CN104484379B - Determine the method and apparatus of music property relationship and inquiry processing method and device - Google Patents

Determine the method and apparatus of music property relationship and inquiry processing method and device Download PDF

Info

Publication number
CN104484379B
CN104484379B CN201410749432.2A CN201410749432A CN104484379B CN 104484379 B CN104484379 B CN 104484379B CN 201410749432 A CN201410749432 A CN 201410749432A CN 104484379 B CN104484379 B CN 104484379B
Authority
CN
China
Prior art keywords
music
information
webpage
property
music property
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
Application number
CN201410749432.2A
Other languages
Chinese (zh)
Other versions
CN104484379A (en
Inventor
雷小强
田振雷
王森
鲁晓莹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Baidu Netcom Science and Technology Co Ltd
Original Assignee
Beijing Baidu Netcom Science and Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Baidu Netcom Science and Technology Co Ltd filed Critical Beijing Baidu Netcom Science and Technology Co Ltd
Priority to CN201410749432.2A priority Critical patent/CN104484379B/en
Publication of CN104484379A publication Critical patent/CN104484379A/en
Application granted granted Critical
Publication of CN104484379B publication Critical patent/CN104484379B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/951Indexing; Web crawling techniques
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/40Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
    • G06F16/48Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Data Mining & Analysis (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Library & Information Science (AREA)
  • Multimedia (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the present invention provides a kind of method and apparatus of determining music property relationship, and this method includes:Original web page is obtained from network side;The first webpage is extracted from original web page, the first webpage is the webpage for including musical relations keyword;Judge that sample sentence is the sentence for the information for including at least two music properties with the presence or absence of sample sentence in the first webpage;By carrying out semantic analysis to sample sentence, determine the relationship between at least two music property, realize determining for the relationship between music property.

Description

Determine the method and apparatus of music property relationship and inquiry processing method and device
Technical field
The present embodiments relate to information search technique field more particularly to a kind of method of determining music property relationship and Device and inquiry processing method and device.
Background technology
With the continuous development of search engine technique, the demand that people search for music by search engine is also more and more.
In the prior art, search engine only provides the search of music the list of each version, does not provide each Relationship between version, if which is original singer, that is to turn over to sing.Wherein, it turns over to sing and is broadly divided into two types:1) lyrics and the music score of Chinese operas Unanimously, but singer is different, and such case is often that different singers have sung same a piece of music in the different periods, so as to generate Multiple and different version;2) music score of Chinese operas is identical, but the lyrics are different, and such case is greatly that domestic singer uses state The music score of Chinese operas of the music of outer singer, the version formed by composing a poem to a given tune of ci again, especially many classical music are all to use state Then the outer music score of Chinese operas is composed a poem to a given tune of ci and is renamed and sung again again.
Currently more the search engine such as Baidu of mainstream, 360 grades all provide the function of music searching, all simply show phase Close list.By taking Fig. 1 as an example, user's query music " asking " in Baidu search engine, wherein " asking " is musical designation, Baidu search The list of each version of music " asking " is simply shown in the search result that engine shows.
Invention content
The embodiment of the present invention provides a kind of method and apparatus of determining music property relationship, with determine different music properties it Between relationship.
The embodiment of the present invention also provides a kind of inquiry processing method and device, with user search for a certain music when, will not User is also provided to the relationship between the music of version.
In a first aspect, an embodiment of the present invention provides a kind of method of determining music property relationship, including:
Original web page is obtained from network side;
The first webpage is extracted from the original web page, first webpage is the net for including musical relations keyword Page;
Judge that the sample sentence is to include at least two music properties with the presence or absence of sample sentence in first webpage The sentence of information;
By carrying out semantic analysis to sample sentence, the relationship between at least two music property is determined.
Second aspect, an embodiment of the present invention provides a kind of device of determining music property relationship, including:
Webpage acquisition module, for obtaining original web page from network side;
First extraction module, for extracting the first webpage from the original web page, first webpage is includes sound The webpage of happy relation keyword;
Sample judgment module, for judging that the sample sentence is includes with the presence or absence of sample sentence in first webpage The sentence of the information of at least two music properties;
Relationship determination module, for by sample sentence carry out semantic analysis, determine at least two music property it Between relationship.
The method and apparatus of determining music property relationship provided in an embodiment of the present invention, after original web page is obtained, pass through Musical relations keyword is filtered original web page, obtains the webpage for including musical relations keyword, passes through music property Information, can be identified from the webpage for including musical relations keyword and obtain the sentence for including the information of different music properties Son, and analyzed by sentence semantics, it is excavated from the sentence of information for including different music properties and obtains corresponding music reality Relationship between body can determine the relationship between different music properties, between the music that can determine different editions Relationship.
The third aspect, an embodiment of the present invention provides a kind of inquiry processing method, including:
Receive the problem of to be retrieved;
When including music information in described problem, searched from music property relationship library according to the music information pair The music property relation information answered, wherein, music property relationship inventory contains the pass between music property data and music property It is information;
Return to the corresponding music property relation information.
Fourth aspect, an embodiment of the present invention provides a kind of query processing device, including:
Problem receiving module, for receiving the problem of to be retrieved;
Relationship searching module, for when including music information in described problem, according to the music information from music Corresponding music property relation information is searched in entity relationship library, wherein, music property relationship inventory contains music property data And the relation information between music property;
Module is returned to, for returning to the corresponding music property relation information.
Inquiry processing method and device provided in an embodiment of the present invention after the problem of to be retrieved is received, are asked when described When including music information in topic, including the method using any embodiment of the present invention offer happy entity relationship of accordatura really Relation information between the music property of formation and include in the music property relationship libraries of music property data described in matching The music information included in problem can provide music property relation information corresponding with the music information in described problem.
Description of the drawings
In order to illustrate more clearly of the present invention, one will be done to attached drawing needed in the present invention below and be simply situated between It continues, it should be apparent that, the accompanying drawings in the following description is some embodiments of the present invention, and those of ordinary skill in the art are come It says, without having to pay creative labor, can also be obtained according to these attached drawings other attached drawings.
Fig. 1 for user in the prior art in a search engine query music when, search result signal that search engine shows Figure;
Fig. 2 a are the flow chart of a kind of method of determining music property relationship that the embodiment of the present invention one provides;
Fig. 2 b are to be extracted from the original web page in the method for determining music property relationship provided in an embodiment of the present invention The flow chart of first webpage;
Fig. 3 a be determining music property relationship provided in an embodiment of the present invention method in judge be in first webpage The no flow chart there are sample sentence;
Fig. 3 b are a kind of flow chart for establishing music libraries provided by Embodiment 2 of the present invention;
Fig. 4 is the structure diagram of the device of a kind of determining music property relationship that the embodiment of the present invention three provides;
Fig. 5 a are the flow chart of a kind of inquiry processing method that the embodiment of the present invention four provides;
Fig. 5 b are a kind of search result displaying figure that the embodiment of the present invention four provides;
Fig. 5 c are another search result displaying figure that the embodiment of the present invention four provides;
Fig. 5 d are another search result displaying figure that the embodiment of the present invention four provides;
Fig. 5 e are another search result displaying figure that the embodiment of the present invention four provides;
Fig. 5 f are another search result displaying figure that the embodiment of the present invention four provides;
Fig. 6 is the structure diagram of a kind of query processing device that the embodiment of the present invention five provides.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to the embodiment of the present invention In technical solution be described in further detail, it is clear that described embodiment is part of the embodiment of the present invention rather than complete The embodiment in portion.It is understood that specific embodiment described herein is only used for explaining the present invention rather than to the present invention's It limits, based on the embodiments of the present invention, those of ordinary skill in the art are obtained without creative efforts Every other embodiment, shall fall within the protection scope of the present invention.It also should be noted that for ease of description, attached drawing In illustrate only part related to the present invention rather than full content.
Embodiment one
The method of determining music property relationship provided in this embodiment can be realized really by being configured with hardware and/or software The device of the happy entity relationship of accordatura performs, which is typically the system for being configured at and being capable of providing music searching service Such as in search engine.
Fig. 2 a are please referred to, a kind of method of determining music property relationship provided in this embodiment includes:Operation 210~operation 240。
In operation 210, original web page is obtained from network side.
In operation 220, the first webpage is extracted from the original web page, first webpage is includes musical relations pass The webpage of keyword.
Wherein, musical relations correspond to multiple music properties, and the relationship between multiple music properties is musical relations.
Relationship between music property belongs to the one of which in problem answers class.Specifically, problem correspondence is led to by user Cross the retrieval type for including music property information of search engine input, answer correspond to the music property that shows in search result it Between relationship.
User is when the answer for finding some problem, it will usually directly scan for, but search in a search engine This original web page of search result that index holds up offer can not meet the search need of user namely search result generally can not be straight It connects and shows the corresponding answer of problem.The webpage (for example, Baidu is known, Sina's love is asked) and profession of existing many question and answer types The answer corresponding with problem that forum generally comprises the problem of user proposes and other users provide.
Similarly, for the relationship between music property, many users directly can not find music property from search engine Between relationship when and putd question to by delivering the modes such as model, demand is in the answer of other users.Therefore, most sounds Relationship between happy entity is all present in the webpage and forum Web pages of question and answer type, it is known that the webpage of both types includes There is musical relations keyword.
Therefore, this operation is specifically and original web page is filtered using musical relations keyword, obtains including music The webpage of relation keyword.
In operation 230, judge that the sample sentence is includes at least two with the presence or absence of sample sentence in first webpage The sentence of the information of music property.
In this operation, the sentence that can be included to first webpage carries out the judgement operation respectively.
Music property refers to music object in itself, and form can be * .mp3, * .mp4, * .wma or * .wav etc., wherein, * Arbitrary string is represented, usually musical designation.
The information of the music property can be included in musical designation, bent author, word author, singer and player extremely One item missing information.
Since the information of music property can include the information of at least two music properties accordingly there are many form Sentence can be able to can be the sound for including at least two music properties there are many form there are many form namely sample sentence The sentence of happy title can also be the musical designation for including at least two music properties and the sentence of singer.In this way, for It is turned over and sings the still constant different music properties of musical designation, can be distinguished by adding in singer.Sample sentence can be with It is the sentence of the other information comprising music property or information combination, the present embodiment is not limited this.
In operation 240, by carrying out semantic analysis to sample sentence, the pass between at least two music property is determined System.
In this operation, the feature of sample sentence can be extracted, the feature mainly includes following several classes:1) grammar property, Refer mainly to the interdependent path among two music properties;2) lexical characteristics, the setting quantity including the music property left side and the right Word and corresponding part of speech, the quantity can be adjusted according to actual needs;3) other features, the main master for including the page The features such as epigraph, the structural similarity of upper and lower sentence.Based on these features, a model is trained using machine learning algorithm, It can determine the relationship between at least two music properties described in sample sentence.
Technical solution provided in this embodiment, after original web page is obtained, by musical relations keyword to original web page It is filtered, obtains the webpage for including musical relations keyword, and by from the webpage for including musical relations keyword The sentence for the information for including different music properties is found out, carries out semantic analysis, it is determined that the different music properties that sentence includes Between relationship, so as to excavate the relationship between the different music properties provided in network, determine related music Relationship between entity, and it is supplied to user.
A kind of preferred embodiment of first webpage this operation is extracted from the original web page, please refers to Fig. 2 b, is had Body includes:Operation 221~operation 222.
In operation 221, question and answer webpage and forum Web pages are identified from original web page.
As previously mentioned, the relationship between most music properties is all present in the webpage and forum Web pages of question and answer type In, therefore this operation is from original web page by identifying question and answer webpage and forum Web pages, can obtain excavating music property it Between relationship needed for web page resources.
It, can be with since the question and answer webpage that same question and answer website provides has one or more of identical web page templates It chooses with any question and answer webpage in the corresponding question and answer webpage of certain same web page template, using based on Webkit, (one is opened The browser engine in source) html parser such as HTMLParser to HTML (the Hyper Text Mark-up of the question and answer webpage Language, hypertext markup language) source code parsed, obtain a DOM for meeting World Wide Web Consortium (W3C) standard (Document Object Model, document dbject model) is set.Wherein, dom tree is a kind of representation method of webpage.
Then, the features such as cutting word feature, structure repeated characteristic and the tag attributes semantic feature of DOM tree structure are extracted, profit These features are trained with machine learning algorithm, obtain the model of question and answer webpage corresponding with this kind of same web page template, Then it is predicted with the model, all question and answer nets corresponding with this kind of same web page template can be identified from original web page Page.
Similarly, it can choose with any forum Web pages in certain corresponding forum Web pages of same web page template, lead to The html source code for parsing the forum Web pages is crossed, dom tree corresponding with the forum Web pages can be obtained, extract DOM tree structure The features such as cutting word feature, structure repeated characteristic and the tag attributes semantic feature of structure, using machine learning algorithm to these spies Sign is trained, and is obtained the model of forum Web pages corresponding with this kind of same web page template, is then predicted with the model, can To identify all forum Web pages corresponding with this kind of same web page template from original web page.
It should be noted that filtering original web page relative to using keyword, question and answer webpage and forum Web pages, technology are obtained Difficulty is, since the information category of every page of question and answer webpage and forum Web pages is various, is difficult to determine suitable keyword, leads It causes to bring impurity in the webpage of screening, the accuracy of question and answer webpage and forum Web pages being obtained by filtration is low.And dom tree is utilized together The general character web page template feature of webpage that one website provides, for example, structure repeated characteristic and tag attributes semantic feature, avoid Make due to keyword is improper with impurity in the webpage of screening, the question and answer webpage that improves and forum Web pages it is accurate Degree.
In operation 222, the musical relations keyword is searched from the question and answer webpage and forum Web pages, will be included State musical relations keyword question and answer webpage or forum Web pages as first webpage.
It should be noted that identifying question and answer webpage and forum Web pages from original web page by operation 221, can obtain The web page resources needed for the relationship between music property are excavated, realize the mistake from original web page to question and answer webpage and forum Web pages Filter;This operation is the further fine screening of the question and answer webpage or forum Web pages to recognizing, and is realized from question and answer webpage or opinion Altar webpage is to the fine filtering for the webpage for including musical relations keyword.
This preferred embodiment, by from original web page to the preliminary identification filtering of question and answer webpage and forum Web pages, energy The web page resources excavated needed for the relationship between music property are accessed, using musical relations keyword, to the question and answer recognized The further fine screening of webpage or forum Web pages realizes crucial from question and answer webpage or forum Web pages to musical relations are included The fine filtering of the webpage of word has been accurately located the web page resources for including musical relations keyword.
After operation 240, the method for determining application entity relationship provided in an embodiment of the present invention can also include:
When between at least two music property for turn over sing, original singer, reorganization or during original work song relationship, will described at least two The solid data and relation information of a music property correspond to storage, establish musical relations library.
It is illustrated by original singer and for turning over the relationship of singing.Assuming that first music property was sung for Liang Jingru《It asks》This is first Song, the information of the music property include musical designation " asking " and singer " Liang Jingru ";Second music property is sung for Chen Shuhua 's《It asks》, the information of the music property includes musical designation " asking " and singer " Chen Shuhua ";Third music property is recalled for woods What lotus was sung《It asks》, the information of the music property includes musical designation " asking " and singer " Lin Yilian ".From including musical relations Identification obtains puing question to sentence " original singer that song is asked " and answers sentence " Chen Shuhua ", and pass through sentence in the webpage of keyword " original singer " Sub- semantic analysis is so as to excavate to obtain the relationship between these three music properties:What Chen Shuhua was sung《It asks》For《It asks》Original singer, What other people sang《It asks》It is sung for Chen Shuhua《It asks》Turn over and sing, so as to which the corresponding solid data of these three music properties " be asked .mp3 ", " ask .wma " and " asking .wma " and relation information " Chen Shuhua is sung《It asks》For original singer " storage is corresponded to, it is added to music pass It is in library.Wherein it is possible to corresponding storage is carried out in the form of list or relational graph.
It should be noted that the solid data of music property can cover a large amount of existing music reality in musical relations library Body.
The musical relations library that present embodiment provides can be configured in special music application, can also be configured at search In engine, present embodiment is not limited this.
Embodiment two
The present embodiment on the basis of above-described embodiment, provide judge in first webpage with the presence or absence of sample sentence this The preferred embodiment of one operation.
Please refer to Fig. 3 a, judge in the method for determining music property relationship provided in this embodiment be in first webpage The no flow there are sample sentence specifically includes:Operate 3a1~operation 3a3.
It operates in 3a1, is sentence by the text dividing in first webpage.
In this operation, if the right side that the rule for cutting sentence can be a line text does not take full line, this conduct of the style of writing One sentence output;Can also using punctuation works symbol such as fullstop, branch and exclamation etc. as the cutting segmentation symbol of sentence, into Row text sentence cutting.
It operates in 3a2, the sentence is matched with the music property in music libraries, wherein, music libraries are stored with sound Happy solid data and music property information.
Multimode matching algorithm may be used herein to match sentence with the music property in music libraries.Wherein, music Solid data refers to music object in itself, and form can be * .mp3, * .mp4, * .wma or * .wav etc., wherein, * represents arbitrary Character string, usually musical designation.
The music property information can be included in musical designation, bent author, word author, singer and player at least One information.
It operates in 3a3, when the sentence matches at least two music properties, judges that there are samples in first webpage This sentence.
For example, two music that can be directly matched there are two the sentence of music property information in music libraries are included simultaneously Entity;Sentence can also at least two music property of indirect matching, for example, including words such as " original singers ", " turn over and sing " or " reorganization " Sentence, only give one of music property information.
The technical solution of the present embodiment, after the webpage for including musical relations keyword is extracted from original web page, by Cover in the sentence of webpage of musical relations keyword is included in music property relationship, therefore pass through the webpage to extracting In text carry out sentence cutting, as unit of sentence, using each sentence of music property information matches in music libraries, can obtain To the sentence for the information for including at least two music properties, it is accurately located the sentence money for including music property relationship Source.
The foundation of above-mentioned music libraries please refers to Fig. 3 b, specifically includes:Operate 3b1~operation 3b3.
It operates in 3b1, music property data and information is extracted from the original web page.
In internet, music property data and information are mainly with the formal distribution of structural data in various offer music On the website of service, since the webpage that same website provides has one or more of identical web page templates, for every A website, can choose the website have certain corresponding webpage of same web page template in any webpage, using based on The HTMLParser of Webkit parses the html source code of the webpage, obtains a dom tree for meeting W3C standards.Profit The xpath that specified data field collection is obtained with dom tree gathers, and specified data field can be singer, musical designation, word author Or bent author etc., so as to quickly position music property data and information in the corresponding webpage of this kind of same web page template.
The set of the set, i.e. dom tree of the web page template that same website provides can specifically be obtained first, and from webpage In template set search obtain with the matched dom tree of the current web page of the website, by traversing the dom tree, in ergodic process The xpath of dom nodes can be spliced into, by comparing xpath, you can obtain music property data and information.
It operates in 3b2, duplicate removal and fusion treatment is carried out to the music property data and information of extraction, obtain each sound The solid data and entity information of happy entity.
In internet, music property data and information are mainly with the formal distribution of structural data in various offer music In the webpage of the website of service, all music property data and information, and each station are covered without an independent website Point provide music property data and information between have overlapping, therefore after music property data and information is extracted, need into The row duplicate removal and fusion treatment are not leaked so that the solid data and entity information of obtained each music property be not heavy, are improved Information integrity, and reduce information redundance.
Specifically, for the same music of different websites, when the entity information of music is identical, arbitrary selection one is first Music property data and corresponding entity information;When the entity information of music is not exactly the same, the net that site quality is higher is taken Entity information of the page music property information as the music property.When the entity information of music has missing, then pass through other stations The value of the identical entity information name of the page is put to supplement.
It operates in 3b3, establishes the index of each music property, generate the music libraries.
Present embodiment, by extracting music property data and information from the original web page that network side is got, due to Music property data and information mainly with the formal distribution of structural data in the webpage of the various websites for providing music services, There is no an independent website to cover all music property data and information, and the music property data that each website provides There is overlapping between information, therefore after music property data and information is extracted, pass through duplicate removal and fusion treatment so that obtain Music libraries in each music property solid data and entity information is not heavy does not leak, improve information integrity, and reduce Information redundance.
Embodiment three
Referring to Fig. 4, a kind of device of determining music property relationship provided in this embodiment includes:Webpage acquisition module 410th, the first extraction module 420, sample judgment module 430 and relationship determination module 440.
Wherein, webpage acquisition module 410 is used to obtain original web page from network side;First extraction module 420 is used for from institute It states and the first webpage is extracted in original web page, first webpage is the webpage for including musical relations keyword;Sample judges mould Block 430 is for judging to whether there is in first webpage sample sentence, and the sample sentence is includes at least two music properties Information sentence;Relationship determination module 440 is used to, by carrying out semantic analysis to sample sentence, determine at least two music Relationship between entity.
The technical solution of the present embodiment after original web page is obtained, carries out original web page by musical relations keyword Filtering, obtains the webpage for including musical relations keyword, by the information of music property, can be closed from musical relations are included Identification obtains the sentence for the information for including different music properties in the webpage of keyword, and is analyzed by sentence semantics, so as to The relationship between corresponding music property is obtained to be excavated from the sentence of information for including different music properties.
In said program, first extraction module 420 specifically can be used for:
Question and answer webpage and forum Web pages are identified from original web page;
The musical relations keyword is searched from the question and answer webpage and forum Web pages, the musical relations will be included The question and answer webpage or forum Web pages of keyword are as first webpage.
In said program, the sample judgment module 430 specifically can be used for:
It is sentence by the text dividing in first webpage;
The sentence is matched with the music property in music libraries, wherein, the music libraries are stored with music property Data and music property information;
When the sentence matches at least two music properties, judge that there are sample sentences in first webpage.
In said program, described device can also include:Second extraction module, data processing module and first establish mould Block.
Wherein, the second extraction module is used to extract music property data and information from the original web page;Data processing Module is used to carry out duplicate removal and fusion treatment to the music property data and information of extraction, obtains the reality of each music property Volume data and entity information;First establishes module for establishing the index of each music property, generates the music libraries.
In said program, described device can also include:Second establishes module, true for working as the relationship determination module Between fixed at least two music property for turn over sing, original singer, reorganization or during original work song relationship, will at least two music it is real The solid data and relation information of body correspond to storage, establish musical relations library.
The device of determining music property relationship provided in an embodiment of the present invention can perform any embodiment of the present invention and be provided Really the method for the happy entity relationship of accordatura has the corresponding function module and advantageous effect of execution method.
Example IV
The method of the embodiment of the present invention can be performed by the query processing device realized with hardware and/or software is configured, The realization device is typically configured in the system such as search engine for being capable of providing music searching service.
Fig. 5 a are please referred to, a kind of inquiry processing method provided in this embodiment includes:Operation 510~operation 530.
In operation 510, the problem of to be retrieved is received.
For example, after terminal receives the problem of user inputs in the search interactive interface of browser, generation message is sent To server end or search engine, server or search engine receive the problem of to be retrieved.Wherein, it is carried in the message of transmission There is the problem of to be retrieved.
In operation 520, when including music information in described problem, according to the music information from music property relationship Corresponding music property relation information is searched in library, wherein, music property relationship inventory contains music property data and music is real Relation information between body.
Wherein, the relationship in music property library between music property can be provided really by embodiment one or embodiment two The method of the happy entity relationship of accordatura obtains, and details are not described herein again.
Wherein, the relationship between music property can be turn over sing, original singer, the relationships such as reorganization or original work song.
This operation can be there are many embodiment, such as including following one of which:
According to musical designation search the original singer of corresponding with musical designation music property, turn over sing, original work are bent or adapt Bent information.
The original work that music property corresponding with the musical designation and composer is searched according to musical designation and composer are bent Or the information of arrangement.
It searches the original singer of song corresponding with the musical designation and composer according to song title and singer or turns over and sing Information.
In operation 530, the corresponding music property relation information is returned.
After the corresponding music property relation information is returned, it can also further be illustrated in search result, be User provides music property relation information corresponding with the retrieval type for including music information.
After corresponding music property relation information is searched from music property relationship library according to the music information, also It can include:Return to the corresponding music property data of the corresponding music property relation information.
It is possible to further show the corresponding music property relation information and corresponding music property data.Tool Body is that existing search result is carried out assembly with the corresponding music property relation information, shows front end user.
The technical solution of the present embodiment, after the problem of to be retrieved is received, when including music information in described problem When, between the music property that the method that the happy entity relationship of accordatura really is provided using any embodiment of the present invention is formed is included Relation information and include the music letter included in matching described problem in the music property relationship libraries of music property data Breath, can obtain music property relation information corresponding with the music information in described problem.
Different exhibition methods is illustrated respectively with reference to Fig. 5 b- Fig. 5 f.
The first exhibition method, when user directly search music original singer when, in search result directly displaying and original singer Corresponding and user can be with the card of audition.
In Fig. 5 b, when user is by Baidu search engine search " the flowers are in blossom by Cai Guoqing in journey original singer ", in search result Show that musical designation be " Hua Misaki く trips road ", audition card of the singer for " reason son ", user can intuitively know and search for The associated musical designation of music is " road that the flowers are in blossom ", the music version that singer is " Cai Guoqing " and musical designation are " flower Misaki く trips The relationship of road ", singer between the music version of " reason ", specifically, the former is not original singer, and the latter is only original singer, displaying Abundant and accurate music property relationship is conducive to direct the user to the search result of user's needs.
Second of exhibition method, when user directly search music turn over sing when, directly show and turn in search result and sing Corresponding and user can be with the card of audition.This exhibition method is similar with the first exhibition method.
In Fig. 5 c, when user is by Baidu search engine search " turning over for Kiroro future へ is sung ", opened up in search result Show the audition card that musical designation is " later ", singer is " Liu Ruoying ", user can intuitively be known closes with the music of search The music version and musical designation that the musical designation of connection is " future ", singer is " Kiroro " are " later ", singer is " Liu Ruo Relationship between the music version of English ", specifically, the latter are that the former one kind is turned over to sing, and illustrate abundant and accurate music Entity relationship is conducive to direct the user to the search result of user's needs.
The third exhibition method, when some music of user's search contains multiple versions, with list in search result Each version is shown, and identifies original singer.Wherein, the multiple version can be with the lyrics and music score of Chinese operas all same, but singer is not Together, can also the music score of Chinese operas it is identical, but the lyrics are different namely turn over singing by composing a poem to a given tune of ci again.
In Fig. 5 d, when user is by Baidu search engine search " asking ", search result shows multiple versions with tabular form, And directly indicate original singer's version of music " asking " in search result, specifically, the singer of music " asking " is " old refined The corresponding original singer's version of birch ", is identified with " original singer " in the version corresponding position, and user can intuitively be known closes with the music of search Relationship between multiple versions of connection illustrates abundant and accurate music property relationship, is conducive to direct the user to user The search result needed.
4th kind of exhibition method when the song of user's search has original singer, provides the link of original singer, user on audition card Can directly on card audition original singer, after the completion of audition, be returned directly to turn over and sing.
Please refer to Fig. 5 e and Fig. 5 f.In Fig. 5 e, when user is by Baidu search engine search " Old Boy's song ", search knot Link " audition original singer's version of original singer is provided on audition card in fruit《ありがとう》Bridge Zhuo Mi ", user can be direct The audition original singer on card provides the link turned over and sung after the completion of audition, on audition card and " is back to《Old Boy》Chopsticks brother Younger brother " (as shown in figure 5f) can be returned directly to turn over to sing.Abundant and accurate music property relationship is illustrated in audition card, Be conducive to direct the user to the search result of user's needs.
Embodiment five
Referring to Fig. 6, a kind of structure diagram of query processing device provided for the embodiment of the present invention five.The device packet It includes:Problem receiving module 610, relationship searching module 620 and return module 630.
Wherein, problem receiving module 610 is used to receive the problem of to be retrieved;Relationship searching module 620 is used to ask when described When including music information in topic, corresponding music property relationship is searched from music property relationship library according to the music information Information, wherein, music property relationship inventory contains the relation information between music property data and music property;Return to module 630 are used to return to the corresponding music property relation information.
The technical solution of the present embodiment, after the problem of to be retrieved is received, when including music information in described problem When, between the music property that the method that the happy entity relationship of accordatura really is provided using any embodiment of the present invention is formed is included Relation information and include the music letter included in matching described problem in the music property relationship libraries of music property data Breath, can obtain music property relation information corresponding with the music information in described problem.
In said program, the relationship searching module 620 specifically can be used for:
According to musical designation search the original singer of corresponding with musical designation music property, turn over sing, original work are bent or adapt Bent information;
Alternatively,
The original work that music property corresponding with the musical designation and composer is searched according to musical designation and composer are bent Or the information of arrangement;
Alternatively,
It searches the original singer of song corresponding with the musical designation and composer according to song title and singer or turns over and sing Information.
In said program, the return module 630 can be also used for:Return to the corresponding music property relation information Corresponding music property data.
Query processing device provided in an embodiment of the present invention can perform the query processing that any embodiment of the present invention is provided Method has the corresponding function module and advantageous effect of execution method.
Finally it should be noted that:The above various embodiments is merely to illustrate technical scheme of the present invention rather than it is limited System;Preferred embodiment in embodiment, is not intended to limit it, to those skilled in the art, the present invention can be with There are various modifications and changes.All any modification, equivalent replacement, improvement and so within spirit and principles of the present invention, It should be included within protection scope of the present invention.

Claims (14)

  1. A kind of 1. method of determining music property relationship, which is characterized in that including:
    The retrieval type for including music property information by search engine input is retrieved, and is obtained from network side as search knot The original web page of fruit;
    The first webpage is extracted from the original web page, first webpage is the webpage for including musical relations keyword, more Relationship between a music property is musical relations, including turn over sing, original singer, reorganization or original work song relationship;
    Judge that the sample sentence is the information for including at least two music properties with the presence or absence of sample sentence in first webpage Sentence;
    By carrying out semantic analysis to sample sentence, the relationship between at least two music property is determined;
    Wherein, the first webpage is extracted from the original web page, including:
    The web page template of question and answer webpage and forum Web pages is chosen, the source code of the web page template is parsed to obtain the net The dom tree of page template;
    Cutting word feature, structure repeated characteristic and the tag attributes semantic feature of DOM tree structure are extracted, utilizes machine learning algorithm pair The cutting word feature, structure repeated characteristic and tag attributes semantic feature are trained, and are obtained corresponding with the web page template Web page model;
    Predicted using the web page model, identified from original web page question and answer webpage corresponding with the web page template and Forum Web pages;
    The musical relations keyword is searched from the question and answer webpage and forum Web pages, it is crucial that the musical relations will be included The question and answer webpage or forum Web pages of word are as first webpage.
  2. 2. according to the method described in claim 1, it is characterized in that, judge in first webpage with the presence or absence of sample sentence, packet It includes:
    It is sentence by the text dividing in first webpage;
    The sentence is matched with the music property in music libraries, wherein, the music libraries are stored with music property data With music property information;
    When the sentence matches at least two music properties, judge that there are sample sentences in first webpage.
  3. 3. according to the method described in claim 2, it is characterized in that, the foundation of the music libraries, including:
    Music property data and information are extracted from the original web page;
    Duplicate removal and fusion treatment are carried out to the music property data and information of extraction, obtain the entity number of each music property According to and entity information;
    The index of each music property is established, generates the music libraries.
  4. 4. according to any methods of claim 1-3, which is characterized in that by carrying out semantic analysis to sample sentence, determine After relationship between at least two music property, further include:
    When between at least two music property for turn over sing, original singer, reorganization or during original work song relationship, by least two sound The solid data and relation information of happy entity correspond to storage, establish musical relations library.
  5. 5. a kind of inquiry processing method, which is characterized in that including:
    Receive the problem of to be retrieved;
    When including music information in described problem, searched from music property relationship library according to the music information corresponding Music property relation information, wherein, music property relationship inventory contains the relationship letter between music property data and music property Breath, wherein, the relation information between the music property data and music property is determined using claim 1-4 is any described The method of music property relationship performs;
    Return to the corresponding music property relation information.
  6. 6. it according to the method described in claim 5, it is characterized in that, is looked into from music property relationship library according to the music information Corresponding music property relation information is looked for, including:
    According to musical designation search the original singer of corresponding with musical designation music property, turn over sing, original work song or arrangement Information;
    Alternatively,
    The original work song of music property corresponding with the musical designation and composer is searched according to musical designation and composer or is changed The information of music;
    Alternatively,
    The original singer of song corresponding with the song title and singer is searched according to song title and singer or turns over the letter sung Breath.
  7. 7. method according to claim 5 or 6, which is characterized in that according to the music information from music property relationship After corresponding music property relation information is searched in library, the method further includes:
    Return to the corresponding music property data of the corresponding music property relation information.
  8. 8. a kind of device of determining music property relationship, which is characterized in that including:
    Webpage acquisition module, the retrieval type for including music property information by search engine input is retrieved, from net Network side obtains the original web page as search result;
    First extraction module, for extracting the first webpage from the original web page, first webpage is includes music pass Be the webpage of keyword, the relationship between multiple music properties is musical relations, including turn over sing, original singer, reorganization or original work it is bent Relationship;
    Sample judgment module, for judging that the sample sentence is includes at least with the presence or absence of sample sentence in first webpage The sentence of the information of two music properties;
    Relationship determination module, for by carrying out semantic analysis to sample sentence, determining between at least two music property Relationship;
    First extraction module is specifically used for:
    The web page template of question and answer webpage and forum Web pages is chosen, the source code of the web page template is parsed to obtain the net The dom tree of page template;
    Cutting word feature, structure repeated characteristic and the tag attributes semantic feature of DOM tree structure are extracted, utilizes machine learning algorithm pair The cutting word feature, structure repeated characteristic and tag attributes semantic feature are trained, and are obtained corresponding with the web page template Web page model;
    Predicted using the web page model, identified from original web page question and answer webpage corresponding with the web page template and Forum Web pages;
    The musical relations keyword is searched from the question and answer webpage and forum Web pages, it is crucial that the musical relations will be included The question and answer webpage or forum Web pages of word are as first webpage.
  9. 9. device according to claim 8, which is characterized in that the sample judgment module is specifically used for:
    It is sentence by the text dividing in first webpage;
    The sentence is matched with the music property in music libraries, wherein, the music libraries are stored with music property data With music property information;
    When the sentence matches at least two music properties, judge that there are sample sentences in first webpage.
  10. 10. device according to claim 9, which is characterized in that described device further includes:
    Second extraction module, for extracting music property data and information from the original web page;
    Data processing module for carrying out duplicate removal and fusion treatment to the music property data and information of extraction, obtains every The solid data and entity information of a music property;
    First establishes module, for establishing the index of each music property, generates the music libraries.
  11. 11. according to any devices of claim 8-10, which is characterized in that described device further includes:
    Second establishes module, for work as the relationship determination module determine between at least two music property for turn over sing, it is former It sings, adapt or during original work song relationship, the solid data of at least two music property and relation information are corresponded into storage, established Musical relations library.
  12. 12. a kind of query processing device, which is characterized in that including:
    Problem receiving module, for receiving the problem of to be retrieved;
    Relationship searching module, for when including music information in described problem, according to the music information from music property Corresponding music property relation information is searched in relationship library, wherein, music property relationship inventory contains music property data and sound Relation information between happy entity, wherein, the relation information between the music property data and music property will using right The device of any happy entity relationships of accordatura really of 8-11 is asked to obtain;
    Module is returned to, for returning to the corresponding music property relation information.
  13. 13. device according to claim 12, which is characterized in that the relationship searching module is specifically used for:
    According to musical designation search the original singer of corresponding with musical designation music property, turn over sing, original work song or arrangement Information;
    Alternatively,
    The original work song of music property corresponding with the musical designation and composer is searched according to musical designation and composer or is changed The information of music;
    Alternatively,
    The original singer of song corresponding with the song title and singer is searched according to song title and singer or turns over the letter sung Breath.
  14. 14. device according to claim 12 or 13, which is characterized in that the return module is additionally operable to:It is described right to return The corresponding music property data of music property relation information answered.
CN201410749432.2A 2014-12-09 2014-12-09 Determine the method and apparatus of music property relationship and inquiry processing method and device Active CN104484379B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201410749432.2A CN104484379B (en) 2014-12-09 2014-12-09 Determine the method and apparatus of music property relationship and inquiry processing method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201410749432.2A CN104484379B (en) 2014-12-09 2014-12-09 Determine the method and apparatus of music property relationship and inquiry processing method and device

Publications (2)

Publication Number Publication Date
CN104484379A CN104484379A (en) 2015-04-01
CN104484379B true CN104484379B (en) 2018-06-12

Family

ID=52758920

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201410749432.2A Active CN104484379B (en) 2014-12-09 2014-12-09 Determine the method and apparatus of music property relationship and inquiry processing method and device

Country Status (1)

Country Link
CN (1) CN104484379B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110717062B (en) * 2018-07-11 2024-03-22 斑马智行网络(香港)有限公司 Music search and vehicle-mounted music playing method, device, equipment and storage medium
CN110245197B (en) * 2019-05-20 2022-01-28 北京百度网讯科技有限公司 Whole-network entity association method and system
CN111552778B (en) * 2020-04-26 2024-05-14 北京达佳互联信息技术有限公司 Audio resource management method, device, computer readable storage medium and equipment
CN112948603B (en) * 2021-03-08 2023-05-05 北方自动控制技术研究所 Transport delivery knowledge question-answering method based on transfer learning
CN113609309B (en) * 2021-08-16 2024-02-06 脸萌有限公司 Knowledge graph construction method and device, storage medium and electronic equipment
CN115840845A (en) * 2021-09-18 2023-03-24 华为技术有限公司 Webpage retrieval method and related equipment

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101071422A (en) * 2006-06-15 2007-11-14 腾讯科技(深圳)有限公司 Musicfile search processing system and method
CN102708100A (en) * 2011-03-28 2012-10-03 北京百度网讯科技有限公司 Method and device for digging relation keyword of relevant entity word and application thereof

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101452470B (en) * 2007-10-18 2012-06-06 广州索答信息科技有限公司 Summary-style network search engine system and search method and uses

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101071422A (en) * 2006-06-15 2007-11-14 腾讯科技(深圳)有限公司 Musicfile search processing system and method
CN102708100A (en) * 2011-03-28 2012-10-03 北京百度网讯科技有限公司 Method and device for digging relation keyword of relevant entity word and application thereof

Also Published As

Publication number Publication date
CN104484379A (en) 2015-04-01

Similar Documents

Publication Publication Date Title
CN104484379B (en) Determine the method and apparatus of music property relationship and inquiry processing method and device
US10706236B1 (en) Applied artificial intelligence technology for using natural language processing and concept expression templates to train a natural language generation system
KR102158352B1 (en) Providing method of key information in policy information document, Providing system of policy information, and computer program therefor
CN107704453B (en) Character semantic analysis method, character semantic analysis terminal and storage medium
US6601026B2 (en) Information retrieval by natural language querying
US7739257B2 (en) Search engine
US20170337261A1 (en) Decision Making and Planning/Prediction System for Human Intention Resolution
CN108304375A (en) A kind of information identifying method and its equipment, storage medium, terminal
US20150286943A1 (en) Decision Making and Planning/Prediction System for Human Intention Resolution
JP2005537532A (en) Comprehensive development tool for building natural language understanding applications
US20200210441A1 (en) System and method of database creation through form design
KR100905744B1 (en) Method and system for providing conversation dictionary service based on user created dialog data
CN109033282A (en) A kind of Web page text extracting method and device based on extraction template
CN111753536A (en) Automatic patent application text writing method and device
CN103020311A (en) Method and system for processing user search terms
Chortaras et al. WITH: human-computer collaboration for data annotation and enrichment
JP5688754B2 (en) Information retrieval apparatus and computer program
CN111949781B (en) Intelligent interaction method and device based on natural sentence syntactic analysis
YesuRaju et al. A language independent web data extraction using vision based page segmentation algorithm
van Hooland et al. Evaluating the success of vocabulary reconciliation for cultural heritage collections
Shakhov User voice overview: topic recognition and sentiment analysis of customer feedback in the B2C sector
Escudero et al. Obtaining knowledge from the web using fusion and summarization techniques
van Hooland et al. Named-entity recognition: a gateway drug for cultural heritage collections to the linked data cloud
AU2012200686B2 (en) Improved search engine
Varga et al. Multilingual extraction and editing of concept strings for the legal domain

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant