KR20200069806A - Method and apparatus for statistical data visualization through voice recognition - Google Patents

Abstract

Disclosed are a method and an apparatus for statistical data visualization through voice recognition. The statistical data visualization method through voice recognition includes the steps of: receiving, from a database, a mapping relationship between a control command of a statistical program for generating statistical data by performing statistical processing on data and a voice command from a user; recognizing a voice from the user based on the received mapping relationship, and extracting a function for visualizing the statistical data from the user′s voice and function-specific factors; generating a control command for visualizing the statistical data by combining the extracted function and the function-specific factors based on the received mapping relationship, and inputting the generated control command to the statistical program; and receiving a result of executing the control command from the statistical program and providing the result to the user. According to the present invention, a voice recognition technology is applied through a mobile device to check statistical information on outliers by using a graph, so that a smart factory is monitored in real time and abnormal behaviors are detected.

Description

Method and apparatus for statistical data visualization through voice recognition}

The present invention relates to a method and apparatus for visualizing statistical data, and in particular, by developing a mapping logic of a user's voice and a statistical graph generating program command and allowing the user to control the statistical program through voice based on the developed logic, the statistical program To users who do not have expertise in managing big data efficiently through voice, and furthermore, it is possible to visualize statistical data through voice recognition method and device to allow smart factories to easily monitor the abnormal operation status and take precautions. It is about.

In the field of smart factories, there is a need to analyze defect diagnosis, which is a main issue in real time, not only in Korea but also worldwide.

The global market for IoT-based smart factories is expected to grow by 6.8% annually, creating economic value of $287.8 billion in 2020. In addition, the device and ICT markets that make up the smart factory supply market are also expected to record annual growth rates of 7.8% and 8.1%, respectively. As a result, it can be expected that in the smart factory and related device markets, economic value of $250 billion from 2018 to $250 billion by 2020 will be created.

In addition, in Asia, the manufacturing factories of major companies in the world are located in developing countries in Asia with low labor costs, so it is expected that the introduction of smart factories to improve the productivity of manufacturing processes will be faster than other regions. Most of the high ICT market share is attributed to China, and China is expected to overtake the European market in 2016 and the Americas in 2019 due to the expansion of the smart factory.

Therefore, the government-wide support policy is expanding to secure the competitiveness of the manufacturing industry due to this social structural change, and large enterprises in the manufacturing sector are also accelerating the introduction of smart factories to secure their own productivity and increase efficiency.

In particular, Korea's smart factory market is expected to grow at an annual average rate of 11.2% by 2020, and the domestic market is expected to generate economic value of $4.12 billion from 2018 to $5.4 billion by 2020. In addition, the German Acatech report pointed to Korea as one of the countries with the size and environment to supply Industry 4.0 solutions, which also emphasized the need to establish a strategy for manufacturing.

Smart factories essentially require systems to manage vast amounts of data from various devices, and these systems enable monitoring and management of smart factories anytime, anywhere via mobile devices. That is, since convenient data analysis is possible with a mobile device, it can be applied to overseas markets according to the needs of countries around the world through interworking with various languages.

In addition, in the era of big data and IoT, analytical languages (R, Python, etc.) and solutions are being developed to support more effective decision making by utilizing various data. Leading the trend of the times.

However, a narrow input device (virtual keyboard) applied to a mobile device has a limitation in analyzing programming languages including various symbols through coding. Therefore, the application of automated programming technology utilizing speech recognition technology can lead to the emergence of a new smart factory source technology for countries around the world.

Republic of Korea Patent Application No. 1020000060732 (published on April 24, 2002) "Method for controlling computer program using speech recognition" provides a method for controlling a computer program using speech recognition. Particularly, this application allows the user to control the computer's operating system and various application programs by using only voice commands of the user by using voice recognition technology without a hand-operated input device such as a keyboard or mouse. It provides a method of controlling a computer program using speech recognition that can conveniently utilize various programs as well as a computer system.

However, this application only deals with programming through voice recognition, and has not disclosed a technique for managing and controlling big data of a smart factory.

Therefore, by applying voice recognition technology through a mobile device (smartphone, tablet PC, etc.), it is recognized by speaking speech (using natural language processing technology) to utilize data to check statistical information about outliers in a graph. Technology is desperately needed.

Republic of Korea Patent Application No. 1020000060732

The object of the present invention is to apply a voice recognition technology through a mobile device (smartphone, tablet PC, etc.), and if it is spoken by voice, it recognizes it and visualizes statistical information about outliers through voice recognition. Is to provide

In addition, another object of the present invention, using a speech recognition technology (SRT: Speech Recognition Technology) and natural language processing technology (NLP: Natural Language Processing) to monitor the smart factory in real time through data visualization with voice and detect abnormal behavior It is to provide a statistical data visualization device through speech recognition.

One aspect of the present invention for achieving the above object, relates to a method of visualizing statistical data through speech recognition. A method of visualizing statistical data through speech recognition according to the present invention comprises: receiving a mapping relationship between a control command of a statistical program that generates statistical data by performing statistical processing on data from a database, and a voice command from a user; Recognizing a voice from a user based on the received mapping relationship, extracting a function for visualizing the statistical data from the user's voice and the function-specific factors; Generating a control command for visualizing statistical data by combining the extracted function and function-specific factors based on the received mapping relationship, and inputting the generated control command to the statistical program; And receiving a result of performing the control command from the statistical program and providing the result to the user. In particular, the database stores a primary mapping relationship between a user's voice command and a standardization command that standardized control commands of a plurality of statistical programs, and a secondary mapping relationship between the standardization command and control commands of each statistical program. In addition, the extracting step may classify a situation in which the control command is used based on the relationship between the number of arguments and arguments of a function constituting the control command, and extract the argument of the function according to the classified situation do. Furthermore, the speech recognition is performed using natural language processing technology installed in a mobile device. In particular, the voice recognition may be performed by utilizing an application programming interface (API) of a voice recognition server connected through a network, and the method is used to control a smart factory. In addition, the method for visualizing statistical data through speech recognition according to the present invention may further include detecting an abnormal operation of the smart factory from a result of performing the control command.

Another aspect of the present invention for achieving the above object relates to a statistical data visualization apparatus through speech recognition. The apparatus for visualizing statistical data through voice recognition according to the present invention includes a database storing a mapping relationship between a control command of a statistical program that performs statistical processing on data and a voice command from a user; A voice recognition unit that recognizes a voice from a user based on the received mapping relationship and extracts a function for visualizing statistical data from the user's voice and the function-specific factors; A visualization control command generator for generating a control command for visualizing statistical data by combining the extracted function and function-specific factors based on the received mapping relationship, and inputting the generated control command to the statistical program; An interface module that receives a result of performing the control command from the statistical program and provides it to a user; And a central control unit controlling the components. In particular, the database stores a primary mapping relationship between a user's voice command and a standardization command that standardized control commands of a plurality of statistical programs, and a secondary mapping relationship between the standardization command and control commands of each statistical program. In addition, the speech recognition unit classifies a situation in which the control command is used based on the relationship between the number of factors of the function constituting the control command and factors, and extracts the function parameter according to the classified situation. . The speech recognition may be performed using a natural language processing technology installed in a mobile device, or may be performed using an application programming interface (API) of a speech recognition server connected through a network. Furthermore, the apparatus for visualizing statistical data through speech recognition may further include an abnormal operation detection unit that detects an abnormal operation of the smart factory from a result of performing the control command, and the statistical program includes an R program and a Python. It may be at least one of.

According to the present invention, by applying a voice recognition technology through a mobile device, it is possible to recognize the spoken words in speech and check statistical information on anomalies, so that the smart factory can be monitored in real time and abnormal behaviors can be detected.

In addition, according to the present invention, a service for generating a statistical graph with speech recognition applied to a mobile device becomes possible, and the service can be applied to various mobile services in the smart factory market.

Furthermore, according to the present invention, it is possible to pre-defect defects because the inexperienced person in data analysis can easily analyze the factory data through voice recognition, and furthermore, it is possible to develop advanced data analysis scenarios and technical ripple through a combination of languages in each country. The effect is great.

