CN110263248A - A kind of information-pushing method, device, storage medium and server - Google Patents
A kind of information-pushing method, device, storage medium and server Download PDFInfo
- Publication number
- CN110263248A CN110263248A CN201910422926.2A CN201910422926A CN110263248A CN 110263248 A CN110263248 A CN 110263248A CN 201910422926 A CN201910422926 A CN 201910422926A CN 110263248 A CN110263248 A CN 110263248A
- Authority
- CN
- China
- Prior art keywords
- information
- point
- designated user
- text
- keyword
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 52
- 238000013507 mapping Methods 0.000 claims abstract description 50
- 238000003058 natural language processing Methods 0.000 claims abstract description 29
- 230000008569 process Effects 0.000 claims abstract description 16
- 238000003062 neural network model Methods 0.000 claims description 25
- 238000012549 training Methods 0.000 claims description 15
- 238000002372 labelling Methods 0.000 claims description 9
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000012217 deletion Methods 0.000 claims description 3
- 230000037430 deletion Effects 0.000 claims description 3
- 238000002716 delivery method Methods 0.000 claims 1
- 239000002699 waste material Substances 0.000 abstract description 8
- 238000012545 processing Methods 0.000 description 7
- 239000000463 material Substances 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 235000013361 beverage Nutrition 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000009412 basement excavation Methods 0.000 description 1
- 230000009194 climbing Effects 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 239000002537 cosmetic Substances 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000002453 shampoo Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000007794 visualization technique 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
Abstract
The present invention relates to field of computer technology, a kind of information-pushing method, device, storage medium and server are proposed.The information-pushing method includes: the text information for obtaining designated user;Natural language processing is carried out to the text information, obtains the point of interest label of the designated user;It is searched from the knowledge mapping constructed in advance and records the correlation between each default object with the associated target object of point of interest label, the knowledge mapping;Product information associated with the target object is pushed to the designated user.The above process carries out natural language processing to text information, to determine the point of interest of user by the text information of acquisition user;Then search from the knowledge mapping constructed in advance can be improved the precision of information push and reduces unnecessary bandwidth resources waste with the associated target object of the point of interest for user's push product information relevant to these target objects.
Description
Technical field
The present invention relates to field of computer technology more particularly to a kind of information-pushing method, device, storage medium and services
Device.
Background technique
The product of newest publication or information can be generally pushed to use when carrying out information push by existing many platforms
Family.However, frequent, blindness information push meeting extreme influence arrives when the product or excessive information updated in the same period
User experience.Moreover, will cause unnecessary bandwidth resources waste.
Therefore, how improving the precision of information push and reducing unnecessary bandwidth resources waste becomes this field skill
Art personnel technical problem in the urgent need to address.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of information-pushing method, device, storage medium and server, energy
The precision for enough improving information push reduces unnecessary bandwidth resources waste.
The embodiment of the present invention in a first aspect, providing a kind of information-pushing method, comprising:
Obtain the text information of designated user;
Natural language processing is carried out to the text information, obtains the point of interest label of the designated user;
It is searched from the knowledge mapping constructed in advance and the associated target object of point of interest label, the knowledge mapping
Record the correlation between each default object;
Product information associated with the target object is pushed to the designated user;
Wherein, the natural language processing process includes:
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating
The phrase of name of product;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, as
One point of interest label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging and the target text deleted after stop words operation will be executed and be converted to term vector, it is defeated
Enter the neural network model constructed in advance, the neural network model is by each corresponding to the text feature of different labels as instruction
Practice training to get;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest mark of the designated user
Label.
The second aspect of the embodiment of the present invention provides a kind of information push-delivery apparatus, comprising:
Text obtains module, for obtaining the text information of designated user;
Natural language processing module obtains the designated user for carrying out natural language processing to the text information
Point of interest label;
Target object searching module, it is associated with the point of interest label for being searched from the knowledge mapping constructed in advance
Target object, the knowledge mapping record the correlation between each default object;
Info push module, for pushing product information associated with the target object to the designated user;
Wherein, the natural language processing process includes:
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating
The phrase of name of product;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, as
One point of interest label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging and the target text deleted after stop words operation will be executed and be converted to term vector, it is defeated
Enter the neural network model constructed in advance, the neural network model is by each corresponding to the text feature of different labels as instruction
Practice training to get;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest mark of the designated user
Label.
The third aspect of the embodiment of the present invention, provides a kind of computer readable storage medium, described computer-readable to deposit
Storage media is stored with computer-readable instruction, and such as the embodiment of the present invention is realized when the computer-readable instruction is executed by processor
First aspect propose information-pushing method the step of.
The fourth aspect of the embodiment of the present invention, provides a kind of server, including memory, processor and is stored in institute
The computer-readable instruction that can be run in memory and on the processor is stated, the processor executes described computer-readable
The step of information-pushing method that the first aspect such as the embodiment of the present invention proposes is realized when instruction.
The information-pushing method that the embodiment of the present invention proposes includes: the text information for obtaining designated user;To the text
Information carries out natural language processing, obtains the point of interest label of the designated user;It is searched from the knowledge mapping constructed in advance
With the associated target object of point of interest label, the knowledge mapping records the correlation between each default object;To
The designated user pushes product information associated with the target object.The text envelope that the above process passes through acquisition user
Breath carries out natural language processing to text information, to determine the point of interest of user;Then from the knowledge mapping constructed in advance
Searching can be improved with the associated target object of the point of interest for user's push product information relevant to these target objects
The precision of information push and the unnecessary bandwidth resources waste of reduction.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without any creative labor, can also be according to these
Attached drawing obtains other attached drawings.
Fig. 1 is a kind of flow chart of one embodiment of information-pushing method provided in an embodiment of the present invention;
Fig. 2 is a kind of flow chart of second embodiment of information-pushing method provided in an embodiment of the present invention;
Fig. 3 is a kind of structure chart of one embodiment of information push-delivery apparatus provided in an embodiment of the present invention;
Fig. 4 is a kind of schematic diagram of server provided in an embodiment of the present invention.
Specific embodiment
The embodiment of the invention provides a kind of information-pushing method, device, storage medium and servers, can be improved information
The precision of push reduces unnecessary bandwidth resources waste.
In order to make the invention's purpose, features and advantages of the invention more obvious and easy to understand, below in conjunction with the present invention
Attached drawing in embodiment, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that disclosed below
Embodiment be only a part of the embodiment of the present invention, and not all embodiment.Based on the embodiments of the present invention, this field
Those of ordinary skill's all other embodiment obtained without making creative work, belongs to protection of the present invention
Range.
Referring to Fig. 1, a kind of one embodiment of information-pushing method includes: in the embodiment of the present invention
101, the text information of designated user is obtained;
Firstly, obtaining the text information of designated user, it is flat which, which can be personal user or enterprise customer,
The object of platform execution information push.The source of text information may include two aspects, and one is that the designated user mentions oneself
Text material of confession, such as brief introduction, product description, supplier's list and the resume of enterprise etc.;Secondly for by climbing
Worm tool crawls the text material on network comprising associative key, for example crawls enterprise's name on network comprising the designated user
The related text material of title or name of product, may include customer service feedback information, the customer evaluation etc. in social media.In order to
The point of interest for more accurately positioning user, can generally obtain multiple and different text informations, such as a system relevant to the user
Column text material.
Further, after step 101, can also include:
(1) the first keyword and the second keyword for including in the text information are extracted, first keyword is pre-
The front evaluation keyword first defined, second keyword is unfavorable ratings keyword predetermined;
(2) quantity of first keyword and the quantity of second keyword are counted respectively;
(3) if the ratio of the quantity of first keyword and the quantity of second keyword is less than preset threshold,
The text information is deleted.
It, can be with for each text information got where more accurately determining the point of interest of the designated user
Extracting certain particular keywords that its content includes respectively, (front evaluation keyword is for example good, likes, appreciates;Unfavorable ratings close
For example bad, bad, detest of keyword etc.), then by calculating between front evaluation keyword and the quantity of unfavorable ratings keyword
That a part of text information that the ratio is less than preset threshold is deleted, is substantially belonged to just to filter out content of text by ratio
Face evaluation, can relatively accurately reflect the text information of user interest point.
102, natural language processing is carried out to the text information, obtains the point of interest label of the designated user;
After obtaining text information, natural language processing is carried out to the text information, obtains the designated user's
Point of interest label.Natural-sounding processing is NLP processing, primarily to the semanteme of parsing text information, so that it is determined that this is specified
The point of interest of user.
Specifically, the process of the natural language processing includes:
(1) body matter of the text information is extracted;
Firstly, extracting the body matter of the text information.Text information is usually a text material, includes mark
Each sections content, the embodiment of the present invention such as topic, abstract and text come out body matter as interested contents extraction, by it
The content of its part is regarded as noise deletion.
(2) the name of product word for including in the body matter is detected, the name of product word is being used for of constructing in advance
Indicate the phrase of name of product;
It is then detected that the name of product word for including in the body matter.Name of product word is to be constructed in advance for table
The phrase for showing name of product can be brand name more well-known on the market, such as " XX is laughable ", " XX " shampoo etc..System
System collects multiple products or brand name word in advance, and specified data is arrived in storage after these title words are classified according to product category
In library.By detection, all name of product words for including in the body matter can all be extracted.
(3) label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, is made
For the first point of interest label;
Then, label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance,
As the first point of interest label.Specified data is arrived in storage after system classifies the name of product word being collected into according to product category
When library, one Product labelling table of comparisons of creation can be synchronized, which records label corresponding to each name of product word, the label
For indicating the product category of name of product word.For example, name of product word " XX is laughable " corresponds to " beverage " label, name of product
Word is certain well-known apparel brand title, then corresponds to " clothes " label.If the name of product word detected includes " XX is laughable "
And certain well-known apparel brand title, then it can determine that the first point of interest label is " beverage " and " clothes ".
(4) the name of product word for including in the body matter is deleted, target text is obtained;
The name of product word for including in the body matter is deleted, target text is obtained.Name of product word is usually certain
The title of famous brand name, these titles execute NLP processing when, cannot be carried out according to common Chinese character or word participle and after
Continuous semantics recognition, therefore need to delete these name of product words, target text is obtained, to improve subsequent execution NLP processing
Accuracy.
(5) participle, part-of-speech tagging are executed to the target text and deletes stop words operation;
Then, participle, part-of-speech tagging are executed to the target text and deletes stop words operation.Participle can generally be adopted
With based on string matching, based on understand, based on statistics and the various segmenting methods such as rule-based.Part-of-speech tagging is to each
Word stamps part of speech label, such as verb, adjective and noun etc., and common part-of-speech tagging method includes the part of speech based on statistics
Mask method, the part-of-speech tagging method based on maximum entropy, based on statistics maximum probability output part of speech and based on the part-of-speech tagging of HMM
Method etc..Stop words is deleted, refers to the words deleted and do not make any contribution to text feature, such as punctuation mark, modal particle and people
Claim pronoun etc..
(6) participle, part-of-speech tagging and the target text deleted after stop words operation will be executed and be converted to term vector,
Input the neural network model that constructs in advance, the neural network model is by each corresponding to the text feature conducts of different labels
Training set training obtains;
(7) the second point of interest label of the designated user is determined according to the output result of the neural network model;
Then, will execute participle, part-of-speech tagging and delete stop words operation after the target text be converted to word to
Amount inputs the neural network model constructed in advance, and determines the specified use according to the output result of the neural network model
The second point of interest label at family.Word is converted into term vector, to be expressed as the data that computer can be identified and be calculated.The mind
It is obtained by the text feature for each corresponding to different labels as training set training through network model, passes through the neural network model
One or more labels can be exported, the second point of interest label as the designated user.Specifically, the neural network model exists
When training, determined as training set by comparing the matching degree of text feature using the corresponding text feature of each difference label
Corresponding second point of interest label.For example, the matching of the text feature and the corresponding text feature of label " exported product " of input
Highest is spent, then can determine that the second point of interest label is " exported product ".
(8) the first point of interest label and the second point of interest label are determined as to the point of interest of the designated user
Label.
Finally, the first point of interest label and the second point of interest label to be determined as to the interest of the designated user
Point label successfully obtains the point of interest label of the designated user to complete the process of natural language processing.
103, it is searched from the knowledge mapping constructed in advance and the associated target object of point of interest label, the knowledge
Correlation between each default object of map record;
After obtaining the point of interest label of the designated user, from the knowledge mapping constructed in advance search with it is described emerging
The interest point associated target object of label, the knowledge mapping record the correlation between each default object.
Knowledge mapping is a series of different figures of explicit knowledge's development process from structural relation, is retouched with visualization technique
State knowledge resource and its carrier, excavation, analysis, building, drafting and explicit knowledge and connecting each other between them.The knowledge graph
Spectrum records the correlation between each default object, such as each enterprise, product, person-to-person correlation.
Specifically, the association that can be constructed in advance in point of interest label and the knowledge mapping between each default object is closed
System, such as point of interest label " imported product " are associated with object " company A ", and " B is public for point of interest label " internet product " and object
Department " and " C company " association.Then, corresponding mesh can be found out from the knowledge mapping by determining point of interest label
Mark object.
Further, step 103 may include:
(1) it is searched and associated first object of the point of interest label from the knowledge mapping;
(2) the second object for having connection relationship in the knowledge mapping with first object is obtained;
(3) first object and second object are determined as the target object.
For example, point of interest label " imported product " is associated with object " company A ", then " company A " is and the point of interest mark
Associated first object is signed, and in knowledge mapping, have the second object of connection relationship including " B is public with object " company A "
" company A ", " B company " and " C product " is then all used as the target object by department " and " C product ".
104, Xiang Suoshu designated user pushes product information associated with the target object.
Finally, push associated with target object product information to the designated user, thus realization according to
The point of interest at family pushes product information, improves the precision of information push.Product information associated with target object can be
Various types of information, such as text, picture, network connection or video etc..
Further, the product information is that the form of network linking can also include: after step 104
(1) after preset duration, judge whether the designated user clicks the network linking;
(2) if the designated user does not click on the network linking, the accuracy score value of the knowledge mapping is deducted
One score value;
(3) if the designated user has clicked the network linking, increase the accuracy score value of the knowledge mapping
Two score values;
(4) if the accuracy score value of the knowledge mapping is less than preset threshold, preset instruction information is exported.
Specifically, server sends the network linking of push at the terminal device of the designated user, if certain
Within time (such as 1 day), which clicks the network linking and accesses corresponding product information, then increases the accurate of knowledge mapping
Property the first score value of score value (such as 2 points) if the user does not click the network linking deduct the accuracy point of the knowledge mapping
The first score value of value (such as 10 points).If the accuracy score value of the knowledge mapping is less than preset threshold value (such as 60 points), generate
Preset police instruction information needs to be adjusted and repair to remind related personnel to notice that the accuracy of the knowledge mapping is too low
Just.It is arranged in this way, the precision of information push can be further increased.
The information-pushing method that the embodiment of the present invention proposes includes: the text information for obtaining designated user;To the text
Information carries out natural language processing, obtains the point of interest label of the designated user;It is searched from the knowledge mapping constructed in advance
With the associated target object of point of interest label, the knowledge mapping records the correlation between each default object;To
The designated user pushes product information associated with the target object.The text envelope that the above process passes through acquisition user
Breath carries out natural language processing to text information, to determine the point of interest of user;Then from the knowledge mapping constructed in advance
Searching can be improved with the associated target object of the point of interest for user's push product information relevant to these target objects
The precision of information push and the unnecessary bandwidth resources waste of reduction.
Referring to Fig. 2, a kind of second embodiment of information-pushing method includes: in the embodiment of the present invention
201, the text information of designated user is obtained;
202, natural language processing is carried out to the text information, obtains the point of interest label of the designated user;
203, it is searched from the knowledge mapping constructed in advance and the associated target object of point of interest label, the knowledge
Correlation between each default object of map record;
Step 201-203 is identical as step 101-103, specifically can refer to the related description of step 101-103.
204, the quantity of the target object is counted;
If 205, the quantity of the target object is more than preset threshold, the user information of the designated user is obtained;
206, it is scored respectively each target object according to the user information;
207, the target object for the minimum preset quantity that scores is deleted;
In some cases, being found from the customer relationship map with the associated target pair of point of interest label
It is many as having, if the corresponding product information of these target objects is all pushed to the designated user, one side data at this time
Amount is big, on the other hand may be mingled with the uninterested information of many users in these information, interference can be brought to client.To understand
Certainly this problem, the embodiment of the present invention can count the quantity of the target object, if the quantity is more than certain threshold value, obtain
The user information of the designated user.If the designated user is personal user, the individual of the available designated user
Information is as the user information;If the designated user is enterprise customer, the buying of the available designated user is remembered
Record is used as the user information.Then, it is scored each target object, will be scored respectively according to the user information
The target object of minimum preset quantity is deleted, to complete the process of screening.
For example, the personal information such as available age, gender, occupation, address are to each if designated user is personal user
Target object scores, all ages and classes, different sexes, and the interested object of the people of different occupation is generally different, interested
The more high then corresponding higher scoring of degree, for example for female user, the scoring of the target objects such as clothes, cosmetics is higher;It is right
It is higher in the scoring of the user of engineer's occupation, various digital products.If designated user is enterprise customer, the available enterprise
The purchase records of industry can count the purchase number and quantity of various product according to the purchase records, then according to purchase number
And quantity, it scores for various product (i.e. target object), buys number and quantity is more, then score higher.
208, Xiang Suoshu designated user pushes product information associated with the target object.
Step 208 is identical as step 104, specifically can refer to the related description of step 104.
The information-pushing method that the embodiment of the present invention proposes includes: the text information for obtaining designated user;To the text
Information carries out natural language processing, obtains the point of interest label of the designated user;It is searched from the knowledge mapping constructed in advance
With the associated target object of point of interest label, the knowledge mapping records the correlation between each default object;System
Count the quantity of the target object;If the quantity of the target object is more than preset threshold, the use of the designated user is obtained
Family information;It is scored respectively each target object according to the user information;By the minimum preset quantity that scores
Target object is deleted;Product information associated with the target object is pushed to the designated user.The embodiment of the present invention exists
It searches with after the associated target object of point of interest label, the quantity of target object can be also counted, if quantity excessively understands root
These target objects are screened according to the user information of the designated user, so as to more accurately be that user pushes product letter
Breath reduces unnecessary bandwidth resources waste.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
A kind of information-pushing method is essentially described above, a kind of information push-delivery apparatus will be described below.
Referring to Fig. 3, a kind of one embodiment of information push-delivery apparatus includes: in the embodiment of the present invention
Text obtains module 301, for obtaining the text information of designated user;
Natural language processing module 302 obtains the specified use for carrying out natural language processing to the text information
The point of interest label at family;
Target object searching module 303 is closed for searching from the knowledge mapping constructed in advance with the point of interest label
The target object of connection, the knowledge mapping record the correlation between each default object;
Info push module 304, for pushing product information associated with the target object to the designated user;
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating
The phrase of name of product;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, as
One point of interest label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging and the target text deleted after stop words operation will be executed and be converted to term vector, it is defeated
Enter the neural network model constructed in advance, the neural network model is by each corresponding to the text feature of different labels as instruction
Practice training to get;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest mark of the designated user
Label.
Further, the information push-delivery apparatus can also include:
Keyword extracting module, for extracting the first keyword and the second keyword that include in the text information, institute
Stating the first keyword is that keyword is evaluated in front predetermined, and second keyword is that unfavorable ratings predetermined are crucial
Word;
Keyword quantity statistics module, for count respectively first keyword quantity and second keyword
Quantity;
Text information removing module, if the ratio of the quantity of the quantity and second keyword for first keyword
Value is less than preset threshold, then deletes the text information.
Further, the target object searching module may include:
First object searching unit, for being searched and associated first pair of the point of interest label from the knowledge mapping
As;
Second object searching unit has the of connection relationship for obtaining with first object in the knowledge mapping
Two objects;
Target object determination unit, for first object and second object to be determined as the target object.
Further, the information push-delivery apparatus can also include:
Target object quantity statistics module, for counting the quantity of the target object;
User profile acquisition module obtains described specified if the quantity for the target object is more than preset threshold
The user information of user;
Target object grading module, for being scored respectively each target object according to the user information;
Target object removing module, the target object for the minimum preset quantity that will score are deleted.
Further, the User profile acquisition module may include:
First information acquiring unit obtains of the designated user if being personal user for the designated user
People's information is as the user information;
Second information acquisition unit obtains adopting for the designated user if being enterprise customer for the designated user
Purchase record is used as the user information.
Further, the product information is the form of network linking, and the information push-delivery apparatus can also include::
Judgment module is clicked, for judging whether the designated user clicks the network linking after preset duration;
Accuracy score value deducts module, if not clicking on the network linking for the designated user, knows described in deduction
Know the first score value of accuracy score value of map;
Accuracy score value increases module, if having clicked the network linking for the designated user, knows described in increase
Know the second score value of accuracy score value of map;
Indicate message output module, if the accuracy score value for the knowledge mapping is less than preset threshold, output is pre-
If instruction information.
The embodiment of the present invention also provides a kind of computer readable storage medium, and the computer-readable recording medium storage has
Computer-readable instruction realizes any one letter indicated such as Fig. 1 or Fig. 2 when the computer-readable instruction is executed by processor
The step of ceasing method for pushing.
The embodiment of the present invention also provides a kind of server, including memory, processor and storage are in the memory
And the computer-readable instruction that can be run on the processor, the processor are realized when executing the computer-readable instruction
The step of any one information-pushing method indicated such as Fig. 1 or Fig. 2.
Fig. 4 is the schematic diagram for the server that one embodiment of the invention provides.As shown in figure 4, the server 4 of the embodiment wraps
It includes: processor 40, memory 41 and being stored in the computer that can be run in the memory 41 and on the processor 40
Readable instruction 42.The processor 40 realizes that above-mentioned each information-pushing method is implemented when executing the computer-readable instruction 42
Step in example, such as step 101 shown in FIG. 1 is to 104.Alternatively, the processor 40 executes the computer-readable instruction
The function of each module/unit in above-mentioned each Installation practice, such as the function of module 301 to 304 shown in Fig. 3 are realized when 42.
Illustratively, the computer-readable instruction 42 can be divided into one or more module/units, one
Or multiple module/units are stored in the memory 41, and are executed by the processor 40, to complete the present invention.Institute
Stating one or more module/units can be the series of computation machine readable instruction section that can complete specific function, the instruction segment
For describing implementation procedure of the computer-readable instruction 42 in the server 4.
The server 4 can be smart phone, notebook, palm PC and cloud server etc. and calculate equipment.It is described
Server 4 may include, but be not limited only to, processor 40, memory 41.It will be understood by those skilled in the art that Fig. 4 is only to take
The example of business device 4, does not constitute the restriction to server 4, may include components more more or fewer than diagram, or combine certain
A little components or different components, such as the server 4 can also include input-output equipment, network access equipment, bus
Deng.
The processor 40 can be central processing unit (CentraL Processing Unit, CPU), can also be
Other general processors, digital signal processor (DigitaL SignaL Processor, DSP), specific integrated circuit
(AppLication Specific Integrated Circuit, ASIC), ready-made programmable gate array (FieLd-
ProgrammabLe Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 41 can be the internal storage unit of the server 4, such as the hard disk or memory of server 4.
The memory 41 is also possible to the External memory equipment of the server 4, such as the plug-in type being equipped on the server 4 is hard
Disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure DigitaL, SD) card, flash card
(FLash Card) etc..Further, the memory 41 can also both include the internal storage unit of the server 4 or wrap
Include External memory equipment.The memory 41 is for storing needed for the computer-readable instruction and the server other
Program and data.The memory 41 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated unit is realized in the form of SFU software functional unit and sells or use as independent product
When, it can store in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially
The all or part of the part that contributes to existing technology or the technical solution can be in the form of software products in other words
It embodies, which is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) executes the complete of each embodiment the method for the present invention
Portion or part steps.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (ROM, Read-OnLy
Memory), random access memory (RAM, Random Access Memory), magnetic or disk etc. are various can store journey
The medium of sequence code.
The above, the above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although referring to before
Stating embodiment, invention is explained in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features;And these
It modifies or replaces, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of information-pushing method characterized by comprising
Obtain the text information of designated user;
Natural language processing is carried out to the text information, obtains the point of interest label of the designated user;
It is searched from the knowledge mapping constructed in advance and the associated target object of point of interest label, the knowledge mapping record
Correlation between each default object;
Product information associated with the target object is pushed to the designated user;
Wherein, the natural language processing process includes:
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating product
The phrase of title;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, it is emerging as first
Interest point label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging will be executed and delete the target text after stop words operation and be converted to term vector, input is pre-
The neural network model first constructed, the neural network model is by each corresponding to the text feature of different labels as training set
Training obtains;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest label of the designated user.
2. information-pushing method according to claim 1, which is characterized in that obtain designated user text information it
Afterwards, further includes:
The first keyword and the second keyword for including in the text information are extracted, first keyword is predetermined
Front evaluation keyword, second keyword are unfavorable ratings keyword predetermined;
The quantity of first keyword and the quantity of second keyword are counted respectively;
If the ratio of the quantity of first keyword and the quantity of second keyword is less than preset threshold, by the text
This information deletion.
3. information-pushing method according to claim 1, which is characterized in that described to be looked into from the knowledge mapping constructed in advance
It looks for and includes: with the associated target object of point of interest label
It is searched and associated first object of the point of interest label from the knowledge mapping;
Obtain the second object for having connection relationship in the knowledge mapping with first object;
First object and second object are determined as the target object.
4. information-pushing method according to claim 1, which is characterized in that searched from the knowledge mapping constructed in advance
After the associated target object of point of interest label, further includes:
Count the quantity of the target object;
If the quantity of the target object is more than preset threshold, the user information of the designated user is obtained;
It is scored respectively each target object according to the user information;
The target object for the minimum preset quantity that scores is deleted.
5. information-pushing method according to claim 4, which is characterized in that the user's letter for obtaining the designated user
Breath includes:
If the designated user is personal user, the personal information of the designated user is obtained as the user information;
If the designated user is enterprise customer, the purchase records of the designated user is obtained as the user information.
6. information-pushing method according to any one of claim 1 to 5, which is characterized in that the product information is net
The form of network link, after pushing product information associated with the target object to the designated user, further includes:
After preset duration, judge whether the designated user clicks the network linking;
If the designated user does not click on the network linking, the first score value of accuracy score value of the knowledge mapping is deducted;
If the designated user has clicked the network linking, increase the second score value of accuracy score value of the knowledge mapping;
If the accuracy score value of the knowledge mapping is less than preset threshold, preset instruction information is exported.
7. a kind of information push-delivery apparatus characterized by comprising
Text obtains module, for obtaining the text information of designated user;
Natural language processing module obtains the emerging of the designated user for carrying out natural language processing to the text information
Interest point label;
Target object searching module, for being searched and the associated target of point of interest label from the knowledge mapping constructed in advance
Object, the knowledge mapping record the correlation between each default object;
Info push module, for pushing product information associated with the target object to the designated user;
Wherein, the natural language processing process includes:
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating product
The phrase of title;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, it is emerging as first
Interest point label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging will be executed and delete the target text after stop words operation and be converted to term vector, input is pre-
The neural network model first constructed, the neural network model is by each corresponding to the text feature of different labels as training set
Training obtains;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest label of the designated user.
8. a kind of computer readable storage medium, the computer-readable recording medium storage has computer-readable instruction, special
Sign is, is realized when the computer-readable instruction is executed by processor as information described in any one of claims 1 to 6 pushes away
The step of delivery method.
9. a kind of server, including memory, processor and storage can transport in the memory and on the processor
Capable computer-readable instruction, which is characterized in that the processor realizes following steps when executing the computer-readable instruction:
Obtain the text information of designated user;
Natural language processing is carried out to the text information, obtains the point of interest label of the designated user;
It is searched from the knowledge mapping constructed in advance and the associated target object of point of interest label, the knowledge mapping record
Correlation between each default object;
Product information associated with the target object is pushed to the designated user;
Wherein, the natural language processing process includes:
Extract the body matter of the text information;
The name of product word for including in the body matter is detected, the name of product word is to be constructed in advance for indicating product
The phrase of title;
Label corresponding to the name of product word detected is inquired from the Product labelling table of comparisons constructed in advance, it is emerging as first
Interest point label;
The name of product word for including in the body matter is deleted, target text is obtained;
Participle, part-of-speech tagging are executed to the target text and delete stop words operation;
Participle, part-of-speech tagging will be executed and delete the target text after stop words operation and be converted to term vector, input is pre-
The neural network model first constructed, the neural network model is by each corresponding to the text feature of different labels as training set
Training obtains;
The second point of interest label of the designated user is determined according to the output result of the neural network model;
The first point of interest label and the second point of interest label are determined as to the point of interest label of the designated user.
10. server according to claim 9, which is characterized in that after the text information for obtaining designated user, also wrap
It includes:
The first keyword and the second keyword for including in the text information are extracted, first keyword is predetermined
Front evaluation keyword, second keyword are unfavorable ratings keyword predetermined;
The quantity of first keyword and the quantity of second keyword are counted respectively;
If the ratio of the quantity of first keyword and the quantity of second keyword is less than preset threshold, by the text
This information deletion.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910422926.2A CN110263248B (en) | 2019-05-21 | 2019-05-21 | Information pushing method, device, storage medium and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910422926.2A CN110263248B (en) | 2019-05-21 | 2019-05-21 | Information pushing method, device, storage medium and server |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110263248A true CN110263248A (en) | 2019-09-20 |
CN110263248B CN110263248B (en) | 2023-11-28 |
Family
ID=67914879
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910422926.2A Active CN110263248B (en) | 2019-05-21 | 2019-05-21 | Information pushing method, device, storage medium and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110263248B (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110926493A (en) * | 2019-12-10 | 2020-03-27 | 广州小鹏汽车科技有限公司 | Navigation method, navigation device, vehicle and computer readable storage medium |
CN111292743A (en) * | 2020-01-22 | 2020-06-16 | 北京松果电子有限公司 | Voice interaction method and device and electronic equipment |
CN111753195A (en) * | 2020-06-17 | 2020-10-09 | 百度在线网络技术(北京)有限公司 | Label system construction method, device, equipment and storage medium |
CN111932131A (en) * | 2020-08-12 | 2020-11-13 | 上海冰鉴信息科技有限公司 | Service data processing method and device |
CN111984876A (en) * | 2020-06-29 | 2020-11-24 | 北京百度网讯科技有限公司 | Interest point processing method, device, equipment and computer readable storage medium |
CN112000884A (en) * | 2020-08-13 | 2020-11-27 | 腾讯音乐娱乐科技(深圳)有限公司 | User content recommendation method and device, server and storage medium |
CN112256943A (en) * | 2020-10-22 | 2021-01-22 | 上海适享文化传播有限公司 | Portal portrait extraction method based on combination of natural language processing and knowledge graph |
KR20210036878A (en) * | 2020-09-15 | 2021-04-05 | 베이징 바이두 넷컴 사이언스 앤 테크놀로지 코., 엘티디. | Method and apparatus for pushing information, device and storage medium |
CN112836126A (en) * | 2021-02-08 | 2021-05-25 | 珠海格力电器股份有限公司 | Recommendation method and device based on knowledge graph, electronic equipment and storage medium |
CN112954025A (en) * | 2021-01-29 | 2021-06-11 | 北京百度网讯科技有限公司 | Information pushing method, device, equipment and medium based on layered knowledge graph |
CN113688164A (en) * | 2021-07-28 | 2021-11-23 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Interest point query method and system based on knowledge graph correlation analysis |
CN114422585A (en) * | 2021-12-27 | 2022-04-29 | 航天信息股份有限公司 | Message pushing method and system for enterprise service platform |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106095939A (en) * | 2016-06-12 | 2016-11-09 | 腾讯科技(深圳)有限公司 | The acquisition methods of account authority and device |
CN106294744A (en) * | 2016-08-11 | 2017-01-04 | 上海动云信息科技有限公司 | Interest recognition methods and system |
CN107291899A (en) * | 2017-06-22 | 2017-10-24 | 努比亚技术有限公司 | A kind of recommendation method and terminal and computer-readable recording medium based on label |
CN107526800A (en) * | 2017-08-20 | 2017-12-29 | 平安科技(深圳)有限公司 | Device, method and the computer-readable recording medium of information recommendation |
CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
CN107832287A (en) * | 2017-09-26 | 2018-03-23 | 晶赞广告(上海)有限公司 | A kind of label identification method and device, storage medium, terminal |
CN108831442A (en) * | 2018-05-29 | 2018-11-16 | 平安科技(深圳)有限公司 | Point of interest recognition methods, device, terminal device and storage medium |
CN109446412A (en) * | 2018-09-25 | 2019-03-08 | 中国平安人寿保险股份有限公司 | Product data method for pushing, device, equipment and medium based on web page tag |
-
2019
- 2019-05-21 CN CN201910422926.2A patent/CN110263248B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106095939A (en) * | 2016-06-12 | 2016-11-09 | 腾讯科技(深圳)有限公司 | The acquisition methods of account authority and device |
CN106294744A (en) * | 2016-08-11 | 2017-01-04 | 上海动云信息科技有限公司 | Interest recognition methods and system |
CN107291899A (en) * | 2017-06-22 | 2017-10-24 | 努比亚技术有限公司 | A kind of recommendation method and terminal and computer-readable recording medium based on label |
CN107526800A (en) * | 2017-08-20 | 2017-12-29 | 平安科技(深圳)有限公司 | Device, method and the computer-readable recording medium of information recommendation |
CN107832287A (en) * | 2017-09-26 | 2018-03-23 | 晶赞广告(上海)有限公司 | A kind of label identification method and device, storage medium, terminal |
CN107784092A (en) * | 2017-10-11 | 2018-03-09 | 深圳市金立通信设备有限公司 | A kind of method, server and computer-readable medium for recommending hot word |
CN108831442A (en) * | 2018-05-29 | 2018-11-16 | 平安科技(深圳)有限公司 | Point of interest recognition methods, device, terminal device and storage medium |
CN109446412A (en) * | 2018-09-25 | 2019-03-08 | 中国平安人寿保险股份有限公司 | Product data method for pushing, device, equipment and medium based on web page tag |
Cited By (19)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110926493A (en) * | 2019-12-10 | 2020-03-27 | 广州小鹏汽车科技有限公司 | Navigation method, navigation device, vehicle and computer readable storage medium |
CN111292743A (en) * | 2020-01-22 | 2020-06-16 | 北京松果电子有限公司 | Voice interaction method and device and electronic equipment |
CN111292743B (en) * | 2020-01-22 | 2023-09-26 | 北京小米松果电子有限公司 | Voice interaction method and device and electronic equipment |
CN111753195A (en) * | 2020-06-17 | 2020-10-09 | 百度在线网络技术(北京)有限公司 | Label system construction method, device, equipment and storage medium |
CN111753195B (en) * | 2020-06-17 | 2024-01-09 | 百度在线网络技术(北京)有限公司 | Label system construction method, device, equipment and storage medium |
CN111984876A (en) * | 2020-06-29 | 2020-11-24 | 北京百度网讯科技有限公司 | Interest point processing method, device, equipment and computer readable storage medium |
CN111932131B (en) * | 2020-08-12 | 2024-03-15 | 上海冰鉴信息科技有限公司 | Service data processing method and device |
CN111932131A (en) * | 2020-08-12 | 2020-11-13 | 上海冰鉴信息科技有限公司 | Service data processing method and device |
CN112000884A (en) * | 2020-08-13 | 2020-11-27 | 腾讯音乐娱乐科技(深圳)有限公司 | User content recommendation method and device, server and storage medium |
KR20210036878A (en) * | 2020-09-15 | 2021-04-05 | 베이징 바이두 넷컴 사이언스 앤 테크놀로지 코., 엘티디. | Method and apparatus for pushing information, device and storage medium |
EP3968185A1 (en) * | 2020-09-15 | 2022-03-16 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for pushing information, device and storage medium |
KR102485129B1 (en) * | 2020-09-15 | 2023-01-06 | 베이징 바이두 넷컴 사이언스 앤 테크놀로지 코., 엘티디. | Method and apparatus for pushing information, device and storage medium |
CN112256943A (en) * | 2020-10-22 | 2021-01-22 | 上海适享文化传播有限公司 | Portal portrait extraction method based on combination of natural language processing and knowledge graph |
CN112256943B (en) * | 2020-10-22 | 2024-01-23 | 上海适享文化传播有限公司 | Portal store image extraction method based on natural language processing combined with knowledge graph |
CN112954025A (en) * | 2021-01-29 | 2021-06-11 | 北京百度网讯科技有限公司 | Information pushing method, device, equipment and medium based on layered knowledge graph |
CN112954025B (en) * | 2021-01-29 | 2023-07-18 | 北京百度网讯科技有限公司 | Information pushing method, device, equipment and medium based on hierarchical knowledge graph |
CN112836126A (en) * | 2021-02-08 | 2021-05-25 | 珠海格力电器股份有限公司 | Recommendation method and device based on knowledge graph, electronic equipment and storage medium |
CN113688164A (en) * | 2021-07-28 | 2021-11-23 | 华东计算技术研究所(中国电子科技集团公司第三十二研究所) | Interest point query method and system based on knowledge graph correlation analysis |
CN114422585A (en) * | 2021-12-27 | 2022-04-29 | 航天信息股份有限公司 | Message pushing method and system for enterprise service platform |
Also Published As
Publication number | Publication date |
---|---|
CN110263248B (en) | 2023-11-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110263248A (en) | A kind of information-pushing method, device, storage medium and server | |
CN109189942B (en) | Construction method and device of patent data knowledge graph | |
CN107515873B (en) | Junk information identification method and equipment | |
CN109657054B (en) | Abstract generation method, device, server and storage medium | |
CN110020422B (en) | Feature word determining method and device and server | |
CN103336766B (en) | Short text garbage identification and modeling method and device | |
US20170004128A1 (en) | Device and method for analyzing reputation for objects by data mining | |
US20130159277A1 (en) | Target based indexing of micro-blog content | |
KR101644817B1 (en) | Generating search results | |
EP2562659A1 (en) | Data mapping acceleration | |
CN107102993B (en) | User appeal analysis method and device | |
CN107544988B (en) | Method and device for acquiring public opinion data | |
CN110287314B (en) | Long text reliability assessment method and system based on unsupervised clustering | |
CN110362689A (en) | A kind of methods of risk assessment, device, storage medium and server | |
CN112989208B (en) | Information recommendation method and device, electronic equipment and storage medium | |
US11495227B2 (en) | Artificial intelligence (AI) based user query intent analyzer | |
CN110134845A (en) | Project public sentiment monitoring method, device, computer equipment and storage medium | |
CN109446393B (en) | Network community topic classification method and device | |
CN114238573A (en) | Information pushing method and device based on text countermeasure sample | |
CN110134844A (en) | Subdivision field public sentiment monitoring method, device, computer equipment and storage medium | |
CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
CN110880142A (en) | Risk entity acquisition method and device | |
CN115147130A (en) | Problem prediction method, apparatus, storage medium, and program product | |
CN110019763B (en) | Text filtering method, system, equipment and computer readable storage medium | |
CN107688594B (en) | The identifying system and method for risk case based on social information |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |