Summary of the invention
The main purpose of the present invention is to provide a kind of questionnaire generation method, device, equipment and storage mediums, it is intended toRealize the questionnaire for flexibly intelligently generating and meeting investigation purpose.
To achieve the above object, the present invention provides a kind of questionnaire generation method, the questionnaire generation method packetInclude following steps:
When receiving questionnaire generation request, obtains the questionnaire and generate the corresponding investigation scene information of requestWith target of investication and study information；
Investigation topic, composition investigation topic collection are extracted from the default questionnaire information library according to the investigation scene informationIt closes；
Target investigation topic is extracted from the investigation topic set according to the target of investication and study information and generation of sorting is adjustedVolume is interrogated, inputs answer information so that user is based on the questionnaire；
When detecting that the questionnaire replies completion, the questionnaire is associated with preservation with the answer information.
Optionally, described to extract investigation topic, group from the default questionnaire information library according to the investigation scene informationThe step of at investigation topic set, comprising:
Extract the scene word in the investigation scene information, obtain in default questionnaire information library with the scene wordThe investigation topic matched forms initial investigation topic set；
Extract the investigation purpose in the investigation scene information, delete in the initial investigation topic set with the investigationTopic is investigated in the incoherent interference of purpose, obtains investigation topic set.
Optionally, described to extract target investigation topic from the investigation topic set according to the target of investication and study information simultaneouslyThe step of sequence generates questionnaire, is based on questionnaire input answer information for user, comprising:
When the information content of the target of investication and study information is no more than presupposed information amount, obtains the questionnaire and generate requestCorresponding questionnaire type；
When the questionnaire type is card form subjective item type, the behavior obtained in the target of investication and study information is specialReference breath；
It obtains and investigates topic with the matched first object of the behavior characteristic information in the investigation topic set and export,Answer information is inputted so that user is based on first object investigation topic；
The answer information received is analyzed by default answer analysis model, is obtained related to the answer informationNew first object investigation topic and export；
When detecting that the quantity of first object investigation topic reaches preset quantity, prompt information is exported, to prompt userQuestionnaire, which generates, to be completed.
Optionally, described that the answer information received is analyzed by default answer analysis model, obtain with it is describedBefore the step of relevant new first object of answer information is investigated topic and is exported, comprising:
The subjective item investigation sample of preset ratio is extracted from default subjective item investigation sample set as the first use-case, andUsing the subjective item investigation sample in addition to first use-case as the second use-case；
By the basic answer analysis model of first use-case training, initial answer analysis model is obtained, passes through described theTwo use-cases verify the initial answer analysis model, and obtain the initial answer analysis model is verified rate；
It is described be verified rate and be higher than default percent of pass when, the initial answer analysis model is divided as default answerAnalyse model.
Optionally, when the information content in the target of investication and study information is no more than presupposed information amount, the investigation is obtainedQuestionnaire generated after the step of requesting corresponding questionnaire type, comprising:
When the questionnaire type is not card form subjective item type, judges whether there is and meet the target of investication and studyThe default virtual target of investication and study of information；
When there is the default virtual target of investication and study for meeting the target of investication and study information, the default virtual investigation pair is obtainedIt as corresponding questionnaire and exports, inputs answer information so that user is based on the questionnaire.
Optionally, described to extract target investigation topic from the investigation topic set according to the target of investication and study information simultaneouslyThe step of sequence generates questionnaire, is based on questionnaire input answer information for user, comprising:
When the information content of the target of investication and study information is more than presupposed information amount, by the target of investication and study information and the tuneEach investigation topic ground in topic set is compared, and it is similar to the target of investication and study information to obtain each investigation topicDegree；
Each investigation topic is arranged by the height of similarity, obtains the second target investigation topic for the forward preset quantity that sortsMesh；
Second target is investigated into topic combination producing questionnaire, is answered so that user is based on questionnaire inputCase information.
Optionally, described when detecting that the questionnaire replies completion, the questionnaire and the answer are believedAfter the step of breath association saves, comprising:
The answer information is divided into preset kind, counts the informational capacity of each default type answer information；
The corresponding conclusion clause of the informational capacity is obtained, and the conclusion clause is input in default template, is generatedThe corresponding research conclusion report of the questionnaire.
In addition, to achieve the above object, the present invention also provides a kind of questionnaire generating means, the questionnaire is generatedDevice includes:
Command reception module, for obtaining the questionnaire and generating request when receiving questionnaire generation requestCorresponding investigation scene information and target of investication and study information；
Problem abstraction module, for extracting investigation topic from the default questionnaire information library according to the investigation scene informationMesh, composition investigation topic set；
Questionnaire generation module, for extracting target investigation from the investigation topic set according to the target of investication and study informationTopic simultaneously sorts and generates questionnaire, inputs answer information so that user is based on the questionnaire；
Answer preserving module, for detect the questionnaire reply complete when, by the questionnaire and describedAnswer information association saves.
In addition, to achieve the above object, the present invention also provides a kind of questionnaire generating devices；
The questionnaire generating device includes: memory, processor and is stored on the memory and can be describedThe computer program run on processor, in which:
The computer program realizes the step of questionnaire generation method as described above when being executed by the processor.
In addition, to achieve the above object, the present invention also provides computer storage mediums；
Computer program, the realization when computer program is executed by processor are stored in the computer storage mediumSuch as the step of above-mentioned questionnaire generation method.
A kind of questionnaire generation method, device, equipment and the storage medium that the embodiment of the present invention proposes, server are connecingWhen receiving questionnaire generation request, obtains the questionnaire and generate the corresponding investigation scene information of request and target of investication and study letterBreath；Investigation topic, composition investigation topic set are extracted from the default questionnaire information library according to the investigation scene information；RootTarget investigation topic is extracted from the investigation topic set according to the target of investication and study information and is sorted and generates questionnaire, forUser is based on the questionnaire and inputs answer information；When detecting that the questionnaire replies completion, the investigation is askedVolume is associated with preservation with the answer information.In the present invention in the generating process of questionnaire, server is according to questionnaireThe target of investication and study information that utility efficiency scene information and questionnaire are directed to generates questionnaire, realizes the customization of questionnaireMetaplasia is at so that the more flexible intelligence of the generation of questionnaire.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
As shown in Figure 1, the server that Fig. 1 is the hardware running environment that the embodiment of the present invention is related to (is called investigation to askRoll up generating device, wherein questionnaire generating device can be to be made of individual questionnaire generating means, be also possible to byOther devices are combined with questionnaire generating means and to be formed) structural schematic diagram.
Server of the embodiment of the present invention refers to a management resource and provides the computer of service for user, is generally divided into fileServer, database server and apps server.The computer or computer system for running the above software are also referred to asServer.For common PC (personal computer) personal computer, server is in stability, safety, propertyEnergy etc. requires higher；As shown in Figure 1, the server may include: processor 1001, such as central processing unit(Central Processing Unit, CPU), network interface 1004, user interface 1003, memory 1005, communication bus1002, hardware such as chipset, disk system, network etc..Wherein, communication bus 1002 is for realizing the connection between these componentsCommunication.User interface 1003 may include cutout screen (Display), input unit such as keyboard (Keyboard), optional userInterface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include having for standardLine interface, wireless interface (such as Wireless Fidelity WIreless-FIdelity, WIFI interface).Memory 1005 can be high speed withMachine accesses memory (random access memory, RAM), is also possible to stable memory (non-volatile), such as magnetic disk storage memory.Memory 1005 optionally can also be the storage dress independently of aforementioned processor 1001It sets.
Optionally, server can also include camera, RF (Radio Frequency, radio frequency) circuit, sensor, soundFrequency circuit, WiFi module；Input unit, than cutout screen, touch screen；Network interface can be blue in blanking wireless interface in addition to WiFiTooth, probe etc..It will be understood by those skilled in the art that server architecture shown in Fig. 1 does not constitute the restriction to server,It may include perhaps combining certain components or different component layouts than illustrating more or fewer components.
As shown in Figure 1, the computer software product, which is stored in a storage medium, (storage medium: is called computer storageMedium, computer media, readable medium, readable storage medium storing program for executing, computer readable storage medium are directly medium etc., storageMedium can be non-volatile readable storage medium, such as RAM, magnetic disk, CD) in, including several requests use is so that an endEnd equipment (can be mobile phone, computer, server, air conditioner or the network equipment etc.) executes each embodiment institute of the present inventionThe method stated, as may include operating system, network communication module, use in a kind of memory 1005 of computer storage mediumFamily interface module and computer program.
In server shown in Fig. 1, network interface 1004 be mainly used for connect background data base, with background data base intoRow data communication；User interface 1003 is mainly used for connection client, and (client, is called user terminal or terminal, and the present invention is implementedExample terminal can be also possible to mobile terminal with fixed terminal, e.g., intelligent air condition, intelligent electric lamp, intelligent power with network savvy,Intelligent sound box, autonomous driving vehicle, PC, smart phone, tablet computer, E-book reader, portable computer etc., are wrapped in terminalContaining sensor such as optical sensor, motion sensor and other sensors, details are not described herein), data are carried out with clientCommunication；And processor 1001 can be used for calling the computer program stored in memory 1005, and it is real to execute the present invention or lessStep in the questionnaire generation method of example offer is provided.
This motion point provides a kind of questionnaire generation method, applied to server as shown in Figure 1.
Referring to Fig. 2, in the first embodiment of questionnaire generation method of the present invention, the questionnaire generation method packetIt includes:
Step S10 obtains the questionnaire and generates the corresponding investigation of request when receiving questionnaire generation requestScene information and target of investication and study information.
User browses a Webpage in terminal, and triggers questionnaire on the Webpage of the terminal and generate requestWhen, terminal obtains the corresponding investigation purpose of Webpage, business scenario (being called scene word), operation flow and business datumDeng the conduct questionnaire generation request correspondence such as investigation purpose, business scenario, operation flow and business datum that terminal will acquireInvestigation scene information, at the same time, terminal acquires the corresponding user behavior data of Webpage, account information and the account pairThe historical behavior data answered, terminal are asked collected user behavior data, account information and historical behavior data as investigationVolume, which generates, requests corresponding target of investication and study information.
Terminal will be investigated scene information and target of investication and study information and is associated with investigation questionnaire generation request, and will be after associationQuestionnaire generate request and be sent to server, server receives questionnaire and generates request, and obtains questionnaire generationRequest corresponding investigation scene information and target of investication and study information.
Step S20, investigation topic is extracted according to the investigation scene information from the default questionnaire information library, and composition is adjustedGrind topic set.
Specifically, comprising:
Step a1 extracts the scene word in the investigation scene information, obtain in default questionnaire information library with the fieldThe investigation topic of scape word match forms initial investigation topic set；
Step b1 extracts the investigation purpose in the investigation scene information, delete in the initial investigation topic set andTopic is investigated in the incoherent interference of investigation purpose, obtains investigation topic set.
Server extracts the scene word in investigation scene information, presets questionnaire information library in server, the default questionnairePreserve different types of investigation topic in information bank, server is by each investigation in scene word and default questionnaire information libraryTopic is compared, and server obtains the investigation topic in default questionnaire information library with scene word match, and server will matchEach investigation topic summarized, form initial investigation topic set；Then, server extracts the tune in investigation scene informationGrind purpose, server obtains the corresponding default investigation purpose of each investigation topic in initial investigation topic set, and server will tuneIt grinds the default investigation purpose corresponding with each investigation topic of the investigation purpose in scene information to be compared, server is deleted initialIt investigates interference different from the investigation purpose in investigation scene information in topic set and investigates topic, obtain investigation topic set.
Server is according to investigation scene information in the present embodiment, and preliminary screening goes out part and investigates from default questionnaire information libraryTopic combines to form investigation topic set, raw so as to the screening investigation topic relevant to target of investication and study from investigation topic setAt questionnaire.
Step S30 extracts target investigation topic according to the target of investication and study information side by side from the investigation topic setSequence generates questionnaire, inputs answer information so that user is based on the questionnaire.
According to the target of investication and study information of acquisition, the screening target investigation topic from investigation topic set generates investigation to serverQuestionnaire can realize by different modes, for example,
Implementation one: server by target of investication and study information user behavior data with investigation topic set in it is eachInvestigation topic is compared, and server, which is obtained, forms questionnaire with 10 high investigation topics of user behavior data similarity；
Implementation two: server obtains historical user's behavioral data in target of investication and study information, and server uses historyFamily behavioral data is compared with each investigation topic in investigation topic set, and server obtains and historical user's behavioral data10 high investigation topics of similarity form questionnaire；
Implementation three: server screens a mesh from investigation topic set by the account information in target of investication and study informationMark investigation topic makes user be based on target investigation topic and replies, and server reception user is based on target investigation topic and repliesAnswer information, server analyzes answer information, obtains second target investigation topic relevant to answer information, directlyThe target investigation topic that setting number is filtered out into server terminates.
Step S40 closes the questionnaire and the answer information when detecting that the questionnaire replies completionUNPROFOR is deposited.
When server detects that questionnaire replies completion, questionnaire is associated with preservation with answer information by server,To generate investigation research conclusion report, that is,
The answer information is divided into preset kind, counts the informational capacity of each default type answer information by step a2；
Step b2 obtains the corresponding conclusion clause of the informational capacity, and the conclusion clause is input to default templateIn, generate the corresponding research conclusion report of the questionnaire.
Server obtains identical investigation topic in each questionnaire, and server obtains the answer letter of the investigation topicBreath, then, each answer information is divided into preset kind by server, and (preset kind refers to that pre-set answer is classified, in advanceIf type can be arranged according to concrete scene, such as preset kind is set as interested, loses interest in), server statistics are each pre-If the informational capacity of type answer information, that is, server obtains the corresponding difference of identical investigation topic in different questionnaires and answersCase information, server classify answer information, and count the informational capacity of each type answer information；Server obtains letterThe corresponding conclusion clause of total amount is ceased, and conclusion clause is input to default template and (presets template and refers to that pre-set investigation is tiedBy blank template) in, generate the corresponding research conclusion report of questionnaire.
In the present embodiment in the generating process of questionnaire, server is believed according to the utility efficiency scene of questionnaireIt ceases target of investication and study information corresponding with questionnaire and generates questionnaire, realize the customization of questionnaire, so that investigation is askedThe more flexible intelligence of the generation of volume, ensure that the comprehensive of investigation.
Further, on the basis of first embodiment of the invention, the of questionnaire generation method of the present invention is proposedTwo embodiments.
The present embodiment is a kind of specific implementation that step S20 is refined in first embodiment, is passed through in the present embodiment pre-If answer analysis model analyzes the answer information of user, questionnaire is generated.
Before the step of the present embodiment executes, answer analysis model is preset in server, presets answer analysis modelSetting steps, comprising:
Step S01 extracts the subjective item investigation sample of preset ratio as the from default subjective item investigation sample setOne use-case, and using the subjective item investigation sample in addition to first use-case as the second use-case.
Subjective item is preset in server and investigates sample set, and default subjective item investigation sample set refers to by different subjectivitiesThe set of topic investigation sample composition, specifically, server obtain the subjective item investigation topic and subjective item tune of magnanimity from networkGrind the corresponding answer information of topic, wherein subjective item investigation topic refers to the investigation topic of question and answer type, and server is by subjective itemTopic and the subjective item corresponding answer information of investigation topic are investigated as a subjective item and investigates sample, server is by each subjectivityTopic investigation sample combines to obtain default subjective item investigation sample set.
Server extracts preset ratio from default subjective item investigation sample set, and (preset ratio refers in server in advanceThe training sample of setting extracts ratio, and preset ratio can be according to concrete scene flexible setting, for example, preset ratio is set as95%) subjective item investigation sample as the first use-case, investigated and remove the first use-case in sample set by server by default subjective itemExcept subjective item investigation sample as the second use-case.
Step S02 is obtained initial answer analysis model, is led to by the basic answer analysis model of first use-case trainingIt crosses second use-case and verifies the initial answer analysis model, obtain the initial answer analysis model is verified rate.
Server is using the first use-case as input information input into basic answer analysis model, wherein basic answer pointAnalysis model refers to the pre-set algorithm with semantics recognition function, the basic answer analysis of the first use-case of server by utilizing trainingModel obtains initial answer analysis model, that is, server is by the first use-case repetitive exercise basis answer analysis model, to adjustThe parameter information of integral basis plinth answer analysis model obtains the initial answer analysis model after parameter optimization；Initially answeredAfter case analysis model, the second use-case is input in initial answer analysis model by server, obtain it is corresponding analysis as a result, withSo that server is verified the percent of pass of initial answer analysis model based on the analysis results, that is, server by obtained analysis result andStandard analysis result is compared, and when obtained analysis result is identical with standard analysis result, server determination is verified,The ratio of use-case quantity and the second use-case of total quantity quantity that server statistics are verified, obtains initial answer analysis modelIt is verified rate.
Step S03, it is described be verified rate and be higher than default percent of pass when, will the initial answer analysis model as in advanceIf answer analysis model.
When server determines that the rate that is verified is higher than default percent of pass, (default percent of pass, which refers in server, to be presetPercent of pass critical value, be verified rate be higher than can be according to concrete scene flexible setting, for example, being set as 98%), serverUsing initial answer analysis model as default answer analysis model, determine that the rate that is verified is not higher than default percent of pass in serverWhen, server extracts subjective item investigation sample again from subjective item investigation sample set and is iterated training.
The present embodiment presets answer analysis model in the server, to carry out answer information using default answer analysis modelAnalysis, determine target investigate topic, and by target investigation topic summarize the method for obtaining questionnaire, specifically, comprising:
Step S21 obtains the questionnaire when the information content of the target of investication and study information is no more than presupposed information amountIt generates and requests corresponding questionnaire type.
Server determine target of investication and study information information content be no more than presupposed information amount (presupposed information amount refers to be set in advanceThe information sum set, such as presupposed information amount are set as 10) when, that is, server determines that target of investication and study information is less, serverIt obtains questionnaire and generates the corresponding questionnaire type of request, the questionnaire type in the present embodiment includes but not only limitsIn: card form subjective item type, card form objective item type, sheet format subjective item type, sheet format objective item type, thin note formulaSubjective item type and thin note formula objective item type.
Step S22 is obtained in the target of investication and study information when the questionnaire type is card form subjective item typeBehavior characteristic information；It obtains in the investigation topic set and investigates topic with the matched first object of the behavior characteristic informationAnd export, answer information is inputted so that user is based on first object investigation topic.
When server determines that questionnaire type is card form subjective item type, server is obtained in target of investication and study informationBehavior characteristic information (behavior characteristic information can be understood as user in the browse operation information of network)；Server obtains investigationWith the matched first object investigation topic of behavior characteristic information (it is understood that card form subjective item type in topic setQuestionnaire refer to that the primary questionnaire for only showing a subjective item type, the first object in the present embodiment investigate topicQuantity be one) and export, first object investigation topic is sent to terminal, the first mesh that terminal displays reception arrives by serverMark investigation topic inputs answer information so that user is based on first object investigation topic.
For example, the behavior characteristic information that server obtains are as follows: browsing course is recommended webpage 1 minute, at Junior Mathematics private tutorClick checks that then server is got and the matched first object investigation topic of behavior characteristic information are as follows: could you tell me to private tutor'sIt is required that is；First object is investigated topic by server: could you tell me to the requirement of private tutor it is what is sent to terminal, forFamily is based on first object investigation topic and inputs answer information.
Step S23 analyzes the answer information received by default answer analysis model, obtains and the answerThe relevant new first object investigation topic of information simultaneously exports.
The answer information for the first object investigation topic that server receiving terminal is sent, then, server is by answer informationIt is input in default answer analysis model, answer information is analyzed by default answer analysis model, obtain believing with answerIt ceases relevant new first object investigation topic and exports, specifically:
Answer information is carried out word segmentation processing by default answer analysis model, obtains the corresponding each participle of answer information, is takenThe noise word being engaged in device removal participle, for example, auxiliary word, server summarizes the participle after denoising to obtain answer information correspondingEach participle in participle set is compared by participle set, server with each investigation topic in investigation topic set,It obtains that the target is investigated topic as new first object with the highest target investigation topic of answer information similarity, serverInvestigation topic simultaneously exports, and inputs answer information so that user is based on first object investigation topic.
Step S24 exports prompt information when detecting that the quantity of first object investigation topic reaches preset quantity, withIt prompts user investigation questionnaire to generate to complete.
As soon as the new first object of the every determination of server investigates topic, server carries out plus a current investigation of processing statisticsThe topic quantity of the investigation topic of questionnaire detects that the quantity of first object investigation topic reaches preset quantity (in advance in serverIf quantity is pre-set questionnaire topic quantity, preset quantity can be arranged according to the Reaction time of each topic, exampleSuch as, total Reaction time of questionnaire was at 10 minutes, and each topic 1 minute, then preset quantity is set as 10) when, serverIt determines output prompt information, is completed with prompting user investigation questionnaire to generate.
The answering information of user is compared by default answer analysis model in the present embodiment, screens next tuneTopic is ground, generates different questionnaires to customize according to different people, so that the generating mode of questionnaire is more intelligent, thisOutside, by customizing different questionnaires in this present embodiment, the information of the available different dimensions to target of investication and study be ensure thatInvestigation and analysis is more comprehensive.
Further, the third embodiment of the present invention is proposed on the basis of second embodiment of the invention.
The present embodiment be in second embodiment after step S21 the step of, specifically illustrate questionnaire in the present embodimentWhen type is not card form subjective item type, the questionnaire generation method includes:
Step S25 is judged whether there is described in meeting when the questionnaire type is not card form subjective item typeThe default virtual target of investication and study of target of investication and study information.
When it is card form subjective item type that server, which determines questionnaire type not, that is, server cannot be according to answerThe semanteme of information determines next target investigation topic, and therefore, server needs to generate questionnaire according to target of investication and study information,But since the information content of the target of investication and study information in server is few, such as determine that target investigation is inscribed according only to target of investication and study informationMesh, not can achieve the demand of investigation, the method that server generates questionnaire, specifically, comprising:
Virtual target of investication and study is preset in server, preset virtual target of investication and study refer to it is pre-set each different types ofTarget of investication and study, for example, pre-setting virtual target of investication and study 1: student enrollment in server, the virtual target of investication and study 1 is correspondingCharacteristic information are as follows: income is lower, and consuming capacity is strong, the age 18 to 22 years old；Virtual target of investication and study 2: company's white collar, the virtual tuneGrind the corresponding characteristic information of object 2 are as follows: middle-income, consuming capacity is strong, and the age 23 to 28 years old；Virtual target of investication and study 3: middle production rankGrade, the corresponding characteristic information of virtual target of investication and study 3 are as follows: it earns a large income, consuming capacity is medium, and the age 29 to 40 years old etc.；ServiceDevice obtain it is each presets virtual target of investication and study, and preset the corresponding characteristic information of virtual target of investication and study, server is by target of investication and studyInformation is compared with the corresponding characteristic information of virtual target of investication and study is preset, and judges whether there is and meet target of investication and study informationPreset virtual target of investication and study, that is, target of investication and study information characteristic information matching corresponding with virtual target of investication and study is preset in server,Then server determines there is the default virtual target of investication and study for meeting target of investication and study information；It is on the contrary.
Step S26 obtains the default void when there is the default virtual target of investication and study for meeting the target of investication and study informationThe quasi- corresponding questionnaire of target of investication and study simultaneously exports, and inputs answer information so that user is based on the questionnaire.
When server determines the default virtual target of investication and study for existing and meeting target of investication and study information, server is by the investigation pairAs with preset virtual target of investication and study and carry out analogy, server acquisition is preset the corresponding questionnaire of virtual target of investication and study and is exported,Answer information is inputted so that user is based on the questionnaire.Determine that there is no meet the default of target of investication and study information in serverWhen virtual target of investication and study, server obtains the corresponding basic survey questionnaire of investigation scene information and exports, so that user is based on basePlinth questionnaire input answer information presets virtual target of investication and study in server in the present embodiment, server according to investigation pairThe target of investication and study information of current target of investication and study is compared image information with the characteristic information for presetting virtual target of investication and study, obtainsWith target of investication and study information similarity is highest presets virtual target of investication and study, and obtain this and preset the corresponding investigation of virtual target of investication and studyQuestionnaire, i.e., by the existing virtual target of investication and study of analogy in the present embodiment, so that the more flexible intelligence of the generation of questionnaire.
Further, the fourth embodiment of the present invention is proposed on the basis of the above embodiment of the present invention.
The present embodiment is another specific implementation that step S20 is refined in first embodiment, and the questionnaire is rawAt method, comprising:
Step S27, when the information content of the target of investication and study information is more than presupposed information amount, by the target of investication and study informationIt is compared with each investigation topic in the investigation topic set, obtains each investigation topic and the target of investication and study informationSimilarity.
The information content that target of investication and study information is determined in server is more than presupposed information amount (the presupposed information amount in the present embodimentThe present embodiment identical as the presupposed information amount in second embodiment does not repeat) when, server is by target of investication and study information and investigatesEach investigation topic in topic set is compared, and obtains the similarity of each investigation topic and target of investication and study information.
Step S28 arranges each investigation topic by the height of similarity, obtains the second target of the forward preset quantity that sortsInvestigate topic.
Server will investigate each investigation topic in topic set, sort by the height of similarity, server obtains sequenceForward preset quantity (do not go to live in the household of one's in-laws on getting married by preset quantity the present embodiment identical as the preset quantity in second embodiment in the present embodimentState) the second target investigation topic.
Second target is investigated topic combination producing questionnaire by step S29, is asked so that user is based on the investigationVolume input answer information.
Second target is investigated topic combination producing questionnaire by server, so that user is inputted based on the questionnaireAnswer information；Server chooses target according to target of investication and study information and investigates topic in the present embodiment, forms questionnaire, so that adjustingIt is more flexible to interrogate volume generation.
In addition, the embodiment of the present invention also proposes a kind of questionnaire generating means referring to Fig. 3, the questionnaire is generatedDevice includes:
Command reception module 10 is asked for when receiving questionnaire generation request, obtaining the questionnaire generationAsk corresponding investigation scene information and target of investication and study information；
Problem abstraction module 20, for extracting investigation from the default questionnaire information library according to the investigation scene informationTopic, composition investigation topic set；
Questionnaire generation module 30, for extracting target tune from the investigation topic set according to the target of investication and study informationIt grinds topic and sorts and generate questionnaire, input answer information so that user is based on the questionnaire；
Answer preserving module 40, for detect the questionnaire reply complete when, by the questionnaire and instituteAnswer information association is stated to save.
Optionally, described instruction receiving module 10, comprising:
First screening unit obtains default questionnaire information library for extracting the scene word in the investigation scene informationIn investigation topic with the scene word match, form initial investigation topic set；
Second screening unit deletes the initial investigation topic for extracting the investigation purpose in the investigation scene informationTopic is investigated with the incoherent interference of the investigation purpose in mesh set, obtains investigation topic set.
Optionally, the questionnaire generation module 30, comprising:
First acquisition unit when being no more than presupposed information amount for the information content in the target of investication and study information, obtains instituteIt states questionnaire and generates the corresponding questionnaire type of request；
Second acquisition unit, for obtaining the investigation when the questionnaire type is card form subjective item typeBehavior characteristic information in object information；
Topic output unit, for obtain in the investigation topic set with matched first mesh of the behavior characteristic informationMark investigation topic simultaneously exports, and inputs answer information so that user is based on first object investigation topic；
Output unit is analyzed, for being analyzed by default answer analysis model the answer information received, is obtainedNew first object investigation topic relevant to the answer information simultaneously exports；
Determination unit is detected, for when detecting that the quantity of first object investigation topic reaches preset quantity, output to be mentionedShow information, is completed with prompting user investigation questionnaire to generate.
Optionally, the questionnaire generating means, comprising:
Sampling module, the subjective item for extracting preset ratio from default subjective item investigation sample set investigate sampleThis investigates sample as the second use-case as the first use-case, and using the subjective item in addition to first use-case；
Model training module, for obtaining initial answer point by the basic answer analysis model of first use-case trainingModel is analysed, the initial answer analysis model is verified by second use-case, obtains testing for the initial answer analysis modelDemonstrate,prove percent of pass；
Model generation module, for it is described be verified rate and be higher than default percent of pass when, will the initial answer analysisModel is as default answer analysis model.
Optionally, the questionnaire generation module 30, comprising:
Object judging unit, for judging whether to deposit when the questionnaire type is not card form subjective item typeIn the default virtual target of investication and study for meeting the target of investication and study information；
Questionnaire output unit, for obtaining when there is the default virtual target of investication and study for meeting the target of investication and study informationIt is described to preset the corresponding questionnaire of virtual target of investication and study and export, so that user is based on questionnaire input answer letterBreath.
Optionally, the questionnaire generation module 30, further includes:
Information comparison unit, when being more than presupposed information amount for the information content in the target of investication and study information, by the tuneIt grinds object information to be compared with each investigation topic in the investigation topic set, obtains each investigation topic and the tuneGrind the similarity of object information；
Topic sequencing unit obtains the forward preset quantity that sorts for arranging each investigation topic by the height of similarityThe second target investigate topic；
Questionnaire generation unit, for second target to be investigated topic combination producing questionnaire, so that user is based onThe questionnaire inputs answer information.
Optionally, the questionnaire generating means, further includes:
Answer statistical module counts each default type answer information for the answer information to be divided into preset kindInformational capacity；
Report generation module is inputted for obtaining the corresponding conclusion clause of the informational capacity, and by the conclusion clauseInto default template, the corresponding research conclusion report of the questionnaire is generated.
Wherein, the step of each Implement of Function Module of questionnaire generating means can refer to questionnaire of the present invention and generateEach embodiment of method, details are not described herein again.
In addition, the embodiment of the present invention also proposes a kind of computer storage medium.
Computer program, the realization when computer program is executed by processor are stored in the computer storage mediumOperation in questionnaire generation method provided by the above embodiment.
It should be noted that, in this document, relational terms such as first and second and the like are used merely to a realityBody/operation/object is distinguished with another entity/operation/object, without necessarily requiring or implying these entity/operations/There are any actual relationship or orders between object；The terms "include", "comprise" or its any other variant meaningCovering non-exclusive inclusion, so that the process, method, article or the system that include a series of elements not only include thatA little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article orThe intrinsic element of system.In the absence of more restrictions, the element limited by sentence "including a ...", is not arrangedExcept there is also other identical elements in process, method, article or the system for including the element.
For device embodiment, since it is substantially similar to the method embodiment, related so describing fairly simplePlace illustrates referring to the part of embodiment of the method.The apparatus embodiments described above are merely exemplary, wherein makingIt may or may not be physically separated for the unit of separate part description.In can selecting according to the actual needsSome or all of the modules realize the purpose of the present invention program.Those of ordinary skill in the art are not making the creative laborIn the case where, it can it understands and implements.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment sideMethod can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many casesThe former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior artThe part contributed out can be embodied in the form of software products, which is stored in one as described aboveIn storage medium (such as ROM/RAM, magnetic disk, CD), including several requests are used so that terminal device (it can be mobile phone,Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hairEquivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skillsArt field, is included within the scope of the present invention.