CN110188163A - Data intelligence processing system based on natural language - Google Patents
Data intelligence processing system based on natural language Download PDFInfo
- Publication number
- CN110188163A CN110188163A CN201910296255.XA CN201910296255A CN110188163A CN 110188163 A CN110188163 A CN 110188163A CN 201910296255 A CN201910296255 A CN 201910296255A CN 110188163 A CN110188163 A CN 110188163A
- Authority
- CN
- China
- Prior art keywords
- subelement
- data
- information
- user
- module
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/335—Filtering based on additional data, e.g. user or group profiles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/338—Presentation of query results
Abstract
The present invention provides a kind of data intelligence processing system based on natural language, and client includes problem input unit, dynamic data visualization, information feedback unit and preference setting unit；Server end includes semantic understanding module, general data processing module and machine learning module；Server end and the information feedback unit of client, preference setting unit are attached；Client passes through problem input unit typing information, information is transmitted by information feedback unit, processing through semantic understanding module, determine the data type read, general data processing module is analyzed, analysis result is shown in dynamic data visualization by information feedback and is shown, while setting the preference query information of client by preference setting unit based on the analysis results.User can obtain key message whenever and wherever possible in processing system provided by the present invention, it is only necessary to which 3-4 weeks initial configuration and testing time do not need exploitation report, greatly reduce the O&M cost of IT.
Description
Technical field
The present invention relates to enterprise data analysis fields, and in particular to a kind of data intelligence processing system based on natural language
And application.
Background technique
Enterprise customer obtains business information and carries out data analysis at present, mainly passes through business intelligence (BI) system
Method, logic are the data warehouses based on operation system database or comprising multiple operation system data, use Report Form Design
Multiple data sheets of software development, then service-user is supplied to by a portal website.
This method meets enterprise customer to a certain extent and obtains the demand of information, but also have the following drawbacks:
1. flexibility is insufficient；At present BI system provide every report show pattern and content be it is fixed, can only meet has
The data requirements scene of limit is unable to satisfy the new analysis and data requirements that user temporarily encounters in use；
2. convenience is insufficient；Current operation user needs first to log in BI system when encountering a traffic issues, from a large amount of reports
In find and may include the report of oneself information needed, then adjust parameters to control data area, finally equal pending datas return
It returns, whole flow process needs multiple mouse clicking operation, and there are time wastes, also cause customer analysis thinking discontinuous；
3. ease for use is insufficient；BI system needs repeatedly selection and clicking operation in use at present, and there are a large amount of technical terms (such as
" drilling through ", " dimension ", " index ", " Zuo Guanlian " etc.).Service-user needs higher learning cost, this also leads to current BI system
The usage degree of system is not generally high；
4. mobility is insufficient；BI system is most of based on PC (PC) end use at present, has to the support of mobile terminal
Limit, and user needs to access inside and outside company and could access in intra-company or by Virtual Private Network (VPN), in mobile terminal
It is inconvenient for use.
UF network technology limited liability company has applied for entitled " data analysis system and a data analysis in 2013
The patent of invention (application number: 201310062951 .7) of method ", proposes the method for having report using voice operating.The party
The purpose of method be " making the simple operation for analyzing data, primary analysis is completed in once-through operation ", " support accurately filter condition,
Woth no need to click filter condition from known list ", be limited in that it only to the operation of existing report (such as open,
Filtering etc.), it cannot directly answer the traffic issues that service-user is expected at any time；The operation of support PC (PC) end, is not propped up
Hold the way access that App is moved by mobile device (mobile phone or tablet computer)；Only support the pattern match based on keyword,
Accurate matching is only supported not support the fuzzy matching based on similarity the inquiry of crucial master data and metadata；Do not mention as
What solves the problems, such as that user can not access company Intranet data and system in external network.
Summary of the invention
For above situation, in order to make up for the deficiencies of the prior art, the purpose of the present invention is to provide a kind of new data
Inquiry, analysis and visual processing system and method, can preferably meet the fast-changing analysis demand of service-user,
It can be suitable for the use in any place, time and different application equipment.
To achieve the goals above, the technical solution used in the present invention is: providing a kind of intelligence based on natural language
Data processing system；It is appreciated that user by PC (PC) or mobile device (mobile phone or tablet computer, including iOS and
Android android system), the problem of being proposed using natural-sounding (including voice input and text input), the language based on understanding
Justice finds out suitable data combination from back-end data platform, and immediately with suitable pattern (pie chart, line chart, thermodynamic chart etc.)
User (supporting the end PC and mobile terminal) is showed, the information casting and exception that speech form can be carried out when showing are interpreted.It should
Technology is understood that the contextual information that user puts question to, and supports more wheel sessions.
A kind of data intelligence processing system based on natural language, structure include two, client and server end
Point, wherein the client includes: problem input unit, dynamic data visualization, information feedback unit and preference setting
Unit；The server end includes semantic understanding module, general data processing module and machine learning module；Wherein, client
Be embodied in user uses interface, and by the information of reading needed for problem input unit typing, information feedback unit passes information
It is defeated by server end, by the processing of server end semantic understanding module, the data type read is determined, by general data processing
Module, which is analyzed and processed, obtains analysis as a result, analysis result is shown in the progress of dynamic data visualization by information feedback
It has been shown that, while the preference query information of client is set by preference setting unit based on the analysis results.
PC, mobile phone and tablet computer can be selected in the client；User can by PC, mobile App,
The modes such as wechat small routine, wechat enterprise use this system.
Described problem input unit is asked including voice input, Characters, preset and based on context the prompt of click
Inscribe 3 kinds of modes.
The server end includes semantic understanding module, general data processing module and machine learning module；Wherein, give
Understanding Module, general data processing module and machine learning module pass sequentially through circuit connection.Server end respectively with client
Information feedback unit, preference setting unit be attached, for data transmission with reading process result.
The semantic understanding module includes: Text similarity computing subelement, semantic generation subelement, fuzzy message processing
Subelement, missing information processing subelement, context processing subelement, information interpret subelement and problem clew subelement；With
It is handled in the analysis to input problem.
The Text similarity computing subelement is led for handling since user pronunciation is nonstandard and speech recognition is not perfect
The identification error of cause, basic algorithm are the editing distances of Chinese character and English word, and plus in input there are wrong word, plus
Word hiatus, each non-type typicalness of region mandarin of China (such as front and back nasal sound, it is flat stick up tongue, L/N, H/F regardless of) tolerance
Property processing；Semanteme generates subelement, for extracting crucial semantic letter from the text that input text or the voice of user change into
Breath, including zero, one or more dimensions, index, condition, sortord, maximum or how many smallest, specified chart exhibition
Show type etc., and based on the semantic generation parsed or updates contextual information；Fuzzy message handles subelement, for handling
Since the inaccurate caused ambiguity of user's input or data collision problem, such as input " growth rate " may be matched to " sale
Income year-on-year growth rate "or" net profit year-on-year growth rate ", contextual information determination is needed to need which is returned；Missing
Information processing subelement, for handle user input information it is imperfect when, how accurately to understand as far as possible user be intended to without
It is to report an error；Context handles subelement, for context to be created or updated according to each information input of user, and based on up and down
Text realizes more accurately semantic understanding and intention assessment, to support more wheel sessions；Information interprets subelement, for according to the pre- purchase of property
Logic of being engaged in carries out voice interpretation to the place that user in returned data pays close attention to, for example voice broadcast " reach by 2018 annual sales revenues
Rate is 89.3% ", " expense in Shandong and Hebei is higher " etc.；Problem clew subelement proposes that the context of problem is given based on user
Targetedly next how prompt problem, guidance client put question to out, take turns session to realize more.
The general data processing module includes: generic data model subelement, general data connection subelement, inquiry language
Sentence generates subelement and data acquisition and encapsulation subelement；Generic data model subelement is for abstracting enterprises respectively
Unite Various types of data complexity, directly by the data model of the general Business User-oriented of these data (including title, title
Meaning, type, calculating logic, unit etc.) it is indicated；General data connects subelement, for being connected to enterprises from outer net
Database, and based on general configuration realize to enterprise's types of databases (including Microsoft SQL Server, Oracle,
The types such as MySql, SAP HANA, Cloudera Impala, Hive, PostgreSQL, AWS Redshift) connection sum number
It is investigated that asking；Query statement generates subelement, for being generated dimension that subelement extracts from user's input based on semantic, being referred to
Mark, condition, sortord, most before/how many of most end, the information such as user right generate corresponding in conjunction with underlying database type
Query statement (such as the SQL statement to relevant database, the MDX statement to multi-dimensional database, and combine specific data
Optimization of the library type to query statement)；Data acquisition and encapsulation subelement, for generating what subelement obtained based on query statement
Query statement calls general data to connect subelement, obtains result data, and the figure that may be specified based on data characteristics and user
Table type is formatted encapsulation, is shown with passing to front end.
The machine learning module, comprising: user behavior preferential learning submodule and generic data model subelement.
The beneficial effects of the present invention are: user can be from any angle problem analysis, as long as bottom based on system of the invention
Layer has data that can obtain result it is not necessary to rely on the limited report of BI system and function；User can propose oneself at any time
Traffic issues, system provides result immediately；System of the invention provide based on mobile device, can with App that outer net uses and
The application of wechat end, user can obtain key message whenever and wherever possible；Based on system of the invention, it is only necessary to 3-4 weeks initially match
It sets and the testing time, does not need exploitation report, greatly reduce the O&M cost of IT.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of data intelligence processing system；
Fig. 2 is the flow diagram of data intelligence processing system；
Fig. 3 is the flow diagram of data intelligence processing system；
Fig. 4 is data intelligence processing system information typing schematic diagram；
Fig. 5 is the processing result schematic diagram of data intelligence processing system；
Fig. 6 is the processing result schematic diagram of data intelligence processing system。
Specific embodiment
As shown in Figure 1-3, the present invention provides a kind of data intelligence processing system based on natural language, structure includes
Client and server end two parts, wherein the client includes: problem input unit, dynamic data visualization, letter
Cease feedback unit and preference setting unit；The server end includes semantic understanding module, general data processing module and machine
Study module；Wherein, what client was embodied in user uses interface, by the information of reading needed for problem input unit typing,
Information feedback unit transmits information to server end, by the processing of server end semantic understanding module, determines the number read
According to type, it is analyzed and processed by general data processing module and obtains analysis as a result, analysis result is shown in by information feedback
Dynamic data visualization is shown, while based on the analysis results by preference setting unit, the preference for setting client is looked into
Ask information.
Technical solution of the present invention can identify and parse the data query and business that user is proposed by natural language
Problem analysis provides intuitive, instant answer by unified data query technique and dynamic data visualization technique, and
It can be continued the interaction more taken turns with user, solve asking in terms of traditional BI system generally existing flexibility and timeliness
Topic.By the machine learning techniques based on user behavior and feedback, technical solution of the present invention can constantly lift pins to every
The recognition accuracy of a user.
Embodiment
Business background: the marketing personnel of certain fast-moving consumer goods enterprise (by taking Coca-Cola Chinese companies as an example) wish neatly
The said firm's product is inquired in the sales volume of each channel, each department.Wherein, channel includes channel major class and sub- channel, and area includes big
Area (such as East) and province (such as Beijing, Hubei), sales volume include actual sales volume and same period sales volume and year-on-year growth rate.
System function: marketing personnel can by mobile device (mobile phone and tablet computer of iOS or android system),
By APP or wechat small routine, the problem of saying oneself with the mode of typewriting or voice, and answer is obtained immediately, concrete application
Steps are as follows.
User's input
As shown in figure 4, user passes through the problem of voice or typewriting input oneself, for example " list the sales volume row of each category in Beijing
Name ".
Semantic understanding
(1) the similarity mode algorithm for including based on this technology, understand user want the data seen dimension, measurement (index) and
Condition.In this example, dimension=category is identified, measure (index)=sales volume, condition is province=Beijing and implied terms: time
=when the year before last；
Note: this technology is not based on the accurate matching of keyword, but is matched based on the longest of short text similarity.For example, such as
Fruit user speech inputs " the smiling face's ranking for listing each category in Beijing ", then still will recognise that " index=pin according to " smiling face "
Amount "；
(2) Secondary Match is carried out to fertility text based on canonical logic, understands sortord and subtype of user's tendency etc.
Information.In this example, both of which is not indicated, therefore subtype is the histogram (column chart) of default；
(3) current sessions information is charged into context, time including this input, whether identify successfully, the dimension that identifies
Degree, index, condition, sortord, subtype, several (Top/Bottom N) that whether only see minimax etc..This example
In, context includes dimension=category, index=sales volume, condition={ province=Beijing;Time=2019 }, subtype=histogram
It is to support to take turns sessions Deng, the purpose of context more.
Data acquisition and encapsulation
(1) it is configured according to the database of system, generates query statement, and execute data base querying to obtain real data；
(2) according to query result and subtype, return information is encapsulated to front end.
Front end is shown
Front end is returned the result according to rear end carries out data visualization and information interpretation.As shown in figure 5, front end shows one in this example
A histogram, abscissa are categories, and ordinate is sales volume, and title shows that current data range is Beijing in 2019.Voice simultaneously
" good " is broadcasted (if recognition failures, to broadcast " sorry, recognition failures "；If what is returned is a monodrome, for example problem is
" this year, sales volume was how many ", then " sales volume in 2019 is xxx " can be broadcasted).
User continues epicycle session
User is seeing each category product this year after Pekinese's sales volume ranking, wants to see down the sales volume situation in Shandong, and think
See ranking.Then he can directly input " ranking for seeing Shandong ".
Semantic understanding
(1) the similarity mode algorithm for including based on this technology identifies that condition is province=Shandong from this user input,
And implied terms (time=when the year before last)；
Pay attention to not indicating dimension and index in this input.In fact, this technology meeting contextual information is judged.This
In example, completion missing information, including dimension=category, index=sales volume from the context；
(2) Secondary Match is carried out to fertility text based on canonical logic, according to " ranking " in input problem, understand user want with
The form of bar chart is seen from big to small, obtains subtype=bar chart；
(3) contextual information is updated.In this example, contextual information be updated to dimension=category, index=sales volume, condition=province=
Shandong;Time=2019 }, subtype=bar chart.
Data acquisition and encapsulation
(1) it is configured according to the database of system, generates query statement, and execute data base querying to obtain real data；
(2) according to query result and subtype, return information is encapsulated to front end.
Front end is shown
Front end is returned the result according to rear end carries out data visualization and information interpretation.As shown in fig. 6, front end shows one in this example
A bar chart, ordinate show the title of each category, and abscissa is corresponding sales volume, and title shows that current data range is 2019
The data in year Shandong.
The APP that user is developed by this technology, or by wechat small routine, into this product.Then by under this product
The problem of input frame of side inputs text or voice, proposes oneself.If user is inputted by voice, since pronunciation is not marked
Standard, dialect, it is flat stick up tongue regardless of, front and back nasal sound regardless of the problems such as, the text received from the background may be ambiguous.For example, language
Sound inputs " sales volume of each category in East ", and system may be identified as " smiling face of East category " (based on current natural language
Handle the level of NLP).For another example, user thinks " the highest provinces and cities of fruit juice sales volume ", but system may be identified as " fruit pin
Measure highest close examination ".The similarity algorithm and longest matching logic for including based on this technology, this technology are supported defeated to these ambiguities
The correction process entered utmostly understands the intention of user.
Except the content that user explicitly points out, this technology can also find out the hiding item in terms of time and permission
Part, such as to input " listing the sales volume of each category in East ", what user should want to see at this time is when the year before last (rather than last year or yesterday
It) data and user can only see the data in oneself responsible category or channel or region.That is, if current
User is only responsible for Zhejiang and two, Jiangsu province, but the business in not responsible Shanghai, then this technology meeting field adds " time=2019 "
" province belongs to { Zhejiang, Jiangsu } " the two conditions.
System provided by the present invention can handle the problems such as information fuzzy, loss of learning are prominent.Information fuzzy, that is, ambiguity, refers to
There are multiple possibility for the content that user says.If such as user input " seeing lower growth rate ", this " growth rate " may be
The indexs such as " sales volume year-on-year growth rate ", " sales volume sequential growth rate ", " selling charges growth rate ", " income from sales growth rate ".This
When need based on context, that is, before user the problem of asking, which data what conjecture user may want to see is；Information lacks
Mistake refers to that user's input information is imperfect, such as " comparing Shandong and Beijing ", can not determine that user thinks actually based on this input
The sales volume for comparing the two provinces still takes in or expense.At this time based on context this technology can be guessed first, if also
It is that can not determine, can allows user feedback, such as prompt user " you are intended to compare Shandong and Pekinese's sales volume or income from sales ",
It is further processed again based on user feedback.
System provided by the present invention supports more wheel sessions, that is to say, that user can help it as a robot
Reason, carries out gradual enquirement.
Innovation of the system provided by the present invention in terms of business is that it eliminates traditional BI system and needs to realize determination
The cost of demand, a large amount of reports of exploitation solves the pain spot that traditional BI system always lags behind business variation.This technology can be straight
It connects and service-user is converted into data using the problem of voice or text input；For service-user, thinks to ask, be asked
That is gained.
System provided by the present invention is in the innovation of technical aspect, is matched using the longest based on short text similarity
Algorithm carries out semantic understanding.Conventional method is carried out based on keyword and search engine technique, to the essence of user's input content
Exactness and integrity degree have higher requirement, not friendly enough to user.
Claims (5)
1. a kind of data intelligence processing system based on natural language, structure includes client and server end two parts,
It is characterized by: the client includes: problem input unit, dynamic data visualization, information feedback unit and preference
Setting unit；The server end includes semantic understanding module, general data processing module and machine learning module；Wherein, objective
Family end is embodied in the use interface of user, and by the information of reading needed for problem input unit typing, information feedback unit will be believed
Breath is transferred to server end, by the processing of server end semantic understanding module, the data type read is determined, by general data
Processing module, which is analyzed and processed, obtains analysis as a result, analysis result is shown in dynamic data visualization by information feedback
It is shown, while the preference query information of client is set by preference setting unit based on the analysis results.
2. a kind of data intelligence processing system based on natural language according to claim 1, it is characterised in that: described to ask
Input unit is inscribed, including voice input, Characters, clicks preset and 3 kinds of modes of prompt problem based on context.
3. a kind of data intelligence processing system based on natural language according to claim 1, it is characterised in that: institute's predicate
Adopted Understanding Module includes: Text similarity computing subelement, semantic generation subelement, fuzzy message processing subelement, missing letter
Breath processing subelement, context processing subelement, information interpret subelement and problem clew subelement；
The Text similarity computing subelement, basic algorithm are the editing distances of Chinese character and English word, and plus to defeated
Enter the inner tolerance there are wrong word plus word hiatus, each non-type typicalness of region mandarin to handle；
The semantic generation subelement, including zero, one or more dimensions, index, condition, sortord, maximum or minimum
How many, specified diagrammatic representation type, and semantic generate or update contextual information based on what is parsed；
The missing information handles subelement, and how processing accurately understands user when user's input information is imperfect as far as possible
It is intended to rather than reports an error；
The context handles subelement, context is created or updated according to each information input of user, and be based on context
More accurately semantic understanding and intention assessment are realized, to support more wheel sessions；
The information interprets subelement, carries out voice solution to the place that user in returned data pays close attention to according to preset service logic
It reads；
Described problem prompts subelement, proposes that the context of problem provides targeted prompt problem, guidance visitor based on user
Next how family puts question to, to realize more wheel sessions.
4. a kind of data intelligence processing system based on natural language according to claim 1, it is characterised in that: described logical
With data processing module include: generic data model subelement, general data connection subelement, query statement generate subelement and
Data acquisition and encapsulation subelement；
The generic data model subelement abstracts the complexity of each system Various types of data of enterprises, converts data to
Data model is indicated；
The general data connects subelement, realized based on general configuration to the connection sum number of enterprise's types of databases it is investigated that
It askes；
The query statement generates subelement, based on semantic dimension, the index, item for generating subelement and extracting from user's input
Part, sortord, most before/how many, user right information of most end in conjunction with underlying database type generate query statement；
The data acquisition and encapsulation subelement generate the query statement that subelement obtains based on query statement and call general data
Subelement is connected, result data is obtained, the subtype that may be specified based on data characteristics and user is formatted encapsulation, with
Front end is passed to be shown.
5. a kind of data intelligence processing system based on natural language according to claim 1, it is characterised in that: the machine
Device study module, including user behavior preferential learning submodule and generic data model subelement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910296255.XA CN110188163A (en) | 2019-04-13 | 2019-04-13 | Data intelligence processing system based on natural language |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910296255.XA CN110188163A (en) | 2019-04-13 | 2019-04-13 | Data intelligence processing system based on natural language |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110188163A true CN110188163A (en) | 2019-08-30 |
Family
ID=67714182
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910296255.XA Pending CN110188163A (en) | 2019-04-13 | 2019-04-13 | Data intelligence processing system based on natural language |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110188163A (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110706699A (en) * | 2019-09-27 | 2020-01-17 | 三星电子（中国）研发中心 | Method and system for realizing interaction task by adopting voice recognition mode |
| CN110825864A (en) * | 2019-11-13 | 2020-02-21 | 北京香侬慧语科技有限责任公司 | Method and device for obtaining answers to questions |
| CN110851484A (en) * | 2019-11-13 | 2020-02-28 | 北京香侬慧语科技有限责任公司 | Method and device for obtaining multi-index question answers |
| CN111125145A (en) * | 2019-11-26 | 2020-05-08 | 复旦大学 | Automatic system for acquiring database information through natural language |
| CN112035506A (en) * | 2019-10-28 | 2020-12-04 | 竹间智能科技（上海）有限公司 | Semantic recognition method and equipment |
| CN112259099A (en) * | 2020-12-23 | 2021-01-22 | 跬云(上海)信息科技有限公司 | Task processing method and device based on voice interaction and storage medium |
| CN112463954A (en) * | 2020-11-11 | 2021-03-09 | 远光软件股份有限公司 | Visual multidimensional data display system and method based on semantic recognition |
| CN115438142A (en) * | 2021-06-02 | 2022-12-06 | 戎易商智(北京)科技有限公司 | Interactive interactive data analysis report system |
| CN117370426A (en) * | 2023-12-04 | 2024-01-09 | 畅捷通信息技术股份有限公司 | Report data generation method, system and storage medium based on artificial intelligence |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| CN108241735A (en) * | 2017-12-05 | 2018-07-03 | 新华智云科技有限公司 | A kind of data analysing method and equipment |
| CN109446306A (en) * | 2018-10-16 | 2019-03-08 | 浪潮软件股份有限公司 | A kind of intelligent answer method of more wheels dialogue of task based access control driving |
| CN109446277A (en) * | 2018-09-21 | 2019-03-08 | 北京翰云时代数据技术有限公司 | Relational data intelligent search method and system based on Chinese natural language |
-
2019
- 2019-04-13 CN CN201910296255.XA patent/CN110188163A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105068661A (en) * | 2015-09-07 | 2015-11-18 | 百度在线网络技术（北京）有限公司 | Man-machine interaction method and system based on artificial intelligence |
| CN108241735A (en) * | 2017-12-05 | 2018-07-03 | 新华智云科技有限公司 | A kind of data analysing method and equipment |
| CN109446277A (en) * | 2018-09-21 | 2019-03-08 | 北京翰云时代数据技术有限公司 | Relational data intelligent search method and system based on Chinese natural language |
| CN109446306A (en) * | 2018-10-16 | 2019-03-08 | 浪潮软件股份有限公司 | A kind of intelligent answer method of more wheels dialogue of task based access control driving |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110706699A (en) * | 2019-09-27 | 2020-01-17 | 三星电子（中国）研发中心 | Method and system for realizing interaction task by adopting voice recognition mode |
| CN112035506A (en) * | 2019-10-28 | 2020-12-04 | 竹间智能科技（上海）有限公司 | Semantic recognition method and equipment |
| CN110825864A (en) * | 2019-11-13 | 2020-02-21 | 北京香侬慧语科技有限责任公司 | Method and device for obtaining answers to questions |
| CN110851484A (en) * | 2019-11-13 | 2020-02-28 | 北京香侬慧语科技有限责任公司 | Method and device for obtaining multi-index question answers |
| CN111125145A (en) * | 2019-11-26 | 2020-05-08 | 复旦大学 | Automatic system for acquiring database information through natural language |
| CN112463954A (en) * | 2020-11-11 | 2021-03-09 | 远光软件股份有限公司 | Visual multidimensional data display system and method based on semantic recognition |
| CN112463954B (en) * | 2020-11-11 | 2024-01-02 | 远光软件股份有限公司 | Visual multidimensional data display system and method based on semantic recognition |
| CN112259099A (en) * | 2020-12-23 | 2021-01-22 | 跬云(上海)信息科技有限公司 | Task processing method and device based on voice interaction and storage medium |
| CN115438142A (en) * | 2021-06-02 | 2022-12-06 | 戎易商智(北京)科技有限公司 | Interactive interactive data analysis report system |
| CN115438142B (en) * | 2021-06-02 | 2023-07-11 | 戎易商智(北京)科技有限公司 | Conversational interactive data analysis report system |
| CN117370426A (en) * | 2023-12-04 | 2024-01-09 | 畅捷通信息技术股份有限公司 | Report data generation method, system and storage medium based on artificial intelligence |
| CN117370426B (en) * | 2023-12-04 | 2024-03-26 | 畅捷通信息技术股份有限公司 | Report data generation method, system and storage medium based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110188163A (en) | Data intelligence processing system based on natural language | |
| CN108027833B (en) | Method for creating structured data language query | |
| US9898554B2 (en) | Implicit question query identification | |
| US20180060410A1 (en) | System and method of applying globally unique identifiers to relate distributed data sources | |
| CN101114294A (en) | Self-help intelligent uprightness searching method | |
| CN109189898A (en) | A kind of intelligent response method and system | |
| CN109918453B (en) | Method and system for searching relational complex management information system data by natural language | |
| CN109492164A (en) | A kind of recommended method of resume, device, electronic equipment and storage medium | |
| US9672490B2 (en) | Procurement system | |
| US20060053096A1 (en) | Natural language query construction using purpose-driven template | |
| CN108829682B (en) | Computer readable storage medium, intelligent question answering method and intelligent question answering device | |
| CN109345282A (en) | A kind of response method and equipment of business consultation | |
| CA2807494C (en) | Method and system for integrating web-based systems with local document processing applications | |
| CN102682118A (en) | Multidimensional data model access method and device | |
| WO2021047169A1 (en) | Information query method and apparatus, storage medium, and smart terminal | |
| CN109343705A (en) | AR auxiliary does tax method and system | |
| TWI743623B (en) | Artificial intelligence-based business intelligence system and its analysis method | |
| CN106326307A (en) | Language interaction method | |
| CN112115252A (en) | Intelligent auxiliary writing processing method and device, electronic equipment and storage medium | |
| CN111767334A (en) | Information extraction method and device, electronic equipment and storage medium | |
| CN114860916A (en) | Knowledge retrieval method and device | |
| CN107748748A (en) | Water conservancy and hydropower technical standard text retrieval system | |
| CN111382256A (en) | Information recommendation method and device | |
| CN111651554A (en) | Insurance question-answer method and device based on natural language understanding and processing | |
| CN103902248B (en) | Intelligent WeChat bank system based on natural language automatic scheduler and intelligent scheduling method for computer system through natural language |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190830 |