CN107908792A - Information-pushing method and device - Google Patents
Information-pushing method and device Download PDFInfo
- Publication number
- CN107908792A CN107908792A CN201711328500.8A CN201711328500A CN107908792A CN 107908792 A CN107908792 A CN 107908792A CN 201711328500 A CN201711328500 A CN 201711328500A CN 107908792 A CN107908792 A CN 107908792A
- Authority
- CN
- China
- Prior art keywords
- text
- word
- detected
- verse
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- 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
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Machine Translation (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the present application discloses information-pushing method and device.One embodiment of this method includes：Receive the text detection request of the subject matter comprising text to be detected and the text to be detected；Determine the target word and target sentence of the text to be detected, and the position based on the target word in the text to be detected and position of the target sentence in the text to be detected, generate the words and phrases distributed intelligence of the text to be detected；The words and phrases distributed intelligence, is inputted the structure evaluation information for the structure detection model, obtaining the text to be detected by extraction and the structure detection model of the type matching of the text to be detected；The text detection result of structure evaluation information of the push comprising the text to be detected.This embodiment improves the flexibility of the detection to text.
Description
Technical field
The invention relates to field of computer technology, and in particular to Internet technical field, more particularly to information push away
Delivery method and device.
Background technology
As the promotion and popularization of Chinese, Chinese writing are more and more important.After Chinese is write, it usually needs to what is write
Article is corrected and checked, to ensure that write content is errorless, has readability.
It is existing that method is usually manual to correct is corrected to article, or the auxiliary using existing Chinese writing field
Product is corrected, these products are usually only enumerated all kinds of writing materials for selection by the user, can not be to text to be detected
The processing such as originally it is reviewed, corrects.Existing English text detection instrument is not suitable for Chinese detection.
The content of the invention
The embodiment of the present application proposes information-pushing method and device.
In a first aspect, the embodiment of the present application provides a kind of information-pushing method, this method includes：Receive comprising to be detected
The text detection request of the subject matter of text and text to be detected；Determine the target word and target sentence of text to be detected, and be based on mesh
Position and target sentence position in be detected text of the word in text to be detected is marked, generates the words and phrases distribution of text to be detected
Information, wherein, the sentence that target sentence matches for clause with any default clause in default clause set；Extraction and subject matter phase
Words and phrases distributed intelligence, is inputted the structure evaluation for structure detection model, obtaining text to be detected by matched structure detection model
Information, structure detection model are used to characterize words and phrases distributed intelligence and the correspondence of structure evaluation information；Push is comprising to be detected
The text detection result of the structure evaluation information of text.
In certain embodiments, text detection result further includes clause testing result, and clause testing result passes through following step
Rapid generation：Cutting text to be detected sentence by sentence；For each sentence after cutting, determine that this whether there is the word of collocation error,
If so, the recommendation word that extraction is mutually arranged in pairs or groups with word, and generate and be used to prompt this to there are problems that the prompt message of collocation, wherein,
Prompt message includes recommending word；Collect generated prompt message, generate clause testing result.
In certain embodiments, generated prompt message is being collected, before generating clause testing result, generation clause inspection
The step of surveying result further includes：For each target sentence, the default clause to match with the target sentence is preset as target
Clause, optimizes the target sentence based on the default principle of optimality that clause matches is preset with target, and generation is used to prompting pair
The prompt message that this optimizes.
In certain embodiments, text detection result further includes word testing result, and word testing result passes through following step
Rapid generation：Extract the keyword and verse in text to be detected；For each keyword extracted, in response to determining key
Word is noun, and collocation information corresponding with the keyword is searched from default keyword collocation information table, in response to determining to be somebody's turn to do
Keyword is adjective, and the synonym of the keyword is searched from default synonym table, and generation includes collocation information or synonymous
The word recommendation information of word；For each verse extracted, searched from default verse storehouse associated with the verse
Verse is associated, generation includes the verse recommendation information of association verse；Collect generated word recommendation information and verse recommendation
Breath, generates word testing result.
In certain embodiments, text detection result further includes wrong word testing result, wrong word testing result by with
Lower step generation：Extract the verse in text to be detected；Text to be detected in addition to verse is segmented, will be each after participle
A word is matched with default word storehouse, and verse is matched with default verse storehouse；Based on matching result, determine to treat
The verse for including wrong word in the word comprising wrong word and text to be detected in text is detected, wrong word is included with generation
The wrong word testing result of information.
Second aspect, the embodiment of the present application provide a kind of information push-delivery apparatus, which includes：Receiving unit, configuration
For receiving the text detection request of the subject matter comprising text to be detected and text to be detected；Generation unit, is configured to determine
The target word and target sentence of text to be detected, and based on position of the target word in text to be detected and target sentence in text to be detected
Position in this, generates the words and phrases distributed intelligence of text to be detected, wherein, target sentence is clause and appointing in default clause set
The sentence that one default clause matches；Input unit, is configured to the structure detection model that extraction matches with subject matter, by words and phrases
Distributed intelligence inputs the structure evaluation information for structure detection model, obtaining text to be detected, and structure detection model is used to characterize
Words and phrases distributed intelligence and the correspondence of structure evaluation information；Push unit, is configured to the knot that push includes text to be detected
The text detection result of structure evaluation information.
In certain embodiments, text detection result further includes clause testing result, and device further includes：Cutting unit, matches somebody with somebody
Put for cutting text to be detected sentence by sentence；First determination unit, is configured to, for each sentence after cutting, whether determine the sentence
There are the word of collocation error, if so, the recommendation word that extraction is mutually arranged in pairs or groups with word, and generate that there are collocation for prompting this to ask
The prompt message of topic, wherein, prompt message includes recommending word；First collection unit, is configured to collect generated prompting letter
Breath, generates clause testing result.
In certain embodiments, device further includes：Optimize unit, be configured to for each target sentence, will be with the mesh
The default clause that matches of mark sentence presets clause as target, based on presetting the default principle of optimality pair that clause matches with target
The target sentence optimizes, and generation is used to prompt the prompt message for optimizing this.
In certain embodiments, text detection result further includes word testing result, and device further includes：First extraction is single
Member, is configured to extract keyword and verse in text to be detected；First searching unit, is configured to every for what is extracted
One keyword, in response to determining that keyword is noun, is searched and the keyword pair from default keyword collocation information table
The collocation information answered, in response to determining that the keyword is adjective, searches the synonymous of the keyword from default synonym table
Word, word recommendation information of the generation comprising collocation information or synonym；Second searching unit, is configured to every for what is extracted
One verse, searches the association verse associated with the verse from default verse storehouse, and generation includes the verse of association verse
Recommendation information；Second collection unit, is configured to collect generated word recommendation information and verse recommendation information, generates word
Testing result.
In certain embodiments, text detection result further includes wrong word testing result, and device further includes：Second extraction is single
Member, is configured to extract the verse in text to be detected；Matching unit, is configured to carry out the text to be detected in addition to verse
Participle, each word after participle is matched with default word storehouse, and verse is matched with default verse storehouse；The
Two determination units, are configured to be based on matching result, determine the word comprising wrong word in text to be detected and text to be detected
The verse for including wrong word in this, the wrong word testing result of wrong word information is included with generation.
The third aspect, the embodiment of the present application provide a kind of server, including：One or more processors；Storage device,
For storing one or more programs, when one or more programs are executed by one or more processors so that one or more
Processor is realized such as the method for any embodiment in information-pushing method.
Fourth aspect, the embodiment of the present application provide a kind of computer-readable recording medium, are stored thereon with computer journey
Sequence, is realized such as the method for any embodiment in information-pushing method when which is executed by processor.
Information-pushing method and device provided by the embodiments of the present application, by the mesh for determining received text to be detected
Word and target sentence are marked, to generate words and phrases distributed intelligence in position based on target word and target sentence, then by words and phrases distributed intelligence
The structure detection model that input matches to the subject matter with text to be detected, obtains the structure evaluation information of text to be detected, most
Text detection of the push comprising the structure evaluation information improves as a result, from without relying on checking for artificial progress text afterwards
The flexibility of detection to text.
Brief description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, the application's is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 is that this application can be applied to exemplary system architecture figure therein；
Fig. 2 is the flow chart according to one embodiment of the information-pushing method of the application；
Fig. 3 is the schematic diagram according to an application scenarios of the information-pushing method of the application；
Fig. 4 is the structure diagram according to one embodiment of the information push-delivery apparatus of the application；
Fig. 5 is adapted for the structure diagram of the computer system of the server for realizing the embodiment of the present application.
Embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
It illustrate only easy to describe, in attached drawing and invent relevant part with related.
It should be noted that in the case where there is no conflict, the feature in embodiment and embodiment in the application can phase
Mutually combination.Describe the application in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the exemplary system architecture of the information-pushing method or information push-delivery apparatus that can apply the application
100。
As shown in Figure 1, system architecture 100 can include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 provide communication link medium.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be interacted with using terminal equipment 101,102,103 by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed, such as text editing class should on terminal device 101,102,103
With, image recognition class application, web browser applications, searching class application, instant messaging tools, mailbox client, social platform
Software etc..
Terminal device 101,102,103 can have a display screen and a various electronic equipments that supported web page browses, bag
Include but be not limited to smart mobile phone, tablet computer, pocket computer on knee and desktop computer etc..
Server 105 can be to provide the server of various services, such as to transmitted by terminal device 101,102,103
The information such as text to be detected provide the Cloud Server for service of checking.Cloud Server can dock the numbers such as received text detection request
According to carrying out the processing such as analyzing, and handling result (such as text detection result) is fed back into terminal device.
It should be noted that the information-pushing method that the embodiment of the present application is provided generally is performed by server 105, accordingly
Ground, information push-delivery apparatus are generally positioned in server 105.
It should be understood that the number of the terminal device, network and server in Fig. 1 is only schematical.According to realizing need
Will, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the flow 200 of one embodiment of information-pushing method according to the application is shown.The letter
Method for pushing is ceased, is comprised the following steps：
Step 201, the text detection request of the subject matter comprising text to be detected and text to be detected is received.
In the present embodiment, the electronic equipment (such as server 105 shown in Fig. 1) of information-pushing method operation thereon
Can be received by wired connection mode or radio connection client (such as terminal device 101 shown in Fig. 1,102,
103) the text detection request of subject matter send, comprising text to be detected and the text to be detected.Wherein, above-mentioned text to be detected
Originally can be the Chinese articles of various subject matters, such as narrative, argumentative writing, expository writing, prose, poem, letter of application, recommendation
Deng.It is pointed out that above-mentioned radio connection can include but is not limited to 3G/4G connections, WiFi connections, bluetooth connection,
WiMAX connections, Zigbee connections, UWB (ultra wideband) connections and other it is currently known or in the future exploitation it is wireless
Connection mode.
It should be noted that other information, such as above-mentioned text to be verified can also be included in above-mentioned text verification request
Label etc..Herein, which can be that user is manually entered, such as " friendship, friend ".
In practice, it can be installed in terminal device (terminal device 101,102,103 as shown in Figure 1) used by a user
There are the application of text detection class or browser application, user can detect class application or the text inspection of browser application in the text
Survey in the page and edit above-mentioned text to be detected, and choose the subject matter of the text to be detected.Then, user can be above-mentioned by clicking on
" starting to detect " button in the page of the application of text detection class or above-mentioned browser application, sends to above-mentioned electronic equipment and includes
The text detection of above-mentioned text to be detected and above-mentioned subject matter is asked.Furthermore, it is necessary to explanation, the application of above-mentioned text detection class or
The text detection page of person's browser application can also support image to import, at this point it is possible to utilize OCR (Optical
Character Recognition, optical character identification) word in image is identified in technology, the word that will identify that
It is determined as word to be verified.
Step 202, the target word and target sentence of text to be detected, and the position based on target word in text to be detected are determined
The position in text to be detected with target sentence is put, generates the words and phrases distributed intelligence of text to be detected.
In the present embodiment, above-mentioned electronic equipment can determine the target word and target sentence of above-mentioned text to be detected first.
It is pointed out that above-mentioned target sentence can be the sentence that clause matches with any default clause in default clause set,
Wherein, default clause can include but is not limited to parallelism sentence, anthropomorphic sentence, put up a question sentence, antithesis sentence etc..On it should be noted that
Stating target word can include being derived from the title of above-mentioned text to be detected, head (such as the first paragragh or the first two paragragh
Deng), the higher one or more names of the frequency of occurrences in afterbody (such as last paragragh or last three paragraghs etc.)
Word (is properly termed as centre word)；It can also include the synonym and near synonym of the above-mentioned centre word in above-mentioned text to be detected；Also
It can include the qualifier for being used to modify above-mentioned centre word, above-mentioned synonym and above-mentioned near synonym in above-mentioned text to be detected.
As an example, if the higher noun of the frequency of occurrences is " friendship " (i.e. centre word) in the title of certain text, head and afterbody, mesh
Mark word can include the following word being derived from the text：" friendship ", " friend ", " boudoir honey ", " bosom friend ", " brother ", " intimate nothing
Between ", " loyal-hearted ", " making something perfect even more perfect ".
After the target word of above-mentioned text to be detected and target sentence is determined, above-mentioned electronic equipment can determine each target
The position of word and target sentence in above-mentioned text to be detected, and based on position of the target word in above-mentioned text to be detected and above-mentioned
Position of the target sentence in above-mentioned text to be detected, generates the words and phrases distributed intelligence of above-mentioned text to be detected.Wherein, this above-mentioned sentence
Distributed intelligence can include being used for the type (such as centre word, near synonym, synonym, qualifier etc.) for indicating each target word
Target word type identification (such as can be the character string being made of various characters, to distinguish the type of target word), be used for
Indicate the target sentence type identification (example of the type (such as parallelism sentence, antithesis sentence, anthropomorphic sentence, rhetoric question sentence etc.) of each target sentence
Such as can be the character string being made of various characters, to distinguish the type of target sentence), for indicating that each target word exists
State the information (such as paragraph label) of the position in text to be detected, for indicating each target word in above-mentioned text to be detected
In the information of position etc..
Step 203, the structure detection model that extraction matches with subject matter, words and phrases distributed intelligence is inputted to structure detection mould
Type, obtains the structure evaluation information of text to be detected.
In the present embodiment, the structure to match with the text of various subject matters can be previously stored with above-mentioned electronic equipment
Detection model.After the words and phrases distributed intelligence of above-mentioned text to be detected is generated, above-mentioned electronic equipment can be extracted to be treated with above-mentioned
The structure detection model that the subject matter of detection text matches, the words and phrases distributed intelligence generated is inputted to the structure detection mould
Type, obtains the structure evaluation information of text to be detected.Wherein, said structure evaluation information can be above-mentioned to be detected for characterizing
The topology layout quality of text scoring (such as 80 points, 90 points, 100 grade) or grading (such as A grades, B grades, C grades etc.) etc..Need
It is noted that said structure detection model can be used for characterizing words and phrases distributed intelligence and the correspondence of structure evaluation information,
For example, said structure detection model can be words and phrases distributed intelligence of the technical staff based on substantial amounts of text and structure evaluation information
Statistics after the words and phrases distributed intelligence formulated and the mapping table of structure evaluation information.
In some optional implementations of the present embodiment, said structure detection model can be given birth to by following steps
Into：It is possible, firstly, to preset training sample is extracted, wherein, above-mentioned training sample can include each text in multiple texts
This words and phrases distributed intelligence and structure evaluation information.Then, above-mentioned electronic equipment can utilize machine learning method, by training sample
Words and phrases distributed intelligence in this, using the structure evaluation information in training sample as output, is used for as input using existing
Disaggregated model (such as decision tree, neutral net, Logic Regression Models etc.) training obtains said structure detection model.Need to illustrate
, the method for above-mentioned training pattern is widely studied at present and application known technology, and details are not described herein.
Step 204, the text detection result of structure evaluation information of the push comprising text to be detected.
In the present embodiment, above-mentioned electronic equipment can include upper to the client push for sending above-mentioned text detection request
State the text detection result of the structure evaluation information of text to be detected.Herein, above-mentioned text detection result can also include other
Information, such as identified centre word, layout description information of structure partial for describing above-mentioned text to be detected etc..
In some optional implementations of the present embodiment, above-mentioned text detection result can also include clause detection knot
Fruit.Wherein, above-mentioned clause testing result can be generated by following steps：The first step, can the above-mentioned text to be detected of cutting sentence by sentence
This.Second step, for each sentence after cutting, it may be determined that this whether there is collocation error word, if so, extraction with it is upper
The recommendation word that predicate language is mutually arranged in pairs or groups, and generate and be used to prompt this to there are problems that the prompt message of collocation, wherein, above-mentioned prompting letter
Breath can include above-mentioned recommendation word.Specifically, for each sentence after cutting, above-mentioned electronic equipment can be determined in the sentence first
Subject, predicate, object, attribute, the adverbial modifier, complement；Then, whether can be arranged in pairs or groups by determining subject and predicate (or verb),
The modes such as whether predicate and the adverbial modifier arrange in pairs or groups, whether object and attribute arrange in pairs or groups determine that this whether there is the word of collocation error；If
Determine the word there are collocation error, word that is preset, mutually arranging in pairs or groups with the word can be extracted, using the word extracted as recommendation
Word, generation include the prompt message of the recommendation word.3rd step, can collect generated prompt message, generation clause detection knot
Fruit.As an example, character string " allowing people to ooze into heart " is included in above-mentioned text to be detected, and " oozing into heart " is verb, subject
Usually " fragrance of a flower ", " delicate fragrance " etc., rather than people, hence, it can be determined that " people " is the word of collocation error, and " spend above-mentioned
The word such as perfume ", " delicate fragrance " is determined as recommending word.
In some optional implementations of the present embodiment, collecting generated prompt message, generation clause detection
As a result before, above-mentioned electronic equipment can also carry out following steps：For each target sentence, by what is matched with the target sentence
Default clause presets clause as target, based on presetting the default principle of optimality that clause matches to the target sentence with above-mentioned target
Optimize, generation is used to prompt the prompt message for optimizing this.As an example, it is parallelism sentence that target, which presets clause,
Parallelism sentence generally comprises the similar short sentence of at least three structures, and verb in three short sentences or adverbial word are identical, such as
" XX, as A, as B, more like C ".The principle of optimality to match with parallelism sentence can be：In each short sentence for determining composition parallelism sentence
It whether there is adjective before noun, if being not present, adjective of the increase for describing the noun before noun.
In some optional implementations of the present embodiment, above-mentioned text detection result can also include word detection knot
Fruit.Wherein, above-mentioned word testing result can be generated by following steps：The first step, can extract in above-mentioned text to be detected
Keyword (such as the word to match with the noun in preset keyword dictionary, adjective) and verse.Second step, for being carried
Each keyword taken, in response to determining that above-mentioned keyword is noun, can look into from default keyword collocation information table
Collocation information (such as modifying the adjective of the keyword) corresponding with the keyword is looked for, is in response to the definite keyword
Adjective, searches the synonym of the keyword from default synonym table, and generation includes above-mentioned collocation information or above-mentioned synonymous
The word recommendation information of word.3rd step, for each verse extracted, can search and the poem from default verse storehouse
The associated association verse of sentence (such as verse of characterization same subject (such as friendship, homesick etc.)), generation includes above-mentioned association
The verse recommendation information of verse.4th step, collects generated word recommendation information and verse recommendation information, generation word detection
As a result.
In some optional implementations of the present embodiment, above-mentioned text detection result can also include wrong word and detect
As a result.Wherein, above-mentioned wrong word testing result can be generated by following steps：The first step, is extracted in above-mentioned text to be detected
Verse.Second step, segments the above-mentioned text to be detected in addition to above-mentioned verse, using string matching mode, will divide
Each word after word is matched with default word storehouse, and above-mentioned verse is matched with default verse storehouse.3rd step,
Based on matching result, determine to include mistake in the word comprising wrong word and above-mentioned text to be detected in above-mentioned text to be detected
The verse of malapropism, the wrong word testing result of wrong word information is included with generation.It should be noted that the electronic equipment may be used also
To carry out wrong word detection using other wrong word detection modes, such as by by word or verse and default wrong word dictionary
Or wrong word verse storehouse carries out string matching, to determine the word comprising wrong word in above-mentioned text to be detected and above-mentioned treat
The verse for including wrong word in text is detected, details are not described herein.
With continued reference to Fig. 3, Fig. 3 is a schematic diagram according to the application scenarios of the upgrade method of the present embodiment.Fig. 3's
In application scenarios, user edits text 301 to be detected first by terminal device, and the subject matter that have selected text to be detected is
" narrative-account ", and by clicking on " starting to detect " button, have sent the text detection for including the subject matter with text to be detected
Request.Server upon receiving the request, performs the text 301 to be detected definite target word and target sentence, generation word
The operation such as sentence distributed intelligence and acquisition structure evaluation information, and the structure for including text to be detected has been pushed to above-mentioned terminal device
The text detection result 302 of evaluation information, the text detection result 302 that above-mentioned terminal device is shown.
The method that above-described embodiment of the application provides, by the target word and mesh that determine received text to be detected
Mark sentence, to generate words and phrases distributed intelligence in position based on target word and target sentence, then by words and phrases distributed intelligence input to
The structure detection model that the subject matter of text to be detected matches, obtains the structure evaluation information of text to be detected, finally push bag
Text detection containing the structure evaluation information is improved to text as a result, from without relying on checking for artificial progress text
The flexibility of detection.
With further reference to Fig. 4, as the realization to method shown in above-mentioned each figure, this application provides a kind of push of information to fill
The one embodiment put, the device embodiment is corresponding with the embodiment of the method shown in Fig. 2, which specifically can be applied to respectively
In kind electronic equipment.
As shown in figure 4, the information push-delivery apparatus 400 described in the present embodiment includes：Receiving unit 401, is configured to receive
The text detection request of subject matter comprising text to be detected and above-mentioned text to be detected；Generation unit 402, is configured to determine
State the target word and target sentence of text to be detected, and based on position of the above-mentioned target word in above-mentioned text to be detected and above-mentioned mesh
Position of the sentence in above-mentioned text to be detected is marked, generates the words and phrases distributed intelligence of above-mentioned text to be detected, wherein, above-mentioned target sentence
The sentence to match for any default clause in clause and default clause set；Input unit 403, be configured to extraction with it is upper
The structure detection model that subject matter matches is stated, above-mentioned words and phrases distributed intelligence is inputted to above-mentioned structure detection model, is obtained above-mentioned
The structure evaluation information of text to be detected, said structure detection model are used to characterize words and phrases distributed intelligence and structure evaluation information
Correspondence；Push unit 404, is configured to the text detection knot of structure evaluation information of the push comprising above-mentioned text to be detected
Fruit.
In the present embodiment, above-mentioned receiving unit 401 can receive client transmission, comprising text to be detected and this treat
Detect the text detection request of the subject matter of text.
In the present embodiment, above-mentioned generation unit 402 can determine the target word and target of above-mentioned text to be detected first
Sentence.After the target word of above-mentioned text to be detected and target sentence is determined, above-mentioned generation unit 402 can determine each target word
With position of the target sentence in above-mentioned text to be detected, and based on position of the target word in above-mentioned text to be detected and above-mentioned mesh
Position of the sentence in above-mentioned text to be detected is marked, generates the words and phrases distributed intelligence of above-mentioned text to be detected.
In the present embodiment, above-mentioned input unit 403 can extract the knot to match with the subject matter of above-mentioned text to be detected
Structure detection model, the words and phrases distributed intelligence generated is inputted to the structure detection model, the structure for obtaining text to be detected is commented
Valency information.It should be noted that said structure detection model can be used for characterizing words and phrases distributed intelligence and structure evaluation information
Correspondence.
In the present embodiment, above-mentioned push unit 404 can be to the client push bag for sending above-mentioned text detection request
The text detection result of structure evaluation information containing above-mentioned text to be detected.
In some optional implementations of the present embodiment, above-mentioned text detection result can also include clause detection knot
Fruit, above device further include cutting unit, the first determination unit and the first collection unit (not shown).Wherein, it is above-mentioned to cut
Subdivision may be configured to the above-mentioned text to be detected of cutting sentence by sentence.Above-mentioned first determination unit may be configured to for cutting
Each sentence afterwards, determines that this whether there is the word of collocation error, if so, the recommendation word that extraction is mutually arranged in pairs or groups with above-mentioned word,
And generate and be used to prompt this to there are problems that the prompt message of collocation, wherein, above-mentioned prompt message includes above-mentioned recommendation word.It is above-mentioned
First collection unit may be configured to collect generated prompt message, generate clause testing result.
In some optional implementations of the present embodiment, above device can also include optimization unit and (not show in figure
Go out).Wherein, above-mentioned optimization unit may be configured to the default sentence that for each target sentence, will be matched with the target sentence
Formula presets clause as target, based on above-mentioned target preset the default principle of optimality that clause matches the target sentence is carried out it is excellent
Change, generation is used to prompt the prompt message for optimizing this.
In some optional implementations of the present embodiment, above-mentioned text detection result can also include word detection knot
Fruit, above device can also include the first extraction unit, the first searching unit, the second searching unit and the second collection unit (figure
Not shown in).Wherein, above-mentioned first extraction unit may be configured to extract the keyword and poem in above-mentioned text to be detected
Sentence.Above-mentioned first searching unit may be configured to each keyword for being extracted, in response to determining above-mentioned keyword
For noun, collocation information corresponding with the keyword is searched from default keyword collocation information table, in response to determining the pass
Keyword is adjective, searches the synonym of the keyword from default synonym table, generation comprising above-mentioned collocation information or on
State the word recommendation information of synonym.Above-mentioned second searching unit may be configured to each verse for being extracted, from
The association verse associated with the verse is searched in default verse storehouse, generation includes the verse recommendation of above-mentioned association verse
Breath.Above-mentioned second collection unit may be configured to collect generated word recommendation information and verse recommendation information, generate word
Language testing result.
In some optional implementations of the present embodiment, above-mentioned text detection result can also include wrong word and detect
As a result, above device can also include the second extraction unit, matching unit and the second determination unit (not shown).Wherein,
Above-mentioned second extraction unit may be configured to extract the verse in above-mentioned text to be detected.Above-mentioned matching unit can configure use
Segmented in the above-mentioned text to be detected in addition to above-mentioned verse, by each word after participle and the progress of default word storehouse
Match somebody with somebody, and above-mentioned verse is matched with default verse storehouse.Above-mentioned second determination unit may be configured to based on matching knot
Fruit, determines the poem for including wrong word in the word comprising wrong word and above-mentioned text to be detected in above-mentioned text to be detected
Sentence, the wrong word testing result of wrong word information is included with generation.
The device that above-described embodiment of the application provides, is determined received by receiving unit 401 by generation unit 402
The target word and target sentence of text to be detected, to generate words and phrases distributed intelligence in position based on target word and target sentence, then
Words and phrases distributed intelligence is inputted the structure detection model to match to the subject matter with text to be detected by input unit 403, is treated
Detect the structure evaluation information of text, text detection of last push unit 404 push comprising the structure evaluation information as a result, from
Checking for text is manually carried out without relying on, improves the flexibility of the detection to text.
Below with reference to Fig. 5, it illustrates suitable for for realizing the computer system 500 of the server of the embodiment of the present application
Structure diagram.Server shown in Fig. 5 is only an example, should not be to the function and use scope band of the embodiment of the present application
Carry out any restrictions.
As shown in figure 5, computer system 500 includes central processing unit (CPU) 501, it can be read-only according to being stored in
Program in memory (ROM) 502 or be loaded into program in random access storage device (RAM) 503 from storage part 508 and
Perform various appropriate actions and processing.In RAM 503, also it is stored with system 500 and operates required various programs and data.
CPU 501, ROM 502 and RAM 503 are connected with each other by bus 504.Input/output (I/O) interface 505 is also connected to always
Line 504.
I/O interfaces 505 are connected to lower component：Importation 506 including keyboard, mouse etc.；Penetrated including such as cathode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loudspeaker etc.；Storage part 508 including hard disk etc.；
And the communications portion 509 of the network interface card including LAN card, modem etc..Communications portion 509 via such as because
The network of spy's net performs communication process.Driver 510 is also according to needing to be connected to I/O interfaces 505.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc., are installed on driver 510, in order to read from it as needed
Computer program be mounted into as needed storage part 508.
Especially, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product, it includes being carried on computer-readable medium
On computer program, the computer program include be used for execution flow chart shown in method program code.In such reality
Apply in example, which can be downloaded and installed by communications portion 509 from network, and/or from detachable media
511 are mounted.When the computer program is performed by central processing unit (CPU) 501, perform what is limited in the present processes
Above-mentioned function.It should be noted that computer-readable medium described herein can be computer-readable signal media or
Computer-readable recording medium either the two any combination.Computer-readable recording medium for example can be --- but
Be not limited to --- electricity, magnetic, optical, electromagnetic, system, device or the device of infrared ray or semiconductor, or it is any more than combination.
The more specifically example of computer-readable recording medium can include but is not limited to：Electrical connection with one or more conducting wires,
Portable computer diskette, hard disk, random access storage device (RAM), read-only storage (ROM), erasable type may be programmed read-only deposit
Reservoir (EPROM or flash memory), optical fiber, portable compact disc read-only storage (CD-ROM), light storage device, magnetic memory
Part or above-mentioned any appropriate combination.In this application, computer-readable recording medium can any be included or store
The tangible medium of program, the program can be commanded the either device use or in connection of execution system, device.And
In the application, computer-readable signal media can include believing in a base band or as the data that a carrier wave part is propagated
Number, wherein carrying computer-readable program code.The data-signal of this propagation can take various forms, including but not
It is limited to electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer
Any computer-readable medium beyond readable storage medium storing program for executing, the computer-readable medium can send, propagate or transmit use
In by instruction execution system, device either device use or program in connection.Included on computer-readable medium
Program code any appropriate medium can be used to transmit, include but not limited to：Wirelessly, electric wire, optical cable, RF etc., Huo Zheshang
Any appropriate combination stated.
Flow chart and block diagram in attached drawing, it is illustrated that according to the system of the various embodiments of the application, method and computer journey
Architectural framework in the cards, function and the operation of sequence product.At this point, each square frame in flow chart or block diagram can generation
The part of one module of table, program segment or code, the part of the module, program segment or code include one or more use
In the executable instruction of logic function as defined in realization.It should also be noted that marked at some as in the realization replaced in square frame
The function of note can also be with different from the order marked in attached drawing generation.For example, two square frames succeedingly represented are actually
It can perform substantially in parallel, they can also be performed in the opposite order sometimes, this is depending on involved function.Also to note
Meaning, the combination of each square frame and block diagram in block diagram and/or flow chart and/or the square frame in flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit can also be set within a processor, for example, can be described as：A kind of processor bag
Include receiving unit, generation unit, input unit and push unit.Wherein, the title of these units not structure under certain conditions
The paired restriction of the unit in itself, for example, receiving unit is also described as, " reception includes text to be detected and text to be detected
The unit of the text detection request of this subject matter ".
As on the other hand, present invention also provides a kind of computer-readable medium, which can be
Included in device described in above-described embodiment；Can also be individualism, and without be incorporated the device in.Above-mentioned calculating
Machine computer-readable recording medium carries one or more program, when said one or multiple programs are performed by the device so that should
Device：Receive the text detection request of the subject matter comprising text to be detected and the text to be detected；Determine the text to be detected
Target word and target sentence, and the position based on the target word in the text to be detected and the target sentence are in the text to be detected
Position, generate the words and phrases distributed intelligence of the text to be detected；Extraction and the structure of the type matching of the text to be detected are examined
Model is surveyed, which is inputted to the structure evaluation information for the structure detection model, obtaining the text to be detected；Push away
Send the text detection result of the structure evaluation information comprising the text to be detected.
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.People in the art
Member should be appreciated that invention scope involved in the application, however it is not limited to the technology that the particular combination of above-mentioned technical characteristic forms
Scheme, while should also cover in the case where not departing from foregoing invention design, carried out by above-mentioned technical characteristic or its equivalent feature
The other technical solutions for being combined and being formed.Such as features described above has similar work(with (but not limited to) disclosed herein
The technical solution that the technical characteristic of energy is replaced mutually and formed.
Claims (12)
1. a kind of information-pushing method, including：
Receive the text detection request of the subject matter comprising text to be detected and the text to be detected；
Determine the target word and target sentence of the text to be detected, and the position based on the target word in the text to be detected
The position in the text to be detected with the target sentence is put, generates the words and phrases distributed intelligence of the text to be detected, wherein,
The sentence that the target sentence matches for clause with any default clause in default clause set；
The structure detection model that extraction matches with the subject matter, the words and phrases distributed intelligence is inputted to the structure detection mould
Type, obtains the structure evaluation information of the text to be detected, and the structure detection model is used to characterize words and phrases distributed intelligence and knot
The correspondence of structure evaluation information；
The text detection result of structure evaluation information of the push comprising the text to be detected.
2. information-pushing method according to claim 1, wherein, the text detection result further includes clause detection knot
Fruit, the clause testing result are generated by following steps：
Text to be detected described in cutting sentence by sentence；
For each sentence after cutting, determine that this whether there is the word of collocation error, if so, extraction is mutually taken with the word
The recommendation word matched somebody with somebody, and generate and be used to prompt this to there are problems that the prompt message of collocation, wherein, the prompt message includes described
Recommend word；
Collect generated prompt message, generate clause testing result.
3. information-pushing method according to claim 2, wherein, collect generated prompt message described, generate sentence
Before formula testing result, described the step of generating the clause testing result, further includes：
For each target sentence, the default clause to match with the target sentence is preset into clause as target, based on it is described
Target presets the default principle of optimality that clause matches and the target sentence is optimized, and generation is used to prompt to optimize the sentence
Prompt message.
4. information-pushing method according to claim 1, wherein, the text detection result further includes word detection knot
Fruit, the word testing result are generated by following steps：
Extract the keyword and verse in the text to be detected；
For each keyword extracted, in response to determining that the keyword is noun, arrange in pairs or groups and believe from default keyword
Cease and collocation information corresponding with the keyword is searched in table, in response to determining that the keyword is adjective, from default synonym
The synonym of the keyword, word recommendation information of the generation comprising the collocation information or the synonym are searched in table；
For each verse extracted, the association verse associated with the verse, generation are searched from default verse storehouse
Include the verse recommendation information of the association verse；
Collect generated word recommendation information and verse recommendation information, generate word testing result.
5. information-pushing method according to claim 1, wherein, the text detection result further includes wrong word detection knot
Fruit, the wrong word testing result are generated by following steps：
Extract the verse in the text to be detected；
The text to be detected in addition to the verse is segmented, each word after participle and default word storehouse are carried out
Matching, and the verse is matched with default verse storehouse；
Based on matching result, the bag in the word comprising wrong word and the text to be detected in the text to be detected is determined
Verse containing wrong word, the wrong word testing result of wrong word information is included with generation.
6. a kind of information push-delivery apparatus, including：
Receiving unit, is configured to receive the text detection request of the subject matter comprising text to be detected and the text to be detected；
Generation unit, is configured to determine the target word and target sentence of the text to be detected, and based on the target word in institute
The position of position and the target sentence in the text to be detected in text to be detected is stated, generates the text to be detected
Words and phrases distributed intelligence, wherein, the sentence that the target sentence matches for clause with any default clause in default clause set；
Input unit, is configured to the structure detection model that extraction matches with the subject matter, and the words and phrases distributed intelligence is defeated
Enter the structure evaluation information for the structure detection model, obtaining the text to be detected, the structure detection model is used for table
Levy words and phrases distributed intelligence and the correspondence of structure evaluation information；
Push unit, is configured to the text detection result of structure evaluation information of the push comprising the text to be detected.
7. information push-delivery apparatus according to claim 6, wherein, the text detection result further includes clause detection knot
Fruit, described device further include：
Cutting unit, is configured to text to be detected described in cutting sentence by sentence；
First determination unit, is configured to for each sentence after cutting, determine the word that this whether there is collocation error, if
It is to extract the recommendation word mutually arranged in pairs or groups with the word, and generates and be used to prompt this to there are problems that the prompt message of collocation, its
In, the prompt message includes the recommendation word；
First collection unit, is configured to collect generated prompt message, generates clause testing result.
8. information push-delivery apparatus according to claim 7, wherein, described device further includes：
Optimize unit, be configured to for each target sentence, the default clause to match with the target sentence is pre- as target
If clause, the target sentence is optimized based on the default principle of optimality that clause matches is preset with the target, generation is used for
Prompt the prompt message optimized to this.
9. information push-delivery apparatus according to claim 6, wherein, the text detection result further includes word detection knot
Fruit, described device further include：
First extraction unit, is configured to extract the keyword and verse in the text to be detected；
First searching unit, is configured to each keyword for being extracted, in response to determining that the keyword is noun,
Collocation information corresponding with the keyword is searched from default keyword collocation information table, in response to determining that the keyword is shape
Hold word, the synonym of the keyword is searched from default synonym table, generation includes the collocation information or the synonym
Word recommendation information；
Second searching unit, is configured to each verse for being extracted, and is searched and the verse from default verse storehouse
Associated association verse, generation include the verse recommendation information of the association verse；
Second collection unit, is configured to collect generated word recommendation information and verse recommendation information, generation word detection
As a result.
10. information push-delivery apparatus according to claim 6, wherein, the text detection result further includes wrong word detection
As a result, described device further includes：
Second extraction unit, is configured to extract the verse in the text to be detected；
Matching unit, is configured to segment the text to be detected in addition to the verse, by each word after participle
Matched with default word storehouse, and the verse is matched with default verse storehouse；
Second determination unit, is configured to be based on matching result, determines the word for including wrong word in the text to be detected
With the verse for including wrong word in the text to be detected, the wrong word testing result of wrong word information is included with generation.
11. a kind of server, including：
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are performed by one or more of processors so that one or more of processors are real
The now method as described in any in claim 1-5.
12. a kind of computer-readable recording medium, is stored thereon with computer program, wherein, when which is executed by processor
Realize the method as described in any in claim 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711328500.8A CN107908792B (en) | 2017-12-13 | 2017-12-13 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711328500.8A CN107908792B (en) | 2017-12-13 | 2017-12-13 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107908792A true CN107908792A (en) | 2018-04-13 |
| CN107908792B CN107908792B (en) | 2021-10-22 |
Family
ID=61854355
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711328500.8A Active CN107908792B (en) | 2017-12-13 | 2017-12-13 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107908792B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113360608A (en) * | 2021-07-08 | 2021-09-07 | 北京阅神智能科技有限公司 | Man-machine combined Chinese composition correcting system and method |
| CN114818732A (en) * | 2022-05-19 | 2022-07-29 | 北京百度网讯科技有限公司 | Text content evaluation method, related device and computer program product |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1886767A (en) * | 2003-11-28 | 2006-12-27 | 语言的森林有限公司 | Composition evaluation device |
| US20130080148A1 (en) * | 2011-09-26 | 2013-03-28 | Fuji Xerox Co., Ltd. | Information processing apparatus, information processing method, and computer readable medium |
| CN105573980A (en) * | 2015-12-10 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Information segment generation method and device |
| CN105653840A (en) * | 2015-12-21 | 2016-06-08 | 青岛中科慧康科技有限公司 | Similar case recommendation system based on word and phrase distributed representation, and corresponding method |
| US10380225B1 (en) * | 2014-07-15 | 2019-08-13 | Google Llc | Systems and methods for layout transformation of document content |
-
2017
- 2017-12-13 CN CN201711328500.8A patent/CN107908792B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1886767A (en) * | 2003-11-28 | 2006-12-27 | 语言的森林有限公司 | Composition evaluation device |
| US20130080148A1 (en) * | 2011-09-26 | 2013-03-28 | Fuji Xerox Co., Ltd. | Information processing apparatus, information processing method, and computer readable medium |
| US10380225B1 (en) * | 2014-07-15 | 2019-08-13 | Google Llc | Systems and methods for layout transformation of document content |
| CN105573980A (en) * | 2015-12-10 | 2016-05-11 | 百度在线网络技术（北京）有限公司 | Information segment generation method and device |
| CN105653840A (en) * | 2015-12-21 | 2016-06-08 | 青岛中科慧康科技有限公司 | Similar case recommendation system based on word and phrase distributed representation, and corresponding method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113360608A (en) * | 2021-07-08 | 2021-09-07 | 北京阅神智能科技有限公司 | Man-machine combined Chinese composition correcting system and method |
| CN113360608B (en) * | 2021-07-08 | 2023-10-20 | 北京阅神智能科技有限公司 | Man-machine combined Chinese composition correcting system and method |
| CN114818732A (en) * | 2022-05-19 | 2022-07-29 | 北京百度网讯科技有限公司 | Text content evaluation method, related device and computer program product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107908792B (en) | 2021-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107729319A (en) | Method and apparatus for output information | |
| CN107220386A (en) | Information-pushing method and device | |
| US20160162467A1 (en) | Methods and systems for language-agnostic machine learning in natural language processing using feature extraction | |
| CN108153901A (en) | The information-pushing method and device of knowledge based collection of illustrative plates | |
| CN107491547A (en) | Searching method and device based on artificial intelligence | |
| CN107577763A (en) | Search method and device | |
| CN108628830A (en) | A kind of method and apparatus of semantics recognition | |
| US10942973B2 (en) | Automatically generating and evaluating candidate terms for trademark clearance | |
| CN109033282A (en) | A kind of Web page text extracting method and device based on extraction template | |
| CN107908615A (en) | A kind of method and apparatus for obtaining search term corresponding goods classification | |
| CN110532567A (en) | Extracting method, device, electronic equipment and the storage medium of phrase | |
| US20200218746A1 (en) | Method and apparatus for presenting information | |
| CN109858045A (en) | Machine translation method and device | |
| CN109840192A (en) | The method and apparatus of automatic test | |
| US11120362B2 (en) | Identifying a product in a document | |
| CN107832298A (en) | Method and apparatus for output information | |
| CN107908792A (en) | Information-pushing method and device | |
| CN110472034A (en) | Detection method, device, equipment and the computer readable storage medium of question answering system | |
| CN107168635A (en) | Information demonstrating method and device | |
| US20180341646A1 (en) | Translated-clause generating method, translated-clause generating apparatus, and recording medium | |
| CN109325227A (en) | Method and apparatus for generating amendment sentence | |
| CN108616413A (en) | Information calibration method and device | |
| CN106775914B (en) | A kind of code method for internationalizing and device for automatically generating key assignments | |
| CN103064967A (en) | Method and device used for establishing user binary relation bases | |
| US11048707B2 (en) | Identifying a product in a document |
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 |