CN108241749B - Method and apparatus for generating information from sensor data - Google Patents

Method and apparatus for generating information from sensor data Download PDF

Info

Publication number
CN108241749B
CN108241749B CN201810032456.4A CN201810032456A CN108241749B CN 108241749 B CN108241749 B CN 108241749B CN 201810032456 A CN201810032456 A CN 201810032456A CN 108241749 B CN108241749 B CN 108241749B
Authority
CN
China
Prior art keywords
sensor data
generating
type
statistical feature
keywords
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201810032456.4A
Other languages
Chinese (zh)
Other versions
CN108241749A (en
Inventor
徐常亮
傅丕毅
李尉冉
张珺
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Xinhua Zhiyun Technology Co ltd
Original Assignee
Xinhua Zhiyun Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Xinhua Zhiyun Technology Co ltd filed Critical Xinhua Zhiyun Technology Co ltd
Priority to CN201810032456.4A priority Critical patent/CN108241749B/en
Publication of CN108241749A publication Critical patent/CN108241749A/en
Application granted granted Critical
Publication of CN108241749B publication Critical patent/CN108241749B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/36Creation of semantic tools, e.g. ontology or thesauri
    • G06F16/374Thesaurus
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01DMEASURING 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
    • G01D21/00Measuring or testing not otherwise provided for
    • G01D21/02Measuring two or more variables by means not covered by a single other subclass
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/34Browsing; Visualisation therefor
    • G06F16/345Summarisation for human users
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/35Clustering; Classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/30Semantic analysis
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06KRECOGNITION OF DATA; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
    • G06K9/00Methods or arrangements for reading or recognising printed or written characters or for recognising patterns, e.g. fingerprints
    • G06K9/62Methods or arrangements for recognition using electronic means
    • G06K9/6267Classification techniques

Abstract

The invention aims to provide a method and equipment for generating information from sensor data.

Description

Method and apparatus for generating information from sensor data
Technical Field
The present invention relates to the field of computers, and more particularly, to a method and apparatus for generating information from sensor data.
Background
With the large-scale deployment of sensors in social infrastructures, data generated by the sensors is uploaded and stored in large quantities. Converting sensor data into text information and news will improve the efficiency of information exchange and use, but at present, the identification process is still deficient.
Disclosure of Invention
An object of the present invention is to provide a method and apparatus for generating information from sensor data, which can solve the problem of how to accurately and efficiently convert sensor data into information such as news.
According to one aspect of the present invention, there is provided a method of generating information from sensor data, the method comprising:
acquiring one or more types of sensor data;
generating a statistical characteristic set corresponding to each type of sensor data;
and generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sequencing of the keywords.
Further, in the above method, after generating the text information by semantic de-duplication and sorting the keywords, the method further includes:
and generating chart information according to the statistical feature set corresponding to each type of sensor data.
Further, in the above method, generating a statistical feature set corresponding to each type of sensor data includes:
generating a statistical characteristic set corresponding to each type of sensor data;
and training the statistical feature set corresponding to the data of each type of sensor to obtain the optimized statistical feature set.
Further, in the above method, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic deduplication and sorting of the keywords, the method includes:
inputting the statistical feature set corresponding to each type of sensor data into a statistical feature semantic template library to generate one or more corresponding keywords, and generating text information after semantic duplication removal and sorting of the keywords.
Further, in the above method, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic deduplication and sorting of the keywords, the method includes:
extracting abnormal values from the statistical feature set corresponding to each type of sensor data;
and inputting the extracted abnormal values into an abnormal value set semantic template library to generate one or more corresponding keywords, and generating abnormal event text information after semantic de-duplication and sequencing of the keywords.
Further, in the above method, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic deduplication and sorting of the keywords, the method includes:
extracting abnormal values from the statistical feature set corresponding to each type of sensor data;
inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, if so,
and inputting the extracted abnormal values into an emergency semantic template library to generate one or more corresponding keywords, and generating emergency text information after semantic de-duplication and sequencing of the keywords.
Further, in the above method, training the statistical feature set corresponding to each type of sensor data to obtain an optimized statistical feature set includes:
inputting the statistical feature set corresponding to various sensor data into a machine learning module for deep learning training to obtain an optimized statistical feature set;
after the abnormal value is extracted from the statistical feature set corresponding to each type of sensor data, the method further comprises the following steps:
and inputting abnormal values corresponding to various sensor data into a machine learning module for deep learning training to obtain optimized abnormal values.
According to another aspect of the present invention, there is also provided an apparatus for generating information from sensor data, the apparatus including:
the acquisition module is used for acquiring one or more types of sensor data;
the statistical analysis module is used for generating a statistical feature set corresponding to each type of sensor data;
the data set-to-text module is used for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data;
and the automatic writing module is used for generating the text information after semantic de-duplication and sequencing of the keywords.
Further, the device further comprises a data visualization module for generating chart information according to the statistical feature set corresponding to each type of sensor data.
Further, in the above device, the statistical analysis module is configured to generate a statistical feature set corresponding to each type of sensor data; and training the statistical feature set corresponding to the data of each type of sensor to obtain the optimized statistical feature set.
Further, in the above device, the data set-to-text module is configured to input the statistical feature set corresponding to each type of sensor data into a statistical feature semantic template library to generate one or more corresponding keywords.
Further, in the above device, the data set-to-text module is configured to extract an abnormal value from a statistical feature set corresponding to each type of sensor data; inputting the extracted abnormal value into an abnormal value set semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating abnormal event text information after semantic de-duplication and sorting of the keywords.
Further, in the above device, the data set-to-text module is configured to extract an abnormal value from a statistical feature set corresponding to each type of sensor data; inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, and if the abnormal value is the emergency, inputting the extracted abnormal value into an emergency semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating the emergency text information after semantic de-duplication and sequencing of the keywords.
Further, in the above device, the statistical analysis module is further configured to input the statistical feature set corresponding to each type of sensor data into the machine learning module for deep learning training to obtain an optimized statistical feature set;
the data set-to-text module is further used for extracting abnormal values from the statistical feature set corresponding to each type of sensor data, and inputting the abnormal values corresponding to each type of sensor data into the machine learning module for deep learning training to obtain optimized abnormal values.
According to another aspect of the present invention, there is also provided a computing-based device, including:
a processor; and
a memory arranged to store computer executable instructions that, when executed, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical characteristic set corresponding to each type of sensor data;
and generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sequencing of the keywords.
According to another aspect of the present invention, there is also provided a computer-readable storage medium having stored thereon computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical characteristic set corresponding to each type of sensor data;
and generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sequencing of the keywords.
Compared with the prior art, the method and the device have the advantages that the problem that the existing massive sensor data is difficult to uniformly schedule and use can be solved by identifying the statistical characteristic set of the sensor data and converting the statistical characteristic set into the text information such as news, the sensor data can be accurately and efficiently converted into the information such as news, the use efficiency of the sensor data is improved, and the information exchange efficiency is improved.
Drawings
Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:
FIG. 1 illustrates a flow diagram of a method for generating information from sensor data, in accordance with one embodiment of the present invention;
FIG. 2 is a flow chart of a method for generating information from sensor data according to another embodiment of the present invention.
The same or similar reference numbers in the drawings identify the same or similar elements.
Detailed Description
The present invention is described in further detail below with reference to the attached drawing figures.
In a typical configuration of the present application, the terminal, the device serving the network, and the trusted party each include one or more processors (CPUs), input/output interfaces, network interfaces, and memory.
The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium.
Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (transient media), such as modulated data signals and carrier waves.
The invention provides a method for generating information by sensor data, which comprises the following steps:
step S1, acquiring one or more types of sensor data;
here, a plurality of types of sensor data are accessed, as shown in fig. 2, the various types of sensor data include, but are not limited to, types of temperature, humidity, pressure, light sensation, speed measurement, and the like (e.g., 101/102/103 and the like);
step S2, generating a statistical feature set corresponding to each type of sensor data;
the statistical feature set is a set of feature statistics for each type of sensor data according to a plurality of preset sensor index dimensions, for example, for a temperature sensor, the plurality of preset sensor index dimensions may include a region where the temperature of the current day exceeds 17 degrees, temperature change conditions of 11-13 points in a certain region, the temperature of the day of the state of the Hangzhou today, and the like;
as shown in fig. 1, a statistical analysis module may generate a statistical feature set corresponding to each type of sensor data;
as shown in fig. 2, the sensor data set 100 may be passed through a statistical analysis module 200 to generate a set of statistical characteristics 110;
step S3, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sorting the keywords.
Here, as shown in fig. 1, a data set-to-text module may generate one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and then an automatic writing module may generate information by semantically de-duplicating and sorting the keywords.
The invention can solve the problem that the existing mass sensor data is difficult to be uniformly scheduled and used by identifying the statistical characteristic set of the sensor data and converting the statistical characteristic set into character information such as news and the like, realizes the accurate and efficient conversion of the sensor data into the information such as news, improves the use efficiency of the sensor data and improves the efficiency of information exchange.
In an embodiment of the method for generating information from sensor data of the present invention, after the step S2 generating the text information by semantic de-duplication and sorting the keywords, the method further includes:
and generating chart information according to the statistical feature set corresponding to each type of sensor data.
In this case, the chart information corresponding to each type of sensor may be generated for the statistical feature set corresponding to each type of sensor data, or the various types of sensor data may be summarized to generate a piece of summarized chart information.
As shown in fig. 1, a data visualization module may generate chart information according to the statistical feature set corresponding to each type of sensor data, and the automatic writing module may combine the text information and the chart information.
As shown in fig. 2, the statistical feature set processed by the statistical analysis module 200 is matched with different types of templates through the visual template library 500, and a data visualization chart is generated as chart information. The visual template library 500 may include various chart forms and data matching rules corresponding to the chart forms.
The embodiment combines the generated chart information on the basis of generating the character information, so that the generated information is richer, more vivid and vivid.
In an embodiment of the method for generating information from sensor data according to the present invention, in step S2, a statistical feature set corresponding to each type of sensor data is generated, including:
generating a statistical characteristic set corresponding to each type of sensor data;
and training the statistical feature set corresponding to the data of each type of sensor to obtain the optimized statistical feature set.
Here, as shown in fig. 1, a machine learning module may train a statistical feature set corresponding to each type of sensor data to obtain an optimized statistical feature set.
In the embodiment, the statistical feature set is optimized so as to generate one or more corresponding keywords for the statistical feature set after optimization corresponding to each type of sensor data, the keywords are subjected to semantic de-duplication and sorting to generate the text information, and the chart information is generated according to the optimized statistical feature set corresponding to each type of sensor data, so that the generated text information and the chart information are more accurate.
In an embodiment of the method for generating information from sensor data of the present invention, in step S3, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic de-duplication and sorting the keywords, the method includes:
inputting the statistical feature set corresponding to each type of sensor data into a statistical feature semantic template library to generate one or more corresponding keywords, and generating text information after semantic duplication removal and sorting of the keywords.
Here, as shown in fig. 2, the statistical feature set 110 may pass through the statistical feature semantic template library 410, generate a plurality of keywords after data comparison and sorting, and generate text information such as news by matching with the data visualization chart.
In the embodiment, one or more corresponding keywords are generated through the statistical characteristic semantic template library, and the text information is generated after the keywords are subjected to semantic de-duplication and sequencing, so that more accurate text information is further generated more efficiently.
In an embodiment of the method for generating information from sensor data of the present invention, in step S3, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic de-duplication and sorting the keywords, the method includes:
extracting abnormal values from the statistical feature set corresponding to each type of sensor data;
and inputting the extracted abnormal values into an abnormal value set semantic template library to generate one or more corresponding keywords, and generating abnormal event text information after semantic de-duplication and sequencing of the keywords.
Here, as shown in fig. 2, the text information may be generated from the abnormal value set 120 through the abnormal value set semantic template library 420, so as to generate the text information reporting the abnormal condition more efficiently and accurately.
In an embodiment of the method for generating information from sensor data of the present invention, in step S3, generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic de-duplication and sorting the keywords, the method includes:
extracting abnormal values from the statistical feature set corresponding to each type of sensor data;
inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, if so,
and inputting the extracted abnormal values into an emergency semantic template library to generate one or more corresponding keywords, and generating emergency text information after semantic de-duplication and sequencing of the keywords.
Here, as shown in fig. 2, the abnormal value set 120 may be judged whether it is an emergency through the multi-sensor abnormal value statistical comparison module 210, if it is, the abnormal value set 120 is output to the emergency semantic template library 430 to generate a plurality of keywords, and the emergency text information is generated after semantic deduplication and sorting, so as to generate the text information reporting the emergency more efficiently and more accurately.
In an embodiment of the method for generating information from sensor data according to the present invention, the training of the statistical feature set corresponding to each type of sensor data to obtain an optimized statistical feature set includes:
inputting the statistical feature set corresponding to various sensor data into a machine learning module for deep learning training to obtain an optimized statistical feature set;
after the abnormal value is extracted from the statistical feature set corresponding to each type of sensor data, the method further comprises the following steps:
and inputting abnormal values corresponding to various sensor data into a machine learning module for deep learning training to obtain optimized abnormal values.
Here, as shown in fig. 2, the statistical feature set and the abnormal value set may be input to the machine learning module 300 for deep learning training, and the generated model training result includes: the updated statistical feature template and the adjusted abnormal value judgment threshold are matched with each other to obtain an optimized statistical feature set, and the adjusted abnormal value judgment threshold and the input abnormal value set are used to obtain an optimized abnormal value, that is, the optimized statistical feature set and the optimized abnormal value are output from the machine learning module 300 as the sub-data set 130/140/150.
The sub data sets 130/140/150 may then be input into the semantic template library 400 and the visualization template library 500, and the sub data sets may be matched with different categories of semantic templates and visualization templates. For example, the keywords/words 1-keys extracted from the sub-data sets 130, 140, 150 may be semantically de-duplicated/sorted and semantically understood to generate text information.
The semantic template library 400 includes, but is not limited to, the statistical characteristic semantic template library 410, the abnormal value semantic template library 420, the emergency semantic template library 430, and the like, and each template library includes a plurality of artificially labeled semantic templates and various different meaning classification dictionaries.
The semantic template library 400 may include:
α) semantic templates of a plurality of classifications, including but not limited to statistical feature semantic template 410, outlier semantic template 420, incident semantic template 430, etc., each classification corresponding to a generated data set.
b) Each semantic template comprises a plurality of marked information/news under classification and a multidimensional word vector group generated by extracting keywords/words.
c) A variety of different meaning classification dictionaries that may include a machine-learned corpus of utterances (nouns/verbs/adjectives/quantifiers, etc.); synonym queries may also be included.
After matching the sub data sets 130/140/150 with the classifications in the semantic template 400, the sub data sets are used to generate corresponding keywords and are filled into the corresponding semantic template.
As shown in fig. 2, the machine learning module 300 includes two parts:
1) the solidified layer 310 is a multidimensional vector set generated by labeling the statistical feature set template 311 and the abnormal value set template 312 by the solidified layer 310 through a sensor data set segment selected by a news professional;
2) and the autonomous learning layer 320 is used for generating a multi-dimensional vector set generated by the statistical feature set and the abnormal value together with the multi-dimensional vector set of the solidified layer 310 after training through a machine learning model, updating the statistical feature template, adjusting the judgment threshold of the abnormal value, and generating a new sub data set 130/140/150 according to the updated statistical feature template and the adjusted judgment threshold of the abnormal value.
In the embodiment, the statistical feature set and the abnormal values are optimized through the machine learning module, so that one or more corresponding keywords are generated for the statistical feature set and the abnormal values after optimization corresponding to each type of sensor data is performed, the text information is generated after semantic de-duplication and sequencing is performed on the keywords, and the chart information is generated according to the optimized statistical feature set corresponding to each type of sensor data, so that the generated text information and the chart information are more accurate and efficient.
According to another aspect of the present invention, there is also provided an apparatus for generating information from sensor data, wherein the apparatus includes:
the acquisition module is used for acquiring one or more types of sensor data;
here, multiple types of sensor data are accessed, as shown in FIG. 2, including but not limited to
Limited to temperature, humidity, pressure, light sensation, speed measurement, etc. (e.g., 101/102/103, etc.);
the statistical analysis module is used for generating a statistical feature set corresponding to each type of sensor data;
the statistical feature set is a set of feature statistics for each type of sensor data according to a plurality of preset sensor index dimensions, for example, for a temperature sensor, the plurality of preset sensor index dimensions may include a region where the temperature of the current day exceeds 17 degrees, temperature change conditions of 11-13 points in a certain region, the temperature of the day of the state of the Hangzhou today, and the like;
as shown in fig. 1, a statistical analysis module may generate a statistical feature set corresponding to each type of sensor data;
as shown in fig. 2, the sensor data set 100 may be passed through a statistical analysis module 200,
generating a statistical feature set 110;
the data set-to-text module is used for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data;
and the automatic writing module is used for generating the text information after semantic de-duplication and sequencing of the keywords.
Here, as shown in fig. 1, a data set-to-text module may generate one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and then an automatic writing module may generate information by semantically de-duplicating and sorting the keywords.
The invention can solve the problem that the existing mass sensor data is difficult to be uniformly scheduled and used by identifying the statistical characteristic set of the sensor data and converting the statistical characteristic set into character information such as news and the like, realizes the accurate and efficient conversion of the sensor data into the information such as news, improves the use efficiency of the sensor data and improves the efficiency of information exchange.
In an embodiment of the apparatus for generating information from sensor data, the apparatus further includes a data visualization module, configured to generate chart information according to a statistical feature set corresponding to each type of sensor data.
In this case, the chart information corresponding to each type of sensor may be generated for the statistical feature set corresponding to each type of sensor data, or the various types of sensor data may be summarized to generate a piece of summarized chart information.
As shown in fig. 1, a data visualization module may generate chart information according to the statistical feature set corresponding to each type of sensor data, and the automatic writing module may combine the text information and the chart information.
As shown in fig. 2, the statistical feature set processed by the statistical analysis module 200 is matched with different types of templates through the visual template library 500, and a data visualization chart is generated as chart information. The visual template library 500 may include various chart forms and data matching rules corresponding to the chart forms.
The embodiment combines the generated chart information on the basis of generating the character information, so that the generated information is richer, more vivid and vivid.
In an embodiment of the apparatus for generating information from sensor data according to the present invention, the statistical analysis module is configured to generate a statistical feature set corresponding to each type of sensor data; and training the statistical feature set corresponding to the data of each type of sensor to obtain the optimized statistical feature set.
Here, as shown in fig. 1, a machine learning module may train a statistical feature set corresponding to each type of sensor data to obtain an optimized statistical feature set.
In the embodiment, the statistical feature set is optimized so as to generate one or more corresponding keywords for the statistical feature set after optimization corresponding to each type of sensor data, the keywords are subjected to semantic de-duplication and sorting to generate the text information, and the chart information is generated according to the optimized statistical feature set corresponding to each type of sensor data, so that the generated text information and the chart information are more accurate.
In an embodiment of the apparatus for generating information from sensor data according to the present invention, the data set-to-text module is configured to input a statistical feature set corresponding to each type of sensor data into a statistical feature semantic template library to generate one or more corresponding keywords.
Here, as shown in fig. 2, the statistical feature set 110 may pass through the statistical feature semantic template library 410, generate a plurality of keywords after data comparison and sorting, and generate text information such as news by matching with the data visualization chart.
In the embodiment, one or more corresponding keywords are generated through the statistical characteristic semantic template library, and the text information is generated after the keywords are subjected to semantic de-duplication and sequencing, so that more accurate text information is further generated more efficiently.
In an embodiment of the apparatus for generating information from sensor data according to the present invention, the data set-to-text module is configured to extract an abnormal value from a statistical feature set corresponding to each type of sensor data; inputting the extracted abnormal value into an abnormal value set semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating abnormal event text information after semantic de-duplication and sorting of the keywords.
Here, as shown in fig. 2, the text information may be generated from the abnormal value set 120 through the abnormal value set semantic template library 420, so as to generate the text information reporting the abnormal condition more efficiently and accurately.
In one embodiment of the device for generating information from sensor data, the data set-to-text module is used for extracting abnormal values from statistical feature sets corresponding to each type of sensor data; inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, and if the abnormal value is the emergency, inputting the extracted abnormal value into an emergency semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating the emergency text information after semantic de-duplication and sequencing of the keywords.
Here, as shown in fig. 2, the abnormal value set 120 may be judged whether it is an emergency through the multi-sensor abnormal value statistical comparison module 210, if it is, the abnormal value set 120 is output to the emergency semantic template library 430 to generate a plurality of keywords, and the emergency text information is generated after semantic deduplication and sorting, so as to generate the text information reporting the emergency more efficiently and more accurately.
In an embodiment of the apparatus for generating information from sensor data, the statistical analysis module is further configured to input the statistical feature set corresponding to each type of sensor data into the machine learning module for deep learning training to obtain an optimized statistical feature set;
the data set-to-text module is further used for extracting abnormal values from the statistical feature set corresponding to each type of sensor data, and inputting the abnormal values corresponding to each type of sensor data into the machine learning module for deep learning training to obtain optimized abnormal values.
Here, as shown in fig. 2, the statistical feature set and the abnormal value set may be input to the machine learning module 300 for deep learning training, and the generated model training result includes: the updated statistical feature template and the adjusted abnormal value judgment threshold are matched with each other to obtain an optimized statistical feature set, and the adjusted abnormal value judgment threshold and the input abnormal value set are used to obtain an optimized abnormal value, that is, the optimized statistical feature set and the optimized abnormal value are output from the machine learning module 300 as the sub-data set 130/140/150.
The sub data sets 130/140/150 may then be input into the semantic template library 400 and the visualization template library 500, and the sub data sets may be matched with different categories of semantic templates and visualization templates. For example, the keywords/words 1-keys extracted from the sub-data sets 130, 140, 150 may be semantically de-duplicated/sorted and semantically understood to generate text information.
The semantic template library 400 includes, but is not limited to, the statistical characteristic semantic template library 410, the abnormal value semantic template library 420, the emergency semantic template library 430, and the like, and each template library includes a plurality of artificially labeled semantic templates and various different meaning classification dictionaries.
The semantic template library 400 may include:
a) semantic templates for multiple classifications, including but not limited to statistical feature semantic template 410, outlier semantic template 420, incident semantic template 430, etc., each classification corresponding to a generated dataset.
b) Each semantic template comprises a plurality of marked information/news under classification and a multidimensional word vector group generated by extracting keywords/words.
c) A variety of different meaning classification dictionaries that may include a machine-learned corpus of utterances (nouns/verbs/adjectives/quantifiers, etc.); synonym queries may also be included.
After matching the sub data sets 130/140/150 with the classifications in the semantic template 400, the sub data sets are used to generate corresponding keywords and are filled into the corresponding semantic template.
As shown in fig. 2, the machine learning module 300 includes two parts:
1) the solidified layer 310 is a multidimensional vector set generated by labeling the statistical feature set template 311 and the abnormal value set template 312 by the solidified layer 310 through a sensor data set segment selected by a news professional;
2) and the autonomous learning layer 320 is used for generating a multi-dimensional vector set generated by the statistical feature set and the abnormal value together with the multi-dimensional vector set of the solidified layer 310 after training through a machine learning model, updating the statistical feature template, adjusting the judgment threshold of the abnormal value, and generating a new sub data set 130/140/150 according to the updated statistical feature template and the adjusted judgment threshold of the abnormal value.
In the embodiment, the statistical feature set and the abnormal values are optimized through the machine learning module, so that one or more corresponding keywords are generated for the statistical feature set and the abnormal values after optimization corresponding to each type of sensor data is performed, the text information is generated after semantic de-duplication and sequencing is performed on the keywords, and the chart information is generated according to the optimized statistical feature set corresponding to each type of sensor data, so that the generated text information and the chart information are more accurate and efficient.
According to another aspect of the present invention, there is also provided a computing-based device, including:
a processor; and
a memory arranged to store computer executable instructions that, when executed, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical characteristic set corresponding to each type of sensor data;
and generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sequencing of the keywords.
According to another aspect of the present invention, there is also provided a computer-readable storage medium having stored thereon computer-executable instructions, wherein the computer-executable instructions, when executed by a processor, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical characteristic set corresponding to each type of sensor data;
and generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic de-duplication and sequencing of the keywords.
For details of the embodiments of the apparatus and the computer-readable storage medium, reference may be made to corresponding parts of the embodiments of the methods, which are not described herein again.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.
It should be noted that the present invention may be implemented in software and/or in a combination of software and hardware, for example, as an Application Specific Integrated Circuit (ASIC), a general purpose computer or any other similar hardware device. In one embodiment, the software program of the present invention may be executed by a processor to implement the steps or functions described above. Also, the software programs (including associated data structures) of the present invention can be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Further, some of the steps or functions of the present invention may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.
In addition, some of the present invention can be applied as a computer program product, such as computer program instructions, which when executed by a computer, can invoke or provide the method and/or technical solution according to the present invention through the operation of the computer. Program instructions which invoke the methods of the present invention may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the invention herein comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or solution according to embodiments of the invention as described above.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims (12)

1. A method of generating information from sensor data, wherein the method comprises:
acquiring one or more types of sensor data;
generating a statistical feature set corresponding to each type of sensor data, wherein the generating of the statistical feature set corresponding to each type of sensor data comprises: generating a statistical characteristic set corresponding to each type of sensor data; training the statistical feature set corresponding to the data of each type of sensor to obtain an optimized statistical feature set;
generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic duplication removal and sorting of the keywords; wherein the information includes: news;
the method for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data and generating text information after semantic de-duplication and sorting of the keywords comprises the following steps: extracting abnormal values from the statistical feature set corresponding to each type of sensor data; and inputting the extracted abnormal values into an abnormal value set semantic template library to generate one or more corresponding keywords, and generating abnormal event text information after semantic de-duplication and sequencing of the keywords.
2. The method of claim 1, wherein after generating the text information by semantic de-duplication and sorting, further comprising:
and generating chart information according to the statistical feature set corresponding to each type of sensor data.
3. The method of claim 1 or 2, wherein generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic deduplication and sorting of the keywords comprises:
inputting the statistical feature set corresponding to each type of sensor data into a statistical feature semantic template library to generate one or more corresponding keywords, and generating text information after semantic duplication removal and sorting of the keywords.
4. The method of claim 1, wherein generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information by semantic deduplication and sorting of the keywords comprises:
extracting abnormal values from the statistical feature set corresponding to each type of sensor data;
inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, if so,
and inputting the extracted abnormal values into an emergency semantic template library to generate one or more corresponding keywords, and generating emergency text information after semantic de-duplication and sequencing of the keywords.
5. The method of claim 1 or 4, wherein training the statistical feature set corresponding to each type of sensor data to obtain the optimized statistical feature set comprises:
inputting the statistical feature set corresponding to various sensor data into a machine learning module for deep learning training to obtain an optimized statistical feature set;
after the abnormal value is extracted from the statistical feature set corresponding to each type of sensor data, the method further comprises the following steps:
and inputting abnormal values corresponding to various sensor data into a machine learning module for deep learning training to obtain optimized abnormal values.
6. An apparatus for generating information from sensor data, wherein the apparatus comprises:
the acquisition module is used for acquiring one or more types of sensor data;
the statistical analysis module is used for generating a statistical feature set corresponding to each type of sensor data, wherein the statistical analysis module is used for generating the statistical feature set corresponding to each type of sensor data; training the statistical feature set corresponding to the data of each type of sensor to obtain an optimized statistical feature set;
the data set-to-text module is used for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data;
the automatic writing module is used for generating text information after semantic de-duplication and sequencing of the keywords; wherein the information includes: news;
the data set-to-text module is used for extracting abnormal values from the statistical feature set corresponding to each type of sensor data; inputting the extracted abnormal value into an abnormal value set semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating abnormal event text information after semantic de-duplication and sorting of the keywords.
7. The apparatus of claim 6, further comprising a data visualization module for generating chart information based on the set of statistical features corresponding to each type of sensor data.
8. The device of claim 6 or 7, wherein the data set-to-text module is configured to input the corresponding statistical feature set for each type of sensor data into a statistical feature semantic template library to generate the corresponding one or more keywords.
9. The device of claim 6, wherein the data set-to-text module is configured to extract outliers from the corresponding statistical feature set for each type of sensor data; inputting the extracted abnormal value into a multi-sensor abnormal value statistic comparison module, judging whether the abnormal value is an emergency or not, and if the abnormal value is the emergency, inputting the extracted abnormal value into an emergency semantic template library to generate one or more corresponding keywords;
and the automatic writing module is used for generating the emergency text information after semantic de-duplication and sequencing of the keywords.
10. The device according to claim 6 or 9, wherein the statistical analysis module is further configured to input the statistical feature set corresponding to each type of sensor data into the machine learning module for deep learning training to obtain an optimized statistical feature set;
the data set-to-text module is further used for extracting abnormal values from the statistical feature set corresponding to each type of sensor data, and inputting the abnormal values corresponding to each type of sensor data into the machine learning module for deep learning training to obtain optimized abnormal values.
11. A computing-based device, comprising:
a processor; and
a memory arranged to store computer executable instructions that, when executed, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical feature set corresponding to each type of sensor data, wherein the generating of the statistical feature set corresponding to each type of sensor data comprises: generating a statistical characteristic set corresponding to each type of sensor data; training the statistical feature set corresponding to the data of each type of sensor to obtain an optimized statistical feature set;
generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic duplication removal and sorting of the keywords; wherein the content of the first and second substances,
the information includes: news;
the method for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data and generating text information after semantic de-duplication and sorting of the keywords comprises the following steps: extracting abnormal values from the statistical feature set corresponding to each type of sensor data; inputting the extracted abnormal value into an abnormal value set semantic template library to generate one or more corresponding key words,
and generating abnormal event text information after semantic de-duplication and sorting of the keywords.
12. A computer-readable storage medium having computer-executable instructions stored thereon, wherein the computer-executable instructions, when executed by a processor, cause the processor to:
acquiring one or more types of sensor data;
generating a statistical feature set corresponding to each type of sensor data, wherein the generating of the statistical feature set corresponding to each type of sensor data comprises: generating a statistical characteristic set corresponding to each type of sensor data; training the statistical feature set corresponding to the data of each type of sensor to obtain an optimized statistical feature set;
generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data, and generating text information after semantic duplication removal and sorting of the keywords; wherein the information includes: news;
the method for generating one or more corresponding keywords according to the statistical feature set corresponding to each type of sensor data and generating text information after semantic de-duplication and sorting of the keywords comprises the following steps: extracting abnormal values from the statistical feature set corresponding to each type of sensor data; and inputting the extracted abnormal values into an abnormal value set semantic template library to generate one or more corresponding keywords, and generating abnormal event text information after semantic de-duplication and sequencing of the keywords.
CN201810032456.4A 2018-01-12 2018-01-12 Method and apparatus for generating information from sensor data Active CN108241749B (en)

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 CN108241749A (en) 2018-07-03
CN108241749B true 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)

Family Cites Families (11)

* Cited by examiner, † Cited by third party
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
JP5904021B2 (en) * 2012-06-07 2016-04-13 ソニー株式会社 Information processing apparatus, electronic device, information processing method, and program
CN103177391B (en) * 2013-01-31 2016-01-27 中国人民解放军后勤工程学院 A kind of facilities and equipment supervisory system based on three-dimensional live and system integration method
CN103390039B (en) * 2013-07-17 2017-02-08 北京建筑工程学院 Urban disaster thematic map real-time generating method based on network information
CN103700054B (en) * 2013-12-10 2017-08-25 中国地质大学武汉 A kind of Sudden Geological Hazards emergency preplan digitization system
CN103678910B (en) * 2013-12-12 2014-12-10 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
CN105791343B (en) * 2014-12-19 2019-09-03 中国移动通信集团公司 Intelligent home information interactive system, method and relevant device
CN105446156B (en) * 2015-12-30 2018-09-07 百度在线网络技术(北京)有限公司 Control method, the device and system of household electrical 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
CN106649455B (en) * 2016-09-24 2021-01-12 孙燕群 Standardized system classification and command set system for big data development

Also Published As

Publication number Publication date
CN108241749A (en) 2018-07-03

Similar Documents

Publication Publication Date Title
AU2016203856B2 (en) System and method for automating information abstraction process for documents
CN104778158B (en) A kind of document representation method and device
US8712926B2 (en) Using rule induction to identify emerging trends in unstructured text streams
CN108319668B (en) Method and equipment for generating text abstract
KR101983538B1 (en) Systems and methods for calculating category proportions
CN107436922A (en) Text label generation method and device
JP5534280B2 (en) Text clustering apparatus, text clustering method, and program
CN105975459B (en) A kind of the weight mask method and device of lexical item
CN106598999B (en) Method and device for calculating text theme attribution degree
CN108573045A (en) A kind of alignment matrix similarity retrieval method based on multistage fingerprint
TW202029079A (en) Method and device for identifying irregular group
US10242107B2 (en) Extraction of quantitative data from online content
Nguyen et al. Real-time event detection using recurrent neural network in social sensors
CN107357895B (en) Text representation processing method based on bag-of-words model
Bafna et al. Hindi Multi-document Word Cloud based Summarization through Unsupervised Learning
CN112559739A (en) Method for processing insulation state data of power equipment
US10956469B2 (en) System and method for metadata correlation using natural language processing
CN108241749B (en) Method and apparatus for generating information from sensor data
CN110674297A (en) Public opinion text classification model construction method, public opinion text classification device and public opinion text classification equipment
CN106599002B (en) Topic evolution analysis method and device
CN111475641B (en) Data extraction method and device, storage medium and equipment
CN108108371B (en) Text classification method and device
US20210004385A1 (en) System and method for analysis of one or more unstructured data
CN108052503B (en) Confidence coefficient calculation method and device
CN107729509B (en) Discourse similarity determination method based on recessive high-dimensional distributed feature representation

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