CN110263248B - Information pushing method, device, storage medium and server - Google Patents
Information pushing method, device, storage medium and server Download PDFInfo
- Publication number
- CN110263248B CN110263248B CN201910422926.2A CN201910422926A CN110263248B CN 110263248 B CN110263248 B CN 110263248B CN 201910422926 A CN201910422926 A CN 201910422926A CN 110263248 B CN110263248 B CN 110263248B
- Authority
- CN
- China
- Prior art keywords
- information
- text
- user
- interest point
- knowledge graph
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 238000003058 natural language processing Methods 0.000 claims abstract description 29
- 230000008569 process Effects 0.000 claims abstract description 13
- 238000003062 neural network model Methods 0.000 claims description 26
- 230000011218 segmentation Effects 0.000 claims description 18
- 238000012217 deletion Methods 0.000 claims description 16
- 230000037430 deletion Effects 0.000 claims description 16
- 238000011156 evaluation Methods 0.000 claims description 14
- 239000000463 material Substances 0.000 claims description 12
- 238000012549 training Methods 0.000 claims description 9
- 239000013598 vector Substances 0.000 claims description 9
- 239000002699 waste material Substances 0.000 abstract description 8
- 238000012545 processing Methods 0.000 description 7
- 230000006870 function Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 235000013361 beverage Nutrition 0.000 description 2
- 230000009193 crawling Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000003491 array Methods 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
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036651 mood Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 239000002453 shampoo Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention relates to the technical field of computers, and provides an information pushing method, an information pushing device, a storage medium and a server. The information pushing method comprises the following steps: acquiring text information of a designated user; performing natural language processing on the text information to obtain the interest point tag of the appointed user; searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects; pushing product information associated with the target object to the specified user. The process comprises the steps of obtaining text information of a user, and carrying out natural language processing on the text information to determine interest points of the user; and then searching target objects associated with the interest points from the pre-constructed knowledge graph, and pushing product information related to the target objects for the user, so that the accuracy of information pushing can be improved, and unnecessary bandwidth resource waste can be reduced.
Description
Technical Field
The present invention relates to the field of computer technologies, and in particular, to an information pushing method, an information pushing device, a storage medium, and a server.
Background
Many existing platforms generally push the latest released products or information to users when pushing information. However, when there are too many products or information updated in the same time period, frequent and blind information pushing can greatly affect the user experience. Moreover, unnecessary waste of bandwidth resources is caused.
Therefore, how to improve the accuracy of information push and reduce unnecessary bandwidth resource waste is a technical problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide an information pushing method, an apparatus, a storage medium, and a server, which can improve the accuracy of information pushing and reduce unnecessary bandwidth resource waste.
In a first aspect of an embodiment of the present invention, there is provided an information pushing method, including:
acquiring text information of a designated user;
performing natural language processing on the text information to obtain the interest point tag of the appointed user;
searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects;
pushing product information associated with the target object to the designated user;
wherein the natural language processing process comprises:
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
In a second aspect of an embodiment of the present invention, there is provided an information pushing apparatus, including:
the text acquisition module is used for acquiring text information of a designated user;
the natural language processing module is used for carrying out natural language processing on the text information to obtain the interest point tag of the appointed user;
the target object searching module is used for searching target objects associated with the interest point labels from a pre-constructed knowledge graph, and the knowledge graph records the interrelationship among all the preset objects;
the information pushing module is used for pushing the product information associated with the target object to the appointed user;
wherein the natural language processing process comprises:
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
In a third aspect of the embodiments of the present invention, there is provided a computer readable storage medium storing computer readable instructions which, when executed by a processor, implement the steps of the information push method as set forth in the first aspect of the embodiments of the present invention.
In a fourth aspect of the embodiments of the present invention, there is provided a server comprising a memory, a processor and computer readable instructions stored in the memory and executable on the processor, the processor implementing the steps of the information push method as set forth in the first aspect of the embodiments of the present invention when executing the computer readable instructions.
The information pushing method provided by the embodiment of the invention comprises the following steps: acquiring text information of a designated user; performing natural language processing on the text information to obtain the interest point tag of the appointed user; searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects; pushing product information associated with the target object to the specified user. The process comprises the steps of obtaining text information of a user, and carrying out natural language processing on the text information to determine interest points of the user; and then searching target objects associated with the interest points from the pre-constructed knowledge graph, and pushing product information related to the target objects for the user, so that the accuracy of information pushing can be improved, and unnecessary bandwidth resource waste can be reduced.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments or the description of the prior art will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a first embodiment of an information pushing method according to an embodiment of the present invention;
fig. 2 is a flowchart of a second embodiment of an information pushing method according to an embodiment of the present invention;
FIG. 3 is a block diagram of one embodiment of an information pushing device according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a server according to an embodiment of the present invention.
Detailed Description
The embodiment of the invention provides an information pushing method, an information pushing device, a storage medium and a server, which can improve the accuracy of information pushing and reduce unnecessary bandwidth resource waste.
In order to make the objects, features and advantages of the present invention more comprehensible, the technical solutions in the embodiments of the present invention are described in detail below with reference to the accompanying drawings, and it is apparent that the embodiments described below are only some embodiments of the present invention, but not all embodiments of the present invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a first embodiment of an information pushing method in an embodiment of the present invention includes:
101. acquiring text information of a designated user;
firstly, text information of a designated user is acquired, wherein the designated user can be an individual user or an enterprise user and is an object for a platform to execute information pushing. The source of the text information may include two aspects, one providing text material for the specified user itself, such as a company profile, a product description, a vendor list, a personal resume, etc.; and secondly, crawling text materials containing related keywords on a network through a crawler tool, for example, crawling related text materials containing enterprise names or product names of the appointed users on the network, wherein the text materials can comprise customer service feedback information, customer evaluation on social media and the like. To more accurately locate a user's point of interest, a number of different textual information, such as a series of textual materials associated with the user, are typically obtained.
Further, after step 101, the method may further include:
(1) Extracting a first keyword and a second keyword contained in the text information, wherein the first keyword is a pre-defined positive evaluation keyword, and the second keyword is a pre-defined negative evaluation keyword;
(2) Counting the number of the first keywords and the number of the second keywords respectively;
(3) And deleting the text information if the ratio of the number of the first keywords to the number of the second keywords is smaller than a preset threshold value.
In order to more accurately determine the interest point of the appointed user, for each piece of acquired text information, certain specific keywords (positive evaluation keywords such as good, like, apprehension and the like; negative evaluation keywords such as bad, disliked and the like) contained in the content of the text information can be respectively extracted, then the text information of which the ratio is smaller than a preset threshold value is deleted by calculating the ratio between the number of the positive evaluation keywords and the number of the negative evaluation keywords, so that the text content approximately belongs to the positive evaluation, and the text information of the interest point of the user can be reflected more accurately.
102. Performing natural language processing on the text information to obtain the interest point tag of the appointed user;
and after obtaining the text information, carrying out natural language processing on the text information to obtain the interest point tag of the appointed user. Natural speech processing, NLP processing, is primarily intended to parse the semantics of text information to determine points of interest for the specified user.
Specifically, the natural language processing includes:
(1) Extracting the text content of the text information;
first, the body content of the text information is extracted. The text information is usually a text material, and comprises contents of each part such as a title, a abstract, a body and the like.
(2) Detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
then, a product name word contained in the text content is detected. The product name words are pre-constructed phrase for representing the product name, and can be brand names which are relatively known in the market, such as XX cola, XX shampoo and the like. The system collects a plurality of product or brand name words in advance, classifies the name words according to product categories and stores the name words in a specified database. Through detection, all product name words contained in the text content can be extracted.
(3) Inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
and then, inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label. When the system classifies the collected product name words according to the product categories and stores the product name words in a specified database, a product tag comparison table can be synchronously created, the table records the tag corresponding to each product name word, and the tag is used for representing the product category of the product name word. For example, the product name word "XX cola" corresponds to a "beverage" label, and the product name word is a well-known clothing brand name, and then corresponds to a "clothing" label. If the detected product name word contains "XX cola" and a well-known clothing brand name, then the first point of interest tag may be determined to be "beverage" and "clothing".
(4) Deleting the product name words contained in the text content to obtain a target text;
and deleting the product name words contained in the text content to obtain a target text. The product name words are generally names of some well-known brands, and when the NLP processing is executed, the names cannot be segmented according to common Chinese characters or words and subsequent semantic recognition, so that the product name words need to be deleted to obtain target texts, and the accuracy of the subsequent NLP processing is improved.
(5) Executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
then, word segmentation, part-of-speech tagging and stop word deletion operations are performed on the target text. The word segmentation can generally adopt various word segmentation methods based on character string matching, understanding, statistics, rules and the like. Part of speech tagging is to label each word with part of speech, such as verbs, adjectives, nouns and the like, and common part of speech tagging methods comprise a part of speech tagging method based on statistics, a part of speech tagging method based on maximum entropy, a part of speech tagging method based on statistics maximum probability output part of speech, a part of speech tagging method based on HMM and the like. Deleting stop words refers to deleting words that do not contribute to text features, such as punctuation marks, mood words, and human pronouns.
(6) Converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
(7) Determining a second interest point label of the appointed user according to an output result of the neural network model;
and then, converting the target text after the word segmentation, the part-of-speech tagging and the stop word deletion operation into word vectors, inputting a pre-constructed neural network model, and determining a second interest point label of the appointed user according to an output result of the neural network model. The words are converted into word vectors to represent data that can be recognized and calculated by the computer. The neural network model is trained by using text features corresponding to different labels as training sets, and one or more labels are output through the neural network model to serve as second interest point labels of the appointed user. Specifically, when the neural network model is trained, text features corresponding to different labels are adopted as training sets, and corresponding second interest point labels are determined by comparing the matching degree of the text features. For example, if the matching degree between the input text feature and the text feature corresponding to the label "export product" is highest, the second interest point label may be determined to be "export product".
(8) And determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
And finally, determining the first interest point tag and the second interest point tag as interest point tags of the appointed user, thereby completing a natural language processing process and successfully obtaining the interest point tags of the appointed user.
103. Searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects;
after obtaining the interest point label of the appointed user, searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelationship among all preset objects.
The knowledge graph is a series of different graphs for displaying the knowledge development process and the structural relationship, the knowledge resource and the carrier thereof are described by using a visualization technology, and knowledge and the interrelation between the knowledge resource and the carrier thereof are mined, analyzed, constructed, drawn and displayed. The knowledge graph records the interrelationship among all preset objects, such as the interrelationship among all enterprises, products and individuals.
Specifically, the association relationship between the interest point tag and each preset object in the knowledge graph may be pre-constructed, for example, the interest point tag "import product" is associated with the object "company a", and the interest point tag "internet product" is associated with the object "company B" and "company C". Then, the corresponding target object can be found out from the knowledge graph through the determined interest point label.
Further, step 103 may include:
(1) Searching a first object associated with the interest point label from the knowledge graph;
(2) Acquiring a second object with a connection relation with the first object in the knowledge graph;
(3) The first object and the second object are determined to be the target object.
For example, if the point of interest tag "import product" is associated with the object "company a", the "company a" is the first object associated with the point of interest tag, and in the knowledge graph, the second object having a connection relationship with the object "company a" includes "company B" and "product C", and then "company a", "company B" and "product C" are all taken as the target objects.
104. Pushing product information associated with the target object to the specified user.
And finally, pushing the product information associated with the target object to the appointed user, so that the product information is pushed according to the interest points of the user, and the accuracy of information pushing is improved. The product information associated with the target object may be various types of information such as text, pictures, network connections, or video, etc.
Further, the product information is in the form of a network link, and after step 104, may further include:
(1) After a preset duration, judging whether the appointed user clicks the network link;
(2) If the appointed user does not click the network link, deducting an accuracy score first score of the knowledge graph;
(3) If the appointed user clicks the network link, increasing the accuracy score second score of the knowledge graph;
(4) And if the accuracy score of the knowledge graph is smaller than a preset threshold value, outputting preset indication information.
Specifically, the server sends the pushed network link to the terminal device of the designated user, if the user clicks the network link to access the corresponding product information within a certain time (for example, 1 day), the accuracy score of the knowledge graph is increased by a first score (for example, 2 scores), and if the user does not click the network link, the accuracy score of the knowledge graph is deducted by the first score (for example, 10 scores). If the accuracy score of the knowledge graph is smaller than a preset threshold (for example, 60 minutes), generating preset alarm indication information to remind related personnel of paying attention to the fact that the accuracy of the knowledge graph is too low, and adjusting and correcting are needed. Through this setting, can further improve the accuracy of information push.
The information pushing method provided by the embodiment of the invention comprises the following steps: acquiring text information of a designated user; performing natural language processing on the text information to obtain the interest point tag of the appointed user; searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects; pushing product information associated with the target object to the specified user. The process comprises the steps of obtaining text information of a user, and carrying out natural language processing on the text information to determine interest points of the user; and then searching target objects associated with the interest points from the pre-constructed knowledge graph, and pushing product information related to the target objects for the user, so that the accuracy of information pushing can be improved, and unnecessary bandwidth resource waste can be reduced.
Referring to fig. 2, a second embodiment of an information pushing method in an embodiment of the present invention includes:
201. acquiring text information of a designated user;
202. performing natural language processing on the text information to obtain the interest point tag of the appointed user;
203. searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects;
steps 201-203 are identical to steps 101-103 and reference is made specifically to the description of steps 101-103.
204. Counting the number of the target objects;
205. if the number of the target objects exceeds a preset threshold, acquiring user information of the appointed user;
206. scoring each target object according to the user information;
207. deleting the target objects with the lowest scores and the preset number;
in some cases, there are many target objects found from the customer relationship map and associated with the interest point tags, and if all corresponding product information of the target objects is pushed to the designated user, on the one hand, the data size is large, and on the other hand, many pieces of information which are not interesting for the user may be included in the information, which may cause interference to the customer. To solve this problem, embodiments of the present invention may count the number of the target objects, and if the number exceeds a certain threshold, acquire the user information of the specified user. If the appointed user is a personal user, personal information of the appointed user can be obtained as the user information; if the appointed user is an enterprise user, a purchase record of the appointed user can be obtained as the user information. And then, scoring each target object according to the user information, and deleting the target objects with the lowest scores and the preset quantity so as to complete the screening process.
For example, if the designated user is a personal user, personal information such as age, gender, occupation, address and the like can be obtained to score each target object, and objects of interest of people of different ages, sexes, occupations and the like are generally different, and the higher the degree of interest is, the higher the score is, for example, the score for target objects such as female users, clothes, cosmetics and the like is; for users of the engineer profession, the scores of various digital products are higher. If the designated user is an enterprise user, the purchasing record of the enterprise can be obtained, the purchasing times and the number of various products can be counted according to the purchasing records, then the various products (namely the target objects) are scored according to the purchasing times and the number, and the score is higher as the purchasing times and the number are more.
208. Pushing product information associated with the target object to the specified user.
Step 208 is identical to step 104, and reference is specifically made to the description associated with step 104.
The information pushing method provided by the embodiment of the invention comprises the following steps: acquiring text information of a designated user; performing natural language processing on the text information to obtain the interest point tag of the appointed user; searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects; counting the number of the target objects; if the number of the target objects exceeds a preset threshold, acquiring user information of the appointed user; scoring each target object according to the user information; deleting the target objects with the lowest scores and the preset number; pushing product information associated with the target object to the specified user. According to the method and the device for searching the target objects associated with the interest point labels, the number of the target objects is counted after the target objects associated with the interest point labels are searched, and if the number is too large, the target objects are screened according to the user information of the appointed user, so that product information can be pushed to the user more accurately, and unnecessary bandwidth resource waste is reduced.
It should be understood that the sequence number of each step in the foregoing embodiment does not mean that the execution sequence of each process should be determined by the function and the internal logic, and should not limit the implementation process of the embodiment of the present invention.
The above mainly describes an information pushing method, and an information pushing apparatus will be described below.
Referring to fig. 3, an embodiment of an information pushing device according to an embodiment of the present invention includes:
a text acquisition module 301, configured to acquire text information of a specified user;
the natural language processing module 302 is configured to perform natural language processing on the text information to obtain a point of interest tag of the specified user;
the target object searching module 303 is configured to search a target object associated with the interest point tag from a pre-constructed knowledge graph, where the knowledge graph records a correlation between preset objects;
an information pushing module 304, configured to push product information associated with the target object to the specified user;
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
Further, the information pushing device may further include:
the keyword extraction module is used for extracting a first keyword and a second keyword which are contained in the text information, wherein the first keyword is a pre-defined positive evaluation keyword, and the second keyword is a pre-defined negative evaluation keyword;
the keyword number counting module is used for counting the number of the first keywords and the number of the second keywords respectively;
and the text information deleting module is used for deleting the text information if the ratio of the number of the first keywords to the number of the second keywords is smaller than a preset threshold value.
Further, the target object searching module may include:
the first object searching unit is used for searching a first object associated with the interest point label from the knowledge graph;
the second object searching unit is used for acquiring a second object with a connection relation with the first object in the knowledge graph;
and a target object determining unit configured to determine the first object and the second object as the target objects.
Further, the information pushing device may further include:
the target object quantity counting module is used for counting the quantity of the target objects;
the user information acquisition module is used for acquiring the user information of the appointed user if the number of the target objects exceeds a preset threshold;
the target object scoring module is used for scoring each target object according to the user information;
and the target object deleting module is used for deleting the target objects with the lowest scores and the preset number.
Still further, the user information obtaining module may include:
a first information obtaining unit, configured to obtain, if the specified user is a personal user, personal information of the specified user as the user information;
and the second information acquisition unit is used for acquiring the purchase record of the appointed user as the user information if the appointed user is an enterprise user.
Further, the product information is in the form of a network link, and the information pushing device may further include: :
the click judging module is used for judging whether the appointed user clicks the network link or not after the preset duration;
the accuracy score deduction module is used for deducting the accuracy score first score of the knowledge graph if the specified user does not click on the network link;
the accuracy score increasing module is used for increasing the accuracy score second score of the knowledge graph if the appointed user clicks the network link;
and the indication information output module is used for outputting preset indication information if the accuracy score of the knowledge graph is smaller than a preset threshold value.
The embodiment of the invention also provides a computer readable storage medium, which stores computer readable instructions, and the computer readable instructions implement steps of any information pushing method as shown in fig. 1 or fig. 2 when being executed by a processor.
The embodiment of the invention also provides a server, which comprises a memory, a processor and computer readable instructions stored in the memory and capable of running on the processor, wherein the steps of any information pushing method shown in fig. 1 or 2 are realized when the processor executes the computer readable instructions.
Fig. 4 is a schematic diagram of a server according to an embodiment of the present invention. As shown in fig. 4, the server 4 of this embodiment includes: a processor 40, a memory 41, and computer readable instructions 42 stored in the memory 41 and executable on the processor 40. The steps of the various information pushing method embodiments described above, such as steps 101 through 104 shown in fig. 1, are implemented when the processor 40 executes the computer readable instructions 42. Alternatively, the processor 40, when executing the computer readable instructions 42, performs the functions of the modules/units of the apparatus embodiments described above, such as the functions of modules 301 through 304 shown in fig. 3.
Illustratively, the computer readable instructions 42 may be partitioned into one or more modules/units that are stored in the memory 41 and executed by the processor 40 to complete the present invention. The one or more modules/units may be a series of computer readable instructions capable of performing a particular function describing the execution of the computer readable instructions 42 in the server 4.
The server 4 may be a computing device such as a smart phone, a notebook, a palm computer, a cloud server, etc. The server 4 may include, but is not limited to, a processor 40, a memory 41. It will be appreciated by those skilled in the art that fig. 4 is merely an example of the server 4 and does not constitute a limitation of the server 4, and may include more or fewer components than shown, or may combine certain components, or different components, e.g., the server 4 may further include input-output devices, network access devices, buses, etc.
The processor 40 may be a central processing unit (CentraL Processing Unit, CPU), but may also be other general purpose processors, digital signal processors (DigitaL SignaL Processor, DSP), application specific integrated circuits (AppLication Specific Integrated Circuit, ASIC), off-the-shelf programmable gate arrays (fierld-ProgrammabLe Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 41 may be an internal storage unit of the server 4, such as a hard disk or a memory of the server 4. The memory 41 may be an external storage device of the server 4, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure DigitaL (SD) Card, a FLash Card (FLash Card) or the like, which are provided on the server 4. Further, the memory 41 may also include both an internal storage unit and an external storage device of the server 4. The memory 41 is used to store the computer readable instructions and other programs and data required by the server. The memory 41 may also be used for temporarily storing data that has been output or is to be output.
It will be clear to those skilled in the art that, for convenience and brevity of description, specific working procedures of the above-described systems, apparatuses and units may refer to corresponding procedures in the foregoing method embodiments, which are not repeated herein.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a usb disk, a removable hard disk, a Read-OnLy Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The above embodiments are only for illustrating the technical solution of the present invention, and not for limiting the same; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (10)
1. An information pushing method is characterized by comprising the following steps:
acquiring text information of a designated user; the text information comprises text materials provided by a user and text materials containing related keywords on a network;
performing natural language processing on the text information to obtain the interest point tag of the appointed user;
searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects;
pushing product information associated with the target object to the designated user;
wherein the natural language processing process comprises:
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
2. The information pushing method according to claim 1, further comprising, after acquiring text information of the specified user:
extracting a first keyword and a second keyword contained in the text information, wherein the first keyword is a pre-defined positive evaluation keyword, and the second keyword is a pre-defined negative evaluation keyword;
counting the number of the first keywords and the number of the second keywords respectively;
and deleting the text information if the ratio of the number of the first keywords to the number of the second keywords is smaller than a preset threshold value.
3. The information pushing method according to claim 1, wherein the searching for the target object associated with the interest point tag from the pre-constructed knowledge graph includes:
searching a first object associated with the interest point label from the knowledge graph;
acquiring a second object with a connection relation with the first object in the knowledge graph;
the first object and the second object are determined to be the target object.
4. The information pushing method according to claim 1, further comprising, after searching for a target object associated with the point of interest tag from a pre-constructed knowledge graph:
counting the number of the target objects;
if the number of the target objects exceeds a preset threshold, acquiring user information of the appointed user;
scoring each target object according to the user information;
and deleting the target objects with the lowest scores and the preset number.
5. The information pushing method according to claim 4, wherein the acquiring the user information of the specified user includes:
if the appointed user is a personal user, personal information of the appointed user is obtained as the user information;
and if the appointed user is an enterprise user, acquiring a purchase record of the appointed user as the user information.
6. The information pushing method according to any one of claims 1 to 5, characterized in that the product information is in the form of a web link, further comprising, after pushing the product information associated with the target object to the specified user:
after a preset duration, judging whether the appointed user clicks the network link;
if the appointed user does not click the network link, deducting an accuracy score first score of the knowledge graph;
if the appointed user clicks the network link, increasing the accuracy score second score of the knowledge graph;
and if the accuracy score of the knowledge graph is smaller than a preset threshold value, outputting preset indication information.
7. An information pushing apparatus, characterized by comprising:
the text acquisition module is used for acquiring text information of a designated user; the text information comprises text materials provided by a user and text materials containing related keywords on a network;
the natural language processing module is used for carrying out natural language processing on the text information to obtain the interest point tag of the appointed user;
the target object searching module is used for searching target objects associated with the interest point labels from a pre-constructed knowledge graph, and the knowledge graph records the interrelationship among all the preset objects;
the information pushing module is used for pushing the product information associated with the target object to the appointed user;
wherein the natural language processing process comprises:
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
8. A computer readable storage medium storing computer readable instructions which, when executed by a processor, implement the steps of the information pushing method according to any of claims 1 to 6.
9. A server comprising a memory, a processor, and computer readable instructions stored in the memory and executable on the processor, wherein the processor, when executing the computer readable instructions, performs the steps of:
acquiring text information of a designated user; the text information comprises text materials provided by a user and text materials containing related keywords on a network;
performing natural language processing on the text information to obtain the interest point tag of the appointed user;
searching a target object associated with the interest point label from a pre-constructed knowledge graph, wherein the knowledge graph records the interrelation between all preset objects;
pushing product information associated with the target object to the designated user;
wherein the natural language processing process comprises:
extracting the text content of the text information;
detecting product name words contained in the text content, wherein the product name words are pre-constructed phrase groups used for representing product names;
inquiring a label corresponding to the detected product name word from a pre-constructed product label comparison table to be used as a first interest point label;
deleting the product name words contained in the text content to obtain a target text;
executing word segmentation, part-of-speech tagging and stop word deletion operations on the target text;
converting the target text after word segmentation, part-of-speech tagging and stop word deletion operation into word vectors, and inputting a pre-constructed neural network model, wherein the neural network model is trained by taking text features corresponding to different labels as training sets;
determining a second interest point label of the appointed user according to an output result of the neural network model;
and determining the first interest point tag and the second interest point tag as interest point tags of the appointed user.
10. The server according to claim 9, further comprising, after acquiring text information of the specified user:
extracting a first keyword and a second keyword contained in the text information, wherein the first keyword is a pre-defined positive evaluation keyword, and the second keyword is a pre-defined negative evaluation keyword;
counting the number of the first keywords and the number of the second keywords respectively;
and deleting the text information if the ratio of the number of the first keywords to the number of the second keywords is smaller than a preset threshold value.
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 CN110263248A (en) | 2019-09-20 |
| CN110263248B true 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) |
Families Citing this family (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 |
| 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 |
| 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 |
| 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 |
| CN112104734B (en) * | 2020-09-15 | 2022-09-02 | 北京百度网讯科技有限公司 | Method, device, equipment and storage medium for pushing information |
| 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 |
| 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 |
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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110263248A (en) | 2019-09-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110263248B (en) | Information pushing method, device, storage medium and server | |
| CN106649818B (en) | Application search intention identification method and device, application search method and server | |
| CN108536852B (en) | Question-answer interaction method and device, computer equipment and computer readable storage medium | |
| US8676730B2 (en) | Sentiment classifiers based on feature extraction | |
| CN108108426B (en) | Understanding method and device for natural language question and electronic equipment | |
| US20130159277A1 (en) | Target based indexing of micro-blog content | |
| KR20200007969A (en) | Information processing methods, terminals, and computer storage media | |
| CN110413787B (en) | Text clustering method, device, terminal and storage medium | |
| CN111538816B (en) | Question-answering method, device, electronic equipment and medium based on AI identification | |
| WO2019041520A1 (en) | Social data-based method of recommending financial product, electronic device and medium | |
| US11893537B2 (en) | Linguistic analysis of seed documents and peer groups | |
| CN112989208B (en) | Information recommendation method and device, electronic equipment and storage medium | |
| CN112883734B (en) | Block chain security event public opinion monitoring method and system | |
| WO2016036345A1 (en) | External resource identification | |
| CN111782793A (en) | Intelligent customer service processing method, system and equipment | |
| CN114238573A (en) | Information pushing method and device based on text countermeasure sample | |
| CN114387061A (en) | Product pushing method and device, electronic equipment and readable storage medium | |
| KR102185733B1 (en) | Server and method for automatically generating profile | |
| CN109753646B (en) | Article attribute identification method and electronic equipment | |
| CN107688594B (en) | The identifying system and method for risk case based on social information | |
| CN116882414B (en) | Automatic comment generation method and related device based on large-scale language model | |
| CN117278675A (en) | Outbound method, device, equipment and medium based on intention classification | |
| CN113392195A (en) | Public opinion monitoring method and device, electronic equipment and storage medium | |
| CN110705285B (en) | Government affair text subject word library construction method, device, server and readable storage medium | |
| CN112364068A (en) | Course label generation method, device, equipment and medium |
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 |