CN108241749A - By the method and apparatus of sensing data generation information - Google Patents
By the method and apparatus of sensing data generation information Download PDFInfo
- Publication number
- CN108241749A CN108241749A CN201810032456.4A CN201810032456A CN108241749A CN 108241749 A CN108241749 A CN 108241749A CN 201810032456 A CN201810032456 A CN 201810032456A CN 108241749 A CN108241749 A CN 108241749A
- Authority
- CN
- China
- Prior art keywords
- statistical nature
- sensing data
- collection
- generation
- exceptional value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/374—Thesaurus
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- 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
- G06F16/34—Browsing; Visualisation therefor
- G06F16/345—Summarisation for human users
-
- 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
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Machine Translation (AREA)
Abstract
The object of the present invention is to provide a kind of method and apparatus by sensing data generation information, the present invention is by identifying the statistical nature collection of sensing data, and statistical nature collection is converted to word information such as news etc., it can solve the problems, such as that existing magnanimity sensing data is difficult to United Dispatching use, realization is accurate, sensing data efficiently is converted into information such as news, the service efficiency of sensing data is improved, improves the efficiency of information interchange.
Description
Technical field
The present invention relates to computer realm more particularly to it is a kind of by sensing data generation information method and set
It is standby.
Background technology
With large scale deployment of the sensor in social infrastructure, the data that sensor generates largely are uploaded and are protected
It deposits.Sensing data is converted to word information and news, the efficiency that information interchange will be improved and used, but this identification at present
Flow is still shortcoming.
Invention content
It is an object of the present invention to provide a kind of method and apparatus by sensing data generation information, can solve
It is certainly how accurate, sensing data is efficiently converted into information such as news.
According to an aspect of the invention, there is provided a kind of method by sensing data generation information, this method
Including：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the pass
Keyword is via generation word information after semantic duplicate removal and sequence.
Further, in the above method, by the keyword via generation word information after semantic duplicate removal and sequence
Later, it further includes：
According to the corresponding statistical nature collection generation chart information of every class sensing data.
Further, it in the above method, generates per the corresponding statistical nature collection of class sensing data, including：
Generation is per the corresponding statistical nature collection of class sensing data；
To various kinds of sensors data, corresponding statistical nature collection is trained, with the statistical nature collection after being optimized.
Further, it is one corresponding according to the corresponding statistical nature collection generation of every class sensing data in the above method
Or multiple keywords, the keyword is generated into word information via semantic duplicate removal and after sorting, including：
Statistical nature semantic template library will be inputted per the corresponding statistical nature collection of class sensing data, to generate corresponding one
A or multiple keywords, and by the keyword via generation word information after semantic duplicate removal and sequence.
Further, it is one corresponding according to the corresponding statistical nature collection generation of every class sensing data in the above method
Or multiple keywords, the keyword is generated into word information via semantic duplicate removal and after sorting, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into exceptional value Ji Yuyimobanku, to generate corresponding one or more keywords,
And by the keyword via generation anomalous event word information after semantic duplicate removal and sequence.
Further, it is one corresponding according to the corresponding statistical nature collection generation of every class sensing data in the above method
Or multiple keywords, the keyword is generated into word information via semantic duplicate removal and after sorting, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into multisensor exceptional value Statistical Comparison module, determines whether accident, if
It is,
The exceptional value extracted is inputted into accident semantic template library, to generate corresponding one or more keywords,
And by the keyword via generation accident word information after semantic duplicate removal and sequence.
Further, in the above method, to various kinds of sensors data, corresponding statistical nature collection is trained, excellent to obtain
Statistical nature collection after change, including：
The corresponding statistical nature collection input machine learning module of various kinds of sensors data is subjected to deep learning training, with
Statistical nature collection after to optimization；
After the corresponding statistical nature of every class sensing data concentrates extraction exceptional value, further include：
The corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, it is excellent to obtain
Exceptional value after change.
According to another aspect of the present invention, a kind of equipment by sensing data generation information is additionally provided, this sets
It is standby to include：
Acquisition module, for obtaining a kind of or multiclass sensor data；
Statistical analysis module, for generating the corresponding statistical nature collection of every class sensing data；
Data set turns text module, for one corresponding according to statistical nature collection generation corresponding per class sensing data
Or multiple keywords；
Automatic writing module, for generating word information by the keyword via semantic duplicate removal and after sorting.
Further, in above equipment, data visualization module is further included, for according to corresponding per class sensing data
Statistical nature collection generates chart information.
Further, in above equipment, the statistical analysis module, for generating the corresponding statistics of every class sensing data
Feature set；To various kinds of sensors data, corresponding statistical nature collection is trained, with the statistical nature collection after being optimized.
Further, in above equipment, the data set turns text module, for will per class sensing data corresponding system
Feature set input statistical nature semantic template library is counted, to generate corresponding one or more keywords.
Further, in above equipment, the data set turns text module, for from the corresponding system of every class sensing data
Exceptional value is extracted in meter feature set；The exceptional value extracted is inputted into exceptional value Ji Yuyimobanku, to generate corresponding one
Or multiple keywords；
The automatic writing module, for generating anomalous event word by the keyword via semantic duplicate removal and after sorting
Information.
Further, in above equipment, data set turns text module, for special from the corresponding statistics of every class sensing data
Exceptional value is extracted in collection；The exceptional value extracted is inputted into multisensor exceptional value Statistical Comparison module, determines whether to dash forward
Hair event, if so, by the exceptional value extracted input accident semantic template library, it is corresponding one or more crucial to generate
Word；
The automatic writing module, for generating accident word by the keyword via semantic duplicate removal and after sorting
Information.
Further, in above equipment, the statistical analysis module is additionally operable to the corresponding statistics of various kinds of sensors data
Feature set input machine learning module carries out deep learning training, with the statistical nature collection after being optimized；
The data set turns text module, is additionally operable to concentrate extraction abnormal from the corresponding statistical nature of every class sensing data
After value, the corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, it is excellent to obtain
Exceptional value after change.
According to another aspect of the present invention, a kind of equipment based on calculating is additionally provided, wherein, including：
Processor；And
The memory of storage computer executable instructions is arranged to, the executable instruction makes the place when executed
Manage device：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the pass
Keyword is via generation word information after semantic duplicate removal and sequence.
According to another aspect of the present invention, a kind of computer readable storage medium is additionally provided, is stored thereon with computer
Executable instruction, wherein, which causes the processor when being executed by processor：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the pass
Keyword is via generation word information after semantic duplicate removal and sequence.
Compared with prior art, the present invention is by identifying the statistical nature collection of sensing data, and by statistical nature collection
Word information such as news etc. is converted to, can solve the problems, such as that existing magnanimity sensing data is difficult to United Dispatching use,
Realization is accurate, sensing data efficiently is converted into information such as news, improves the service efficiency of sensing data, improves
The efficiency of information interchange.
Description of the drawings
By reading the detailed description made to non-limiting example made with reference to the following drawings, of the invention is other
Feature, objects and advantages will become more apparent upon：
Fig. 1 shows the method flow diagram according to an embodiment of the invention by sensing data generation information；
Fig. 2 shows the method flow diagrams by sensing data generation information of another embodiment of the present invention.
The same or similar reference numeral represents the same or similar component in attached drawing.
Specific embodiment
The present invention is described in further detail below in conjunction with the accompanying drawings.
In a typical configuration of this application, terminal, the equipment of service network and trusted party include one or more
Processor (CPU), input/output interface, network interface and memory.
Memory may include computer-readable medium in volatile memory, random access memory (RAM) and/or
The forms such as Nonvolatile memory, such as read-only memory (ROM) or flash memory (flash RAM).Memory is computer-readable medium
Example.
Computer-readable medium includes permanent and non-permanent, removable and non-removable media can be by any method
Or technology come realize information store.Information can be computer-readable instruction, data structure, the module of program or other data.
The example of the storage medium of computer includes, but are not limited to phase transition internal memory (PRAM), static RAM (SRAM), moves
State random access memory (DRAM), other kinds of random access memory (RAM), read-only memory (ROM), electric erasable
Programmable read only memory (EEPROM), fast flash memory bank or other memory techniques, CD-ROM read-only memory (CD-ROM),
Digital versatile disc (DVD) or other optical storages, magnetic tape cassette, magnetic tape disk storage or other magnetic storage apparatus or
Any other non-transmission medium, available for storing the information that can be accessed by a computing device.It is defined according to herein, computer
Readable medium does not include the data-signal and carrier wave of non-temporary computer readable media (transitory media), such as modulation.
The present invention provides a kind of method by sensing data generation information, including：
Step S1 obtains a kind of or multiclass sensor data；
Here, access multiclass sensor data, as shown in Fig. 2, various kinds of sensors data include but not limited to temperature, wet
Degree, pressure, light sensation, the types (such as 101/102/103) such as test the speed；
Step S2 is generated per the corresponding statistical nature collection of class sensing data；
Here, the statistical nature collection is to carry out feature by multiple default sensor indices dimensions to every class sensing data
The set of statistics, for example, for temperature sensor, multiple default sensor indices dimensions can include temperature today more than 17 degree
Area, the temperature Change situation of some area 11~13 point, the Hangzhou today temperature of one day etc.；
As shown in Figure 1, it can be generated by a statistical analysis module per the corresponding statistical nature collection of class sensing data；
As shown in Fig. 2, statistical nature can be generated by the sensor data set 100 by statistical analysis module 200
Collection 110；
Step S3 generates corresponding one or more keywords according to the corresponding statistical nature collection of every class sensing data,
By the keyword via generation word information after semantic duplicate removal and sequence.
Here, as shown in Figure 1, text module can be turned by a data set according to statistics spy corresponding per class sensing data
Collection generates the corresponding multiple keywords of one or more, is then gone the keyword via semanteme by an automatic writing module
Information is generated after weight and sequence.
Statistical nature collection is converted to word information letter by the present invention by identifying the statistical nature collection of sensing data
Such as news is ceased, can solve the problems, such as that existing magnanimity sensing data is difficult to United Dispatching use, realization is accurate, efficiently will
Sensing data is converted into information such as news, improves the service efficiency of sensing data, improves the efficiency of information interchange.
In one embodiment of method by sensing data generation information of the present invention, step S2, by the keyword
After generation word information after semantic duplicate removal and sequence, further include：
According to the corresponding statistical nature collection generation chart information of every class sensing data.
Here, the corresponding figure of such sensor can be generated respectively to the corresponding statistical nature collection of every class sensing data
Table information can also summarize various kinds of sensors data and generate a chart information summarized.
As shown in Figure 1, it can be given birth to by a data visualization module according to per the corresponding statistical nature collection of class sensing data
Into chart information, the word information and chart information are being combined by the automatic writing module.
As shown in Fig. 2, treated the statistical nature collection of statistical analysis module 200 is by visual template library 500, with difference
Template types are matched, and generate data visualization chart as chart information.Wherein, visual template library 500 can
To include multiple types diagrammatic form and corresponding Data Matching rule.
The present embodiment is the information of generation with reference to generation chart information on the basis of word information is generated
Information is more rich, vivid, lively.
In one embodiment of method by sensing data generation information of the present invention, step S2 is generated and is sensed per class
The corresponding statistical nature collection of device data, including：
Generation is per the corresponding statistical nature collection of class sensing data；
To various kinds of sensors data, corresponding statistical nature collection is trained, with the statistical nature collection after being optimized.
Here, as shown in Figure 1, can by a machine learning module to various kinds of sensors data corresponding statistical nature collection
It is trained, with the statistical nature collection after being optimized.
The present embodiment is by optimizing statistical nature collection, in order to follow-up according to corresponding excellent per class sensing data
One or more keywords corresponding to the generation of statistical nature collection after change, by the keyword via raw after semantic duplicate removal and sequence
The statistical nature collection after optimization is corresponded into word information and according to every class sensing data and generates chart information, is made
The word information letter and chart information of generation are more accurate.
In one embodiment of method by sensing data generation information of the present invention, step S3 is sensed according to every class
The corresponding one or more keywords of the corresponding statistical nature collection generation of device data, by the keyword via semantic duplicate removal and row
Word information is generated after sequence, including：
Statistical nature semantic template library will be inputted per the corresponding statistical nature collection of class sensing data, to generate corresponding one
A or multiple keywords, and by the keyword via generation word information after semantic duplicate removal and sequence.
It here, as shown in Fig. 2, can be by statistical nature collection 110 by statistical nature semantic template library 410, by data
Multiple keywords are generated after comparison and sequence, and coordinate above-mentioned data visualization chart generation word information such as news.
The present embodiment generates corresponding one or more keywords by statistical nature semantic template library, and by the key
Word is further ensured that via word information is generated after semantic duplicate removal and sequence and more efficiently generates more accurate word information
Information.
In one embodiment of method by sensing data generation information of the present invention, step S3 is sensed according to every class
The corresponding one or more keywords of the corresponding statistical nature collection generation of device data, by the keyword via semantic duplicate removal and row
Word information is generated after sequence, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into exceptional value Ji Yuyimobanku, to generate corresponding one or more keywords,
And by the keyword via generation anomalous event word information after semantic duplicate removal and sequence.
Here, as shown in Fig. 2, exceptional value collection 120 can be generated text information by exceptional value collection semantic template library 420
Information, with word information that is more efficient, more accurately generating report exceptions.
In one embodiment of method by sensing data generation information of the present invention, step S3 is sensed according to every class
The corresponding one or more keywords of the corresponding statistical nature collection generation of device data, by the keyword via semantic duplicate removal and row
Word information is generated after sequence, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into multisensor exceptional value Statistical Comparison module, determines whether accident, if
It is,
The exceptional value extracted is inputted into accident semantic template library, to generate corresponding one or more keywords,
And by the keyword via generation accident word information after semantic duplicate removal and sequence.
Here, as shown in Fig. 2, exceptional value collection 120 can be judged by multisensor exceptional value Statistical Comparison module 210
Whether it is accident, if exporting exceptional value collection 120 to accident semantic template library 430, generates multiple keywords, and
Via semantic duplicate removal and sequence after generate accident word information, with it is more efficient, more accurately generate report burst thing
The word information of part.
In one embodiment of method by sensing data generation information of the present invention, various kinds of sensors data are corresponded to
Statistical nature collection be trained, with the statistical nature collection after being optimized, including：
The corresponding statistical nature collection input machine learning module of various kinds of sensors data is subjected to deep learning training, with
Statistical nature collection after to optimization；
After the corresponding statistical nature of every class sensing data concentrates extraction exceptional value, further include：
The corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, it is excellent to obtain
Exceptional value after change.
Here, as shown in Fig. 2, the statistical nature collection and exceptional value collection can be input to machine learning module 300
Deep learning training is carried out, the model training result of generation includes：Newer statistical nature masterplate and the exceptional value after adjustment
Newer statistical nature masterplate is matched the statistical nature after being optimized by judgment threshold with the statistical nature collection inputted
Collection by the judgment threshold of the exceptional value after adjustment and the exceptional value collection of input, can obtain the exceptional value after generation optimization, i.e., excellent
The exceptional value after statistical nature collection and optimization after change is exported as Sub Data Set 130/140/150 from machine study module 300.
130/140/150 grade of Sub Data Set can subsequently be inputted to semantization template library 400 and visual template library 500, and
Sub Data Set and the semantic template of different classifications, visual template are subjected to matched process.It for example, can be to above-mentioned subdata
Crucial character/word keyword1~keywordN of 130,140,150 extraction of collection carries out semantic duplicate removal/sequence, sentence meaning understands, with life
Into text information.
Wherein, semantization template library 400 includes but not limited to above-mentioned statistical nature semantic template library 410, exceptional value language
Adopted template library 420, accident semantic template library 430 etc., each template library contain multiple semantization templates manually marked respectively
And a variety of different meaning category dictionaries.
It can include in semantic template library 400：
α) the semantic template of multiple classification, including but not limited to statistical nature semantic template 410, exceptional value semantic template
420th, accident semantic template 430 etc., each classification are corresponding with the data set generated.
B) information/news and its multidimensional of the crucial character/word generation of extraction marked under each semantic template classification containing more
Term vector group.
C) a variety of different meaning category dictionaries can include identification part of speech (noun/verb/adjective/quantifier etc.)
The corpus of machine learning；It can also include the inquiry of synonym.
By by after the classification and matching in Sub Data Set 130/140/150 and semantic template 400, being generated using Sub Data Set
Corresponding keyword, and be filled into corresponding semantic template.
As shown in Fig. 2, machine learning module 300 includes two parts：
1) cured layer 310 are by sensor data set segment, including statistical nature collection masterplate by Journalism person
311 and exceptional value collection masterplate 312, after cured layer 310 is labeled statistical nature collection masterplate 311 and exceptional value collection masterplate 312
The multi-C vector collection of generation；
2) autonomous learning layer 320, for will be by the multi-C vector collection that statistical nature collection and exceptional value generate and cured layer 310
Multi-C vector collection trained together by machine learning model after generate, update statistical nature masterplate, and adjust sentencing for exceptional value
Disconnected threshold value, and according to update statistical nature masterplate and the judgment threshold of adjustment exceptional value, generate new Sub Data Set 130/140/
150。
The present embodiment optimizes statistical nature collection and exceptional value by machine learning module, in order to follow-up according to every
One or more keywords corresponding to statistical nature collection and exceptional value generation after the corresponding optimization of class sensing data, by described in
Keyword corresponds to the system after optimization via generation word information after semantic duplicate removal and sequence and according to every class sensing data
Feature set generation chart information is counted, the word information of generation is made to believe with chart information more accurately, efficiently.
According to another aspect of the present invention, a kind of equipment by sensing data generation information is additionally provided, wherein,
The equipment includes：
Acquisition module, for obtaining a kind of or multiclass sensor data；
Here, access multiclass sensor data, as shown in Fig. 2, various kinds of sensors data are included but not
It is limited to that temperature, humidity, pressure, light sensation, the types (such as 101/102/103) such as test the speed；
Statistical analysis module, for generating the corresponding statistical nature collection of every class sensing data；
Here, the statistical nature collection is to carry out feature by multiple default sensor indices dimensions to every class sensing data
The set of statistics, for example, for temperature sensor, multiple default sensor indices dimensions can include temperature today more than 17 degree
Area, the temperature Change situation of some area 11~13 point, the Hangzhou today temperature of one day etc.；
As shown in Figure 1, it can be generated by a statistical analysis module per the corresponding statistical nature collection of class sensing data；
As shown in Fig. 2, can by the sensor data set 100 by statistical analysis module 200,
Generate statistical nature collection 110；
Data set turns text module, for one corresponding according to statistical nature collection generation corresponding per class sensing data
Or multiple keywords；
Automatic writing module, for generating word information by the keyword via semantic duplicate removal and after sorting.
Here, as shown in Figure 1, text module can be turned by a data set according to statistics spy corresponding per class sensing data
Collection generates the corresponding multiple keywords of one or more, is then gone the keyword via semanteme by an automatic writing module
Information is generated after weight and sequence.
Statistical nature collection is converted to word information letter by the present invention by identifying the statistical nature collection of sensing data
Such as news is ceased, can solve the problems, such as that existing magnanimity sensing data is difficult to United Dispatching use, realization is accurate, efficiently will
Sensing data is converted into information such as news, improves the service efficiency of sensing data, improves the efficiency of information interchange.
In one embodiment of equipment by sensing data generation information of the present invention, data visualization mould is further included
Block, for according to statistical nature collection generation chart information corresponding per class sensing data.
Here, the corresponding figure of such sensor can be generated respectively to the corresponding statistical nature collection of every class sensing data
Table information can also summarize various kinds of sensors data and generate a chart information summarized.
As shown in Figure 1, it can be given birth to by a data visualization module according to per the corresponding statistical nature collection of class sensing data
Into chart information, the word information and chart information are being combined by the automatic writing module.
As shown in Fig. 2, treated the statistical nature collection of statistical analysis module 200 is by visual template library 500, with difference
Template types are matched, and generate data visualization chart as chart information.Wherein, visual template library 500 can
To include multiple types diagrammatic form and corresponding Data Matching rule.
The present embodiment is the information of generation with reference to generation chart information on the basis of word information is generated
Information is more rich, vivid, lively.
In one embodiment of equipment by sensing data generation information of the present invention, the statistical analysis module is used
In generation per the corresponding statistical nature collection of class sensing data；To various kinds of sensors data, corresponding statistical nature collection is instructed
Practice, with the statistical nature collection after being optimized.
Here, as shown in Figure 1, can by a machine learning module to various kinds of sensors data corresponding statistical nature collection
It is trained, with the statistical nature collection after being optimized.
The present embodiment is by optimizing statistical nature collection, in order to follow-up according to corresponding excellent per class sensing data
One or more keywords corresponding to the generation of statistical nature collection after change, by the keyword via raw after semantic duplicate removal and sequence
The statistical nature collection after optimization is corresponded into word information and according to every class sensing data and generates chart information, is made
The word information letter and chart information of generation are more accurate.
In one embodiment of equipment by sensing data generation information of the present invention, the data set turns text mould
Block, for statistical nature semantic template library will to be inputted per the corresponding statistical nature collection of class sensing data, to generate corresponding one
A or multiple keywords.
It here, as shown in Fig. 2, can be by statistical nature collection 110 by statistical nature semantic template library 410, by data
Multiple keywords are generated after comparison and sequence, and coordinate above-mentioned data visualization chart generation word information such as news.
The present embodiment generates corresponding one or more keywords by statistical nature semantic template library, and by the key
Word is further ensured that via word information is generated after semantic duplicate removal and sequence and more efficiently generates more accurate word information
Information.
In one embodiment of equipment by sensing data generation information of the present invention, the data set turns text mould
Block, for concentrating extraction exceptional value from the corresponding statistical nature of every class sensing data；The exceptional value extracted is inputted abnormal
Value Ji Yuyimobanku, to generate corresponding one or more keywords；
The automatic writing module, for generating anomalous event word by the keyword via semantic duplicate removal and after sorting
Information.
Here, as shown in Fig. 2, exceptional value collection 120 can be generated text information by exceptional value collection semantic template library 420
Information, with word information that is more efficient, more accurately generating report exceptions.
In one embodiment of equipment by sensing data generation information of the present invention, data set turns text module, uses
Exceptional value is extracted in being concentrated from the corresponding statistical nature of every class sensing data；The exceptional value extracted input multisensor is different
Constant value Statistical Comparison module, determines whether accident, if so, the exceptional value extracted is inputted accident semantic template
Library, to generate corresponding one or more keywords；
The automatic writing module, for generating accident word by the keyword via semantic duplicate removal and after sorting
Information.
Here, as shown in Fig. 2, exceptional value collection 120 can be judged by multisensor exceptional value Statistical Comparison module 210
Whether it is accident, if exporting exceptional value collection 120 to accident semantic template library 430, generates multiple keywords, and
Via semantic duplicate removal and sequence after generate accident word information, with it is more efficient, more accurately generate report burst thing
The word information of part.
In one embodiment of equipment by sensing data generation information of the present invention, the statistical analysis module, also
It is excellent to obtain for the corresponding statistical nature collection input machine learning module of various kinds of sensors data to be carried out deep learning training
Statistical nature collection after change；
The data set turns text module, is additionally operable to concentrate extraction abnormal from the corresponding statistical nature of every class sensing data
After value, the corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, it is excellent to obtain
Exceptional value after change.
Here, as shown in Fig. 2, the statistical nature collection and exceptional value collection can be input to machine learning module 300
Deep learning training is carried out, the model training result of generation includes：Newer statistical nature masterplate and the exceptional value after adjustment
Newer statistical nature masterplate is matched the statistical nature after being optimized by judgment threshold with the statistical nature collection inputted
Collection by the judgment threshold of the exceptional value after adjustment and the exceptional value collection of input, can obtain the exceptional value after generation optimization, i.e., excellent
The exceptional value after statistical nature collection and optimization after change is exported as Sub Data Set 130/140/150 from machine study module 300.
130/140/150 grade of Sub Data Set can subsequently be inputted to semantization template library 400 and visual template library 500, and
Sub Data Set and the semantic template of different classifications, visual template are subjected to matched process.It for example, can be to above-mentioned subdata
Crucial character/word keyword1~keywordN of 130,140,150 extraction of collection carries out semantic duplicate removal/sequence, sentence meaning understands, with life
Into text information.
Wherein, semantization template library 400 includes but not limited to above-mentioned statistical nature semantic template library 410, exceptional value language
Adopted template library 420, accident semantic template library 430 etc., each template library contain multiple semantization templates manually marked respectively
And a variety of different meaning category dictionaries.
It can include in semantic template library 400：
A) semantic template of multiple classification, including but not limited to statistical nature semantic template 410, exceptional value semantic template
420th, accident semantic template 430 etc., each classification are corresponding with the data set generated.
B) information/news and its multidimensional of the crucial character/word generation of extraction marked under each semantic template classification containing more
Term vector group.
C) a variety of different meaning category dictionaries can include identification part of speech (noun/verb/adjective/quantifier etc.)
The corpus of machine learning；It can also include the inquiry of synonym.
By by after the classification and matching in Sub Data Set 130/140/150 and semantic template 400, being generated using Sub Data Set
Corresponding keyword, and be filled into corresponding semantic template.
As shown in Fig. 2, machine learning module 300 includes two parts：
1) cured layer 310 are by sensor data set segment, including statistical nature collection masterplate by Journalism person
311 and exceptional value collection masterplate 312, after cured layer 310 is labeled statistical nature collection masterplate 311 and exceptional value collection masterplate 312
The multi-C vector collection of generation；
2) autonomous learning layer 320, for will be by the multi-C vector collection that statistical nature collection and exceptional value generate and cured layer 310
Multi-C vector collection trained together by machine learning model after generate, update statistical nature masterplate, and adjust sentencing for exceptional value
Disconnected threshold value, and according to update statistical nature masterplate and the judgment threshold of adjustment exceptional value, generate new Sub Data Set 130/140/
150。
The present embodiment optimizes statistical nature collection and exceptional value by machine learning module, in order to follow-up according to every
One or more keywords corresponding to statistical nature collection and exceptional value generation after the corresponding optimization of class sensing data, by described in
Keyword corresponds to the system after optimization via generation word information after semantic duplicate removal and sequence and according to every class sensing data
Feature set generation chart information is counted, the word information of generation is made to believe with chart information more accurately, efficiently.
According to another aspect of the present invention, a kind of equipment based on calculating is additionally provided, wherein, including：
Processor；And
The memory of storage computer executable instructions is arranged to, the executable instruction makes the place when executed
Manage device：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the pass
Keyword is via generation word information after semantic duplicate removal and sequence.
According to another aspect of the present invention, a kind of computer readable storage medium is additionally provided, is stored thereon with computer
Executable instruction, wherein, which causes the processor when being executed by processor：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the pass
Keyword is via generation word information after semantic duplicate removal and sequence.
The detailed content of above equipment and each embodiment of computer readable storage medium for details, reference can be made to each method embodiment
Corresponding part, details are not described herein.
Obviously, those skilled in the art can carry out the application essence of the various modification and variations without departing from the application
God and range.In this way, if these modifications and variations of the application belong to the range of the application claim and its equivalent technologies
Within, then the application is also intended to include these modifications and variations.
It should be noted that the present invention can be carried out in the assembly of software and/or software and hardware, for example, can adopt
With application-specific integrated circuit (ASIC), general purpose computer or any other realized similar to hardware device.In one embodiment
In, software program of the invention can perform to realize steps described above or function by processor.Similarly, it is of the invention
Software program can be stored in computer readable recording medium storing program for performing (including relevant data structure), for example, RAM memory,
Magnetic or optical driver or floppy disc and similar devices.In addition, hardware can be used to realize in some steps or function of the present invention, example
Such as, as coordinating with processor so as to perform the circuit of each step or function.
In addition, the part of the present invention can be applied to computer program product, such as computer program instructions, when its quilt
When computer performs, by the operation of the computer, it can call or provide according to the method for the present invention and/or technical solution.
And the program instruction of the method for the present invention is called, it is possibly stored in fixed or moveable recording medium and/or passes through
Broadcast or the data flow in other signal loaded mediums and be transmitted and/or be stored according to described program instruction operation
In the working storage of computer equipment.Here, including a device according to one embodiment of present invention, which includes using
Memory in storage computer program instructions and processor for executing program instructions, wherein, when the computer program refers to
When order is performed by the processor, method and/or skill of the device operation based on aforementioned multiple embodiments according to the present invention are triggered
Art scheme.
It is obvious to a person skilled in the art that the present invention is not limited to the details of above-mentioned exemplary embodiment, Er Qie
In the case of without departing substantially from spirit or essential attributes of the invention, the present invention can be realized in other specific forms.Therefore, no matter
From the point of view of which point, the present embodiments are to be considered as illustrative and not restrictive, and the scope of the present invention is by appended power
Profit requirement rather than above description limit, it is intended that all by what is fallen within the meaning and scope of the equivalent requirements of the claims
Variation includes within the present invention.Any reference numeral in claim should not be considered as to the involved claim of limitation.This
Outside, it is clear that one word of " comprising " is not excluded for other units or step, and odd number is not excluded for plural number.That is stated in device claim is multiple
Unit or device can also be realized by a unit or device by software or hardware.The first, the second grade words are used for table
Show title, and do not represent any particular order.
Claims (16)
1. a kind of method by sensing data generation information, wherein, this method includes：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the keyword
Via generation word information after semantic duplicate removal and sequence.
2. according to the method described in claim 1, wherein, the keyword is provided via generation word after semantic duplicate removal and sequence
After interrogating information, further include：
According to the corresponding statistical nature collection generation chart information of every class sensing data.
3. according to the method described in claim 1, wherein, generate per the corresponding statistical nature collection of class sensing data, including：
Generation is per the corresponding statistical nature collection of class sensing data；
To various kinds of sensors data, corresponding statistical nature collection is trained, with the statistical nature collection after being optimized.
4. method according to any one of claims 1 to 3, wherein, according to the corresponding statistical nature of every class sensing data
The corresponding one or more keywords of collection generation, by the keyword via generation word information letter after semantic duplicate removal and sequence
Breath, including：
Will per class sensing data corresponding statistical nature collection input statistical nature semantic template library, with generate corresponding one or
Multiple keywords, and by the keyword via generation word information after semantic duplicate removal and sequence.
5. it according to the method described in claim 3, wherein, is corresponded to according to the corresponding statistical nature collection generation of every class sensing data
One or more keywords, by the keyword via semantic duplicate removal and sequence after generate word information, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into exceptional value Ji Yuyimobanku, to generate corresponding one or more keywords, and will
The keyword is via generation anomalous event word information after semantic duplicate removal and sequence.
6. it according to the method described in claim 3, wherein, is corresponded to according to the corresponding statistical nature collection generation of every class sensing data
One or more keywords, by the keyword via semantic duplicate removal and sequence after generate word information, including：
Extraction exceptional value is concentrated from the corresponding statistical nature of every class sensing data；
The exceptional value extracted is inputted into multisensor exceptional value Statistical Comparison module, determines whether accident, if so,
The exceptional value extracted is inputted into accident semantic template library, to generate corresponding one or more keywords, and will
The keyword is via generation accident word information after semantic duplicate removal and sequence.
7. method according to claim 5 or 6, wherein, to various kinds of sensors data, corresponding statistical nature collection is instructed
Practice, with the statistical nature collection after being optimized, including：
The corresponding statistical nature collection input machine learning module of various kinds of sensors data is subjected to deep learning training, it is excellent to obtain
Statistical nature collection after change；
After the corresponding statistical nature of every class sensing data concentrates extraction exceptional value, further include：
The corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, after obtaining optimization
Exceptional value.
8. a kind of equipment by sensing data generation information, wherein, which includes：
Acquisition module, for obtaining a kind of or multiclass sensor data；
Statistical analysis module, for generating the corresponding statistical nature collection of every class sensing data；
Data set turns text module, for one or more corresponding according to statistical nature collection generation corresponding per class sensing data
A keyword；
Automatic writing module, for generating word information by the keyword via semantic duplicate removal and after sorting.
9. equipment according to claim 8, wherein, data visualization module is further included, for according to per class sensor number
Chart information is generated according to corresponding statistical nature collection.
10. equipment according to claim 8, wherein, the statistical analysis module, for generating every class sensing data pair
The statistical nature collection answered；To various kinds of sensors data, corresponding statistical nature collection is trained, special with the statistics after being optimized
Collection.
11. according to claim 8 to 10 any one of them equipment, wherein, the data set turns text module, for will be per class
The corresponding statistical nature collection input statistical nature semantic template library of sensing data, it is corresponding one or more crucial to generate
Word.
12. equipment according to claim 10, wherein, the data set turns text module, for from every class sensor number
Extraction exceptional value is concentrated according to corresponding statistical nature；The exceptional value extracted is inputted into exceptional value Ji Yuyimobanku, with generation
Corresponding one or more keyword；
The automatic writing module, for generating anomalous event word information by the keyword via semantic duplicate removal and after sorting
Information.
13. equipment according to claim 10, wherein, data set turns text module, for from every class sensing data pair
The statistical nature answered concentrates extraction exceptional value；The exceptional value extracted is inputted into multisensor exceptional value Statistical Comparison module, is sentenced
Whether disconnected is accident, if so, by the exceptional value extracted input accident semantic template library, to generate corresponding one
Or multiple keywords；
The automatic writing module, for generating accident word information by the keyword via semantic duplicate removal and after sorting
Information.
14. equipment according to claim 12 or 13, wherein, the statistical analysis module is additionally operable to various kinds of sensors
Data corresponding statistical nature collection input machine learning module carries out deep learning training, with the statistical nature after being optimized
Collection；
The data set turns text module, be additionally operable to from the corresponding statistical nature of every class sensing data concentrate extraction exceptional value it
Afterwards, the corresponding exceptional value input machine learning module of various kinds of sensors data is subjected to deep learning training, after obtaining optimization
Exceptional value.
15. a kind of equipment based on calculating, wherein, including：
Processor；And
The memory of storage computer executable instructions is arranged to, the executable instruction makes the processing when executed
Device：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the keyword
Via generation word information after semantic duplicate removal and sequence.
16. a kind of computer readable storage medium, is stored thereon with computer executable instructions, wherein, which can perform
Instruction causes the processor when being executed by processor：
Obtain a kind of or multiclass sensor data；
Generation is per the corresponding statistical nature collection of class sensing data；
According to the corresponding one or more keywords of the corresponding statistical nature collection generation of every class sensing data, by the keyword
Via generation word information after semantic duplicate removal and sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810032456.4A CN108241749B (en) | 2018-01-12 | 2018-01-12 | Method and apparatus for generating information from sensor data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810032456.4A CN108241749B (en) | 2018-01-12 | 2018-01-12 | Method and apparatus for generating information from sensor data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108241749A true CN108241749A (en) | 2018-07-03 |
| CN108241749B CN108241749B (en) | 2021-03-26 |
Family
ID=62699597
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810032456.4A Active CN108241749B (en) | 2018-01-12 | 2018-01-12 | Method and apparatus for generating information from sensor data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108241749B (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103177391A (en) * | 2013-01-31 | 2013-06-26 | 李生林 | Equipment monitoring system based on three-dimensional real scenes and system integration method based on three-dimensional real scenes |
| CN103390039A (en) * | 2013-07-17 | 2013-11-13 | 北京建筑工程学院 | Urban disaster thematic map real-time generating method based on network information |
| US20130332410A1 (en) * | 2012-06-07 | 2013-12-12 | Sony Corporation | Information processing apparatus, electronic device, information processing method and program |
| CN103678910A (en) * | 2013-12-12 | 2014-03-26 | 河海大学 | Cloud system structure pre-warning system and method for riverway type reservoir tributary bay water bloom |
| CN103700054A (en) * | 2013-12-10 | 2014-04-02 | 中国地质大学武汉 | Sudden-onset geological disaster emergency plan digitization system |
| CN103955873A (en) * | 2014-05-20 | 2014-07-30 | 成都汇资聚源科技有限公司 | Innovation resource information integration service platform |
| CN102708096B (en) * | 2012-05-29 | 2014-10-15 | 代松 | Network intelligence public sentiment monitoring system based on semantics and work method thereof |
| CN105446156A (en) * | 2015-12-30 | 2016-03-30 | 百度在线网络技术（北京）有限公司 | Method, device and system for controlling household electric appliance based on artificial intelligence |
| CN105791343A (en) * | 2014-12-19 | 2016-07-20 | 中国移动通信集团公司 | Intelligent home information interaction system, intelligent home information interaction method and related equipment |
| CN106095101A (en) * | 2016-06-16 | 2016-11-09 | 福建师范大学 | Human bodys' response method based on power-saving mechanism and client |
| CN106649455A (en) * | 2016-09-24 | 2017-05-10 | 孙燕群 | Big data development standardized systematic classification and command set system |
-
2018
- 2018-01-12 CN CN201810032456.4A patent/CN108241749B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102708096B (en) * | 2012-05-29 | 2014-10-15 | 代松 | Network intelligence public sentiment monitoring system based on semantics and work method thereof |
| US20130332410A1 (en) * | 2012-06-07 | 2013-12-12 | Sony Corporation | Information processing apparatus, electronic device, information processing method and program |
| CN103177391A (en) * | 2013-01-31 | 2013-06-26 | 李生林 | Equipment monitoring system based on three-dimensional real scenes and system integration method based on three-dimensional real scenes |
| CN103390039A (en) * | 2013-07-17 | 2013-11-13 | 北京建筑工程学院 | Urban disaster thematic map real-time generating method based on network information |
| CN103700054A (en) * | 2013-12-10 | 2014-04-02 | 中国地质大学武汉 | Sudden-onset geological disaster emergency plan digitization system |
| CN103678910A (en) * | 2013-12-12 | 2014-03-26 | 河海大学 | Cloud system structure pre-warning system and method for riverway type reservoir tributary bay water bloom |
| CN103955873A (en) * | 2014-05-20 | 2014-07-30 | 成都汇资聚源科技有限公司 | Innovation resource information integration service platform |
| CN105791343A (en) * | 2014-12-19 | 2016-07-20 | 中国移动通信集团公司 | Intelligent home information interaction system, intelligent home information interaction method and related equipment |
| CN105446156A (en) * | 2015-12-30 | 2016-03-30 | 百度在线网络技术（北京）有限公司 | Method, device and system for controlling household electric appliance based on artificial intelligence |
| CN106095101A (en) * | 2016-06-16 | 2016-11-09 | 福建师范大学 | Human bodys' response method based on power-saving mechanism and client |
| CN106649455A (en) * | 2016-09-24 | 2017-05-10 | 孙燕群 | Big data development standardized systematic classification and command set system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108241749B (en) | 2021-03-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108319668B (en) | Method and equipment for generating text abstract | |
| CN109472033B (en) | Method and system for extracting entity relationship in text, storage medium and electronic equipment | |
| US8457950B1 (en) | System and method for coreference resolution | |
| CN113051371B (en) | Chinese machine reading understanding method and device, electronic equipment and storage medium | |
| WO2020114100A1 (en) | Information processing method and apparatus, and computer storage medium | |
| US9009029B1 (en) | Semantic hashing in entity resolution | |
| CN112256822A (en) | Text search method and device, computer equipment and storage medium | |
| CN105589941A (en) | Emotional information detection method and apparatus for web text | |
| CN110852056B (en) | Method, device and equipment for obtaining text similarity and readable storage medium | |
| US20200364216A1 (en) | Method, apparatus and storage medium for updating model parameter | |
| CN109597983A (en) | A kind of spelling error correction method and device | |
| US20130304468A1 (en) | Contextual Voice Query Dilation | |
| CN112395412B (en) | Text classification method, apparatus and computer readable medium | |
| CN116932730B (en) | Document question-answering method and related equipment based on multi-way tree and large-scale language model | |
| CN112287100A (en) | Text recognition method, spelling error correction method and voice recognition method | |
| CN111813925A (en) | Semantic-based unsupervised automatic summarization method and system | |
| CN112434533A (en) | Entity disambiguation method, apparatus, electronic device, and computer-readable storage medium | |
| CN109597982B (en) | Abstract text recognition method and device | |
| CN118245568A (en) | Question and answer method and device based on large model, electronic equipment and storage medium | |
| CN105095826B (en) | A kind of character recognition method and device | |
| CN107861950A (en) | The detection method and device of abnormal text | |
| CN109902162B (en) | Text similarity identification method based on digital fingerprints, storage medium and device | |
| CN112765976A (en) | Text similarity calculation method, device and equipment and storage medium | |
| US20240232245A1 (en) | Method, device, and medium for consistency detection of a document and an abstract | |
| CN111209724A (en) | Text verification method and device, storage medium and processor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |