CN109947908A - The building method and construction system of robot knowledge base - Google Patents
The building method and construction system of robot knowledge base Download PDFInfo
- Publication number
- CN109947908A CN109947908A CN201711172532.3A CN201711172532A CN109947908A CN 109947908 A CN109947908 A CN 109947908A CN 201711172532 A CN201711172532 A CN 201711172532A CN 109947908 A CN109947908 A CN 109947908A
- Authority
- CN
- China
- Prior art keywords
- knowledge point
- gauge outfit
- static
- attribute information
- answer
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 66
- 238000010276 construction Methods 0.000 title claims abstract description 14
- 230000003068 static effect Effects 0.000 claims abstract description 84
- 230000015654 memory Effects 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 abstract description 2
- 238000012937 correction Methods 0.000 description 11
- 238000010586 diagram Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 238000004364 calculation method Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000011218 segmentation Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000013139 quantization Methods 0.000 description 2
- 241001672694 Citrus reticulata Species 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013481 data capture Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000010129 solution processing Methods 0.000 description 1
- 239000004575 stone Substances 0.000 description 1
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses the building method and construction system of a kind of robot knowledge base, the building method includes: to obtain static two dimensional table data；The corresponding more than two attribute informations of the table body are determined from gauge outfit；One or more automatic question answerings knowledge point is generated according to the attribute information, each automatic question answering knowledge point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic；By the storage of the static two dimensional table data, the automatic question answering knowledge point and the attribute information into knowledge base.The embodiment of the present invention, which is realized, automatically generates knowledge point according to static two dimensional table data, and establishes corresponding knowledge base, reduce operator's workload and reduce the possibility artificially made a mistake, and improves the accuracy and formation efficiency of the knowledge point of generation.
Description
Technical field
The present embodiments relate to automatic question answering technology more particularly to a kind of building method, the machines of robot knowledge base
People's knowledge base, automatic question-answering method, automatically request-answering system, the construction system of robot knowledge base, terminal device and computer
Storage medium.
Background technique
In automatically request-answering system, some knowledge points are not and to come from one from simple automatic question answering pair
The structural data that a little static two dimensional table structures are realized, the data volume of structural data is very huge, financing table pair as shown in Table 1
About 8*5 knowledge point (as increased benefit 90 days Annual Percentage Rates of series are how many) is answered, knowledge point includes problem and answer, knowledge quantity
It is very big, and every knowledge point requires operator's Manual arranging.
The financing table of table 1
If content in table is once high-volume is modified, operator just needs the answer for finding corresponding knowledge point to do one by one
It changes, not only heavy workload, and is easy to make mistakes.
Summary of the invention
In view of this, the embodiment of the present invention provides the building method of a kind of robot knowledge base, robot knowledge base, automatic
Answering method, automatically request-answering system, the construction system of robot knowledge base, terminal device and computer storage medium, with reality
Knowledge point is now automatically generated, operator's workload is reduced and improves the accuracy of knowledge point.
The embodiment of the invention provides a kind of building methods of robot knowledge base, comprising:
Static two dimensional table data are obtained, the static two dimensional table data include topic, gauge outfit and table body, and the gauge outfit is the
A line, the table body are other rows except the first row；
The corresponding more than two attribute informations of the table body are determined from gauge outfit, when the gauge outfit content of multi-column data is corresponding
When attribute is identical, the gauge outfit content of the multi-column data is summarized as the attribute information, when the gauge outfit of an only column data
When content corresponds to an attribute, directly using the gauge outfit content of the column data as an attribute information；
One or more automatic question answerings knowledge point is generated according to the attribute information, each automatic question answering knowledge point includes asking
It inscribes expression formula and answer expression formula, the answer expression formula includes the topic；
By the storage of the static two dimensional table data, the automatic question answering knowledge point and the attribute information into knowledge base.
Optionally, the method also includes:
Establish the inclusion relation of the attribute information with perhaps gauge outfit content interior in corresponding table body；
By inclusion relation storage into knowledge base.
Optionally, the method also includes:
Part of speech is established for the word in the gauge outfit or/and the table body, part of speech name of the word as corresponding part of speech,
The part of speech includes the synonym of the word and the word；
Establish the attribute information and the inclusion relation of content in corresponding table body include: establish the attribute information with it is right
The inclusion relation of part of speech name in the table body or gauge outfit answered；
By inclusion relation storage into knowledge base further include: by part of speech storage into knowledge base.
Optionally, generating one or more automatic question answerings knowledge point according to the attribute information includes:
An initial knowledge point is automatically generated according at least two attribute informations；
Each initial knowledge point is adjusted, the automatic question answering knowledge point is obtained.
The embodiment of the invention also provides a kind of robots that the building method using above-mentioned robot knowledge base is built
Knowledge base.
The embodiment of the invention also provides a kind of automatic question-answering methods based on above-mentioned robot knowledge base, comprising:
In the solicited message for receiving user, the automatic question answering knowledge in knowledge base is matched according to the solicited message
Point；
According to the corresponding topic in the automatic question answering knowledge point being matched to, corresponding static two dimensional table is searched, and obtain
The static two dimensional table found；
Search corresponding answer in the static two dimensional table according to the solicited message, according to the answer found and
Determining answer expression formula generates final result, and the final result is returned to the user.
The embodiment of the invention also provides a kind of automatically request-answering systems based on above-mentioned robot knowledge base, comprising:
Matching module is requested, for matching knowledge base according to the solicited message in the solicited message for receiving user
In question and answer knowledge point；
Data acquisition module, for searching corresponding static state according to the corresponding topic in question and answer knowledge point being matched to
Bivariate table, and obtain the static two dimensional table found；
Answer generation module searches corresponding answer in the static two dimensional table according to the solicited message, according to looking into
The answer expression formula of the answer and determination found generates final result；
Answer return module, for the final result to be returned to the user.
The present invention also provides a kind of construction systems of robot knowledge base, comprising:
Data acquisition module, for obtaining static two dimensional table data, the static two dimensional table data include topic, gauge outfit and
Table body, the gauge outfit are the first row, and the table body is other rows except the first row；
Attribute determination module determines the corresponding more than two attribute informations of the table body, when multi-column data from gauge outfit
When the corresponding attribute of gauge outfit content is identical, the gauge outfit content of the multi-column data is summarized as the attribute information, when only
When the gauge outfit content of one column data corresponds to an attribute, directly believe the gauge outfit content of the column data as an attribute
Breath；
Knowledge point generation module generates one or more automatic question answerings knowledge point according to the attribute information, each automatic
Question and answer knowledge point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic；
Memory module, for depositing the static two dimensional table data, the automatic question answering knowledge point and the attribute information
It stores up in knowledge base.
The embodiment of the invention also provides a kind of terminal devices, comprising:
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes the building method or above-mentioned automatic question-answering method of above-mentioned robot knowledge base.
The embodiment of the invention also provides a kind of computer storage mediums, are stored thereon with computer program, the program quilt
Processor realizes the building method or above-mentioned automatic question-answering method of above-mentioned robot knowledge base when executing.
The technical solution of the embodiment of the present invention, by obtaining static two dimensional table data, from the gauge outfit of static two dimensional table data
The corresponding more than two attribute informations of middle determining table body, generate one or more automatic question answering knowledge according to the attribute information
Point, each automatic question answering knowledge point include problem expression formula and answer expression formula, and the answer expression formula includes the topic, will
Static two dimensional table data, automatic question answering knowledge point and attribute information storage are realized into knowledge base according to static two dimensional table number
According to automatically generating knowledge point, and establish corresponding knowledge base, it is no longer necessary to which operator is whole manually according to static two dimensional table data
Knowledge point is managed, reduces the workload of operator, and reduce the possibility artificially made a mistake, improves the standard of the knowledge point of generation
True property and formation efficiency.And when modifying to static two dimensional table data, it may not be necessary to manual as the prior art
It arranges the every knowledge point generated to modify, need to only modify the corresponding automatic question answering knowledge point of attribute information of variation, significantly
Reduce the workload of operator.
Detailed description of the invention
Fig. 1 is a kind of flow chart of the building method for robot knowledge base that the embodiment of the present invention one provides；
Fig. 2 is that the primary key column to table 1 in the building method for the robot knowledge base that the embodiment of the present invention one provides is established
The schematic diagram of part of speech；
Fig. 3 is the column of other column to table 1 in the building method for the robot knowledge base that the embodiment of the present invention one provides
Topic establishes the schematic diagram of part of speech；
Fig. 4 is a kind of flow chart for automatic question-answering method that the embodiment of the present invention three provides；
Fig. 5 is a kind of structural schematic diagram for automatically request-answering system that the embodiment of the present invention four provides；
Fig. 6 is a kind of structural schematic diagram of the construction system for robot knowledge base that the embodiment of the present invention five provides；
Fig. 7 is a kind of structural schematic diagram for terminal device that the embodiment of the present invention six provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
In description, only some but not all contents related to the present invention are shown in the drawings.
The content of embodiment to facilitate the understanding of the present invention first introduces commonly used noun in automatic question answering:
1 knowledge point
Basic knowledge point most original and simplest form in knowledge base are exactly usually common FAQ, general form
It is that " ask-answer " is right.For example, " rate of CRBT " are exactly that clearly standard asks description for expression.Here " asking " should not be by narrowly
It is interpreted as " inquiring ", and should broadly understand one " input ", being somebody's turn to do " input " has corresponding " output ".For example, for being used for
For the semantics recognition of control system, the instruction of user, such as " opening radio " should also be understood to be one
" asking ", corresponding at this time " answering " can be the calling for executing the control program accordingly controlled.
User to machine when inputting, the most ideal situation is that asked using standard, then the intelligent semantic identifying system of machine
At once it will be appreciated that the meaning of user.However, user often not uses standard to ask, but some deformations for asking of standard
Form.For example, if being " changing a radio station " for the standard form of asking of the radio station switching of radio, then what user may use
Order is " switching one radio station ", and what machine was also required to can to identify user's expression is the same meaning.
For intelligent semantic identification, the extension that the standard that needs in knowledge base is asked is asked, which, which asks, asks table with standard
There is slight difference up to form, but expresses identical meaning.
Therefore, include multiple knowledge points in knowledge base, each knowledge point includes problem and answer, problem include standard ask and
Multiple extensions are asked.
2 parts of speech
Part of speech is divided according to the semanteme of word, and one group of relevant phrase is woven in together to form a tree
Part of speech library, any one n omicronn-leaf child node in this tree is referred to a part of speech (broad sense part of speech), wherein directly
It connects the first order part of speech comprising word and is known as narrow sense part of speech.The purpose of part of speech is defined primarily to participle, constructing semantic expression formula
And Semantic Similarity Measurement is carried out using the semantic information that it is carried.
2.1 the composition of part of speech
Part of speech (narrow sense part of speech) is summarized to one group of related term, and part of speech is made of part of speech name and one group of related term.Word
Class name is the word in this group of related term with label effect, the i.e. representative of part of speech.Word is contained at least one (i.e. in one part of speech
Part of speech name itself).Part of speech name, which generally requires, meets following rule: part of speech name answer it is simple and clear, it is easy-to-understand, can give top priority to what is the most important；
There are the words such as phonetic, English, network popular word, dialect and written word in one group of word, mandarin conduct can be commonly used in this case
Part of speech name；If in part of speech name should not with any symbol (as/,? Deng).
The semantic tagger of 2.2 parts of speech
Part of speech with a kind of word if only come using meaning is also just had a greatly reduced quality.Part of speech is preferably answered
With, it is necessary to the semantic information of its default is defined on it, and marks other semantic informations.There are these semantic informations,
Various operations can be just carried out in subsequent semantic analysis.The semanteme of mark has inherited characteristics, i.e. subclass inherits parent
Mark is semantic.
2.2.1 without mark part of speech
It is " similar " that definition is semantic when default is without any mark, it will be appreciated that is synonym, such part of speech is to subsequent
Semantic computation plays very big help, and what is largely used in semantic formula is all this kind of part of speech.Such as: " open-minded " this word
Class includes the words such as " open-minded ", " customization ", " unlatching ".
2.2.2 mass-word (#)
" # " can be used and be labeled as " dissmilarity ".At this moment the main function of part of speech is the expression for carrying out semantic formula,
Same type of word is normally behaved as, there is certain semantic dependency, but and non-synonymous, referred to as set part of speech.Such as:
" operating system " this part of speech includes the words such as " wince ", " linux ", " IOS ", " Android ", " palm ", " Symbian ".
2.2.3 primary word (* n)
" * " or " * n " can be used and be labeled as " important " (wherein n indicates different degree).This kind of word should be with respect to other words
It calculates similarity and improves weight in the process.General business class noun requires even more important similar to reinforce what is semantically marked
Weight in degree calculating.Such as: " CRBT " this part of speech includes the words such as " seven color bell sounds ", " CRBT ".
2.2.4 title (%n)
The part of speech of some professional words of " %n " mark can be used.This kind of word generally has the specific meaning, part of speech in field
It is easy to judge incorrectly when mark, it is therefore desirable to correct by manually marking.It can be in subsequent sentence pattern by accurate part of speech
It is played an important role in analysis and similarity calculation.Such as: " multimedia message collection folder " this part of speech, " the multimedia message collection for including
The words such as folder ", " multimedia message collection ".
2.2.5 verb (%v)
The part of speech that " %v " marks some verbs can be used.
2.2.6 phonetic error correction term (@)
The part of speech that some professional words of "@" mark can be used can carry out phonetic error correction.If carried out in whole dictionaries
Phonetic error correction, often due to the accuracy of homonym reason error correction is lower.Though carrying out part error correction by the word to specified portions
Right error correction range becomes smaller, but error correction accuracy rate can be promoted greatly.The mark of " error correction " is generally be directed in field
Professional term, length is often longer than everyday words, and typically no homonym, error correction effect are obvious.Due in professional term
Some other words are usually contained, therefore the principle of automatic error-correcting is that influence after error correction to word segmentation result number cannot increase.
Such as " happy to read meeting member packet " this word, user is often due to the input of input method reason mistake passes through spelling for " reading meeting member packet "
Sound error correction can be to avoid can not understand to customer problem caused by this kind of input error as user.
It should be noted that marked thereon is merely illustrative of, and such as: label symbol and/or corresponding relationship therein are all
It can change, not influence protection scope of the present invention.
3 semantic formulas
Semantic formula is mainly made of word, part of speech and their "or" relationship, and core depends on " part of speech ", part of speech
Simple to understand to be one group of word for having general character, these words can be similar or dissimilar semantically, these words can also be with
It is noted as important or inessential.Semantic formula and user's question sentence relationship and traditional template matching have very big difference,
In conventional template matching, template and user's question sentence are only matched and not matched relationship, and semantic formula and user's question sentence
Between relationship be indicated by the value (similarity) of quantization, while this quantization value between similar question sentence and user's question sentence
Similarity can mutually compare.Since semantic formula will participate in together similarity calculation with similar question sentence, so mould
The definition of plate grammer is unsuitable complicated, but has enough ability expression semantic again, semantic formula illustrated below it is specific
The expression of composition and symbol.
Symbol in 3.1 semantic formulas
3.1.1 the expression ([]) of part of speech
To distinguish the word in expression formula with part of speech, it is specified that part of speech must be present in square brackets " [] ", occur in square brackets
Part of speech be generally " narrow sense part of speech ", but " broad sense part of speech " can also be supported by configuring system parameter.Here is some letters
The example of single expression formula: [Fetion] [how] [open-minded], [introduction] [multimedia message] [business], [login] [method] of [Fetion], [come
Electricity is reminded] [how] [charge].
Or the expression (|) of relationship 3.1.2
Part of speech in square brackets can occur repeatedly by "or" relationship, and the part of speech of these "or" relationships can calculate phase
It is individually calculated in a manner of " expansion " when seemingly spending.Semantic formula is mainly launched into according to the meaning of "or" by " expansion "
The process of multiple structures.Such as: [CRBT] [open-minded] [method | step] it is deployable at the " [step of [CRBT] [open-minded]
Suddenly] " and " [method] of [CRBT] [open-minded] " two simple semantic formulas.The example of this kind of semantic formula is as follows: [color
Bell] [open-minded] [method | step], [how] [inquiry | know] [PUK code], [quit the subscription of | revocation | close | deactivate] [IP |
17951] [the preferential packet of national distance], [call reminding] [function is taken | Monthly Fee | information charge | communication expense].
3.1.3 it is non-essential indicate (?)
Part of speech in square brackets can be added at the end of "? " expression, which may occur in which, to be occurred, i.e., non-essential pass
System, the part of speech of this inessential relationship similarly can individually be calculated in a manner of " expansion " when calculating similarity." exhibition
Open " it is mainly that will be launched into include and do not include this containing non-essential part of speech (or " or combination " of part of speech) in semantic formula
The process of the simple semantic formula of two of a part of speech.
Such as: [introduction] [mobile video] [military column] [content] [what? ] deployable at " [introduction] [mobile video]
[military column] [content] " and " [introduction] [mobile video] [military column] [content] [what] " two simple semantic meaning representations
Formula.
The problem of the problems in present invention expression formula and answer expression formula are exactly indicated with expression formula form and answer.
Embodiment one
Fig. 1 is a kind of flow chart of the building method for robot knowledge base that the embodiment of the present invention one provides, the present embodiment
It is applicable to the case where robot knowledge base is built to static two dimensional table data, this method can be by construction robot knowledge base
Device executes, which can be realized by software and/or hardware, can generally be integrated in terminal device, this method is specific
Include the following steps:
Step 110, static two dimensional table data are obtained, the static two dimensional table data include topic, gauge outfit and table body.
Wherein, static two dimensional table data are by two-dimentional table structure come the data of logical expression and realization, it then follows data format
With length specification, storage and management are mainly carried out by relevant database comprising the topic (topic " financing of such as table 1
Table "), gauge outfit (such as the first row of table 1) and table body (data in addition to the first row).
Static two dimensional table is directly imported and is shown by the function of load bivariate table by the present embodiment.
Step 120, determine the corresponding more than two attribute informations of the table body from gauge outfit, each attribute information with
The gauge outfit content of one or more columns per page is corresponding.
The attribute of every column data is illustrated in the gauge outfit of static two dimensional table data, generally, primary key column has an attribute,
He may be the same or different the attribute of column data, according to the attribute of every column data, can determine table body corresponding at least two
A attribute information, and each attribute information is corresponding with the gauge outfit content of one or more columns per page.
Wherein, primary key column is the column that can be used to unique identification data line in a table, as the first row in table 1 can
To identify the specific finance product that every row indicates, which is primary key column.
When it is implemented, determining the attribute of every column data according to gauge outfit content, primary key column therein (the of such as table 1 is selected
The first row of one column and table 2), an attribute information is determined according to primary key column, determines one further according to the gauge outfit content of other column
Or multiple attribute informations (the gauge outfit content that other in such as table 1 arrange is specific number of days, can be summarized as an attribute information, and
The gauge outfit content that other in table 2 arrange cannot be summarized as an attribute information, can regard each column as an attribute information), also need
To judge that the static two dimensional table data of load could be configured to a knowledge point according to determining attribute information, e.g., have two
A knowledge point can only be constructed when a attribute information, and multiple knowledge points can be constructed when with more than two attribute information.
For example, table 1 can be configured such that a knowledge point, table 2 can be configured such that multiple knowledge points, comprising: the price of xx book, the author of xx book and
The brief introduction etc. of xx book, wherein xx is referred to as all books.
2 books table of table
Optionally, determine that the corresponding more than two attribute informations of the table body include: from gauge outfit
When the corresponding attribute of gauge outfit content of multi-column data is identical, the gauge outfit content of the multi-column data is summarized as one
A attribute information；
When only gauge outfit content one attribute of correspondence of a column data, directly using the gauge outfit content of the column data as one
The attribute information.
The gauge outfit content for comparing multi-column data judges whether the corresponding attribute of gauge outfit content of multi-column data is identical, in phase
Meanwhile the gauge outfit content of multi-column data is summarized as an attribute information (other gauge outfits respectively arranged in such as table 1 in addition to first row
Content is specific number of days, can be summarized as an attribute information, which can be the financing time limit)；When only one column
When the gauge outfit content of data corresponds to an attribute, directly using the gauge outfit content of the column data as an attribute information, for example, main
Key column can have an individual attribute, and the gauge outfit content of the column can be used as an attribute information, and (first row in such as table 1 is corresponding
One attribute information, the attribute information can be finance product), it certainly, may also be only in the gauge outfit of a column data in other column
Holding a corresponding attribute, (attribute of the secondary series in such as table 2 is author, and tertial attribute is price, cannot be summarized as same
Attribute information, i.e. each column respectively correspond an attribute information).By concluding attribute information, to automatically generate automatic question answering knowledge
Point provides foundation, so as to automatically generate less automatic question answering knowledge point, improves the efficiency for generating knowledge point, and save
Memory space.
Step 130, one or more automatic question answerings knowledge point, each automatic question answering knowledge are generated according to the attribute information
Point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic.
When only there are two attribute information, an automatic question answering knowledge point is generated；When there is M attribute information (M be greater than etc.
In 3), then an automatic question answering knowledge point can be generated according to any two attribute information therein, it can also be according to therein
Any three attribute informations generate an automatic question answering knowledge point, can also generate one according to wherein any N number of attribute information
Automatic question answering knowledge point (N is greater than 3 and is less than or equal to M).That is, can be generated one according to more than two attribute informations and ask automatically
Knowledge point is answered, the multiple automatic question answering knowledge points of combination producing to attribute information can also be passed through.
The answer expression formula in automatic question answering knowledge point in the embodiment of the present invention includes the topic of static two dimensional table data,
For and when user's interaction search corresponding with topic static two dimensional table data and obtain answer.
Step 140, the static two dimensional table data, the automatic question answering knowledge point and the attribute information are stored to knowing
Know in library.
It further include common in knowledge base other than above-mentioned static two dimensional table data, automatic question answering knowledge point and attribute information
Knowledge point, general knowledge point include problem expression formula and answer expression formula, and the topic is not present in answer expression formula.
The technical solution of the present embodiment, by obtaining static two dimensional table data, from the gauge outfit of static two dimensional table data really
Determine the corresponding more than two attribute informations of table body, one or more automatic question answerings knowledge point is generated according to the attribute information, often
A automatic question answering knowledge point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic, will be static
Bivariate table data, automatic question answering knowledge point and attribute information storage are realized according to static two dimensional table data certainly into knowledge base
It is dynamic to generate knowledge point, and corresponding knowledge base is established, reduce operator's workload, and reduce the possibility artificially made a mistake,
Improve the accuracy and efficiency of the knowledge point of generation.
Based on the above technical solution, also optional to include:
Establish the inclusion relation of the attribute information with perhaps gauge outfit content interior in corresponding table body；
By inclusion relation storage into knowledge base.
Using the gauge outfit content of a column data as when an attribute information, establish the attribute information in corresponding table body
The inclusion relation (such as establishing the inclusion relation of content in attribute information and the table body of primary key column) of appearance；By the gauge outfit of multi-column data
When content is summarized as an attribute information, which can be the general character of the gauge outfit content of multi-column data, to establish category
The inclusion relation of property information and corresponding gauge outfit content.By the attribute information and the inclusion relation of content in corresponding table body with
And attribute information is stored with the inclusion relation of corresponding gauge outfit content into knowledge base, convenient for carrying out automatic question answering friendship with user
Corresponding knowledge point is quickly searched in mutually.
Based on the above technical solution, optionally, further includes:
Part of speech is established for the word in the gauge outfit or/and the table body, part of speech name of the word as corresponding part of speech,
The part of speech includes the synonym of the word and the word；
Establish the attribute information and the inclusion relation of content in corresponding table body include: establish the attribute information with it is right
The inclusion relation of part of speech name in the table body or gauge outfit answered；
By inclusion relation storage into knowledge base further include: by part of speech storage into knowledge base.
It can also be the word in gauge outfit or/and table body to more quickly be matched to knowledge point when interacting with user
Language establishes part of speech, using the word as the part of speech name of corresponding part of speech, and determines the synonym of the word, the word is connected
It is placed under the part of speech together with corresponding synonym.In the inclusion relation for establishing attribute information with content in corresponding table body
When, the inclusion relation of attribute information with part of speech name in corresponding table body or gauge outfit is established, the inclusion relation of record is more simple
It is single.Part of speech when storing inclusion relation together with foundation stores in knowledge base together.
It is exemplified by Table 1, firstly, being believed according to gauge outfit content selection primary key column using the column topic of primary key column as an attribute
Breath, attribute information is finance product, and using the value under the column as the part of speech name under attribute information, and add corresponding synonym
Enrich its semantic information, generate part of speech, namely part of speech is established to the word in the first row in table body, and establish attribute information with
The inclusion relation of part of speech name in corresponding table body, is illustrated in figure 2 the construction of robot knowledge base provided in an embodiment of the present invention
The schematic diagram that part of speech is established to the primary key column of table 1 in method.By judgement, the gauge outfit content of the column in addition to primary key column is determined
It is specific number of days, corresponding attribute is identical, conclusion name is carried out to column topics of other column, as an attribute information,
Attribute information is the financing time limit, using the column topic of other column as the part of speech name under the attribute information, and is added corresponding synonymous
Word generates part of speech identical with the number of columns of other column, namely establishes part of speech to the word in the gauge outfit of other column, and establish category
The inclusion relation of property information and part of speech name in corresponding gauge outfit, is illustrated in figure 3 robot knowledge provided in an embodiment of the present invention
The column topic of other column to table 1 in the building method in library establishes the schematic diagram of part of speech.By the part of speech and inclusion relation of foundation
It stores in knowledge base together.
By taking table 2 as an example, firstly, according to gauge outfit content selection primary key column, using first row as primary key column, by the column of primary key column
Topic is as an attribute information, and using the value under the column as the part of speech name under attribute information, and adds corresponding synonym
Enrich its semantic information, generate part of speech, namely part of speech is established to the word in the first row in table body, and establish attribute information with
The inclusion relation of part of speech name in corresponding table body.By judgement, determine that other corresponding attributes of column in addition to primary key column are different,
That is the corresponding attribute of each column carries out conclusion name to the column topic of each column in other column, as an attribute information, such as
The attribute information of " author " this column-generation can be " author's title " or " author ", and corresponding column topic is as the attribute information
Under part of speech name, while also regarding word in corresponding table body as the part of speech name under the attribute information, establish multiple parts of speech, namely
Part of speech is established to the word in each column in gauge outfit and table body, and establishes attribute information and part of speech name in corresponding gauge outfit and table body
Inclusion relation.Four part of speech names as shown in table 3: price, brief introduction, Zuo Zhehe can be additionally generated according to the column topic of table 2
Book.
3 part of speech of table
Based on the above technical solution, optionally, one or more automatic question answerings are generated according to the attribute information
Knowledge point includes:
An initial knowledge point is automatically generated according at least two attribute informations；
Each initial knowledge point is adjusted, the automatic question answering knowledge point is obtained.
At least two attribute informations are merged, automatically generate an initial knowledge point, then to each initial knowledge point
It is adjusted, obtains automatic question answering knowledge point.For example, two attribute informations, i.e. finance product and financing can be summarized as in table 1
Time limit automatically generates an initial knowledge point, as follows:
Initial problem expression formula: [finance product] [financing time limit] is generated for the initial knowledge point
Generate initial answer expression formula for the initial knowledge point: [the financing time limit] of [finance product] is { ds. financing
Table }
For the above-mentioned initial knowledge point automatically generated, operator it can be modified adjustment, asked automatically
Knowledge point is answered, as follows:
Modified problem expression formula: be [finance product] [financing time limit] [Annual Percentage Rate] [how many? ]
Modified answer expression formula: [financing time limit] [Annual Percentage Rate] of [finance product] is { ds. financing table }
For example, multiple attribute informations can be summarized as in table 2, it can be according to one in the attribute information of primary key column and other column
The attribute information of column generates initial knowledge point, and the initial knowledge point of generation is following (answer expression formula is omitted): knowledge point 1, life
At the problem of expression formula: [books title] [author]
Knowledge point 2, generate the problem of expression formula: [books title] [price]
Knowledge point 3, generate the problem of expression formula: [books title] [brief introduction]
For above-mentioned knowledge point, operator simply modifies it, as knowledge point 3 it is not necessary to modify, knowledge point 1 and is known
Knowing point 2 can be amended as follows:
Knowledge point 1, modified problem expression formula: [books title] [author] [whose? ]
Knowledge point 2, modified problem expression formula: be [books title] [price] [how many? ]
Initial knowledge point can also be generated according to the attribute information of a column in other column and the attribute information of primary key column, it can be with
For (answer expression formula is omitted):
Sample: road it is distant went out which book?
The problem of generation expression formula: [author's title] [book]
Operator can be amended as follows:
Modified problem expression formula: [author's title] [write | publish] [book]
As it can be seen that initial knowledge point, operator can be automatically generated according to corresponding attribute information in static two dimensional table data
Member only need to simply adjust the automatic question answering knowledge point for the standard of can be obtained by, and greatly reduce the workload of operator.
When using the present embodiment technical solution processing table 1, an automatic question answering knowledge point only need to be generated, thus not
It needs to generate 40 knowledge points again, memory space is greatly saved, and reduce the workload of operator.
Embodiment two
Second embodiment of the present invention provides robot knowledge base, which passes through the machine that embodiment one provides
The building method construction of people's knowledge base, that is, robot knowledge base includes:
One or more static two dimensional table data；
More than two attribute informations corresponding with the static two dimensional table data；
Multiple automatic question answering knowledge points, each automatic question answering knowledge point include problem expression formula and answer expression formula, described
Answer expression formula includes a topic.
The knowledge base is also optional to include:
The inclusion relation of the attribute information and perhaps gauge outfit content interior in corresponding table body.
Optionally, when establishing the knowledge base, part of speech is established for the word in the gauge outfit or/and the table body, institute
Part of speech name of the predicate language as corresponding part of speech, the part of speech includes the synonym of the word and the word；
The inclusion relation is the inclusion relation of the attribute information and part of speech name in corresponding table body or gauge outfit；
The knowledge base further include: the part of speech.
Robot knowledge base also stores common automatic question answering knowledge point, and common automatic question answering knowledge point includes problem expression formula
With answer expression formula, the topic of static two dimensional table data not will include in answer expression formula.
The technical solution of the present embodiment, directly storage static two dimensional table data and corresponding automatic question answering knowledge point, automatically
Question and answer knowledge point be by summarize static two dimensional table data shared attribute obtain, cover multiple problems and answer, with
When user's interaction, corresponding answer can be returned as the case may be.
Embodiment three
Fig. 4 is a kind of flow chart for automatic question-answering method that the embodiment of the present invention three provides, automatic described in the present embodiment
The basis of answering method is that robot knowledge base, the present embodiment described in any embodiment of the present invention are applicable to static two dimensional
The case where data of table data storage carry out automatic question answering, this method can be executed by automatic call answering arrangement, which can be with
It is realized, can generally be integrated in the server, this method specifically comprises the following steps: by software and/or hardware
Step 410, in the solicited message for receiving user, asking in knowledge base automatically is matched according to the solicited message
Answer knowledge point.
The solicited message can be voice messaging, be also possible to text information.When the solicited message is voice messaging
When, voice messaging first can be converted into corresponding text information.
It, will be in the solicited message and knowledge base of user according to Semantic Similarity Measurement in the solicited message for receiving user
Automatic question answering knowledge point matched, choose that similarity is greater than preset threshold and similarity highest is one or more asking automatically
Knowledge point is answered, for the automatic question answering knowledge point being matched to.It is worth noting that automatic question answering knowledge point here can be with it is quiet
The corresponding automatic question answering knowledge point of state bivariate table data includes static two dimensional in the answer expression formula of automatic question answering knowledge point at this time
The topic of table data, automatic question answering knowledge point can also be general knowledge point, at this time the answer expression formula of automatic question answering knowledge point
In do not include the topic.
When specific progress similarity calculation, word segmentation processing first can be carried out to solicited message, obtain word segmentation result, will segment
As a result similarity is calculated based on the part of speech of foundation.
When the automatic question answering knowledge point being matched to is general knowledge point, the answer for the knowledge point being matched to is directly acquired simultaneously
The answer is returned into user.It should be noted that both directly the text can be believed when answer is text information
Breath returns to user, returns again to after text information can also be converted to voice messaging to user, does not influence of the invention
Protection scope.
And when the automatic question answering knowledge point being matched to is automatic question answering knowledge point corresponding with static two dimensional table data, then
Need to continue to execute following step.
Step 420, according to the corresponding topic in the automatic question answering knowledge point being matched to, corresponding static two dimensional table is obtained
Data.
Step 420 can specifically include the following contents:
According to the topic, corresponding static two dimensional table is searched；
Obtain the static two dimensional table found.
Step 430, corresponding answer is searched in the static two dimensional table data according to the solicited message, according to lookup
The answer expression formula of the answer and determination arrived generates final result, and the final result is returned to the user.
The part of speech name being matched to according to the solicited message in similarity calculation searches the institute in static two dimensional table data
Predicate class name obtains corresponding data, as answer, and according to the answer found and the automatic question answering knowledge point being matched to
In answer expression formula generate final result, final result is returned into the user.
For example, in conjunction with table 1, the solicited message of user is " 90 days Annual Percentage Rates of series of turning a crude essay into a literary gem are how many? ", pass through
Similarity calculation is matched to that problem expression formula is " [finance product] [financing time limit] [Annual Percentage Rate] [how many ?] ", answer expression formula
For the automatic question answering knowledge point of " [financing time limit] [Annual Percentage Rate] of [finance product] is { ds. financing table } ", answer expression formula
In include static two dimensional table data topic " financing table ", according to the topic of static two dimensional table data, directly acquire the static state two
Dimension table is searched corresponding with the 90 days specific data of series of turning a crude essay into a literary gem in the static two dimensional table, is searched, obtain stone at
Golden serial specific data corresponding with 90 days are 4.46, substitute into the answer expression formula, and obtaining final result is " to turn a crude essay into a literary gem
90 days Annual Percentage Rates of series are 4.46 ", which is returned to user or the final result is converted to corresponding language
Message ceases and returns to user.
The technical solution of the present embodiment, by the solicited message for receiving user, matching asking in knowledge base automatically
Knowledge point is answered, it, can if the answer expression formula for the automatic question answering knowledge point being matched to does not include the topic of static two dimensional table data
Directly to return to corresponding answer, if the answer expression formula for the automatic question answering knowledge point being matched to includes the topic, obtain
Corresponding static two dimensional table data are taken, corresponding answer are searched in static two dimensional table data, and generate according to answer expression formula
Final result returns to the user, thus according to the static two dimensional table data and corresponding automatic question answering that store in knowledge base
Knowledge point finds final result corresponding with the solicited message of user, thus corresponding automatic in storage static two dimensional table data
When question and answer knowledge point, automatic question answering knowledge point corresponding with the attribute information of static two dimensional table data need to be only stored, it is not necessary to as existing
There is technology to store corresponding many knowledge points like that, the memory space of knowledge base is greatly saved, and according to user's
When solicited message matches the automatic question answering knowledge point in knowledge base, since automatic question answering knowledge point to be matched is less, improve
Matching speed, to improve the speed for obtaining answer.
Example IV
Fig. 5 is a kind of structural schematic diagram for automatically request-answering system that the embodiment of the present invention four provides, described in the present embodiment
The basis of automatic call answering arrangement is that robot knowledge base, the present embodiment described in any embodiment of the present invention are applicable to static state
The case where data of bivariate table data storage carry out automatic question answering, which can be realized by software and/or hardware, generally may be used
It integrates in the server, which includes: request matching module 510, data acquisition module 520, answer generation module 530 and answer
Case return module 540.
Wherein, matching module 510 is requested, in the solicited message for receiving user, according to the solicited message
With the automatic question answering knowledge point in knowledge base；
Data acquisition module 520, for obtaining and corresponding to according to the corresponding topic in automatic question answering knowledge point being matched to
Static two dimensional table data；
Answer generation module 530, it is corresponding for being searched in the static two dimensional table data according to the solicited message
Answer generates final result according to the answer expression formula of the answer and determination found；
Answer return module 540, for the final result to be returned to the user.
Optionally, the data acquisition module includes:
Searching unit, for searching corresponding static two dimensional table according to the topic；
Data capture unit, for obtaining the static two dimensional table found.
Automatic question-answering method provided by any embodiment of the invention can be performed in above-mentioned automatically request-answering system, has the side of execution
The corresponding functional module of method and beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to the present invention is implemented
The automatic question-answering method that example three provides.
Embodiment five
Fig. 6 is a kind of structural schematic diagram of the construction system for robot knowledge base that the embodiment of the present invention five provides, this reality
It applies example and is applicable to the case where robot knowledge base is built to static two dimensional table data, which can be by software and/or hardware
It realizes, can generally be integrated in terminal device, which includes: data acquisition module 610, attribute determination module 620, knowledge
Point generation module 630 and memory module 640.
Wherein, data acquisition module 610, for obtaining static two dimensional table data, the static two dimensional table data include topic
Mesh, gauge outfit and table body；
Attribute determination module 620, for determining the corresponding more than two attribute informations of the table body, Mei Gesuo from gauge outfit
It is corresponding with the gauge outfit content of one or more columns per page to state attribute information；
Knowledge point generation module 630, for generating one or more automatic question answerings knowledge point according to the attribute information, often
A automatic question answering knowledge point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic；
Memory module 640 is used for the static two dimensional table data, the automatic question answering knowledge point and the attribute information
It stores in knowledge base.
Optionally, the memory module includes:
Static two dimensional table storage unit, for storing the static two dimensional table.
Optionally, the attribute determination module includes:
Whether judging unit, the corresponding attribute of gauge outfit content for judging multi-column data are identical；
Unit is concluded, for concluding the gauge outfit content of the identical multi-column data of attribute according to the judging result of judging unit
For an attribute information；
Output unit, for will conclude attribute information that unit obtains and ought an only column data gauge outfit content corresponding one
The gauge outfit content of column data when a attribute is exported.
Optionally, the system also includes:
Inclusion relation establishes module, for establishing the inclusion relation of the attribute information with content in corresponding table body；
Inclusion relation memory module, for storing the inclusion relation into knowledge base.
Optionally, the system also includes:
Part of speech establishes module, for establishing part of speech, the word conduct for the word in the gauge outfit or/and the table body
The part of speech name of corresponding part of speech, the part of speech includes the synonym of the word and the word；
The inclusion relation is established module and is specifically used for: establishing the attribute information and part of speech in corresponding table body or gauge outfit
The inclusion relation of name；
The inclusion relation memory module is also used to: by part of speech storage into knowledge base.
Optionally, the knowledge point generation module includes:
Initial knowledge generation unit, for automatically generating an initial knowledge point according at least two attribute informations；
Knowledge point adjustment unit obtains the automatic question answering knowledge for being adjusted to each initial knowledge point
Point.
Robot knowledge base provided by any embodiment of the invention can be performed in the construction system of above-mentioned robot knowledge base
Building method, have the corresponding functional module of execution method and beneficial effect.The not technology of detailed description in the present embodiment
Details, reference can be made to the building method for the robot knowledge base that the embodiment of the present invention one provides.
Embodiment six
Fig. 7 is a kind of structural schematic diagram for terminal device that the embodiment of the present invention six provides, as shown in fig. 7, the terminal is set
Standby includes processor 710, storage device 720, input unit 730 and output device 740；The number of processor 710 in terminal device
It measures and can be one or more, in Fig. 7 by taking a processor 710 as an example；Processor 710, storage device in terminal device
720, input unit 730 can be connected with output device 740 by bus or other modes, to be connected as by bus in Fig. 7
Example.
Storage device 720 is used as a kind of computer readable storage medium, and it is executable to can be used for storing software program, computer
Program and module, the building method and/or automatic question-answering method such as the robot knowledge base in the embodiment of the present invention are corresponding
Program instruction/module (for example, data acquisition module 610, attribute determination module 620 in the construction device of robot knowledge base,
Knowledge point generation module 630 and memory module 640).The software journey that processor 710 is stored in storage device 720 by operation
Sequence, instruction and module realize above-mentioned robot thereby executing the various function application and data processing of terminal device
The building method and/or automatic question-answering method of knowledge base.
Storage device 720 can mainly include storing program area and storage data area, wherein storing program area can store operation
Application program needed for system, at least one function；Storage data area, which can be stored, uses created data etc. according to terminal.
In addition, storage device 720 may include high-speed random access memory, it can also include nonvolatile memory, for example, at least
One disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, storage device 720
It can further comprise the memory remotely located relative to processor 710, these remote memories can be by being connected to the network extremely
Terminal device.The example of above-mentioned network includes but is not limited to internet, intranet, local area network, mobile radio communication and its group
It closes.
Input unit 730 can be used for receiving the number or character information of input, and generates and set with the user of terminal device
It sets and the related key signals of function control inputs.Output device 740 may include that display screen etc. shows equipment.
Embodiment seven
The embodiment of the present invention seven also provides a kind of storage medium comprising computer executable instructions, and the computer can be held
Row instruction by computer processor when being executed for executing building method and/or the automatic question answering side of a kind of robot knowledge base
Method.
The building method of the robot knowledge base specifically can be with reference implementation example one, and details are not described herein.
The automatic question-answering method specifically can be with reference implementation example three, and details are not described herein.
Certainly, a kind of storage medium comprising computer executable instructions, computer provided by the embodiment of the present invention
The method operation that executable instruction is not limited to the described above, can also be performed robot provided by any embodiment of the invention and knows
The building method for knowing library and/or the relevant operation in automatic question-answering method.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but in many cases, the former is more
Good embodiment.Based on this understanding, technical solution of the present invention substantially in other words contributes to the prior art
Part can be embodied in the form of software products, which can store in computer readable storage medium
In, floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (Random such as computer
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes method described in each embodiment of the present invention.
It is worth noting that, included each unit and module are only patrolled according to function in the embodiment of above-mentioned apparatus
It volume is divided, but is not limited to the above division, as long as corresponding functions can be realized；In addition, each function list
The specific name of member is also only for convenience of distinguishing each other, the protection scope being not intended to restrict the invention.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of building method of robot knowledge base characterized by comprising
Static two dimensional table data are obtained, the static two dimensional table data include topic, gauge outfit and table body, and the gauge outfit is first
Row, the table body are other rows except the first row；
The corresponding more than two attribute informations of the table body are determined from gauge outfit, when the corresponding attribute of gauge outfit content of multi-column data
When identical, the gauge outfit content of the multi-column data is summarized as the attribute information, when the gauge outfit content of an only column data
When a corresponding attribute, directly using the gauge outfit content of the column data as an attribute information；
One or more automatic question answerings knowledge point is generated according to the attribute information, each automatic question answering knowledge point includes problem table
Up to formula and answer expression formula, the answer expression formula includes the topic；
By the storage of the static two dimensional table data, the automatic question answering knowledge point and the attribute information into knowledge base.
2. the method as described in claim 1, which is characterized in that further include:
Establish the inclusion relation of the attribute information with perhaps gauge outfit content interior in corresponding table body；
By inclusion relation storage into knowledge base.
3. according to the method described in claim 2, it is characterized by further comprising:
Part of speech is established for the word in the gauge outfit or/and the table body, part of speech name of the word as corresponding part of speech is described
Part of speech includes the synonym of the word and the word；
The inclusion relation for establishing content in the attribute information and corresponding table body include: establish the attribute information with it is corresponding
The inclusion relation of part of speech name in table body or gauge outfit；
By inclusion relation storage into knowledge base further include: by part of speech storage into knowledge base.
4. being asked automatically the method according to claim 1, wherein generating one or more according to the attribute information
Answering knowledge point includes:
An initial knowledge point is automatically generated according at least two attribute informations；
Each initial knowledge point is adjusted, the automatic question answering knowledge point is obtained.
5. a kind of robot knowledge that the building method by the robot knowledge base as described in claim 1-4 is any is built
Library.
6. a kind of automatic question-answering method based on the robot knowledge base described in claim 5 characterized by comprising
In the solicited message for receiving user, the automatic question answering knowledge point in knowledge base is matched according to the solicited message；
According to the corresponding topic in the automatic question answering knowledge point being matched to, corresponding static two dimensional table is searched, and obtain lookup
The static two dimensional table arrived；
Corresponding answer is searched in the static two dimensional table according to the solicited message, according to the answer and determination found
Answer expression formula generate final result, the final result is returned into the user.
7. a kind of automatically request-answering system based on the robot knowledge base described in claim 5 characterized by comprising
Matching module is requested, for being matched in knowledge base according to the solicited message in the solicited message for receiving user
Question and answer knowledge point；
Data acquisition module, for searching corresponding static two dimensional according to the corresponding topic in question and answer knowledge point being matched to
Table, and obtain the static two dimensional table found；
Answer generation module searches corresponding answer in the static two dimensional table according to the solicited message, according to finding
Answer and determination answer expression formula generate final result；
Answer return module, for the final result to be returned to the user.
8. a kind of construction system of robot knowledge base characterized by comprising
Data acquisition module, for obtaining static two dimensional table data, the static two dimensional table data include topic, gauge outfit and table
Body, the gauge outfit are the first row, and the table body is other rows except the first row；
Attribute determination module determines the corresponding more than two attribute informations of the table body, when the gauge outfit of multi-column data from gauge outfit
When the corresponding attribute of content is identical, the gauge outfit content of the multi-column data is summarized as the attribute information, when only one arranges
When the gauge outfit content of data corresponds to an attribute, directly using the gauge outfit content of the column data as an attribute information；
Knowledge point generation module generates one or more automatic question answerings knowledge point, each automatic question answering according to the attribute information
Knowledge point includes problem expression formula and answer expression formula, and the answer expression formula includes the topic；
Memory module, for arriving the static two dimensional table data, the automatic question answering knowledge point and the attribute information storage
In knowledge base.
9. a kind of terminal device characterized by comprising
One or more processors；
Storage device, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
The now method as described in any in claim 1-4 or 6.
10. a kind of computer storage medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
Method of the Shi Shixian as described in any in claim 1-4 or 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711172532.3A CN109947908A (en) | 2017-11-22 | 2017-11-22 | The building method and construction system of robot knowledge base |
| US16/106,533 US20190147100A1 (en) | 2017-11-10 | 2018-08-21 | Method and apparatus for establishing intelligent question answering repository, and intelligent question answering method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711172532.3A CN109947908A (en) | 2017-11-22 | 2017-11-22 | The building method and construction system of robot knowledge base |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109947908A true CN109947908A (en) | 2019-06-28 |
Family
ID=67003923
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711172532.3A Pending CN109947908A (en) | 2017-11-10 | 2017-11-22 | The building method and construction system of robot knowledge base |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947908A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543550A (en) * | 2019-09-04 | 2019-12-06 | 上海智臻智能网络科技股份有限公司 | Method and device for automatically generating test questions |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080288313A1 (en) * | 2007-05-18 | 2008-11-20 | Morris & Gunter Associates Llc | Systems and methods for evaluating enterprise issues, structuring solutions, and monitoring progress |
| CN101470602A (en) * | 2007-12-25 | 2009-07-01 | 新奥特（北京）视频技术有限公司 | Software development method depending on relational database |
| CN101556606A (en) * | 2009-05-20 | 2009-10-14 | 同方知网(北京)技术有限公司 | Data mining method based on extraction of Web numerical value tables |
| US20120077178A1 (en) * | 2008-05-14 | 2012-03-29 | International Business Machines Corporation | System and method for domain adaptation in question answering |
| CN105528437A (en) * | 2015-12-17 | 2016-04-27 | 浙江大学 | Question-answering system construction method based on structured text knowledge extraction |
| CN105653716A (en) * | 2015-12-31 | 2016-06-08 | 北京奇艺世纪科技有限公司 | Database construction method and system based on classification-attribute-value |
| CN105677864A (en) * | 2016-01-08 | 2016-06-15 | 国网冀北电力有限公司 | Retrieval method and device for power grid dispatching structural data |
| CN105808614A (en) * | 2014-12-31 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Method and server for establishing specialty product knowledge bases, and method and server fro providing specialty product information |
| CN105912600A (en) * | 2016-04-05 | 2016-08-31 | 上海智臻智能网络科技股份有限公司 | Question-answer knowledge base and establishing method thereof, intelligent question-answering method and system |
| CN105912645A (en) * | 2016-04-08 | 2016-08-31 | 上海智臻智能网络科技股份有限公司 | Intelligent question and answer method and apparatus |
| CN106407377A (en) * | 2016-09-12 | 2017-02-15 | 北京百度网讯科技有限公司 | Search method and device based on artificial intelligence |
-
2017
- 2017-11-22 CN CN201711172532.3A patent/CN109947908A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080288313A1 (en) * | 2007-05-18 | 2008-11-20 | Morris & Gunter Associates Llc | Systems and methods for evaluating enterprise issues, structuring solutions, and monitoring progress |
| CN101470602A (en) * | 2007-12-25 | 2009-07-01 | 新奥特（北京）视频技术有限公司 | Software development method depending on relational database |
| US20120077178A1 (en) * | 2008-05-14 | 2012-03-29 | International Business Machines Corporation | System and method for domain adaptation in question answering |
| CN101556606A (en) * | 2009-05-20 | 2009-10-14 | 同方知网(北京)技术有限公司 | Data mining method based on extraction of Web numerical value tables |
| CN105808614A (en) * | 2014-12-31 | 2016-07-27 | 阿里巴巴集团控股有限公司 | Method and server for establishing specialty product knowledge bases, and method and server fro providing specialty product information |
| CN105528437A (en) * | 2015-12-17 | 2016-04-27 | 浙江大学 | Question-answering system construction method based on structured text knowledge extraction |
| CN105653716A (en) * | 2015-12-31 | 2016-06-08 | 北京奇艺世纪科技有限公司 | Database construction method and system based on classification-attribute-value |
| CN105677864A (en) * | 2016-01-08 | 2016-06-15 | 国网冀北电力有限公司 | Retrieval method and device for power grid dispatching structural data |
| CN105912600A (en) * | 2016-04-05 | 2016-08-31 | 上海智臻智能网络科技股份有限公司 | Question-answer knowledge base and establishing method thereof, intelligent question-answering method and system |
| CN105912645A (en) * | 2016-04-08 | 2016-08-31 | 上海智臻智能网络科技股份有限公司 | Intelligent question and answer method and apparatus |
| CN106407377A (en) * | 2016-09-12 | 2017-02-15 | 北京百度网讯科技有限公司 | Search method and device based on artificial intelligence |
Non-Patent Citations (1)
| Title |
|---|
| 张喜: "基于语义模板与知识库的智能导购机器人系统的研究与实现", 硕士学位论文, 15 December 2012 (2012-12-15), pages 10 - 80 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110543550A (en) * | 2019-09-04 | 2019-12-06 | 上海智臻智能网络科技股份有限公司 | Method and device for automatically generating test questions |
| CN110543550B (en) * | 2019-09-04 | 2022-05-06 | 上海智臻智能网络科技股份有限公司 | Method and device for automatically generating test questions |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107798123A (en) | Knowledge base and its foundation, modification, intelligent answer method, apparatus and equipment | |
| US9213758B2 (en) | Method and apparatus for responding to an inquiry | |
| WO2019169858A1 (en) | Searching engine technology based data analysis method and system | |
| TWI685760B (en) | Method for analyzing semantics of natural language | |
| CN106610990B (en) | Method and device for analyzing emotional tendency | |
| US20190147100A1 (en) | Method and apparatus for establishing intelligent question answering repository, and intelligent question answering method | |
| CN111813961B (en) | Data processing method and device based on artificial intelligence and electronic equipment | |
| CN109299289B (en) | Query graph construction method and device, electronic equipment and computer storage medium | |
| CN111553138B (en) | Auxiliary writing method and device for standardizing content structure document | |
| CN116737915B (en) | Semantic retrieval method, device, equipment and storage medium based on knowledge graph | |
| CN112115252A (en) | Intelligent auxiliary writing processing method and device, electronic equipment and storage medium | |
| CN110287286B (en) | Method and device for determining similarity of short texts and storage medium | |
| WO2022083132A1 (en) | Animation draft generation method and apparatus based on character paragraph | |
| CN110147358A (en) | The building method and construction system of automatic question answering knowledge base | |
| CN111324626B (en) | Search method and device based on voice recognition, computer equipment and storage medium | |
| CN113190692A (en) | Self-adaptive retrieval method, system and device for knowledge graph | |
| CN109947908A (en) | The building method and construction system of robot knowledge base | |
| EP4322066A1 (en) | Method and apparatus for generating training data | |
| KR102492008B1 (en) | Apparatus for managing minutes and method thereof | |
| CN110895924B (en) | Method and device for reading document content aloud, electronic equipment and readable storage medium | |
| CN110750989A (en) | Statement analysis method and device | |
| CN113593543B (en) | Intelligent loudspeaker voice service system, method, device and equipment | |
| CN111161706A (en) | Interaction method, device, equipment and system | |
| CN116595156B (en) | Knowledge management system and knowledge management method | |
| JP2018181250A (en) | Device, program, and method for generating interaction scenario corresponding to context |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |