Convenient acquisition and use of isolated data in automated environments

By receiving natural language input and using natural language processing tools, the problem of data silos in automated environments has been solved, enabling efficient acquisition and use of isolated data, improving data sharing and collaboration capabilities, and ensuring data quality consistency and an overall view.

CN123125589APending Publication Date: 2026-11-17SIEMENS AG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202580028758.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-04-18
Filing Date
2025-03-13
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

In an automated environment, data silos create obstacles to information sharing and collaboration, inconsistent data quality, and make it difficult for leaders to form a holistic view of the company's data.

Method used

By receiving natural language input, extracting parameters, acquiring and storing datasets from multiple independent repositories, using natural language processing tools and chatbots to assist in identifying and acquiring the required data, and utilizing predefined configurable parameter types for data transformation and storage.

Benefits of technology

It enables efficient acquisition and use of siloed data in an automated environment, enhances data sharing and collaboration capabilities, ensures data quality consistency, and provides a holistic view of the company's data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123125589A_ABST
    Figure CN123125589A_ABST
Patent Text Reader

Abstract

This invention relates to a method for acquiring and using isolated data in an automated environment. To facilitate the acquisition and use of isolated data in an automated environment, the following steps are proposed: receiving natural language input (120A) instructing the acquisition of datasets (122A, 122B) stored in at least two independent repositories (124A, 124B); extracting at least two acquisition parameters (126A, 126B) from the received input (120A) according to configurable parameter types (128A, 128B); acquiring datasets (122A, 122B) corresponding to the corresponding, extracted acquisition parameters (126A, 126B) from the repositories (124A, 124B); and storing the corresponding, acquired datasets (122A, 122B) in a data storage device (130).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application generally relates to the acquisition and use of siloed data in automated environments. Systems used to process such siloed data, particularly the acquisition and use of siloed data in automated environments, are collectively referred to herein as product systems. Background Technology

[0002] Data silos are a common problem in manufacturing. The problem arises when different departments, teams, and production lines operate independently and rely on different systems for their varying needs, resulting in a fragmented view of the data. This structure breeds information silos, collections of data held by one group that are not easily or fully accessible to other groups within the same organization. As a result, data silos create barriers to information sharing and collaboration between departments, and data quality is often compromised due to inconsistencies that may exist across overlapping silos. Siloed data also makes it difficult for leaders to form a holistic view of the company's data, which is not healthy data. Summary of the Invention

[0003] The disclosed implementations include methods and computer systems that can be used to facilitate the acquisition and use of isolated data in automated environments.

[0004] According to a first aspect of the present invention, a computer-implemented method may include: Receive natural language input instructing the user to retrieve the dataset from at least two independent repositories; Extract at least two acquisition parameters from the received input based on the predefined parameter types; Retrieve the dataset corresponding to the retrieved parameters from the repository; and The acquired dataset is stored in a data repository.

[0005] According to a second aspect of the invention, a computer system may be arranged and configured to perform the steps of the method implemented by the computer according to the first aspect.

[0006] According to a third aspect, a computer program product may include computer program code that, when executed by a computer system according to a second aspect, causes the computer system to perform the method according to a first aspect.

[0007] According to the fourth aspect, a computer-readable medium may include a computer program product according to the third aspect. For example, the described computer-readable medium may be non-transient and may further be a software component on a storage device.

[0008] The technical features of this application have been broadly outlined above to enable those skilled in the art to better understand the following detailed description. Additional features and advantages of this application, which form the subject matter of the claims, will be described below. Those skilled in the art will recognize that they can readily use the disclosed concepts and specific embodiments as a basis for facilitating the acquisition and use of isolated data in automated environments or for achieving the same purpose as this application.

[0009] Furthermore, before proceeding with the following detailed description, it should be understood that various definitions of certain words and phrases are provided throughout this patent document, and those skilled in the art will understand that these definitions apply in most (if not most) cases to the prior and future uses of these defined words and phrases. While some terms may encompass multiple embodiments, the appended claims expressly limit these terms to specific embodiments.

[0010] The implementation method will be described in more detail below. Attached Figure Description

[0011] Figures 1 to 7 Functional block diagrams of example systems are shown to facilitate the acquisition and use of isolated data in an automated environment within a product system.

[0012] Figure 8 A flowchart illustrates an example method for acquiring and using isolated data in an automated environment within a product system.

[0013] Figure 9 An implementation of an artificial neural network is shown that can be used in a context of acquiring and using isolated data in an automated environment within a product system.

[0014] Figure 10 A block diagram of a data processing system in which one implementation can be carried out is shown. Detailed Implementation

[0015] Various techniques relating to systems and methods for acquiring and using isolated data in an automated environment within a product system will now be described with reference to the accompanying drawings, wherein the same reference numerals denote the same elements throughout. The drawings discussed below, as well as the various embodiments used to describe the principles of this application in this patent document, are for illustrative purposes only and should not be construed as limiting the scope of this application in any way. Those skilled in the art will understand that the principles of this application can be implemented in any suitably arranged device. It should be understood that functions described as being performed by certain system elements can be performed by multiple elements. Similarly, for example, an element may be configured to perform functions described as being performed by multiple elements. Numerous innovative teachings of this patent document will be described with reference to exemplary, non-limiting embodiments.

[0016] refer to Figure 1 This illustration shows a functional block diagram of a first example computer system or data processing system 100 for facilitating the acquisition and use of isolated data in an automated environment. The processing system 100 may include a Data Silo Assistant (DSC) system 118, which in some examples may include at least one processor 102 configured to execute at least one application software component 106 from memory 104 accessed by the processor 102. The application software component 106 may be configured (i.e., programmed) to cause the processor 102 to perform various actions and functions described herein. For example, the application software component 106 may include and / or correspond to one or more components for applications facilitating the acquisition and use of isolated data in an automated environment, wherein the application software component 106 may, for example, be configured to generate and store product data in a data repository 108 (such as a database).

[0017] For example, the DSC system 118 may be cloud-based, internet-based, and / or operated by a provider that facilitates the acquisition and use of siloed data in an automated environment. In some examples, the user may be located near or far from the DSC system 118, for example, in other locations, such as using a mobile device connected to the DSC system 118 via the internet, where the mobile device may include an input device 110 and a display device 112. For example, a speaker may be used instead of or attached to the display device 112 to deliver information to the user. In some examples, the DSC system 118 may be installed and run on the user's device, such as a computer, laptop, tablet, local computing facility, etc.

[0018] It should be recognized that acquiring and using siloed data in automated environments can be a challenging and time-consuming process, potentially requiring highly skilled engineers with years of training. For example, large industrial complexes may contain thousands or even millions of electronic components under a single roof, if this also includes Internet of Things (IoT) or Industrial Internet of Things (IIoT) devices. Many of these devices generate logs, but interpreting the status codes and error codes appearing in these logs is a daunting task, for example, due to the sheer volume and variety of data formats, or because each component may have its own set of documentation. Therefore, the information available within an industrial complex is often siloed and thus unsuitable for optimal use within the corresponding automated environment.

[0019] To enable enhanced acquisition and use of isolated data in an automated environment, the product system or processing system 100 may include at least one input device 110 and at least one display device 112 (such as a display screen) or at least one speaker. The processor 102 may be configured to generate a graphical user interface (GUI) 114 via the display device 112. This GUI 114 may include GUI elements such as buttons, links, search boxes, lists, text boxes, images, and scrollbars, which users can use via the input device 110 to provide input that leads to the acquisition and use of isolated data in the automated environment. For example, the GUI 114 may include a DSC user interface (UI) 116 provided to the user.

[0020] In one exemplary implementation, for example for acquiring and using isolated data in an automated environment, application software component 106 and / or processor 102 may be configured to receive instructions to acquire natural language input 120A of datasets 122A, 122B stored in at least two independent repositories 124A, 124B.

[0021] Natural language input 120A may include input information provided, for example, in human language, which is used by humans to convey meaning orally, in writing, or through sign language. Therefore, natural language input 120A may be provided by a human user in the form of speech, text, or gestures. Input device 110 may include, for example, a microphone, computer keyboard, smartphone, camera, etc., to receive the provided natural language input 120A. In some examples, natural language input 120A may be recorded and forwarded to DSC system 118.

[0022] Natural language input 120A may, for example, instruct the user to retrieve a first dataset 122A from a first repository 124A and a second dataset 122B from a second repository 124B. In other examples, multiple datasets 122C, 122D, ... may exist in multiple repositories 124C, 124D, ... where the user may also be interested in retrieving these datasets 122C, 122D, ... For example, the user may want to retrieve all or selected data available in one or more repositories 124. In other examples, the user may want to retrieve a selected dataset 122, where in some examples the user may not know where these datasets 122 are stored. Furthermore, for example, the various datasets 122 may be stored in at least two different data types or formats, and in some examples, in multiple different data types or formats.

[0023] In some examples, application software component 106 and / or processor 102 may be further configured to extract at least two acquisition parameters 126A and 126B from received input 120A according to configurable parameter types 128A and 128B.

[0024] The type of the first acquisition parameter 126A for the first dataset 122A from the first repository 124A and the type of the second acquisition parameter 126B for the second dataset 122B from the second repository 124B can be predefined and can, for example, indicate the data type or data format that the corresponding datasets 122A and 122B should have. Therefore, the corresponding acquisition parameter type 128 can, in some examples, help correctly identify the corresponding dataset 122 that the user wishes to acquire, or help correctly interpret the corresponding dataset 122.

[0025] In some examples, the received natural language input 120A may include information about the corresponding acquisition parameters 126, which can then be derived or extracted from the received natural language input 120A using a natural language processing (NLP) tool. In this paper, the NLP tool may, for example, use rule-based or probabilistic (i.e., statistical, and more recently, neural network-based) machine learning methods, where the tool may be able to understand the content of the received natural language input 120A. In some examples, the NLP tool may involve natural language understanding and optionally speech recognition. Such NLP tools are currently available from various vendors and include, for example, large language models (LLMs), such as the GPT (Generative Pre-trained Transformer) family of models from OpenAI in San Francisco, California. It should be recognized that the DSC system 118 may, for example, include LLM or chatbot functionality to provide the functionality required by the proposed method. In this paper, LLM can be, for example, a hybrid of multilayer perceptron (MLP) layers and transformer layers, as explained in more detail in the paper "Large Language Models for Mathematicians" by Simon Frieder, Julius Berner, Philipp Petersen, and Thomas Lukasiewicz (see https: / / doi.org / 10.48550 / arXiv.2312.04556) and "A mathematical perspective on Transformers" by Borjan Geshkovski, Cyril Letrouit, Yury Pollanskiy, and Philippe Rigollet (see https: / / doi.org / 10.48550 / arXiv.2312.10794). In some examples, such chatbots using one or more LLMs can also be used to provide further functionality to the DSC system 118, as described below.

[0026] For example, the corresponding configurable parameter types 128A and 128B may reside in internal data storage area 108 or data storage 130, wherein the corresponding configurable parameter types 128A and 128B may be received by a user or from another data source, such as via an API (Application Programming Interface). In some examples, default types may be provided for the corresponding configurable parameter types 128A and 128B, for example, for typical scenarios of device 138 in automation system 140, such as for an electric motor: identifier, physical observation 136 (e.g., speed, power consumption, voltage, or current, etc.), and optionally further documentation 140, as described in more detail below. Similar default types may be defined for other devices 138 and corresponding typical scenarios in automation system 140.

[0027] For example, if a user queries DSC system 118, “What are the temperatures of the different conveyor drives in my factory, and what is the status of the conveyors?”, the first acquire parameter type 128A may include an identifier (e.g., name) and a temperature (e.g., in °C) for the corresponding conveyor drive in the factory, and the second acquire parameter type 128B may include an identifier (e.g., name) and an operating status of the conveyor. In this document, the corresponding first dataset 122A may exist in the factory's PLC (Programmable Logic Controller), corresponding to the first repository 124A, and the corresponding second dataset 122B may exist in the factory's Manufacturing Execution System (MES), corresponding to the second repository 124B. In this document, the corresponding conveyor, conveyor drive, PLC, or MES may be understood as the corresponding device 138, such as the automation equipment of automation system 140, which is explained in more detail below.

[0028] In another example, application software component 106 and / or processor 102 may be further configured to retrieve datasets 122A and 122B from repositories 124A and 124B corresponding to the respective extracted retrieval parameters 126A and 126B.

[0029] The first dataset 122A can then be obtained from the first repository 124A, for example, by retrieving, transmitting, or downloading, while the second dataset 122B can be obtained accordingly from the second repository 124B. In some examples, the corresponding retrieval parameters 126A and 126B can be used to identify and / or retrieve the corresponding datasets 122A and 122B.

[0030] For example, application software component 106 and / or processor 102 may be further configured to store the corresponding acquired datasets 122A, 122B in data memory 130.

[0031] The acquired datasets 122A and 122B can then be stored in data storage 130, such as a database, wherein data storage 130 may, for example, be included in the DSCDSC system 118. In another example, data storage 130 may correspond to internal data storage area 108. In yet another example, data storage 130 may be included in a user's terminal device, or may be cloud-based or internet-based, wherein the user may be able to access data storage 130.

[0032] If you want to obtain multiple datasets 122C, 122D, ... from multiple repositories 124C, 124D, ..., the method described for the first dataset 122A and the second dataset 122B can be used similarly.

[0033] For example, application software component 106 and / or processor 102 may be further configured to: determine that at least one aspect 126-I of the corresponding extracted acquisition parameter 126 is invalid if the corresponding extracted acquisition parameter 126 does not conform to the configurable parameter type 128; prompt the user to provide user input related to the corresponding invalid aspect 126-I in natural language; receive natural language response input 120B related to the corresponding invalid aspect 126-I; and extract the corresponding acquisition parameter 126, including the corresponding valid aspect 126-V related to the previously invalid aspect 126-I, from the received response input 120B.

[0034] In some examples, extracting the corresponding acquisition parameter 126 from the received input 120A may result in the extracted corresponding acquisition parameter 126 having at least one invalid aspect 126-I. Here, invalid aspect 126-I may, for example, mean that the corresponding aspect of the corresponding acquisition parameter 126 is different from or incompatible with the expected configurable parameter type 128. Referring to the example above where a user asks the DSCDSC system 118, “How hot are the different conveyor drives in my plant and what is the condition of the conveyors?”, the first acquisition parameter type 128A may, for example, include an identifier (e.g., name) for the corresponding conveyor drive in the plant and a temperature expressed in °C, while the corresponding first acquisition parameter 126A may be used to obtain temperature data for the corresponding conveyor drive expressed in °Fahrenheit, instead of the expected temperature expressed in °C. Therefore, the temperature aspect of the first acquisition parameter 126A may constitute such an invalid aspect 126-I in some examples. Furthermore, the second acquired parameter type 128B may include an identifier (e.g., name) and the operating status of the conveyor, while the corresponding second acquired parameter 126B may be used to acquire the speed data of the electric pump, rather than the expected conveyor name and the expected conveyor operating status. Therefore, the conveyor name aspect and the operating status aspect of the second acquired parameter 126B may each constitute such invalid aspect 126-I in some examples.

[0035] In other examples, the corresponding invalid aspect 126-I of the relevant get parameter 126 may involve incompatible or different data types (or data formats), such as integers for numbers, double-precision floating-point formats, boolean or string data types. In some examples, the corresponding invalid aspect 126-I of the relevant get parameter 126 can be determined by comparing the relevant get parameter 126 with the relevant configurable parameter type 128.

[0036] If invalid aspect 126-I has been identified, the user can be prompted to provide corresponding input using natural language. In some examples, this can be accomplished using one or more LLMs, such as the GPT (Generative Pre-trained Transformer) family mentioned above. In other examples, some kind of discussion can be performed, including several prompts to the user and several responses from the user, for example, until sufficient information about invalid aspect 126-I has been collected. Thus, in some examples, the user can be guided to focus on invalid aspect 126-I so that the user can more easily understand and identify potential problems, inconsistencies, or contradictions between the corresponding extracted acquisition parameter 126 and the configurable parameter type 128.

[0037] The user can provide a natural language response input 120B to explain the user's intent and provide clarification related to the invalid aspect 126-I. Then, a corresponding acquisition parameter 126, including the corresponding valid aspect 126-V related to the previous invalid aspect 126-I, can be determined. In some examples, this can be accomplished by modifying the corresponding acquisition parameter 126 that can be extracted from the user-provided response input 120B. In other examples, this can be accomplished by modifying the corresponding configurable parameter type 128 that can be extracted from the user-provided response input 120B.

[0038] For example, application software component 106 and / or processor 102 may be further configured to determine that at least one aspect 126-ID of the corresponding acquired datasets 122A and 122B is invalid if the corresponding acquired datasets 122A and 122B do not conform to the corresponding acquisition parameters 126A and 126B or do not conform to the configurable parameter type 128. A similar method involving prompting the user and receiving user input in natural language form may then be applied, wherein, for example, the corresponding acquisition parameters 126A and 126B or the configurable parameter type 128 may be modified, wherein alternatively or additionally, different independent repositories 124A and 124B may be selected.

[0039] In some examples, configurable parameter type 128 represents the name, data type, and optional description of the corresponding retrieved parameter 126.

[0040] Referring to the example above where a user asks the DSCDSC system 118, "How hot are the different conveyor drives in my factory and what is the status of the conveyors?", the first acquisition parameter type 128A can be characterized by including a string data type, such as the name of the corresponding conveyor drive in the factory. Furthermore, the first acquisition parameter type 128A can be characterized by including a "floating-point" data type, for example, in °C. In another example, the second acquisition parameter type 128B can again be characterized by including a string data type, such as the name of the corresponding conveyor drive in the factory. Furthermore, the second acquisition parameter type 128B can be characterized by including a "boolean" data type, for example, to express the normal operating status of the corresponding conveyor drive with a "true" statement.

[0041] Configurable parameter type 128 may, for example, further characterize a description of the corresponding acquisition parameter, for example by including a string data type, which may be used for, for example, the description of the corresponding acquisition parameter 126 or the dataset 122 to be acquired or the corresponding device 138 (such as the automation device of the automation system 140), the corresponding acquisition parameter 126 or the corresponding dataset 122 may be directly or indirectly related to the device, for example via a detectable observation 136 of the corresponding device 138, which can be detected as dataset 122. Therefore, in some examples, the string data type used for the description may be used to provide a (electronic version) user manual (also known as an instruction manual or user guide), which may include product specifications, safety instructions, maintenance instructions, troubleshooting instructions, setup instructions, etc., for example for the corresponding device 138 and optionally for the automation system 140 including the corresponding device 138. Therefore, in some examples, the description included in the corresponding configurable parameter type 128 may correspond to document 142 described below.

[0042] It should be understood that, in some examples, the configurable parameter type 128 and thus the corresponding acquisition parameters 126 and dataset 122 can provide important information for operating, analyzing, controlling and / or monitoring the corresponding device 138, multiple devices 138 or an automation system 140 including such devices 138.

[0043] For example, application software component 106 and / or processor 102 may be further configured to receive a list of available independent repositories 124.

[0044] For example, the list may be received by internal data storage 108 or data storage 130, such as from a user or from another data source, such as via an API. The list may include the name of the corresponding repository 124 and how data can be transferred from and optionally to the corresponding repository 124, for example by providing a reference to the corresponding repository 124, specifying its location on a computer network or the Internet, and the method for retrieving the corresponding repository 124 or data from the corresponding repository 124, for example in the form of a Uniform Resource Locator (URL), commonly known as an address on the network. Here, the reference, location, or URL may also be understood as a connection 132 to the corresponding repository 124, as described below.

[0045] In some examples, a user may be aware of repositories 124 of interest, for example, since datasets 122 related to device 138 of interest may be stored in these repositories 124. Therefore, the user may input a list of repositories 124, for example, as an electronic document or as natural language input, where, in some examples, the aforementioned chatbot may be used to guide and support the user. In other examples, a passive or active network scan may be performed to identify the device 138 of interest and / or repository 124 and to receive the necessary information to obtain a list of available independent repositories 124, wherein the device 138 and / or repository 124 may be communicatively connected to the DSCDSC system 118, for example via a network, or the entities constitute at least a part of the network.

[0046] In another example, a default parameter type 128 can be predefined, and the corresponding parameter type 128 can only be determined if no default parameter type 128 is available. Users can, for example, provide a separate data retrieval module for each repository 124 and optionally for each dataset 122 stored in the respective repository 124. This retrieval module can, for example, correspond to or be included in the list, and can, for example, return so-called pandas dataframe objects. Here, pandas is a software library written for the Python programming language for data processing, manipulation, and analysis. In particular, it provides data structures and operations for processing or manipulating numerical tables and time series. Its name derives from the term "panel data," an econometric term used for datasets comprising observations of the same object across multiple time periods, and is also a parody of the phrase "Python data analysis." In some examples, pandas data can be two-dimensional, variable-size, and potentially heterogeneous tabular data. The data retrieval module can, for example, be written in any programming language, with a Python wrapper on top of it, for example. In some examples, the parser can read the Python wrapper and prepare the specification of the input parameters, which may correspond, for example, to the corresponding configurable parameter type 128 and / or a list.

[0047] In another example, application software component 106 and / or processor 102 may be further configured to determine each repository 124 and optionally a corresponding configurable parameter type 128 for each dataset 122 stored in the respective repository 124.

[0048] For example, the first configurable parameter type 128A associated with the first retrieval parameter 126A and the first dataset 122A to be retrieved from the first repository 124A can be determined based on information provided in the received list. In some examples, the first configurable parameter type 128A can be determined based on information stored in the first repository 124A and to be queried from it, such information may indicate, for example, what type of information is stored in the first repository 124A and how the information stored in the first repository 124A is organized, for example, in terms of name, data type, description, etc. Here, the first configurable parameter type 128A can be repository-specific, such that the same first configurable parameter type 128A can be applied to all data stored in the first repository 124A and retrieved according to the proposed method. A similar approach can be applied to other repositories in multiple repositories 124 and corresponding other configurable parameter types 128. A corresponding repository-specific configurable parameter type 128 may be advantageous in some examples if the datasets 122 (all) stored in the corresponding repository 124 are similarly organized, for example, in terms of name, data type, description, etc. For example, this could happen if a similar dataset 122 of a similar device 138 (such as an electric motor, a pond, a conveyor, etc.) is stored in the corresponding repository 124.

[0049] In another example, the first configurable parameter type 128A may be dataset-specific, such that the same first configurable parameter type 128A may be applicable to similar datasets 122 stored in the first repository 124A and retrievable according to the suggested method, or to only a single dataset 122. For example, this may occur if datasets 122 of different types and / or different devices 138 (such as motors, pumps, conveyors, etc.) are stored in their respective repositories 124. In some examples, the first configurable parameter type 128A may be determined based on a first dataset 122A stored in the first repository 124A and from which it is to be queried, wherein said information may, for example, indicate what type of first dataset 122A is stored in the first repository 124A and how the first dataset 122A is organized, for example, in terms of name, data type, description, etc. Similar methods may be applied to other datasets 122 and corresponding other configurable parameter types 128. In some examples, one dataset 122 may be associated with the measured temperature of the pump, another dataset 122 may be associated with the measured power consumption of the same pump, and yet another dataset 122 may be associated with the operating status of a switch connected to the pump, wherein both the switch and the pump are devices 138 of the automation system 140.

[0050] In some examples, a user may be aware of dataset 122 of interest, for example, because the user is interested in certain devices 138, and therefore interested in dataset 122 associated with those devices 138, wherein dataset 122 may be stored in a repository 124 in the list. Thus, the user can input a list of repositories 124 and optionally, the devices 138 and / or datasets 122 of interest, for example, as an electronic document or as natural language input, wherein in some examples, the aforementioned chatbot may be used to guide and support the user. In some examples, passive or active scanning may be performed to identify the devices 138 of interest, repository 124, and / or dataset 122, and to receive the information needed for a list of available, independent repositories 124, wherein the devices 138, repository 124, and / or dataset 122 may be communicatively connected to the DSC system 118, for example, via a network, or the mentioned entities constitute at least a part of said network.

[0051] For example, application software component 106 and / or processor 102 may also be configured to: prompt the user for user input related to connection 132 to the corresponding repository 124 and / or dataset 122 to be obtained using natural language; receive natural language response input 120C related to connection 132 to the corresponding repository 124 and / or dataset 122 to be obtained using natural language; and extract the corresponding connection 132 from the received response input 120C.

[0052] Inputs associated with connection 132 may include, for example, how data is transferred from and optionally to the corresponding repository 124, for example, by providing a reference to the corresponding repository 124, specifying its location on a computer network or the Internet, and the method for retrieving data from or from the corresponding repository 124, for example, in the form of a URL. In some examples, inputs associated with connection 132 may specify both a reference to the corresponding repository 124 and a separate corresponding dataset 122.

[0053] Here, the user can be prompted to provide input related to connection 132 using natural language, and the user can provide corresponding response input 120C using natural language. In some examples, it may be sufficient if the user provides a textual description of the location of repository 124 or dataset 122 and / or a corresponding reference for communication purposes. In other examples, providing the name or location of device 138 may be sufficient to deduce the corresponding connection 132. For example, the user can indicate that they want to retrieve dataset 122 from a certain repository 124 or a certain device 138. The chatbot can then determine information related to connection 132 based on user interaction, for example, by asking the user questions related to connection 132, such as predefined questions, where, based on the user's response input 120C, the LLM can then develop and create the necessary code to establish connection 132 for transferring data from or to the corresponding repository 124 and optionally from or to the corresponding dataset 122. In some examples, the predefined questions mentioned may include querying the user for the following information: 1) whether a connection 132 to the SQL database or repository 124 should be established, 2) which datasets 122 (e.g., in tabular form) from the connection 132 or the repository 124 are of interest, 3) whether a special query should be generated for the connection 132, and 4) what the parameters of the connection 132 are, such as the identifier or location of the repository 124 or dataset 122, or other metadata of the repository 124 or dataset 122, etc.

[0054] In some examples, the chatbot described above is able to extract the URL or other technical communication details required for connection 132 from the received response input 120C. In other examples, different repositories 124 or datasets 122 may have different data types or data formats. The corresponding connection 132 is able to convert the corresponding data types or data formats of different repositories 124 or datasets 122 into predefined standard data types or formats, for example, using Python or the pandas DataFrame described above.

[0055] In another example, information related to the corresponding connection 132 may be included in the list above and / or in the corresponding patch parameter 126 for the corresponding configurable parameter type 128.

[0056] In some examples, application software component 106 and / or processor 102 may also be configured to: receive natural language input 120D indicating to activate or deactivate a corresponding stored acquired dataset 122; receive natural language input 120E indicating to process one or more of the corresponding stored acquired datasets 122; and process one or more of the corresponding stored acquired datasets 122 only when the corresponding stored acquired dataset 122 is activated or not deactivated.

[0057] Using the proposed method, large amounts of data can be stored in data storage 130, especially over time, and if the acquired and stored datasets 122 are not discarded, for example, after a certain period of time, or if they are not used for data analysis. Therefore, in some examples, analyzing or processing the acquired and stored datasets 122 may consume increasing, in some examples, excessive, computational resources and, for example, memory resources. Therefore, in some examples, the corresponding stored acquired datasets 122 can be deactivated, for example, to make the corresponding deactivated datasets 122 unavailable for analysis or processing. Correspondingly, the corresponding stored acquired datasets 122 can be activated, for example, to make the corresponding activated datasets 122 available for analysis or processing. For example, the activation or deactivation of a corresponding dataset 122 can be achieved by accordingly marking the corresponding dataset 122. In another example, the corresponding dataset 122 may be stored in a dedicated portion of the data memory 130, which indicates the deactivation or activation of the corresponding dataset 122. Optionally, the activated dataset 122 may be stored in a hot storage portion of the data memory 130, while the deactivated dataset 122 may be stored in a cold storage portion of the data memory 130. Here, hot storage may, for example, mean using fast data access and fast transfer protocols for the hot storage portion of the data memory 130, and / or positioning the hot storage portion of the data memory 130 near the processor 102 of the DSC system 118. Cold storage may, for example, mean using relatively slow data access and slower transfer protocols for the cold storage portion of the data memory 130, and / or positioning the cold storage portion of the data memory 130 relatively far from the processor 102 of the DSC system 118. Additionally, hot storage may, for example, be more expensive than the considerably cheaper cold storage. In some examples, hot storage may, for example, be implemented using a tiered or hybrid storage system.

[0058] In some examples, processing the corresponding dataset 122 can be understood as the collection and manipulation of digital data to produce meaningful information. Here, data processing can be a form of information processing, which is the modification (processing) of information in any way that can be detected by an observer. The functionality of processing the corresponding dataset 122 can include processing data on one or more datasets 122. ●Verification: Ensure that the data provided is correct and relevant. ● Sorting: Arrange items in a certain order and / or by different sets. ●Summary (Statistical) or (Automatic): Reduces detailed data to its key points. ● Aggregation: Combining multiple data fragments ●Analysis: The collection, organization, analysis, interpretation, and presentation of data, including plotting or graphing. ●Report: Lists detailed data or a summary of the data or calculated information, and ●Categorization: Divide the data into various categories.

[0059] In another example, one or more of the processing aspects described above can also be understood as enhancing, combining, or transforming one or more datasets 122.

[0060] It should be understood that, in some examples, activating and deactivating selected stored acquired datasets 122 can significantly impact the processing speed and quality of the stored acquired datasets 122. For example, an excessive amount of acquired datasets 122 stored in data storage 130 may lead to excessively long processing times and, in some examples, may even result in erroneous processing results. Therefore, selectively activating or deactivating datasets 122 can facilitate the meaningful collection and processing of datasets 122, which, in some examples, can be used for analysis, monitoring, control, and / or operation of the corresponding devices 138 and / or automation systems 140.

[0061] In another example, in order to process one or more of the corresponding stored acquired datasets, application software component 106 and / or processor 102 may also be configured to: draw the corresponding stored acquired dataset 122, optionally as a function of the same corresponding stored acquired dataset 122 and / or as a function of one or more other corresponding stored acquired datasets 122; and create one or more illustrations 134 or one or more video sequences of illustration 134 from the drawing function.

[0062] Here, plotting can be understood as a graphical technique used to represent dataset 122, typically as a graph showing the relationship between two or more variables. Plotting can be done manually or by computer, for example, with computer-generated plotting considered in the context of this invention. A graph can be a visual representation of the relationship between variables, such as one or more datasets 122, where in some examples, variables may include timestamps, one or more detectable observations 136 included in one or more datasets 122, etc. For example, one or more datasets 122 can be plotted using the plotting library "matplotlib" for the Python programming language. Matplotlib provides an object-oriented API for embedding plotting into applications using common GUI toolkits such as Tkinter, wxPython, Qt, or GTK.

[0063] In another example, the created plot may correspond to a corresponding illustration 134, and in some examples, multiple plots may be combined into a single illustration 134. Additionally, for example, multiple plots or illustrations 134 may be combined to create one or more video sequences of illustration 134, for example, displaying the temporal evolution of one or more detectable observations or quantities derived therefrom through data processing.

[0064] In some examples, the chatbot receives input from the user to plot the corresponding stored acquired dataset 122. In some examples, the chatbot may suggest different plotting and graphing functions, and the user may select the desired function accordingly, such as using natural language.

[0065] For example, the corresponding acquired dataset 122 may be associated with the detectable observation 136 of the corresponding device 138, such as the automated device of the automation system 140, wherein the application software component 106 and / or processor 102 may also be configured to: receive a document 142 associated with the corresponding acquired dataset 122 and / or the corresponding device 138, wherein the document 142 represents at least one, or any combination thereof, of device attributes, device components, device configuration, device performance parameters, device operating parameters, device status parameters, and device condition parameters; identify at least one aspect 142-122 in the received corresponding document 142 that is associated with the corresponding acquired dataset 122; and convey the corresponding acquired dataset 122 and the associated corresponding aspect 142-122 of the received corresponding document 142 to a user, for example, via a display device 112.

[0066] For example, the detectable observation 136 can be a physical, chemical, or other property or quantity that can be directly or indirectly measured, such as position, velocity or rate, current or voltage, electrical power, acidity, volumetric flow rate, etc. In another example, the detectable observation can be understood more broadly to include the state or condition of the corresponding device 138 and / or automation system 140, or a quantity derived from the detected observation 136, for example, through data processing performed before or during storing the corresponding dataset 122 in the corresponding repository 124, before retrieving the corresponding dataset 122 from the corresponding repository 124, and before storing the corresponding dataset 122 in the data storage 130.

[0067] In some examples, the corresponding device 138 may be or include sensors, actuators (such as electric motors, valves, or robots), inverters powering electric motors, gearboxes, programmable logic controllers (PLCs), communication gateways, and / or other components related to industrial automation products and general industrial automation. The corresponding device 138 may be part of an automation system 140, such as a complex production line or manufacturing plant, such as a bottle-filling machine, conveyor, welding machine, welding robot, etc. Additionally, for example, the automation system 140 may include or be part of an IT system, such as a manufacturing operations management (MOM) system, a manufacturing execution system (MES), an enterprise resource planning (ERP) system, a monitoring and data acquisition (SCADA) system, or any combination thereof.

[0068] The detectable observation 136 and / or document 142 may, for example, be included in the corresponding acquired dataset 122. In other examples, i.e., if the detectable observation 136 and / or document 142 are not included in the corresponding acquired dataset 122, the detectable observation 136 and / or document 142 may be received or retrieved from the corresponding repository 124 where the corresponding dataset 122 is stored, or from other data repositories. For example, the document may be retrieved in response to a user query (e.g., in natural language). In some examples, the document 142 may include valuable metadata about the corresponding device 138, which in some examples generates or detects the detectable observation 136. Therefore, the document 142 can help provide more insights into the capabilities of the corresponding device 138 and the corresponding acquired dataset 122. These insights may, for example, allow for improved analysis and interpretation of the corresponding acquired dataset 122, or improved analysis, monitoring, operation, and / or control of the corresponding device 138 or the automation system 140. Here, various aspects of document 142 may indicate to the corresponding device 138 and / or automation system 140: functionality and limitations, the individual components of the corresponding device 138 or automation system 140 if it consists of multiple components, the configuration parameters used by the corresponding device 138 or automation system 140 for operation, the rated power of the corresponding device 138 and / or automation system 140, throughput, etc.

[0069] It should be understood that the aforementioned aspects of the detectable observation 136, the corresponding equipment 138, the automation system 140, and / or the document 142 may, in some examples, be equally applicable to other aspects of the above or below proposed solutions.

[0070] In some examples, document 142 may be comprehensive and cover multiple different aspects of the corresponding dataset 122 or the corresponding device 138. For example, core aspects 142-122 of the corresponding document 142 may be selected or identified and then transmitted to the user along with the corresponding acquired dataset 122, for example, by displaying the corresponding aspects 142-122 via display device 112, or by displaying them acoustically via a speaker, for example. The aspects 142-122 to be transmitted to the user may be provided by the user, for example, through natural language input, which has been analyzed to extract the corresponding aspects 142-122. In some examples, the aspects 142-122 may be predefined and may, for example, focus on key performance parameters of the corresponding device 138 or the automation system 140.

[0071] For example, detectable observations 136, which may be included in the corresponding acquired dataset 122, can be compared with corresponding information in document 142, for example, to check whether recent values ​​of detectable observations 136 remain within parameter ranges that frequently occur during normal operation of the corresponding device 138 or automation system 140. Here, the corresponding detectable observations 136 may, for example, include or be related to key performance indicators of the corresponding device 138 or automation system 140, where recent measurement results or determinations of detectable observations 136 can be compared with the corresponding document 142, for example, as described above. For example, both recent measurement values ​​of detectable observations 136 and the parameter ranges can be transmitted to the user. This allows the user to obtain, for example, data related to the operation of the corresponding device 138 or automation system 140 quickly and reliably, where optionally, the aforementioned data processing functions can be applied to the values ​​of the corresponding acquired dataset 122 or detectable observations 136.

[0072] For example, a user or other data repository can provide document 142 to DSC system 118, for instance, by interacting with a chatbot. The user can then, for example, ask the chatbot about the current configuration of a particular robot or the function of a specific component of that robot, where the robot corresponds to a corresponding device 138 in an automated system 140 used for, for example, manufacturing a car body. Here, the robot can provide current operational data as dataset 122 via a corresponding repository 124, which can be acquired and stored in data storage 130 according to a suggested scheme. The chatbot can then analyze the provided document 142 to answer the user's query and provide an answer to the user. In some examples, the user can ask the chatbot about the current temperature of the robot or a component of the robot, and optionally, the ideal operating temperature of the robot or a component of the robot. Here, the current temperature can be a detectable observation 136 included in the dataset 122, allowing the chatbot to acquire the corresponding dataset 122 and provide the relevant information to the user. In addition, the current temperature of the robot or robot parts can be compared with the ideal operating temperature included in document 142, and the results can also be transmitted to the user, optionally accompanied by a warning or alarm if the robot temperature is too high.

[0073] For example, application software component 106 and / or processor 102 may be further configured to identify at least one difference 144 between relevant aspects 142-122 of the corresponding acquired dataset 122 and the received corresponding document 142; and to transmit information 144-I related to the corresponding difference 144 to the user, for example via display device 112, and / or to change the operating state of the corresponding device 138 or automation system 140 including the corresponding device 138.

[0074] In some examples, the corresponding difference 144 can be related to the aforementioned comparison between the detectable observation 136 (e.g., related to key performance indicators of the corresponding device 138 or automation system 140) and the corresponding document 140. Therefore, if the corresponding device 138 or automation system 140 is operating outside of normal operating conditions, the difference 144 can be identified, and the corresponding information 144-I can be transmitted to the user.

[0075] In other examples, a discrepancy 144 can also be identified if the corresponding device 138 or automation system 140 changes (e.g., by replacing a component) and the document 142 is not updated accordingly. Such a discrepancy 144 can indicate that the relevant parties 142-122 have made unauthorized changes to the corresponding device 138 or automation system 140. In some examples, a discrepancy 144 can indicate that the relevant parties 142-122 have experienced a malfunction or severe wear and tear on the corresponding device 138 or automation system 140, which may require repair, replacement, overhaul, or maintenance of the corresponding device 138, automation system 140, or a component thereof. Various other industrial scenarios may exist in which the relevant parties 142-122 can be identified by comparing the corresponding dataset 122 with the corresponding document 142.

[0076] In some examples, if difference 144 is identified, the operation of the corresponding device 138 or automation device 140 may be altered, causing the operation to return to normal operating conditions, or, if not, shutting down the corresponding device 138 of the automation system 140 or causing it to operate in a safe mode. For example, information 144-I related to the corresponding difference 144 may include a warning or alarm. For example, the DSC system 118 may send a corresponding message or command to the corresponding device 138 of the automation system 140, causing the operation of the corresponding device 138 of the automation system 140 to be altered.

[0077] For example, the acquired dataset 122 may include time-series data associated with the detectable observation 136 of the corresponding device 138. Here, a time series can be understood as a series of data points indexed (or listed or plotted) in chronological order, wherein the time-series data may, for example, be machine-generated. For example, a time series may be a sequence collected at consecutive equally spaced time points. Thus, time-series data may, for example, be a sequence of discrete-time data. Examples of time series in this context may include values ​​of physical, chemical, or other properties or quantities that can be measured indirectly or directly, or quantities derived from the detected observation 136.

[0078] The acquired dataset 122, including time-series data, may appear in some instances in highly heterogeneous data silos. These silos may be difficult to access, analyze, or interpret, for example, because the time-series data may contain a large number of data points, or because it may only contain a series of values ​​without corresponding units of measurement, such as SI units, including seconds (s, a unit of time), meters (m, a unit of length), kilograms (kg, a unit of mass). Utilizing the contextual information of the time-series data in document 142 can, for example, help to gain a deeper understanding of the acquired dataset 122 and, for example, to correctly interpret the time-series data. Therefore, using document 142 can make it possible to determine the operational status of the corresponding device 138, or to determine whether the corresponding device 138 needs maintenance or replacement. For example, this can be achieved by comparing the time-series data with the device's operating parameters, where if the time-series data exceeds the range of values ​​or thresholds included in the device's operating parameters, an abnormal operational status of the corresponding device 138 can be determined, for example.

[0079] In other examples, document 142 includes at least two distinct aspects 142-122, wherein each aspect 142-122 represents a corresponding device attribute, a corresponding device component, a corresponding device configuration, a corresponding device performance parameter, a corresponding device operating parameter, a corresponding device status parameter, a corresponding device condition parameter, or any combination thereof, and wherein application software component 106 and / or processor 102 may be further configured to: ●The analysis results are determined by analyzing the corresponding acquired dataset 122 using the relevant aspects 142-122; and ● Transmit the confirmed analysis results to the user.

[0080] For example, one of aspects 142-122 can characterize a device operating parameter by indicating a range or threshold of values ​​related to the corresponding observation 136 (e.g., the rotational speed of the corresponding device 138). For instance, if time-series data included in the corresponding acquired dataset 122 indicates that the rotational speed remains within the provided rotational speed value range, the determined corresponding analysis result can include that the corresponding device 138 is in normal operating condition.

[0081] Furthermore, another aspect of 142-122 can characterize equipment performance parameters, for example, by indicating the quantity of products produced within a specific time period, such as in a steel mill or similar location, where the corresponding observation 136 could be, for example, the length of steel plate produced by means of the corresponding equipment 138 within a specific time period. For example, the corresponding equipment 138 could be a rolling mill stand or a drive for a rolling mill stand.

[0082] For example, if the time series data included in the corresponding acquired dataset 122 indicates that the rotational speed within a specific time period corresponds to a specific length of the produced steel plate, the determined corresponding analysis result may include the length of the produced steel plate in the equipment performance parameters of the corresponding equipment 138, or the determined corresponding analysis result may include the length of the produced steel plate remaining within a specific production performance.

[0083] In some examples, using machine-generated time-series data to enhance the definition of heterogeneity allows for the coverage of far more data silos than ever before. The ability to enhance machine-generated data (such as time-series data) with document 142 and present deeper insights has never been achieved before. Combining this approach with enabling and disabling certain data sources (e.g., using only natural language) allows users to change the context of the generated data. For example, a user can request analysis of a specific document 1, interpreting data from a specific machine based on the boundaries and thresholds listed in document 1. If the user subsequently realizes that the machine should have been analyzed based on another document 2, they can disable document 1 and analyze the machine data again, this time using document 2. This is just one example of how users can change the context and obtain completely different analytical results and reports based on different contexts. In some cases, using the wrong document to obtain context might obscure alarming situations, but in these cases, users can ensure that only relevant information is presented to them at the time of query.

[0084] For example, application software component 106 and / or processor 102 may be further configured to: store a sequence of steps in data memory 130; provide a user with a corresponding sequence of steps to select; and execute the sequence of steps if the user selects a corresponding sequence of steps, wherein the sequence of steps includes at least three of the following steps: o Retrieve dataset 122 from repository 124, corresponding to the extraction parameter 126; o Determine a corresponding configurable parameter type 128 for each repository 124 and optionally for each dataset 122 stored in the respective repository 124; o One or more of the corresponding stored retrieved datasets 122 are processed only if the corresponding stored retrieved dataset 122 is activated or not deactivated; o transmits the corresponding stored acquired dataset 122 and / or the processed stored acquired dataset 122 to the user, for example via display device 112; o Draw the corresponding stored retrieved dataset 122 as the same corresponding stored retrieved dataset 122 function and / or one or more other corresponding stored retrieved dataset 122 functions; o Create one or more Figure 134 or one or more Figure 134 video sequences based on the drawing function; o Display the corresponding illustration 134 or the video sequence of illustration 134 created via display device 112; o transmits the relevant aspects 142-122 of the acquired dataset 122 and the received corresponding document 142 to the user, for example via display device 112; o transmits information 144-I related to the corresponding difference 144 to the user, for example via display device 112, and / or o Change the operating state of the corresponding device 138 or the automation system 140 including the corresponding device 138; ● Provide users with a corresponding sequence of steps to choose from; and ●If the user selects the appropriate step sequence, then execute that step sequence.

[0085] In some examples, the sequences may correspond to the basic building blocks of the steps involved in processing dataset 122. These sequences can be established by, for example, more advanced or administrator users, and can help facilitate and accelerate standard data processing sequences. In some examples, a user may simply provide input to the chatbot to execute one of the available sequences, which in some examples may be provided to the user acoustically or via a display for selection. Therefore, these sequences can also help avoid erroneous data processing (e.g., caused by inexperienced users providing incorrect or insufficient input data or queries), and thus can contribute to achieving high-quality output data even for inexperienced users.

[0086] In other examples, if the user has performed several steps in the data processing steps described above, such as involving the same dataset 122, the same device 138, the same automation system 140, and / or the same document 142, the chatbot may ask the user whether the corresponding sequence should be stored as a sequence in the data storage. If the user confirms, for example after modifying the sequence, the (modified) sequence may be stored in the data storage 130.

[0087] Application software component 106 and / or processor 102 may then be further configured to: provide a user with a corresponding sequence of steps to choose from; and execute the corresponding sequence of steps when the user selects it.

[0088] It should be understood that the application software component 106 and / or processor 102 can perform similar methods for acquiring and using isolated data in an automated environment. Furthermore, Figure 1The image shows a computer-readable medium 160, which may include a computer program product 162, wherein the computer program product 162 may be encoded with executable instructions that, when executed, cause the computer system 100 and / or the DSC system 118 to perform the method.

[0089] refer to Figure 2 This illustrates a functional block diagram of a second example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0090] According to the second example, if the corresponding extracted acquisition parameter 126 does not conform to the configurable parameter type 128, the DSC system 118 can identify at least one aspect 126-I of the corresponding extracted acquisition parameter 126 as invalid. Subsequently, the DSC system 118 can prompt the user to provide user input related to the corresponding invalid aspect 126-I in natural language, whereby the prompt can be displayed, for example, on the display device 112. The user can then provide a natural language response input 120B related to the corresponding invalid aspect 126-I, which is received by the DSC system 118. The DSC system can then extract the corresponding acquisition parameter 126 from the received response input 120B, including the corresponding valid aspect 126-V related to the previously invalid aspect 126-I.

[0091] refer to Figure 3 This illustrates a functional block diagram of a third example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0092] According to the third example, the DSC system 118 can prompt the user to provide user input in natural language related to the connection 132 to the corresponding repository 124 and / or the dataset 122 to be retrieved. The user can provide a natural language response input 120C related to the connection 132 to the corresponding repository 124 and / or the dataset 122 to be retrieved, wherein the DSC system 118 receives the response input 120C and extracts the corresponding connection 132 and / or dataset 122 from the received response input 120C.

[0093] refer to Figure 4 This illustrates a functional block diagram of a fourth example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0094] According to the fourth example, the DSC system 118 can receive instructions from the user to activate or deactivate the corresponding stored natural language input 120D of the acquired dataset 122. Figure 4In the data storage 130, the dashed line of the first dataset 122A indicates that the first dataset 122A is deactivated, while the second dataset 122B is activated, as shown below. Figure 4 The solid line in the diagram indicates that the DSC system 118 may further receive natural language input 120E instructing the processing of one or more of the corresponding stored and retrieved datasets 122. The DSC system 118 may process one or more of the corresponding stored and retrieved datasets 122 only when the corresponding stored and retrieved dataset 122 is activated or not deactivated. That is, in the fourth example, only the activated second dataset 122B can be processed, while the deactivated first dataset 122A cannot be processed.

[0095] refer to Figure 5 This illustrates a functional block diagram of a fifth example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0096] According to the fifth example, the DSC system 118 can draw a corresponding stored acquired dataset 122, optionally as a function of the same corresponding stored acquired dataset 122 and / or as a function of one or more other corresponding stored acquired datasets 122, and create one or more illustrations 134 or a video sequence of one or more illustrations 134 according to the drawn function. The DSC system 118 can then display the corresponding illustration 134 to a user via a display device 112.

[0097] refer to Figure 6 This illustrates a functional block diagram of a sixth example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0098] According to the sixth example, a first dataset 122A is associated with a first device 138A, and a second dataset 122B is associated with a second device 138B, wherein both the first device 138A and the second device 138B are included in the automation system 140. The DSC system 118 can receive a document 142 associated with the corresponding acquired dataset 122 and / or the corresponding device 138, wherein the document 142 can characterize at least one of the following: device attributes, device components, device configuration, device performance parameters, device operating parameters, device status parameters, device condition parameters, or any combination thereof for the corresponding device 138 or the automation system 140. The DSC system 118 can identify at least one aspect 142-122 of the received corresponding document 142 associated with the corresponding acquired dataset 122, and convey the corresponding acquired dataset 122 and the associated corresponding aspect 142-122 of the received corresponding document 142 to the user via the display device 112.

[0099] refer to Figure 7This illustrates a functional block diagram of a seventh example computer system or data processing system 100 that facilitates the acquisition and use of isolated data in an automated environment.

[0100] The seventh example is consistent with the explanation above and in... Figure 6 The sixth example shown has several similarities. According to the seventh example, the DSC system 118 can identify at least one difference 144 between relevant corresponding aspects 142-122 of the corresponding acquired dataset 122 and the received corresponding document 142, and convey information 144-I about the corresponding difference 144 to the user via the display device 112. Furthermore, the DSC system 118 can cause a change in the operating state of the corresponding device 138 or the automation system 140 including the corresponding device 138, as indicated by the arrow from the processor 102 of the DSC system 118 to the second device 138B of the automation system 140.

[0101] In another embodiment, the main benefits of the proposed solution may include: ● A natural language interface that covers all types of isolated data.

[0102] ● Provides a single interface to access real-time data generated by different devices, which may include logs.

[0103] ● It can expand and interpret data based on documents provided by users in the document index.

[0104] ● It can create mechanisms that use plans or sequences to obtain answers to complex queries.

[0105] ● It can plot the generated data when needed.

[0106] In some examples, chatbots may not have a control flow like typical applications. Instead, chatbots may have only two main components: 1. an agent, and 2. various tools.

[0107] An agent is an algorithm that understands user queries and uses different tools to answer the questions posed by users. When a user provides a query to the agent, the agent generates a plan and selects the first tool from the plan. If the agent cannot generate a plan, it can use the "search related plans" tool to search for a plan provided by the user. After a plan is generated or selected, the agent executes the tool and collectively reflects on all results. It then checks if the final answer has been achieved; if so, it presents the answer; otherwise, it returns to the plan and executes the next tool. Note that the term "skill" is sometimes used instead of "tool," although both have the same purpose and function. It should be noted that the agent may change its plan while reflecting on the results from the tools. Secondly, different tools may generate exceptions, in which case the agent is notified that an exception has occurred. The agent autonomously decides what to do in such cases. Each step taken by the agent queries the LLM after preparing an appropriate hint. Hint generation is templated, which is common practice. Special factors are not considered in hint generation, so their details are not mentioned in this document. Common tools for generating hints via templates are widely used.

[0108] Tools can play a crucial role in the implementation of chatbots. Tools can be designed through a meticulous thought process, following an analysis of the robustness and scalability of the design. Tools may include: ●Dataset-related tools: The "Get Dataset" tool is the most important of these. Users can issue queries like "Get from <connection name>, where <parameter1 = value1> and <parameter2 = value2> and <parameter3 = value3>". The agent will infer from this query that it must invoke the Get Dataset tool along with the user-specified parameters.

[0109] The "Get Loaded Dataset Names" tool is primarily used to allow users to see which datasets have been loaded so they can query related datasets. Agents can also use it to retrieve the names of loaded datasets.

[0110] The “Show Dataset” tool is primarily for individual user use, which is why it includes a user label next to it in the graph. It takes the dataset name as input and outputs the first few lines to give you an overview of the dataset.

[0111] Similar to the above is the "Activate Dataset" tool. This tool is also used individually by the user. When a user activates a dataset, the chatbot focuses on it, and subsequent queries are considered relevant to the activated dataset unless the user explicitly specifies the dataset's name.

[0112] Similarly, the "Deactivate Dataset" tool is used to deactivate a previously activated dataset.

[0113] From the user's perspective, the "Query Dataset" tool is the most important because it allows users to perform common database operations on datasets. For example, a user can say, "Join <Dataset1> with <Dataset2> on column name <column>". Users don't need to write an exact SQL query to get the answer; instead, the tool automatically converts the user's query into SQL. This tool is not proprietary and is open source, so its details can be found in online resources. The authors added a feature to name and save the query results. Therefore, a user can say, "Join <Dataset1> with <Dataset2> on column name <column> and save the results to <result>". Afterward, users can perform all operations on <result> just like they would on any other dataset obtained by the agent.

[0114] The "Plot Data" tool is used to view data. Users can perform numerous operations and save the results in a table. They can then provide the table name to the plot data tool to view the data.

[0115] ● Planning-related tools: The plan is used to answer complex queries that the agent cannot generate a plan. It uses the following tools to search for a plan and then execute it.

[0116] The "Load All Schedules" tool is used to load schedules at runtime. If a user updates their schedule after the application has started, they can use this tool to load the updated schedule.

[0117] The "Search Related Plans" tool is used by the agent when it is unable to generate a suitable plan on its own. It is important to note that the agent can override or extend the plan at any time based on the tool's output.

[0118] ●Connect related tools: The "Load All Connections" tool is used to load connection and data retrieval components at runtime. If users add new connections and related components after the application starts, they can use this tool to load the updated connections.

[0119] The "Get Connection Name" tool is used to load all available connection strings and the retriever component. The connection name is derived from the component's filename, while parameter-related information is extracted from the code.

[0120] ●Other tools: The "Document Search" tool is used to search user-provided document indexes. The solution also provides tools for creating such indexes.

[0121] The "Get Chat History" tool is used to search a user's question-and-answer history to create relevant context.

[0122] The "Get User Input" tool is used to ask the user if there is any ambiguity. The agent formulates appropriate questions and then uses the answers to continue in the already executed thought chain.

[0123] To access different data silos, abstractions of data retrieval components can be created. Users must provide a separate data retrieval module for each table in the data silo. This module must return Pandas DataFrame objects. The module itself can be written in any programming language, but it must have a Python wrapper on top of it. The parser reads the Python wrapper and prepares the specifications for the input arguments.

[0124] After extraction, all parameters can be used in a Python dictionary. Each parameter must have the following attributes.

[0125] ●Name: The name of the parameter

[0126] ●Type: The data type of the parameter

[0127] ●Description: This is used to create a prompt for the user when the agent asks the user for the value of a parameter.

[0128] When a user issues a query to retrieve data from the connection, parameters can be extracted from the query. These parameters are then converted to the types specified in the parameter specification. Afterward, the user-provided data retrieval module is invoked to retrieve the data. The data is then stored in a local database for further processing.

[0129] Now that all the components of the chatbot have been explained, it's helpful to provide an overview of the user experience. When a user starts the chatbot, there is no data available except for the document index. If the user wants to query information from the index, they can, but if they want to query real-time data or data stored in data silos, they will first have to invoke the relevant data retrieval component. Users can achieve this by giving commands in natural language, such as "Get data from connection <connection name>", or simply "Get <connection name>". Issuing this query will invoke the Get Dataset tool. To prepare queries of the above type, users may need to use many other tools, such as first using Load All Connections to load connections, and then using Get Connection Name to view all connection names and select one. After retrieving data from one or more connections, users can perform any common data manipulation commands on them, and the results are saved in intermediate tables. These intermediate tables are available for users to perform queries and apply transformations. Users can propose complex queries that may lead to calls to nested queries. The Query Dataset tool will be invoked in the background, and the results will be saved to the user for further processing. In short, this is a complete interface for data retrieval and transformation in natural language. It should be noted that in the process mentioned above, if the user makes any mistakes, such as failing to provide the required parameters to any function, the chatbot will guide the user to provide the parameters and their types, possibly with examples. This is also a key difference compared to the previous scheme. Furthermore, it is possible for the user to augment the data from the document index. For example, the command "Get all status codes from the past hour and give their descriptions" will retrieve the status codes from the active dataset and then search the document index for their descriptions.

[0130] Figure 8 A flowchart illustrating an example method for facilitating the acquisition and use of isolated data in an automated environment within a product system is shown. This method may begin at M02 and may include multiple actions performed by the operation of at least one processor.

[0131] These actions may include: an action M04 of receiving natural language input instructing the retrieval of datasets from at least two independent repositories; an action M06 of extracting at least two retrieval parameters from the received input based on configurable parameter types; an action M08 of retrieving the dataset corresponding to the extracted retrieval parameters from the repositories; and an action M10 of storing the retrieved dataset in the data repository. The method may end at M12.

[0132] It should also be understood that method M may include other actions and features previously discussed for computer-implemented methods of acquiring and using siloed data in automated environments.

[0133] In some examples, the method may further include: prompting the user to provide user input in natural language related to a connection to the corresponding repository and / or a dataset to be retrieved; receiving natural language response input related to a connection to the corresponding repository and / or a dataset to be retrieved; and extracting the corresponding connection from the received response input.

[0134] In other examples, the method may further include: receiving an action of natural language input indicating activation or deactivation of a corresponding stored acquired dataset; receiving an action of natural language input indicating processing one or more of the corresponding stored acquired datasets; and processing one or more of the corresponding stored acquired datasets only when the corresponding stored acquired dataset is activated or not deactivated, wherein optionally, it may also include: drawing the corresponding stored acquired dataset as a function of the same corresponding stored acquired dataset and / or as a function of one or more other corresponding stored acquired datasets; and creating one or more graphs or a video sequence of one or more graphs according to the drawing function.

[0135] For example, if the acquired dataset is related to detectable observations of a corresponding device, such as an automated device in an automated system, the method may further include the following actions: receiving a document related to the acquired dataset and / or the corresponding device, wherein the document characterizes at least one of device attributes, device components, device configuration, device performance parameters, device operating parameters, device status parameters, device condition parameters, or any combination thereof; identifying at least one aspect of the received document related to the acquired dataset; and displaying the relevant corresponding aspects of the acquired dataset and the received document via a display device. Optionally, the method may further include the following actions: identifying at least one difference between the relevant corresponding aspects of the acquired dataset and the received document; displaying information about the difference via a display device; and / or changing the operating state of the device or the automated system including the device.

[0136] Figure 9 An embodiment of an artificial neural network 2000 is shown, which can be used in a product system in the context of acquiring and using isolated data in an automated environment. Alternative terms for "artificial neural network" are "neural network," "artificial neural mesh," or "neural mesh."