1 is a flowchart schematically showing a method of visualizing statistical data through speech recognition according to an aspect of the present invention.
2 is a view for explaining the operating environment of the statistical data visualization method through speech recognition according to the present invention.
3 illustrates a visualization function of a statistical program used in a method for visualizing statistical data through speech recognition according to the present invention.
4 illustrates an ordering system of factors of a statistical program used in a method of visualizing statistical data through speech recognition according to the present invention.
5 illustrates a mapping relationship between a function and a voice command that enables a user to issue a standardized command in a method for visualizing statistical data through voice recognition according to the present invention.
FIG. 6 is a view showing the mapping relationship illustrated in FIG. 5 in more detail.
7 is a block diagram schematically showing an apparatus for visualizing statistical data through speech recognition according to another aspect of the present invention.

In order to fully understand the present invention, the operational advantages of the present invention, and the objects achieved by the practice of the present invention, reference should be made to the accompanying drawings and the contents described in the accompanying drawings, which illustrate preferred embodiments of the present invention.

Hereinafter, the present invention will be described in detail by explaining preferred embodiments of the present invention with reference to the accompanying drawings. However, the present invention can be implemented in various different forms, and is not limited to the described embodiments. And, in order to clearly describe the present invention, parts irrelevant to the description are omitted, and the same reference numerals in the drawings indicate the same members.

1 is a flowchart schematically showing a method of visualizing statistical data through speech recognition according to an aspect of the present invention.

The method of visualizing statistical data through speech recognition according to an aspect of the present invention makes it easy to analyze a diagnosis that is a main issue in a smart factory. That is, the method of visualizing statistical data through speech recognition uses scenario-specific mapping logic that maps visualization commands of speech and statistical programs and generates graphs. The operating environment of the present invention will be briefly described with reference to FIG. 2.

2 is a view for explaining the operating environment of the statistical data visualization method through speech recognition according to the present invention.

Referring to FIG. 2, first, data transmitted and received between the MES system and the U-FPS server is transmitted to the monitoring system. The method of visualizing statistical data through speech recognition may operate as a module of the monitoring system.

MES (Manufacturing Execution System) is called an integrated production management system, and is a system that manages all production activities, from initiation by product order to quality inspection of finished products. Using the MES system, it is possible to collect, analyze, and collect in real time various types of information on production sites, such as production performance, worker activities, facility operation, and product quality information.

In addition, MES interfaces with automation systems that enable automation by controlling equipment, transportation facilities, PLCs, and various devices below, and the top is an enterprise-wide interface with a planning system or an enterprise resource planning (ERP) system. It plays a role of integrating control from human resource management to the lowest production equipment.

Human-machine interface (HMI) refers to the operator interface system, and acts as an interface that connects the world of human cognition associated with vision or hearing to the world of machines that process digital computers or communication through human machine interfaces. Do it. The HMI of FIG. 2 may serve to collect sensor data.

The U-FPS server is composed of a process program, a data mining program, a machine learning model, and the like, and collects and stores operation performance data of various facilities of a smart factory, and also stores operation recipes. When signs of abnormal operation of the facilities of the smart factory are found, the U-FPS server also serves to send an alarm alarm to the smart factory. For this, a machine learning model can be applied. The machine learning model can generate an output file and perform machine learning using the selected model to increase diagnostic accuracy.

As can be seen in FIG. 2, the method of visualizing statistical data through speech recognition of the present invention creates and visualizes a scenario that even a beginner can easily analyze through speech recognition. In addition, since the statistical data can be easily visualized through voice recognition, even users who do not have expert knowledge of the statistical program can easily analyze the big data and extract outliers.

In general, when learning big data-related languages such as R, practice is essential, but when learning on mobile, it is difficult to practice the program because the keyboard is inconvenient. In order to overcome these shortcomings, a technology capable of programming through voice is required. Most of the high-level programming languages (HighLevel Program Language) are developed to be easily recognized by humans, and in particular, in the case of functional languages, they are often structured to be programmable similar to what is said. That is, the method of visualizing statistical data through speech recognition creates standardization rules for mapping between natural language processing technology and R program visualization function, and builds a data visualization scenario DB that even beginners can utilize.

Returning to FIG. 1 again, the method of visualizing statistical data through speech recognition of the present invention will be described in detail.

First, a mapping relationship between a control command of a statistical program that generates statistical data by performing statistical processing on data and a voice command from a user is received from the database (S110).

Although the R program and Python are exemplified as statistical programs in the present invention, it will be understood by those skilled in the art that this is for illustration only, and the present invention is not limited to them and can be applied to various statistical programs. In particular, R is an open source program and is a language for statistics/data mining and graphing. The R program is mainly used for research and industry-specific applications, and recently it has been used by many companies. In particular, it is receiving attention for the purpose of analyzing big data, and more than 5000 packages (sort of applications) support various functions and are updated frequently.

This R program, which is being developed as an open source project, is based on the S language, a statistical programming language, and is suitable for statistical calculation and result generation graphics. In addition, as a developer, R is compatible with other development languages and can be processed in real time in conjunction with the web, so it is useful for developing new applications or providing them as web services, which are accompanied by economic benefits due to cost reduction. .

In order to build an analysis system in the present invention, first of all, research and DBization are required, and scenarios must be designed. Then, based on each scenario, an ordering system must be created and standardized logic should be developed using it. Each step is described in detail as follows.

-Research and DB

At this stage, the main functions and factors of R for data visualization should be investigated and studied. In addition, functions and parameters necessary for research suitable for use in the operation of the smart factory are selected.

To this end, a survey can be conducted on the visualization function. For example, about 100 programmers using R programs can be selected to provide a script created to check the recognition range of visualization functions and parameters. In addition, the programmer can build a database by collecting data that directly reads the script.

-Scenario design

In order to build a database, the number of required parameters for each visualization function selected in the R program is grasped, and a user's command scenario is designed according to the number of these parameters. For example, first, after recognizing the name of the visualization function, the user may be instructed to issue a voice command to the user sequentially as many as necessary factors. The scenario design process is preferably performed through a bidirectional interface.

3 illustrates a visualization function of a statistical program used in a method for visualizing statistical data through speech recognition according to the present invention.

Referring to FIG. 3, visualization functions plot, barplot, boxplot, and necessary parameters for each function are illustrated. For example, the argument of the plot is x, y, and the description shows that these x, y represent the xy coordinates for the picture. In addition, it can be seen that the height, which is an argument of the barplot function, represents a target data object to be represented by a bar, and the argument col is a color of a bar. In this way, the outline argument of the bloxplot function serves to print outliers as dots.

When the visualization functions used in the R program and the parameters of each visualization function are arranged as shown in FIG. 3, a scenario that is input according to the number and nature of each parameter may be configured differently. In addition, in order to construct a differentiated scenario, essential and optional factors for the visualization function must be investigated. Therefore, in the case of the selection factor, even if it is not necessarily input, it is possible to prevent further prompting of the voice command in the method of visualizing statistical data through voice recognition.

In this specification, input scenarios for a total of 18 functions are disclosed, and descriptions of each function (types and examples of applicable data, function functions, and parameter classification) and scenarios for function use in the form of an algorithmic system diagram. Introduce. However, this is for illustration only and does not limit the invention.

-Order system DB creation

When the input scenario of the visualization function is prepared, it is determined that the order of using the parameters of each function is different according to the characteristics of each user, and the main factor order system of the visualization function for each context is constructed based on the designed scenario.

4 illustrates an ordering system of factors of a statistical program used in a method of visualizing statistical data through speech recognition according to the present invention.

Referring to FIG. 4, essential and optional factors are systematically classified for each selected visualization function. For example, in the case of the pie function, there are no required parameters, and it can be seen that main, xlab, and ylab are optional parameters. Similarly, in the case of the dotchart function, there are five main, xlab, ylab, color, and pch as optional arguments. As can be seen in Figure 4, each selection factor is not necessarily input in order. Therefore, an input scenario is appropriately used to accurately distinguish the inputted factors regardless of the order.

According to this input scenario, in the case of the pie function, it is possible to sequentially input the selection factors of main, xlab, and ylab for input. In addition, in the case of the dotchart function, it is possible to induce to input the input parameters of main, xlab, ylab, color, and pch as a selection factor.

Referring back to FIG. 1, a voice from a user is recognized based on a mapping relationship received from a database, and functions and function-specific factors for visualizing statistical data are extracted from the user's voice (S130 ).

In this specification, a total of 343 ordering systems were constructed for a total of 18 functions and about 50 factors, but this is for illustration only and does not limit the present invention.

-Visualization function standardization logic

When the input system is built, the voice standardization logic of the function is defined by limiting various speech expression methods of the visualization function of the R program to about three expression methods that are most expressed in order to increase the speech recognition rate of the mobile device. As described above, in the method of visualizing statistical data through voice recognition according to the present invention, voice commands for functions and parameters may be sequentially received through a two-way interface with a user.

5 illustrates a mapping relationship between a function and a voice command that enables a user to issue a standardized command in a method for visualizing statistical data through voice recognition according to the present invention, and FIG. 6 illustrates the mapping relationship illustrated in FIG. 5. It is a drawing to be displayed in more detail.

5 and 6, it is possible to know the standardized command for each function. Some of the standardized commands in FIG. 5 are listed as follows.

Function pie:'one graph' or'pie'

Function dotchart:'dot graph' or'dot chart'

Function hist:'Histogram' or'Hist'

Function plot3d: '3D graph' or'Dynamic 3D graph'

Referring to FIG. 5, in the absence of such a standardized command, the function plot3d may be read as'Plot Three-Dee', in which case the desired function may not be properly called. Therefore, the operation accuracy of the voice command is significantly improved through the standardization command.

Speech recognition may be performed using a natural language processing technology installed in a mobile device, or may be performed using an application programming interface (API) of a voice recognition server connected through a network.

6, standardized commands for factors commonly used in many functions will be described in more detail.

Referring to FIG. 6, it can be seen that the argument x used in functions pie, dotchart, heatmap, boxplot, etc. is read as'data' or'x' as a standardization command. On the other hand, it can be seen that the argument x used in the functions slpom, cloud, xyplot, and densityplot reads as'individual item name','item name', or'x'. Since the use of the same argument is different in the function used, the input scenario described above is needed to distinguish this use.

The database may operate the voice command by storing a primary mapping relationship between a user's voice command and a standardized command that standardized control commands of a plurality of statistical programs and a secondary mapping relationship between a standardized command and control commands of each statistical program. .

When such a standardized command is recognized, a control command for visualizing statistical data by combining the extracted function and function-specific arguments is generated based on the mapping relationship received from the database, and the generated control command is input to the statistical program (S150)

The control command is constructed with reference to the input scenario as described above. That is, the argument x used in functions pie, dotchart, heatmap, boxplot, etc. and the argument x used in functions slpom, cloud, xyplot, and densityplot can be input through other standardized commands. Then, each factor is distinguished according to the input scenario, and control commands according to the function to be used are constructed and input to the statistical program. Now, a statistical program, such as an R program, receives each standardized control command and performs normal statistical operations. Then, statistical data is provided to the user.

Then, the result of performing the control command from the statistical program is received and provided to the user (S170).

The received result can be used to control the smart factory.

In addition, the method of visualizing statistical data through speech recognition may further include a step (S190) of detecting an abnormal operation of the smart factory from a result of performing a control command. As described above, according to the present invention, it is possible to control the statistical program by performing speech recognition through standardized commands without having to know different commands for each statistical program. Therefore, it is possible to easily identify the abnormal operation of the smart factory through visualized information without specialized personnel and take precautions.

7 is a block diagram schematically showing an apparatus for visualizing statistical data through speech recognition according to another aspect of the present invention.

Referring to FIG. 7, the statistical data visualization apparatus 700 through voice recognition includes a statistical program module 710, an interface 720, a voice recognition unit 730, a central control unit 750, and a visualization command generation unit 760. , An abnormal motion detection unit 770, and a database 790.

As described above, the database 790 may be built through a research study in which a statistical program expert is queried for key functions and parameters for data visualization. Visualization functions and parameters are stored in the database 790 through the results obtained in this process. Then, the statistical program discriminates input scenarios according to the number of parameters for each visualization function, and induces input of functions and factors according to the differentiated scenarios. When the input scenario is prepared, it is possible to sequentially input the main factors of the contextual visualization function based on the designed scenario. Therefore, the database 790 also stores input scenarios for each function.

In addition, the database 790 also constructs mapping standardization logic between voice and statistical program commands. As described above, the database 790 may store a primary mapping relationship between a user's voice command and a standardization command that standardizes control commands of a plurality of statistical programs, and a secondary mapping relationship between the standardization command and control commands of each statistical program. have.

The voice recognition unit 730 recognizes the voice from the user based on the mapping relationship received from the database 790. Since there is a voice recognition unit 730, it is possible to overcome the limitations of the narrow input device applied to the mobile device. The speech recognition unit 730 may utilize speech recognition technology (SRT) and natural language processing (NLP). In particular, the voice recognition unit 730 may utilize an application programming interface (API) of a voice recognition server connected through a network. For example, using the speech recognition API provided by Google, the speech recognition function can be easily implemented. In this case, the user's voice can be recorded and transmitted to the Google Cloud server, and the recognized text can be received from the server and used. In particular, when an external API is used, processes for complex speech recognition, such as noise processing, speech signal processing, and speech signal modeling, which the speech recognition unit 730 needs to process, can be easily processed using an external API. In this way, the voice recognition unit 730 extracts a function for visualizing statistical data from the user's voice and the function-specific factors.

Then, the visualization command generation unit 760 generates a control command for visualizing statistical data by combining the extracted function and function-specific arguments based on the received mapping relationship, and generates the generated control command in the interface 720. And transmit it to the statistical program module 710.

When the statistical program module 710 executes the instruction, the result is transmitted to the central control unit 750. The central control unit 750 may transmit the result to the user's device through the interface 720 so that the user can refer to the visualized result. That is, the interface 720 may serve to receive a result of performing a control command by the statistical program module 710 and provide it to a user device.

In addition, the abnormal motion detection unit 770 of the statistical data visualization apparatus 700 through voice recognition may detect an abnormal operation early from the visualized result. To this end, machine learning based learning may be performed in the abnormal motion detection unit 770.

As described above, when the statistical data visualization apparatus 700 through voice recognition of FIG. 7 is used, even a beginner can easily analyze and check the operation status of the smart factory through voice.

As described above, the statistical data visualization apparatus 700 through voice recognition creates standardization rules and scenarios of software for big data analysis (SW), which is the most important technology in the fourth industrial revolution, and programming technology using voice, and is smart. It is possible to analyze the factory operation status in real time and take preemptive action.

Therefore, the statistical data visualization apparatus 700 through voice recognition is an important core technology of IoT and big data research business using smart factories, and the technological ripple effect is very high through the combination of advanced data analysis scenario development and language of each country. Big.

The present invention has been described with reference to the embodiments shown in the drawings, but these are merely exemplary, and those skilled in the art will understand that various modifications and other equivalent embodiments are possible therefrom. For example, the standardization commands stored in the database may vary depending on the statistical program used and the selected visualization function.

Further, the method according to the present invention can be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium may include any kind of recording device in which data readable by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disks, optical data storage devices, etc., and also implemented in the form of carrier waves (for example, transmission via the Internet). Includes. In addition, the computer-readable recording medium may store computer-readable code that can be executed in a distributed manner by a networked distributed computer system.

In the terminology used herein, a singular expression should be understood to include a plurality of expressions, unless the context clearly interprets otherwise, and terms such as “comprises” describe features, numbers, steps, operations, and components described. It is to be understood that it means that a part or a combination thereof is present, and does not exclude the presence or addition possibility of one or more other features or numbers, step operation components, parts or combinations thereof. In addition, terms such as "... unit", "... group", "module", and "block" described in the specification mean a unit that processes at least one function or operation, which is hardware, software, or hardware. And software.

Therefore, the drawings attached to the present embodiment and the present specification are merely a part of the technical spirit included in the present invention, and those skilled in the art within the scope of the technical spirit included in the specification and the drawings of the present invention can easily It will be apparent that all examples and specific examples that can be inferred are included in the scope of the present invention.

The present invention is in the field of smart factory monitoring to allow users who do not have expertise in statistical programs to efficiently manage big data through voice, and to further monitor the abnormal operation conditions of the smart factory so that they can take precautions. Can be applied.

710: Statistical program module 720: Interface
730: speech recognition unit 750: central control unit
760: visualization command generation unit 770: abnormal motion detection unit
790: database

Claims (16)

As a method of visualizing statistical data through speech recognition,
Receiving a mapping relationship between a control command of a statistical program that generates statistical data by performing statistical processing on data from a database, and a voice command from a user;
Recognizing a voice from a user based on the received mapping relationship, extracting a function for visualizing the statistical data from the user's voice and the function-specific factors;
Generating a control command for visualizing statistical data by combining the extracted function and function-specific factors based on the received mapping relationship, and inputting the generated control command to the statistical program; And
And receiving a result of performing the control command from the statistical program and providing it to a user. According to claim 1,
The database stores a primary mapping relationship between a user's voice command and a standardized command that standardizes control commands of a plurality of statistical programs and a secondary mapping relationship between the standardized command and control commands of each statistical program, through voice recognition. How to visualize statistical data. According to claim 2,
In the step of extracting, the situation in which the control command is used is classified based on the relationship between the number of factors of the function constituting the control command and factors, and the factor of the function is extracted according to the classified situation. How to visualize statistical data through speech recognition. According to claim 1,
The speech recognition is performed using a natural language processing technology installed in a mobile device, a method of visualizing statistical data through speech recognition. According to claim 1,
The speech recognition is performed by utilizing an application programming interface (API) of a speech recognition server connected through a network, a method of visualizing statistical data through speech recognition. According to claim 1,
The method is used to control the smart factory, statistical data visualization method through speech recognition. The method of claim 6,
The above method,
Further comprising the step of detecting the abnormal operation of the smart factory from the result of performing the control command, statistical data visualization method through speech recognition. According to claim 1,
The statistical program is at least one of the R program and Python (Python), a method of visualizing statistical data through speech recognition. As a device for visualizing statistical data through speech recognition,
A database storing a mapping relationship between a control command of a statistical program that performs statistical processing on data and a voice command from a user;
A voice recognition unit that recognizes a voice from a user based on the received mapping relationship and extracts a function for visualizing statistical data from the user's voice and the function-specific factors;
A visualization control command generator for generating a control command for visualizing statistical data by combining the extracted function and function-specific factors based on the received mapping relationship, and inputting the generated control command to the statistical program;
An interface module that receives a result of performing the control command from the statistical program and provides it to a user; And
Statistical data visualization device through speech recognition, including a central control unit for controlling the components. The method of claim 9,
The database stores a primary mapping relationship between a user's voice command and a standardized command that standardizes control commands of a plurality of statistical programs and a secondary mapping relationship between the standardized command and control commands of each statistical program, through voice recognition. Statistical data visualization device. The method of claim 10,
The speech recognition unit classifies a situation in which the control command is used based on the relationship between the number of factors of the function constituting the control command and factors, and extracts the function parameter according to the classified situation. Statistical data visualization device through recognition. The method of claim 9,
The speech recognition is performed using a natural language processing technology installed in a mobile device, a statistical data visualization apparatus through speech recognition. The method of claim 9,
The speech recognition is performed by utilizing an application programming interface (API) of a speech recognition server connected through a network, and a statistical data visualization device through speech recognition. The method of claim 9,
The device is used to control the smart factory, statistical data visualization device through speech recognition. The method of claim 14,
The device,
A statistical data visualization device through speech recognition, further comprising an abnormal motion detection unit that detects an abnormal operation of the smart factory from a result of performing the control command. The method of claim 9,
The statistical program is at least one of the R program and Python (Python), statistical data visualization apparatus through speech recognition.

Images