be pushed.
In one possible implementation, the method also includes:
Ontology model, the user tag and the knowledge mark are established according to the incidence relation of user tag and knowledge label
Label are the node in the ontology model, each one concept of the node on behalf or example.
In one possible implementation, it is described calculate in the user tag set i-th of user tag with it is described
The semantic similarity and semantic association degree of j-th of knowledge label in knowledge tag set, comprising:
J-th in node where judging in the user tag set i-th of user tag and the knowledge tag set
Node where knowledge label, if there are common father nodes；
If there is no common father node, by i-th of user tag and the knowledge mark in the user tag set
The semantic similarity zero setting of j-th of knowledge label in label set, and calculate in the user tag set i-th of user tag with
The semantic association degree of j-th of knowledge label in the knowledge tag set；
If there is common father node, i-th of user tag and the knowledge mark in the user tag set are calculated
The semantic similarity and semantic association degree of j-th of knowledge label in label set.
In one possible implementation, i-th of user tag and the knowledge in the user tag set are calculated
The semantic similarity of j-th of knowledge label in tag set, comprising:
J-th in node where calculating in the user tag set i-th of user tag and the knowledge tag set
Path length of the node on categorical attribute path apart where knowledge label, as i-th of use in the user tag set
The semantic distance of j-th of knowledge label in family label and the knowledge tag set；
The node set passed through according to node where i-th user tag in the user tag set to root node and
The node set that node where j-th of knowledge label passes through to root node in the knowledge tag set calculates user's mark
The semantic registration of i-th of user tag and j-th of knowledge label in the knowledge tag set in label set；
Calculate the level depth of node and the knowledge tally set where i-th of user tag in the user tag set
The difference of the level depth of node where j-th of knowledge label in conjunction, as i-th of user tag in the user tag set with
The level of j-th of knowledge label is poor in the knowledge tag set；
According to j-th of knowledge label in i-th of user tag in the user tag set and the knowledge tag set
Semantic distance, semantic registration and level it is poor, determine i-th of user tag and the knowledge in the user tag set
The semantic similarity of j-th of knowledge label in tag set.
In one possible implementation, i-th of user tag and the knowledge in the user tag set are calculated
The semantic association degree of j-th of knowledge label in tag set, comprising:
J-th in node where calculating in the user tag set i-th of user tag and the knowledge tag set
Path length of the node on relating attribute path apart where knowledge label；
According to j-th in node where i-th of user tag in the user tag set and the knowledge tag set
Path length of the node on relating attribute path apart where knowledge label determines i-th of use in the user tag set
The semantic association degree of j-th of knowledge label in family label and the knowledge tag set.
In one possible implementation, described according to each user tag in the user tag set and the knowledge
The semantic similarity and semantic association degree of each knowledge label in tag set determine the user and the knowledge to be pushed
The degree of correlation, comprising:
Calculate the weight of i-th of user tag and the weight of j-th of the knowledge label；
According to the semanteme of each knowledge label in each user tag in the user tag set and the knowledge tag set
In similarity and semantic association degree, the user tag set in the weight of each user tag and the knowledge tag set
The weight of each knowledge label determines the degree of correlation of the user Yu the knowledge to be pushed.
In one possible implementation, the weight of i-th of user tag in the user tag set is calculated, is wrapped
It includes:
Go out occurrence in the user tag set of each user according to i-th of user tag in the user tag set
Number, calculates the label frequency of i-th of user tag in the user tag set；
According to number of users and total number of users amount comprising i-th of user tag in the user tag set, calculate
The inverse label frequency of i-th of user tag in the user tag set；
According in the user tag set in the label frequency and the user tag set of i-th of user tag
The inverse label frequency of i user tag calculates the weight of i-th of user tag in the user tag set.
In one possible implementation, the weight of j-th of knowledge label in the knowledge tag set is calculated, is wrapped
It includes:
Go out occurrence in the knowledge tag set of each knowledge according to j-th of knowledge label in the knowledge tag set
Number, calculates the label frequency of j-th of knowledge label in the knowledge tag set；
According to knowledge quantity and knowledge total quantity comprising j-th of knowledge label in the knowledge tag set, calculate
The inverse label frequency of j-th of knowledge label in the knowledge tag set；
According in the label frequency of j-th knowledge label in the knowledge tag set and the knowledge tag set
The inverse label frequency of j knowledge label calculates the weight of j-th of knowledge label in the knowledge tag set.
A kind of knowledge driving means, described device include:
Acquiring unit, it is described for obtaining the user tag set of user and the knowledge tag set of knowledge to be pushed
User tag set includes at least one user tag, and the knowledge tag set includes at least one knowledge label；
First computing unit, for according to the user tag set of the user and the knowledge mark of the knowledge to be pushed
Label set, calculates the degree of correlation of the user Yu the knowledge to be pushed；
Push unit, the knowledge to be pushed for the degree of correlation to be met preset condition are pushed to the user.
A kind of computer readable storage medium is stored with instruction in the computer readable storage medium storing program for executing, works as described instruction
When running on the terminal device, so that the terminal device executes above-mentioned knowledge method for pushing.
A kind of computer program product, when the computer program product is run on the terminal device, so that the terminal
Equipment executes above-mentioned knowledge method for pushing.
It can be seen that the embodiment of the present application has the following beneficial effects:
Before pushing knowledge to user, the user tag set of the user is obtained first, and each knowledge to be pushed
Knowledge tag set.Wherein, user tag set can react the personal feature of the user, and knowledge tag set is for reflecting
The classification of the knowledge to be pushed.Then, according to the knowledge tag set of the user tag set of user and knowledge to be pushed, meter
It calculates user and is each pushed to use wait push the degree of correlation between knowledge, then by the knowledge to be pushed that the degree of correlation meets preset condition
Family.It is, calculating user with each wait push the correlation between knowledge, by correlation before pushing knowledge to user
Biggish knowledge to be pushed is pushed to user, to realize accurate to user's push and have personalized knowledge, improves knowledge
The accuracy of push.
Detailed description of the invention
Fig. 1 is a kind of application scenarios schematic diagram provided by the embodiments of the present application；
Fig. 2 is ontology model schematic diagram provided by the embodiments of the present application；
Fig. 3 is a kind of flow chart of knowledge method for pushing provided by the embodiments of the present application；
Fig. 4 a is a kind of flow chart for obtaining user and knowledge degree of correlation method to be pushed provided by the embodiments of the present application；
Fig. 4 b is a kind of categorical attribute path provided by the embodiments of the present application and relating attribute path schematic diagram；
Fig. 5 is the flow chart of another knowledge method for pushing provided by the embodiments of the present application；
Fig. 6 is a kind of structure chart of knowledge driving means provided by the embodiments of the present application.
Specific embodiment
In order to make the above objects, features, and advantages of the present application more apparent, with reference to the accompanying drawing and it is specific real
Mode is applied to be described in further detail the embodiment of the present application.
Inventor has found that traditional health knowledge pushes main basis in traditional health knowledge method for pushing research
Preset simple rule is pushed, can not be according to essential information of user itself, such as disease condition, surgery situation etc.
Information carries out personalized push, influences the accuracy of push.
Based on this, the embodiment of the present application provides a kind of knowledge method for pushing, specifically, before carrying out knowledge push,
The user tag set of user is obtained first, which may be constructed the residents ' health portrait of the user, and characterization should
The personalization of user.Then, the knowledge tag set that each knowledge to be pushed is obtained from health knowledge library, according to the use of user
The knowledge tag set of family tag set and each knowledge to be pushed calculates the degree of correlation of user with each knowledge to be pushed.Again
Judge user and each wait push whether the degree of correlation between knowledge meets preset condition, so as to will meet preset condition wait push away
It send knowledge to be pushed to the user, to realize the personalized push of health knowledge, improves the accuracy of push.
Knowledge method for pushing provided by the embodiments of the present application for ease of understanding, referring to Fig. 1, which is that the embodiment of the present application mentions
A kind of application scenarios schematic diagram supplied.Wherein, knowledge method for pushing provided by the embodiments of the present application can be applied to server 20
In.
The user tag set of a user is obtained in library in practical application, server 20 can draw a portrait from residents ' health
And the knowledge tag set of every knowledge to be pushed is obtained from health knowledge library, and according to the user tag set of the user
And the knowledge tag set of knowledge to be pushed, calculate the degree of correlation of the user Yu every knowledge to be pushed.Finally, by the degree of correlation
The knowledge to be pushed for meeting preset condition is pushed to the corresponding client 10 of user.
It will be understood by those skilled in the art that block schematic illustration shown in FIG. 1 is only that presently filed embodiment can be
An example being wherein achieved.The scope of application of the application embodiment is not limited by any aspect of the frame.
It should be noted that client 10 can be carried on terminal, which can be existing, researching and developing or incites somebody to action
Come it is researching and developing, can pass through it is any type of wiredly and/or wirelessly connect (for example, Wi-Fi, LAN, honeycomb, coaxial cable etc.)
Any user equipment interacted, including but not limited to: existing, researching and developing or research and development in the future intelligence are wearable to be set
Standby, smart phone, non-smart mobile phone, tablet computer, laptop PC, desktop personal computer, minicomputer,
Medium-size computer, mainframe computer etc..Presently filed embodiment is unrestricted in this regard.It is also to be noted that
In the embodiment of the present application server 20 can be it is existing, researching and developing or in the future research and development, push can be provided a user
One example of the equipment of service.Presently filed embodiment is unrestricted in this regard.
The technical solution of the application for ease of understanding will first carry out the preparation for implementing the knowledge method for pushing below
Illustrate, is below illustrated the foundation respectively to residents ' health portrait library and health knowledge library.
It in the present embodiment, may include the residents ' health portrait of resident in certain regional scope in residents ' health portrait library,
Wherein, residents ' health portrait is the information overall picture that each resident is taken out by resident information labeling, to obtain each residence
The individualized feature of the people, especially health characteristics.Residents ' health portrait may include user basic information, disease condition, hand
Art situation etc..Wherein, resident's essential information, disease condition and surgery situation can be embodied by user tag, the use
Family label is that standard label has following two important feature, first is that semantization, facilitates the meaning for understanding each label；Second is that short
Text only indicates a kind of meaning, is convenient for computer extraction standard information.
In practical applications, user tag may include the essential information of user, such as identification card number, gender, age, the people
Race, marital status etc. can also include the disease condition of user, can be with for example, type-II diabetes, hypertension, coronary heart disease etc.
Surgery situation including user, for example, bypass surgery, mucous membrane of rectum ring cutting operation etc..
In specific implementation, the essential information of user, disease condition and surgery situation can be flat from area health information
Platform, medical center, health control mechanism, other government affairs departments etc. obtain, and then establish every user's according to the data of acquisition
Residents ' health portrait, to obtain residents ' health portrait library.
Health knowledge library may include the diversified health knowledge for pushing to user, wherein health knowledge
Type can be the diversified forms such as text, picture, audio, video.When to health knowledge library typing health knowledge, need to strong
Kang Zhishi labeling, the as health knowledge add knowledge label, push the health knowledge to realize to user.Wherein, know
Label is known for indicating correlation type corresponding to this health knowledge, for example, disease type (diabetes, hypertension, heart disease
Deng), easily ill crowd (such as women, the elderly), so that health knowledge is pushed to corresponding situation according to knowledge label
User.It is understood that each health knowledge may include one or more labels, for example, health knowledge
Knowledge label both includes heart disease, myocardial infarction or may include male etc., so that the health knowledge is carried out comprehensive push away
It send.Wherein, the health knowledge in health knowledge library can obtain through a variety of ways, for example, crawling from healthy class website, buy
Existing knowledge base is simultaneously imported, expert's craft typing etc..
In specific implementation, the knowledge label of the user tag of user and health knowledge can be obtained from ontology model,
It may include all user tag and knowledge label in the ontology model, which can not only embody between label
Semantic Similarity can also embody the incidence relation between label, ontology model schematic diagram as shown in Figure 2, the ontology model
In may include various user tags and knowledge label, establish resident draw a portrait library when, use can be obtained from ontology model
User tag corresponding to family；When establishing health knowledge library, it can obtain from ontology model and know corresponding to health knowledge
Know label.It should be noted that user tag and knowledge label can be same label, for example, the user tag packet of certain user
Hypertension is included, the knowledge label of certain health knowledge also includes hypertension.
It should be noted that Fig. 2 is only the schematic diagram of ontology model, the specific structure of ontology model and composition are not carried out
It limits, when actually establishing ontology model, which may include various user tags and knowledge label.
Based on the description of above-mentioned basic knowledge, carried out below in conjunction with method of the attached drawing to the knowledge push that the application proposes
Explanation.
Referring to Fig. 3, which is a kind of flow chart of knowledge method for pushing provided by the embodiments of the present application, as shown in figure 3, should
Method may include:
S301: the user tag set of user and the knowledge tag set of knowledge to be pushed are obtained.
In the present embodiment, after server, which is established, completes resident portrait library and health knowledge library, it can draw a portrait for resident
Every user in library carries out the push of health knowledge.To realize personalized push, before push, the use of user is obtained first
The knowledge tag set of family tag set and every knowledge to be pushed.Wherein, user tag set includes at least one user
Label, knowledge tag set include at least one knowledge label.
In specific implementation, the user tag set of every user generally includes multiple user tags, for example, gender mark
Label, age label, marital status label etc., convenient for subsequent descriptions, to indicate the user tag set of user with T (u), T (u)=
T1, t2 ..., ti ..., tm, m be the corresponding user tag number of active user；T (k) indicates the knowledge mark of knowledge to be pushed
Label set, T (k)=t1, t2 ..., tj ..., tn, n is the current corresponding knowledge label number of knowledge to be pushed.
S302: according to the knowledge tag set of the user tag set of user and knowledge to be pushed, calculate user with to
Push the degree of correlation of knowledge.
S303: the knowledge to be pushed that the degree of correlation meets preset condition is pushed to user.
In the present embodiment, after the user tag set for obtaining user and the knowledge tag set wait push knowledge,
According to user tag set and knowledge tag set, the user and every are calculated wait push the degree of correlation between knowledge, thus
Obtain the user and a plurality of wait push the degree of correlation between knowledge.Then, judge whether each degree of correlation meets preset condition, such as
Fruit meets, then corresponding knowledge to be pushed is pushed to user.Wherein, preset condition can be specific relevance threshold, when
When calculating the degree of correlation obtained more than or equal to default relevance threshold, corresponding knowledge to be pushed is pushed to user, thus
It realizes to user's push and the biggish health knowledge of its own correlation, realizes personalized push.
In specific implementation, it is calculated and is used according to the knowledge tag set of the user tag set of user and knowledge to be pushed
The degree of correlation at family and knowledge to be pushed is specifically as follows and is combined each user tag and each knowledge label, obtains more
Group label pair, calculate the degree of correlation of every group of label pair, then according between all labels pair the degree of correlation obtain user with wait push away
Send the degree of correlation of knowledge.Wherein, it will be illustrated in subsequent embodiment about calculating user and the degree of correlation of knowledge to be pushed.
Through the foregoing embodiment it is found that obtaining the user tag set of the user first before pushing knowledge to user,
And each knowledge tag set of knowledge to be pushed.Wherein, user tag set can react the personal feature of the user, know
Know the classification that tag set is used to reflect the knowledge to be pushed.Then, know according to the user tag set of user and wait push
The knowledge tag set of knowledge calculates user and each meets preset condition wait push the degree of correlation between knowledge, then by the degree of correlation
Knowledge to be pushed be pushed to user.It is, before pushing knowledge to user, calculate user and each knowledge to be pushed it
Between correlation, the biggish knowledge to be pushed of correlation is pushed to user, to realize accurate to user's push and have
Property knowledge, improve knowledge push accuracy.
In practical applications, the embodiment of the present application provides a kind of total according to user tag set and knowledge tally set
The specific implementation for calculating user and the knowledge degree of correlation to be pushed, is illustrated the calculation method below in conjunction with attached drawing.
A referring to fig. 4, the figure are a kind of acquisition user provided by the embodiments of the present application and knowledge degree of correlation method to be pushed
Flow chart, as shown in fig. 4 a, this method may include:
S401: the language of i-th of user tag and j-th of knowledge label in knowledge tag set in user tag set is calculated
Adopted similarity.
In the present embodiment, any one in any one user tag and knowledge tag set for user tag collection
Knowledge label, by its combination of two, to calculate the semantic similarity between i-th of user tag and j-th of knowledge label.Its
In, the value of i is 1 to m, and m is the number of user tag in user tag set, and the value of j is 1 to n, and n is knowledge tally set
The number of knowledge label in conjunction.It is similar to the semanteme of tj in knowledge tag set T (k) to calculate ti in user tag set T (u)
Degree.Wherein, the semantic similarity of user tag and knowledge label refers to two labels in similitude semantically.
Based on the above embodiment it is found that user tag and knowledge label can be obtained from ontology model, and ontology mould
Type can embody the semantic relation between label.Therefore, calculate the semantic similarity between user tag and knowledge label it
Before, it can establish ontology model.Specifically, establishing ontology model, user according to the incidence relation of user tag and knowledge label
Label and knowledge label are a node in ontology model, one concept of each node on behalf or example.That is, according to all
Label establishes the ontology model, which can embody the semantic relation between each label, and label can be marked for user
Label or knowledge label, the node in ontology model can be user tag, or knowledge label.
Wherein, example can be understood as the occurrence of some concept, such as node " gender " is a concept, the node with
It is node " male " and node " female " respectively that two subordinate's nodes, which are connected, and " male " and " female " is the example of " gender " this concept.
Each user tag is a node in ontology model, and each knowledge label is a node in ontology model, Mei Geyong
Family label or knowledge label can be concept or example.That is, the ontology model is considered as a hierarchical tree, user tag and know
Knowing label is a node in tree.
In practical applications, the semantic similarity directly calculated between user tag and knowledge label may be stranded very much
It is difficult, it is generally the case that first to calculate the distance between two labels, and then be converted to semantic similarity.It is understood that in base
When ontology model calculates the distance between two labels, it is thus necessary to determine that it whether there is common father node between two labels,
If there is no father node, i.e., there is no from user tag to the path of knowledge label, also can not just obtain between two labels
Distance.Such as in Fig. 2, there are common father node " people " between user tag male and knowledge label diabetes B, that is, exist
Path from user tag male to diabetes B；The corresponding father node of user tag male is " people ", and knowledge label acupuncture is corresponding
Father node be " medicine ", common father node is not present therebetween, i.e., there is no from user tag male to knowledge label needle
The path of moxibustion.
Therefore, the language of i-th of user tag and j-th of knowledge label in knowledge tag set in user tag set is calculated
Adopted similarity and semantic association degree may include: to judge node and knowledge where i-th of user tag in user tag set
Node where j-th of knowledge label in tag set, if there are common father nodes.If there is no common father node,
By the semantic similarity zero setting of j-th of knowledge label in i-th of user tag in user tag set and knowledge tag set, and
Calculate the semantic association degree of i-th of user tag and j-th of knowledge label in knowledge tag set in user tag set.If
There are common father node, i-th of user tag and j-th of knowledge label in knowledge tag set in user tag set are calculated
Semantic similarity and semantic association degree.
In practical application, if i-th of user tag and j-th of knowledge label are calculated there are common father node
Semantic similarity between i-th of user tag and j-th of knowledge label.In specific implementation, the embodiment of the present application provides
A method of the semantic similarity calculating user tag and knowledge label can specifically be realized by following steps:
1) node where i-th of user tag and j-th of knowledge mark in knowledge tag set in user tag set are calculated
Path length of the node on categorical attribute path apart where label, as i-th of user tag in user tag set with know
Know the semantic distance of j-th of knowledge label in tag set.
In the present embodiment, it is first determined node where each user tag and the node where each knowledge label, so
Path of the node on categorical attribute path apart where node where calculating i-th of user tag afterwards and j-th of knowledge label
Length, thus using the path length as the semantic distance of i-th user tag and j-th of knowledge label.Wherein, categorical attribute
Path refers in ontology model that the path there are the path of hyponymy (i.e. classification relation), between two neighboring node is 1.
As in Fig. 4 b 1. shown in path be categorical attribute path, diagnosis there are hyponymies with diabetes B.For example, in Fig. 4 b
User tag is diabetes B, and knowledge label is hypertension, then the path length between two labels is 4.In practical application
When, it can indicate that i-th user tag ti to j-th knowledge label tj paths traversed is long using Distance (ti, tj)
Degree.
2) node set passed through according to node where i-th user tag in user tag set to root node and know
Know the node set that node where j-th of knowledge label passes through to root node in tag set, calculates i-th in user tag set
The semantic registration of j-th of knowledge label in a user tag and knowledge tag set.
In the present embodiment, the node set that node where obtaining i-th of user tag first is passed through to root node, and
The node set that node where j-th of knowledge label is passed through to root node, then, according to the corresponding node set of user tag
Node set corresponding with knowledge label calculates the semantic registration of i-th user tag and j-th of knowledge label.
In specific implementation, the node set and knowledge mark that the node where obtaining user tag is passed through to root node
Node where label to root node by node set after, available two node sets there are the node number of intersection,
And the node number of the union of two node sets.Again by the corresponding node number of intersection divided by the corresponding node of union
Number, obtained quotient is the semantic registration of i-th of user tag Yu j-th of knowledge label.
For example, i-th of user tag ti is up to the node set that root node is passed through from present node
NodeSet (ti), j-th of knowledge label tj reach up to the node set that root node is passed through from present node and are
NodeSet (tj), the intersection of two node sets are NodeSet (ti) ∩ NodeSet (tj)；The union of two node sets is
NodeSet (ti) ∪ NodeSet (tj), then the semantic registration of i-th of user tag and j-th of knowledge label is (NodeSet
(ti)∩NodeSet(tj))/(NodeSet(ti)∪NodeSet(tj))。
3) the is calculated in user tag set in the level depth of node where i-th user tag and knowledge tag set
The difference of the level depth of node where j knowledge label, as i-th of user tag and knowledge tally set in user tag set
The level of j-th of knowledge label is poor in conjunction.
In the present embodiment, the level depth of node where each user tag in user tag set can be obtained first,
And in knowledge tag set node where each knowledge label level depth, that is, determine user tag and knowledge respectively
Hierachy number where label in ontology model.When the level depth for determining i-th of user tag place node and know for j-th
After the level depth of node where knowing label, two user tag level depth differences are calculated, so that it is poor to obtain level.Wherein, it uses
The level depth of node where family label or knowledge label can for ontology model is used as when hierarchical tree, label place node
Depth, wherein the level depth of node where root node is 1 in ontology model.
In specific implementation, the level depth of node, Level (tj) table where Level (ti) indicates i-th of user tag
The level depth of node where showing j-th of knowledge label, | Level (ti)-Level (tj) | indicate user tag ti and knowledge mark
The level for signing tj is poor.
It should be noted that in the present embodiment, user tag and knowledge label can be retouched based on ontology model
It states, ti may be a concept in ontology model, it is also possible to for an example in ontology model.Similarly, tj may be this
A concept in body Model, it is also possible to for an example in ontology model.It, can be with when i-th user tag is concept
Node node where its own being determined as where i-th of user tag；It, can be with when i-th user tag is example
The node its corresponding concept being determined as where i-th of user tag.Similarly, if when j-th of knowledge label is concept,
The node that node where its own can be determined as where j-th of knowledge label；When j-th of knowledge label is example,
The node that its corresponding concept can be determined as where j-th of knowledge label.Section where determining i-th of user tag
After node where point and j-th of knowledge label, path length of two labels on categorical attribute path apart, language are calculated
Adopted registration and level are poor.
4) according to the semanteme of j-th of knowledge label in i-th of user tag in user tag set and knowledge tag set
Distance, semantic registration and level are poor, determine i-th of user tag and jth in knowledge tag set in user tag set
The semantic similarity of a knowledge label.
In the present embodiment, when determining in user tag set j-th of knowledge in i-th of user tag and knowledge tag set
After the semantic distance of label, semantic registration and level difference, i-th of user tag and jth are determined according to above three parameter
The semantic similarity of a knowledge label.
In specific implementation, for any one user tag in user tag set with it is any in knowledge tag set
One knowledge label is formed by combination, and it is similar to the semanteme of knowledge label user tag can be obtained by the above method
Degree.Semantic similarity is obtained in practical application, can calculate by formula (1):
Wherein, α, β, γ are respectively pre-set weight coefficient, and specific value can be set according to the actual situation.
When i-th of user tag is identical as j-th of knowledge label, corresponding semantic similarity is 1；When i-th user tag with
When jth knowledge label difference, calculated between two labels according to acquired path length, semantic registration and level difference
Semantic similarity.
S402: the language of i-th of user tag and j-th of knowledge label in knowledge tag set in user tag set is calculated
The adopted degree of association.
In the present embodiment, to obtain the degree of correlation between user and knowledge label, it can also calculate in user tag set
The semantic association degree of each knowledge label in each user tag and knowledge tag set.Wherein, semantic association degree is for reflecting
Incidence relation (non-hyponymy) in ontology model between two concepts.For example, user tag is hypertension, knowledge label
For hypertensive cardiopathy heart failure, the non-hyponymy of the two, but there are correlation, hypertension can cause hypertension
Property Heart Failure.
In specific implementation, the side of a kind of calculating user tag and the semantic association degree of knowledge label is present embodiments provided
Method specifically includes:
1) node where i-th of user tag and j-th of knowledge mark in knowledge tag set in user tag set are calculated
Path length of the node on relating attribute path apart where label.
In the present embodiment, it is first determined node where node where user tag and knowledge label, then in relevance
Determined on path two labels path length among the nodes.In practical application, can with ShortestPath (ti,
Tj the shortest path length from ti to tj) is indicated, when ti, tj are not connected to, the value of ShortestPath (ti, tj) is ∞.Its
In, relating attribute path refers to that there are the paths of incidence relation (non-hyponymy) in ontology model.Such as 2. institute in Fig. 4 b
Show that path is relating attribute path, hypertension and hypertensive cardiopathy are same level-one, there is incidence relation therebetween.Example
Such as, user tag is hypertension in Fig. 4 b, and knowledge label is hypertensive cardiopathy, then path length between the two is 1.
It should be noted that node where user tag can be divided into two kinds of situations, when user tag is example, by this
The corresponding concept of example is determined as the node where the user tag, when user tag is concept, by the section where its own
Point is determined as the node where user tag.Similarly, when knowledge label is example, the corresponding concept of the example is determined as this
Node where its own is determined as where knowledge label by the node where knowledge label when knowledge label is concept
Node.
2) according to j-th of knowledge mark in node where i-th of user tag in user tag set and knowledge tag set
Path length of the node on relating attribute path apart where label, determines in user tag set i-th of user tag and knows
Know the semantic association degree of j-th of knowledge label in tag set.
In the present embodiment, node is being associated with node where j-th of knowledge label where calculating i-th of user tag
When path length on tree path, the semanteme of i-th of user tag Yu j-th of knowledge label is determined according to above-mentioned path length
The degree of association.
In specific implementation, it can be calculated and be obtained by formula (2):
Wherein, λ is pre-set weight coefficient, and specific value can be determined according to the actual situation.It is used when i-th
When family label ti and j-th of knowledge label tj identical or of equal value, the semantic association degree of the two is 1；When i-th user tag ti with
When j-th of knowledge label tj non-equivalence, the semanteme of the two is determined using path length of two labels on relating attribute path
The degree of association.When ti, tj are not connected to, the value of ShortestPath (ti, tj) is that the semantic association degree of ∞, ti and tj are 0.
S403: similar to the semanteme of knowledge label each in knowledge tag set according to user tag each in user tag set
Degree and semantic association degree, determine the degree of correlation of user Yu knowledge to be pushed.
In the present embodiment, when the semantic similarity for obtaining each user tag Yu each knowledge label by S401 and S402
And after semantic association degree, the Technique Using Both Text degree of association of i-th of user tag Yu j-th of knowledge label can be determined.Then, root
According to the Technique Using Both Text degree of association of each user tag and each knowledge label, determine that user is related to currently knowledge to be pushed
Degree.
In specific implementation, it can be calculated by formula (3) and obtain the comprehensive of i-th of user tag and j-th knowledge label
Close semantic association degree:
Sim_Rel(ti, tj)=Sim (t, tj)+Rel(ti, tj)-Sim (t, t) × Rel (ti, tj) (3)
Wherein, Sim_Rel (t, tj) indicate i-th user tag and j-th of user tag the Technique Using Both Text degree of association,
Sim(ti, tj) indicate i-th user tag and j-th of knowledge label semantic similarity, Rel (ti, tj) indicate i-th of user
The semantic association degree of label and j-th of knowledge label.
It, can be with for each knowledge label in each user tag in user tag set and knowledge tag set
The Technique Using Both Text degree of association for obtaining each user tag and each knowledge label is calculated by formula (3), then by all synthesis
Semantic association degree is added, and obtains the degree of correlation of user and knowledge to be pushed.Specifically can by formula (4) obtain user with wait push away
Send the degree of correlation of knowledge.
Wherein, p (u, k) indicates the degree of correlation between user u and knowledge k to be pushed, Sim_Rel (ti, tj) indicate i-th
User tag ti, the Technique Using Both Text degree of association with j-th of knowledge label tj.
The method provided through this embodiment can calculate and obtain user and every wait push the degree of correlation between knowledge,
So as to the degree of correlation according to user and wait push between knowledge determine whether should knowledge be pushed be pushed to user, to realize
The higher knowledge to be pushed of the degree of correlation is pushed to user, improves the accuracy and personalized service of push.
In a kind of possible implementation of the embodiment of the present application, to embody any one user tag for the weight of user
The property wanted and personalization level propose a kind of to calculate each user tag for the power of user based on word frequency weight TF-ITF algorithm
Weight.Similarly, know to embody knowledge label and the importance degree of knowledge to be pushed and the knowledge label being waited pushing for this
The novel degree of knowledge calculates each knowledge label for the weight of knowledge to be pushed based on word frequency weight TF-ITF algorithm, so as to
User is calculated with when pushing the degree of correlation of knowledge, is calculated in conjunction with two weighted values.
For ease of understanding according to the weight of user tag and the weight calculation user of knowledge label in the embodiment of the present application
With the specific implementation of the degree of correlation of knowledge to be pushed, the realization is illustrated below in conjunction with attached drawing.
Referring to Fig. 5, which is the flow chart of another knowledge method for pushing provided by the embodiments of the present application, as shown in figure 5,
This method may include:
S501: the weight of i-th of user tag in user tag set is calculated.
In the present embodiment, for user tag set corresponding to active user, calculate each in the user tag set
The weight of user tag, will pass through the importance and personalization level that the weight embodies each label for user.
In specific implementation, it the present embodiment provides a kind of implementation for calculating user tag weight, can specifically include
Following steps:
1) occurrence is gone out in the user tag set of each user according to i-th of user tag in user tag set
Number calculates the label frequency of i-th of user tag in user tag set.
In the present embodiment, for any user label in user tag set corresponding to active user, the use is obtained
Label frequency of occurrence in the user tag set of each user in family includes the number of users of the user tag.Then, according to this
The label frequency of frequency of occurrence calculating active user's label.
In practical application, can obtain the label frequency of user tag by following two calculation, one is first
First, obtain resident draw a portrait all labels occur included by all users in library number and；Then, active user's label is existed
Number that frequency of occurrence in the user tag set of each user occurs divided by all user tags and, using its quotient as working as
The label frequency of preceding user tag.In specific implementation, formula (5) be can use:
Wherein, U-TFtiIndicate the label frequency of i-th of user tag, pb-ntiIndicate i-th of user tag in each use
Frequency of occurrence in the user tag set at family, pb_n indicate all user tags included by all users in resident's portrait library
The number of appearance and.
Another kind is, firstly, obtaining the corresponding frequency of occurrence of user tag that frequency of occurrence is most in resident's portrait library；So
Afterwards, by active user's label the frequency of occurrence in the user tag set of each user divided by resident draw a portrait library in frequency of occurrence
The corresponding frequency of occurrence of most user tags, using its quotient as the label frequency of active user's label.In specific implementation,
It can use formula (6):
Wherein, U-TFtiIndicate the label frequency of i-th of user tag, pb-ntiIndicate i-th of user tag in each use
Frequency of occurrence in the user tag set at family, max (pb_nt) indicate the user tag that frequency of occurrence is most in resident's portrait library
The frequency of occurrence of t.
Each user tag in user tag set is obtained in practical application, can calculate by formula (5) or (6)
Label frequency.
2) it according to number of users and total number of users amount comprising i-th of user tag in user tag set, calculates and uses
The inverse label frequency of i-th of user tag in the tag set of family.
In the present embodiment, obtaining includes use included in the number of users and resident's portrait library of i-th of user tag
Then family total quantity according to number of users and total number of users amount including i-th of user tag, calculates i-th of user tag
Inverse label frequency.
In practical application, if only characterizing the user tag with label frequency to the weight of the user, when a certain use
Family label resident draw a portrait library in frequency of occurrence it is less when, the less user tag of the frequency of occurrence for user weight compared with
It is small.And sometimes, the less user tag of frequency of occurrence can more embody the individual demand of user, therefore, for for embodying user
Property, the inverse label frequency for calculating user tag is also needed, the label frequency to avoid certain user tag is smaller and cannot reflect residence
The individual demand of the people.
In specific implementation, the inverse label frequency of i-th of user tag can be calculated by formula (7):
Wherein, U-ITFtiIndicate the inverse label frequency of i-th of user tag, sum_u indicates total number of users amount.sum_uti
Indicate the number of users comprising i-th of user tag.As it can be seen that denominator is bigger, U-ITF when a user tag is more commontiMore
It is small, i.e., closer to 0.In order to avoid denominator is 0 (i.e. all resident's portraits is not all comprising user tag), denominator will add 1, log
Expression takes logarithm to obtained value.
In practical application, can by formula (7) calculate obtain user tag set in each user tag it is inverse
Label frequency.
3) according in user tag set in the label frequency and user tag set of i-th of user tag i-th use
The inverse label frequency of family label calculates the weight of i-th of user tag in user tag set.
In the present embodiment, after the label frequency for obtaining i-th of user tag and inverse label frequency, label frequency is utilized
And inverse label frequency calculates the weight of i-th of user tag.When specific implementation, it may refer to formula (8):
Wherein, wti,uIndicate the weight of i-th of user tag, U-TFtiIndicate the label frequency of i-th of user tag, U-
ITFtiIndicate inverse label frequency.
It is understood that if only calculating the weight of i-th of user tag with label frequency, when the power of user tag
When weight is larger, the user tag can be embodied for the importance of user, but cannot reflect the individual demand of user, therefore,
Label frequency is adjusted using inverse label frequency, avoids the larger individual demand that cannot reflect resident of user tag weight.
S502: the weight of j-th of knowledge label in calculation knowledge tag set.
In the present embodiment, for currently wait push knowledge tag set corresponding to knowledge, in calculation knowledge tag set
The weight of j-th of knowledge label embodies each knowledge label for the importance journey of the knowledge to be pushed will pass through the weight
Degree, and embody the novel degree of the knowledge label.
In specific implementation, the implementation method for present embodiments providing a kind of calculation knowledge label weight, specifically can wrap
Include following steps:
1) occurrence is gone out in the knowledge tag set of each knowledge according to j-th of knowledge label in knowledge tag set
It counts, the label frequency of j-th of knowledge label in calculation knowledge tag set.
In the present embodiment, for currently wait push any knowledge label in knowledge tag set corresponding to knowledge, obtaining
Knowledge label frequency of occurrence in the knowledge tag set of each knowledge to be pushed is taken, that is, includes the knowledge number of the knowledge label
Amount.Then, the label frequency of current knowledge label is calculated according to the frequency of occurrence.
In practical application, can obtain the label frequency of knowledge label by following two calculation, one is first
First obtain all knowledge labels occur in health knowledge library number and；Then, current knowledge label is known each wait push
Frequency of occurrence occurs divided by all knowledge labels in the knowledge tag set of knowledge number and, using its quotient as current knowledge mark
The label frequency of label.In specific implementation, formula (9) be can use:
Wherein, K-TFtjIndicate the label frequency of j-th of knowledge label, kb_ntjIt indicates to include j-th of knowledge in knowledge base
The knowledge number of label tj, kb_n indicate the number that all knowledge labels occur in knowledge base and.
Another kind is to obtain the corresponding frequency of occurrence of knowledge label that frequency of occurrence is most in health knowledge library；Then, will
Current knowledge label is most divided by frequency of occurrence in health knowledge library in the frequency of occurrence in the knowledge tag set of each knowledge
The corresponding frequency of occurrence of knowledge label, using its quotient as the label frequency of current knowledge label.When specific implementation, Ke Yili
With formula (10):
Wherein, K-TFtjIndicate the label frequency of j-th of knowledge label, kb_ntjIt indicates to include j-th of knowledge in knowledge base
The knowledge number of label tj, max (kb_nt) indicate the frequency of occurrence of the most knowledge label t of frequency of occurrence in health knowledge library.
In specific implementation, it can be calculated with formula (9) or (10) and obtain each knowledge label in knowledge tag set
Label frequency.
2) known according to the knowledge quantity comprising j-th knowledge label in knowledge tag set and knowledge total quantity, calculating
Know the inverse label frequency of j-th of knowledge label in tag set.
In the present embodiment, knowledge quantity and health knowledge including j-th of knowledge label in knowledge tag set are obtained
Included knowledge total quantity to be pushed in library, then, according to knowledge quantity and the knowledge sum for including j-th of knowledge label
Amount calculates the inverse label frequency of j-th of knowledge label.
In practical application, if only characterizing the knowledge label with label frequency for the weight of the knowledge to be pushed,
When a certain knowledge label is newly added in health knowledge library, frequency of occurrence is less, the less knowledge label of the frequency of occurrence
It is smaller for the weight of knowledge to be pushed, the novelty of the knowledge can not be embodied.Therefore, to embody the knowledge label for wait push away
The novel degree for sending knowledge also needs the inverse label frequency of calculation knowledge label, to avoid knowledge label label frequency it is smaller without
It can reflect the novelty of knowledge.
In specific implementation, the inverse label frequency of j-th of knowledge label can be calculated by formula (11):
Wherein, K-ITFtjIndicate the inverse label frequency of j-th of knowledge label, sum_k indicates the knowledge in health knowledge library
Total quantity, sum_ktjIndicate the knowledge quantity comprising j-th of knowledge label.
In practical applications, can by formula (11) calculate obtain knowledge tag set in each knowledge label it is inverse
Label frequency.By formula (11) it is found that if a knowledge label is not common, denominator is smaller, K-ITFtjIt is bigger.
3) known according to j-th in the label frequency of j-th knowledge label in knowledge tag set and knowledge tag set
Know the inverse label frequency of label, the weight of j-th of knowledge label in calculation knowledge tag set.
In the present embodiment, after the label frequency for obtaining j-th of knowledge label and inverse label frequency, label frequency is utilized
And inverse label frequency calculates the weight of j-th of knowledge label.When specific implementation, it may refer to formula (12):
Wherein, wtj,kIndicate the weight of j-th of knowledge label, K-TFtjIndicate the label frequency of j-th of knowledge label, K-
ITFtjIndicate inverse label frequency.
S503: similar to the semanteme of knowledge label each in knowledge tag set according to user tag each in user tag set
The weight of each user tag and each knowledge label in knowledge tag set in degree and semantic association degree, user tag set
Weight determines the degree of correlation of user Yu knowledge to be pushed.
In the present embodiment, when calculated by S501 and S502 obtain the weight of each user tag in user tag set,
The weight of each knowledge label in knowledge tag set, and each user tag obtained is calculated by above method embodiment
After the semantic similarity of each knowledge label, semantic association degree, the degree of correlation of user Yu knowledge to be pushed are determined.
In specific implementation, it can be calculated and be obtained by formula (13):
Wherein, p (u, k) indicates the degree of correlation between user u and knowledge k to be pushed, and Sim_Rel (ti, tj) is indicated i-th
The Technique Using Both Text degree of association between user tag and j-th of knowledge label, the Technique Using Both Text degree of association can pass through i-th of user
Semantic similarity, semantic association degree between label and j-th of knowledge label, which calculate, to be obtained, and specifically may refer to formula (3),
wti,uIndicate the weight of i-th of user tag, wtj,kIndicate the weight of j-th of knowledge label.
Through the foregoing embodiment it is found that the weight of the label can be reacted by the label frequency of user tag or knowledge label
The property wanted, and label frequency is adjusted by inverse label frequency, can be excessive and cannot reflect user to avoid the weight of user tag
Individual demand；And avoid the weight of knowledge label excessive and cannot reflect the novelty of knowledge, it is pushed away to promote knowledge
The accuracy sent.
Based on above method embodiment, this application provides a kind of knowledge driving means, below in conjunction with attached drawing to the dress
It sets and is illustrated.
Referring to Fig. 6, which is a kind of knowledge driving means structure chart provided by the embodiments of the present application, as shown in fig. 6, the dress
It sets and may include:
Acquiring unit 601, for obtaining the user tag set of user and the knowledge tag set of knowledge to be pushed, institute
Stating user tag set includes at least one user tag, and the knowledge tag set includes at least one knowledge label；
First computing unit 602, for according to the user tag set of the user and knowing for the knowledge to be pushed
Know tag set, calculates the degree of correlation of the user Yu the knowledge to be pushed；
Push unit 603, the knowledge to be pushed for the degree of correlation to be met preset condition are pushed to the user.
In one possible implementation, first computing unit, comprising:
First computation subunit, for calculating i-th of user tag and the knowledge label in the user tag set
The semantic similarity and semantic association degree of j-th of knowledge label in set；
First determines subelement, for according to each user tag in the user tag set and the knowledge tag set
In each knowledge label semantic similarity and semantic association degree, determine the degree of correlation of the user Yu the knowledge to be pushed.
In one possible implementation, described device further include:
Establish unit, for establishing ontology model according to the incidence relation of user tag and knowledge label, user tag and
Knowledge label is the node in ontology model, each one concept of the node on behalf or example.
In one possible implementation, first computation subunit includes:
Judgment sub-unit, for judging node and the knowledge where i-th of user tag in the user tag set
Node where j-th of knowledge label in tag set, if there are common father nodes；
Zero setting subelement is if there is no common father node, by institute for the judging result when the judging unit
I-th of user tag and the semantic similarity of j-th of knowledge label in the knowledge tag set in user tag set is stated to set
Zero, and trigger first computation subunit and calculate semantic association degree；
Second computation subunit, for when the judging result of the judging unit is to calculate there are when common father node
The semantic similarity of i-th of user tag and j-th of knowledge label in the knowledge tag set in the user tag set
And semantic management degree.
In one possible implementation, first computation subunit is specifically used for calculating the user tag collection
Node where i-th of user tag is with node where j-th of knowledge label in the knowledge tag set in categorical attribute in conjunction
Path length on path apart, as in i-th user tag in the user tag set and the knowledge tag set
The semantic distance of j-th of knowledge label；
The node set passed through according to node where i-th user tag in the user tag set to root node and
The node set that node where j-th of knowledge label passes through to root node in the knowledge tag set calculates user's mark
The semantic registration of i-th of user tag and j-th of knowledge label in the knowledge tag set in label set；
Calculate the level depth of node and the knowledge tally set where i-th of user tag in the user tag set
The difference of the level depth of node where j-th of knowledge label in conjunction, as i-th of user tag in the user tag set with
The level of j-th of knowledge label is poor in the knowledge tag set；
According to j-th of knowledge label in i-th of user tag in the user tag set and the knowledge tag set
Semantic distance, semantic registration and level it is poor, determine i-th of user tag and the knowledge in the user tag set
The semantic similarity of j-th of knowledge label in tag set.
In one possible implementation, first computation subunit is specifically used for calculating the user tag collection
Node where i-th of user tag is with node where j-th of knowledge label in the knowledge tag set in relating attribute in conjunction
Path length on path apart；
According to j-th in node where i-th of user tag in the user tag set and the knowledge tag set
Path length of the node on relating attribute path apart where knowledge label determines i-th of use in the user tag set
The semantic association degree of j-th of knowledge label in family label and the knowledge tag set.
In one possible implementation, described first determine that subelement includes:
Third computation subunit, for calculate in the user tag set i-th of user tag weight and j-th
The weight of knowledge label；
4th computation subunit is specifically used for according to each user tag in the user tag set and the knowledge mark
Sign the semantic similarity of each knowledge label in set, each user tag and the knowledge label in the user tag set
The semantic association degree of each knowledge label in set the weight of each user tag and described is known in the user tag set
The weight for knowing each knowledge label in tag set, determines the degree of correlation of the user Yu the knowledge to be pushed.
In one possible implementation, the third computation subunit is specifically used for according to the user tag collection
Frequency of occurrence of i-th of user tag in the user tag set of each user in conjunction, calculates in the user tag set
The label frequency of i-th of user tag；
According to number of users and total number of users amount comprising i-th of user tag in the user tag set, calculate
The inverse label frequency of i-th of user tag in the user tag set；
According in the user tag set in the label frequency and the user tag set of i-th of user tag
The inverse label frequency of i user tag calculates the weight of i-th of user tag in the user tag set.
In one possible implementation, the third computation subunit is specifically used for according to the knowledge tally set
Frequency of occurrence of j-th of knowledge label in the knowledge tag set of each knowledge in conjunction, calculates in the knowledge tag set
The label frequency of j-th of knowledge label；
According to knowledge quantity and knowledge total quantity comprising j-th of knowledge label in the knowledge tag set, calculate
The inverse label frequency of j-th of knowledge label in the knowledge tag set；
According in the label frequency of j-th knowledge label in the knowledge tag set and the knowledge tag set
The inverse label frequency of j knowledge label calculates the weight of j-th of knowledge label in the knowledge tag set.
It should be noted that the realization of each unit may refer to above method example in the present embodiment, the present embodiment exists
This is repeated no more.
In addition, the embodiment of the present application also provides a kind of computer readable storage medium, the computer readable storage medium
Instruction is stored in matter, when described instruction is run on the terminal device, so that the terminal device executes the knowledge and pushes away
Delivery method.
The embodiment of the present application also provides a kind of computer program product, the computer program product is on the terminal device
When operation, so that the terminal device executes the knowledge method for pushing.
In this way, obtaining the user tag set of the user first, and each wait push before pushing knowledge to user
The knowledge tag set of knowledge.Wherein, user tag set can react the personal feature of the user, and knowledge tag set is used for
Reflect the classification of the knowledge to be pushed.Then, according to the knowledge tally set of the user tag set of user and knowledge to be pushed
It closes, calculate user and is each pushed away wait push the degree of correlation between knowledge, then by the knowledge to be pushed that the degree of correlation meets preset condition
Give user.It is, calculating user before pushing knowledge to user with each wait push the correlation between knowledge, inciting somebody to action
The biggish knowledge to be pushed of correlation is pushed to user, to realize accurate to user's push and have personalized knowledge, mentions
The accuracy of high knowledge push.
It should be noted that each embodiment in this specification is described in a progressive manner, each embodiment emphasis is said
Bright is the difference from other embodiments, and the same or similar parts in each embodiment may refer to each other.For reality
For applying system or device disclosed in example, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, phase
Place is closed referring to method part illustration.
It should be appreciated that in this application, " at least one (item) " refers to one or more, and " multiple " refer to two or two
More than a."and/or" indicates may exist three kinds of relationships, for example, " A and/or B " for describing the incidence relation of affiliated partner
It can indicate: only exist A, only exist B and exist simultaneously tri- kinds of situations of A and B, wherein A, B can be odd number or plural number.Word
Symbol "/" typicallys represent the relationship that forward-backward correlation object is a kind of "or"." at least one of following (a) " or its similar expression, refers to
Any combination in these, any combination including individual event (a) or complex item (a).At least one of for example, in a, b or c
(a) can indicate: a, b, c, " a and b ", " a and c ", " b and c ", or " a and b and c ", and wherein a, b, c can be individually, can also
To be multiple.
It should also be noted that, herein, relational terms such as first and second and the like are used merely to one
Entity or operation are distinguished with another entity or operation, without necessarily requiring or implying between these entities or operation
There are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant are intended to contain
Lid non-exclusive inclusion, so that the process, method, article or equipment including a series of elements is not only wanted including those
Element, but also including other elements that are not explicitly listed, or further include for this process, method, article or equipment
Intrinsic element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that
There is also other identical elements in process, method, article or equipment including the element.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.
Claims (10)
1. a kind of knowledge method for pushing, which is characterized in that the described method includes:
The user tag set of user and the knowledge tag set of knowledge to be pushed are obtained, the user tag set includes extremely
A few user tag, the knowledge tag set includes at least one knowledge label；
According to the knowledge tag set of the user tag set of the user and the knowledge to be pushed, calculate the user with
The degree of correlation of the knowledge to be pushed；
The knowledge to be pushed that the degree of correlation meets preset condition is pushed to the user.
2. the method according to claim 1, wherein the user tag set and institute according to the user
The knowledge tag set for stating knowledge to be pushed calculates the degree of correlation of the user Yu the knowledge to be pushed, comprising:
Calculate the language of i-th of user tag and j-th of knowledge label in the knowledge tag set in the user tag set
Adopted similarity and semantic association degree；
It is similar to the semanteme of each knowledge label in the knowledge tag set according to each user tag in the user tag set
Degree and semantic association degree, determine the degree of correlation of the user Yu the knowledge to be pushed.
3. according to the method described in claim 2, it is characterized in that, the method also includes:
Ontology model is established according to the incidence relation of user tag and knowledge label, the user tag and the knowledge label are
Node in the ontology model, each one concept of the node on behalf or example.
4. according to the method described in claim 3, it is characterized in that, described calculate i-th of user in the user tag set
The semantic similarity and semantic association degree of j-th of knowledge label in label and the knowledge tag set, comprising:
Judge node where i-th of user tag and j-th of knowledge in the knowledge tag set in the user tag set
Node where label, if there are common father nodes；
If there is no common father node, by i-th of user tag and the knowledge tally set in the user tag set
The semantic similarity zero setting of j-th of knowledge label in conjunction, and calculate in the user tag set i-th of user tag with it is described
The semantic association degree of j-th of knowledge label in knowledge tag set；
If there is common father node, i-th of user tag and the knowledge tally set in the user tag set are calculated
The semantic similarity and semantic association degree of j-th of knowledge label in conjunction.
5. the method according to claim 3 or 4, which is characterized in that calculate i-th of user's mark in the user tag set
The semantic similarity of label and j-th of knowledge label in the knowledge tag set, comprising:
Calculate node where i-th of user tag and j-th of knowledge in the knowledge tag set in the user tag set
Path length of the node on categorical attribute path apart where label, as i-th of user's mark in the user tag set
The semantic distance of label and j-th of knowledge label in the knowledge tag set；
The node set passed through according to node where i-th user tag in the user tag set to root node and described
The node set that node where j-th of knowledge label passes through to root node in knowledge tag set, calculates the user tag collection
The semantic registration of i-th of user tag and j-th of knowledge label in the knowledge tag set in conjunction；
It calculates in the user tag set in the level depth and the knowledge tag set of i-th of user tag place node
The difference of the level depth of node where j-th of knowledge label, as i-th of user tag in the user tag set with it is described
The level of j-th of knowledge label is poor in knowledge tag set；
According to the language of j-th of knowledge label in i-th of user tag in the user tag set and the knowledge tag set
Justice distance, semantic registration and level are poor, determine i-th of user tag and the knowledge label in the user tag set
The semantic similarity of j-th of knowledge label in set.
6. the method according to claim 3 or 4, which is characterized in that calculate i-th of user's mark in the user tag set
The semantic association degree of label and j-th of knowledge label in the knowledge tag set, comprising:
Calculate node where i-th of user tag and j-th of knowledge in the knowledge tag set in the user tag set
Path length of the node on relating attribute path apart where label；
According to j-th of knowledge in node where i-th of user tag in the user tag set and the knowledge tag set
Path length of the node on relating attribute path apart where label determines i-th of user's mark in the user tag set
The semantic association degree of label and j-th of knowledge label in the knowledge tag set.
7. according to the method described in claim 2, it is characterized in that, described according to each user tag in the user tag set
With the semantic similarity and semantic association degree of knowledge label each in the knowledge tag set, determine the user and it is described to
Push the degree of correlation of knowledge, comprising:
Calculate the weight of i-th of user tag and the weight of j-th of the knowledge label；
It is similar to the semanteme of each knowledge label in the knowledge tag set according to each user tag in the user tag set
Respectively know in the weight of each user tag and the knowledge tag set in degree and semantic association degree, the user tag set
The weight for knowing label, determines the degree of correlation of the user Yu the knowledge to be pushed.
8. a kind of knowledge driving means, which is characterized in that described device includes:
Acquiring unit, for obtaining the user tag set of user and the knowledge tag set of knowledge to be pushed, the user
Tag set includes at least one user tag, and the knowledge tag set includes at least one knowledge label；
First computing unit, for according to the user tag set of the user and the knowledge tally set of the knowledge to be pushed
It closes, calculates the degree of correlation of the user Yu the knowledge to be pushed；
Push unit, the knowledge to be pushed for the degree of correlation to be met preset condition are pushed to the user.
9. a kind of computer readable storage medium, which is characterized in that it is stored with instruction in the computer readable storage medium storing program for executing, when
When described instruction is run on the terminal device, so that the terminal device perform claim requires the described in any item knowledge of 1-7 to push away
Delivery method.
10. a kind of computer program product, which is characterized in that when the computer program product is run on the terminal device, make
It obtains the terminal device perform claim and requires the described in any item knowledge method for pushing of 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910419506.9A CN110147498A (en) | 2019-05-20 | 2019-05-20 | A kind of knowledge method for pushing, device and storage equipment, program product |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910419506.9A CN110147498A (en) | 2019-05-20 | 2019-05-20 | A kind of knowledge method for pushing, device and storage equipment, program product |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110147498A true CN110147498A (en) | 2019-08-20 |
Family
ID=67592312
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910419506.9A Pending CN110147498A (en) | 2019-05-20 | 2019-05-20 | A kind of knowledge method for pushing, device and storage equipment, program product |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110147498A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111180080A (en) * | 2019-12-30 | 2020-05-19 | 四川华迪信息技术有限公司 | Accurate pushing method for health knowledge base of old people |
| CN111741104A (en) * | 2020-06-18 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method for determining response message, method for configuring response message, device, equipment and storage medium |
| CN112331339A (en) * | 2020-10-10 | 2021-02-05 | 于军 | Automatic and personalized health management system and method for large population |
| CN116628456A (en) * | 2023-07-26 | 2023-08-22 | 北京点聚信息技术有限公司 | Layout light reading recommendation method and system based on data analysis |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177093A (en) * | 2013-03-13 | 2013-06-26 | 北京开心人信息技术有限公司 | General recommendation method and system based on object tags |
| CN107491479A (en) * | 2017-07-05 | 2017-12-19 | 上海大学 | A kind of label management method based on ontology library |
| CN109543189A (en) * | 2018-11-28 | 2019-03-29 | 重庆邮电大学 | Robot data interoperability domain body mapping method based on semantic similarity |
-
2019
- 2019-05-20 CN CN201910419506.9A patent/CN110147498A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177093A (en) * | 2013-03-13 | 2013-06-26 | 北京开心人信息技术有限公司 | General recommendation method and system based on object tags |
| CN107491479A (en) * | 2017-07-05 | 2017-12-19 | 上海大学 | A kind of label management method based on ontology library |
| CN109543189A (en) * | 2018-11-28 | 2019-03-29 | 重庆邮电大学 | Robot data interoperability domain body mapping method based on semantic similarity |
Non-Patent Citations (2)
| Title |
|---|
| 唐晓波: "基于本体和标签的个性化推荐", 《信息系统》 * |
| 贺海宏: "一种基于语义关联度计算的查询扩展", 《石家庄铁路职业技术学院学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111180080A (en) * | 2019-12-30 | 2020-05-19 | 四川华迪信息技术有限公司 | Accurate pushing method for health knowledge base of old people |
| CN111741104A (en) * | 2020-06-18 | 2020-10-02 | 腾讯科技（深圳）有限公司 | Method for determining response message, method for configuring response message, device, equipment and storage medium |
| CN112331339A (en) * | 2020-10-10 | 2021-02-05 | 于军 | Automatic and personalized health management system and method for large population |
| CN116628456A (en) * | 2023-07-26 | 2023-08-22 | 北京点聚信息技术有限公司 | Layout light reading recommendation method and system based on data analysis |
| CN116628456B (en) * | 2023-07-26 | 2023-10-20 | 北京点聚信息技术有限公司 | Layout light reading recommendation method and system based on data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Yang et al. | Yum-me: a personalized nutrient-based meal recommender system | |
| CN110147498A (en) | A kind of knowledge method for pushing, device and storage equipment, program product | |
| Yang et al. | Like like alike: joint friendship and interest propagation in social networks | |
| Jiang et al. | User recommendation in healthcare social media by assessing user similarity in heterogeneous network | |
| WO2020192460A1 (en) | Data processing method, terminal-side device, cloud-side device, and terminal-cloud collaboration system | |
| US20150032751A1 (en) | Methods and Systems for Utilizing Subject Matter Experts in an Online Community | |
| White et al. | Evaluation of the feasibility of screening patients for early signs of lung carcinoma in web search logs | |
| JP2011513802A (en) | A recommendation system using social behavior analysis and vocabulary classification | |
| Li et al. | A social appraisal mechanism for online purchase decision support in the micro-blogosphere | |
| US20160189218A1 (en) | Systems and methods for sponsored search ad matching | |
| Hasan et al. | Thingsonomy: Tackling variety in internet of things events | |
| Yang et al. | Enriching user experience in online health communities through thread recommendations and heterogeneous information network mining | |
| US20150112706A1 (en) | System and Methods for Personal health Analytics Technical Field | |
| Wang et al. | A multidimensional network link prediction algorithm and its application for predicting social relationships | |
| Roy et al. | Supporting pervasive computing applications with active context fusion and semantic context delivery | |
| CN107809370B (en) | User recommendation method and device | |
| WO2021139477A1 (en) | Intelligent medical access method and apparatus, and computer device and storage medium | |
| EP2613275B1 (en) | Search device, search method, search program, and computer-readable memory medium for recording search program | |
| Kim et al. | Topic-Driven SocialRank: Personalized search result ranking by identifying similar, credible users in a social network | |
| Niu et al. | Predicting image popularity in an incomplete social media community by a weighted bi-partite graph | |
| Shirgahi et al. | Trust estimation of the semantic web using semantic web clustering | |
| CN103136309A (en) | Method for carrying out modeling on social intensity through learning based on core | |
| Abdaoui et al. | Collaborative content-based method for estimating user reputation in online forums | |
| US20150339602A1 (en) | System and method for modeling health care costs | |
| Jiang et al. | Personalized Recommendation in online health communities with heterogeneous network mining |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190820 |