[0137] The artificial neural network 2000 includes nodes 2020, ..., 2032 and edges 2040, ..., 2042, where each edge 2040, ..., 2042 is a directed connection from a first node 2020, ..., 2032 to a second node 2020, ..., 2032. Typically, the first nodes 2020, ..., 2032 and the second nodes 2020, ..., 2032 are different nodes; however, it is also possible for the first nodes 2020, ..., 2032 to be the same as the second nodes. For example, in... Figure 6 In the diagram, edge 2040 is a directed connection from node 2020 to node 2023, and edge 2042 is a directed connection from node 2030 to node 2032. Edges 2040, ..., 2042 from the first node 2020, ..., 2032 to the second node 2020, ..., 2032 are also called "incoming edges" for the second node 2020, ..., 2032, and "outgoing edges" for the first node 2020, ..., 2032.

[0138] In this embodiment, the nodes 2020, ..., 2032 of the artificial neural network 2000 can be arranged in layers 2010, ..., 2013, wherein each layer may include an inherent order introduced by edges 2040, ..., 2042 between nodes 2020, ..., 2032. In some examples, edges 2040, ..., 2042 may exist only between adjacent node layers. In the illustrated embodiment, there is an input layer 2010 that includes only nodes 2020, ..., 2022 without incoming edges, an output layer 2013 that includes only nodes 2031, 2032 without outgoing edges, and hidden layers 2011, 2012 located between the input layer 2010 and the output layer 2013. Typically, the number of hidden layers 2011, 2012 can be arbitrarily chosen. The number of nodes 2020, ..., 2022 in the input layer 2010 is usually related to the number of input values ​​of the neural network, and the number of nodes 2031, 2032 in the output layer 2013 is usually related to the number of output values ​​of the neural network.

[0139] In some examples, a (real) number can be assigned as the value to each node 2020, ..., 2032 of the neural network 2000. Here, x(n)i represents the value of the i-th node 2020, ..., 2032 of the n-th layer 2010, ..., 2013. The values ​​of nodes 2020, ..., 2022 of the input layer 2010 are equivalent to the input values ​​of the neural network 2000, and the values ​​of nodes 2031, 2032 of the output layer 2013 are equivalent to the output values ​​of the neural network 2000. Furthermore, each edge 2040, ..., 2042 may include a weight that is a real number, for example, a real number within the interval [-1, 20] or the interval [0, 20]. Here, w(m,n)i,j represents the weight of the edge between the i-th node 2020, ..., 2032 of layer m (2010, ..., 2013) and the j-th node 2020, ..., 2032 of layer n (2010, ..., 2013). Furthermore, the abbreviation w(n)i,j is defined for the weight w(n,n+1)i,j.

[0140] In some examples, to compute the output value of neural network 2000, the input values ​​are propagated through the neural network. For instance, the value of node 2020, ..., 2032 in layer (n+1) can be computed based on the values ​​of nodes 2020, ..., 2032 in layer n (2010, ..., 2013) as follows:

[0141] Here, function f is the transfer function (another term is "activation function"). Known transfer functions include step functions, sigmoid functions (e.g., logic functions, generalized logic functions, hyperbolic tangent functions, arctangent functions, error functions, smoothed step functions), or rectifier functions. This transfer function is primarily used for normalization purposes.

[0142] In some examples, these values ​​are propagated layer by layer through the neural network, where the value of input layer 2010 is given by the input of neural network 2000, where the value of first hidden layer 2011 can be calculated based on the value of input layer 2010 of the neural network, where the value of second hidden layer 2012 can be calculated based on the value of first hidden layer 2011, and so on.

[0143] To set the values ​​of the edges w(m,n)i,j, a neural network 2000 must be trained using training data. For example, the training data includes training input data and training output data (denoted as ti). For the training step, the neural network 2000 is applied to the training input data to generate the computed output data. In some examples, the training data and computed output data include a number of values ​​equal to the number of nodes in the output layer.

[0144] In some examples, the weights within the neural network are recursively adjusted using a comparison between the calculated output data and the training data (backpropagation algorithm). For example, the weights are changed as follows:

[0145] Where γ is the learning rate, and the number δ(n)j can be recursively calculated as follows:

[0146] Based on δ(n+1)j, if the (n+1)th layer is not an output layer, and

[0147] If the (n+1)th layer is the output layer 2013, where f' is the first derivative of the activation function, and y(n+1)j is the comparison training value of the j-th node of the output layer 2013.

[0148] Figure 10 A block diagram of a data processing system 1000 (also referred to as a computer system) is shown, in which one implementation may be carried out, for example, as part of a product system, and / or as another system operatively configured by software or otherwise to perform the processes described herein, for example, as an implementation of the chatbot described above. The data processing system 1000 may include, for example, the computer or IT system or data processing system 100 described above. The illustrated data processing system includes at least one processor 1002 (e.g., a CPU) that may be connected to one or more of bridge / controller / buses 1004 (e.g., a northbridge, a southbridge). For example, one of the buses 1004 may include one or more I / O buses, such as a PCI Express bus. In the illustrated example, main memory 1006 (RAM) and a graphics controller 1008 may also be connected to the various buses. The graphics controller 1008 may be connected to one or more display devices 1010. It should also be noted that in some implementations, one or more controllers (e.g., a graphics controller, a southbridge) may be integrated with the CPU (on the same chip or wafer). Examples of CPU architectures include IA-32, x86-64, and ARM processor architectures.

[0149] Other peripheral devices connected to one or more buses may include a communication controller 1012 (Ethernet controller, WiFi controller, cellular controller) operably connected to a local area network (LAN), wide area network (WAN), cellular network and / or other wired or wireless network 1014 or communication devices.

[0150] Other components connected to various buses may include one or more I / O controllers 1016, such as USB controllers, Bluetooth controllers, and / or dedicated audio controllers (connected to speakers and / or microphones). It should also be understood that various peripheral devices may be connected to the I / O controller (via various ports and connections), including input devices 1018 (e.g., keyboards, mice, pointers, touchscreens, touchpads, drawing tablets, trackballs, buttons, keypads, game controllers, gamepads, cameras, microphones, scanners, motion sensing devices capturing motion gestures), output devices 1020 (e.g., printers, speakers), or any other type of device operable to provide input to or receive output from the data processing system. It should also be understood that many devices referred to as input devices or output devices can both provide input to and receive communication output from the data processing system. For example, processor 1002 may be integrated into a housing (such as a tablet computer) that includes a touchscreen that functions as both an input device and a display device. Furthermore, it should be understood that some input devices (such as laptops) may include multiple different types of input devices (e.g., touchscreens, touchpads, keyboards). It should also be understood that other peripheral hardware 1022 connected to the I / O controller 1016 may include any type of device, machine, or component configured to communicate with the data processing system.

[0151] Additional components connected to various buses may include one or more storage controllers 1024 (e.g., SATA). The storage controller may be connected to storage devices 1026, such as one or more storage drives and / or any associated removable media, which may be any suitable non-transitory machine-usable or machine-readable storage media. Examples include non-volatile devices, volatile devices, read-only devices, writable devices, ROMs, EPROMs, magnetic tape storage, floppy disk drives, hard disk drives, solid-state drives (SSDs), flash memory, optical disc drives (CD, DVD, Blu-ray), and other known optical, electrical, or magnetic storage devices and / or computer media. Furthermore, in some examples, storage devices such as SSDs may be directly connected to I / O buses 1004, such as the PCI Express bus.

[0152] A data processing system according to one embodiment of this application may include an operating system 1028, software / firmware 1030, and a data storage area 1032 (which may be stored on storage device 1026 and / or memory 1006). This operating system may employ a command-line interface (CLI) shell and / or a graphical user interface (GUI) shell. The GUI shell allows multiple display windows to be presented simultaneously in the graphical user interface, each providing an interface to different applications or different instances of the same application. The cursor or pointer in the graphical user interface can be manipulated by the user via a pointing device such as a mouse or a touchscreen. The position of the cursor / pointer can be changed and / or events such as clicking a mouse button or touching a touchscreen can be generated to initiate a desired response. Examples of operating systems that can be used in the data processing system include Microsoft Windows, Linux, UNIX, iOS, and Android operating systems. Furthermore, examples of data storage repositories include data files, data tables, relational databases (e.g., Oracle, Microsoft SQL Server), database servers, or any other structure and / or device capable of storing data that can be retrieved by a processor.

[0153] The communication controller 1012 may be connected to a network 1014 (not part of the data processing system 1000), which may be any public or private data processing system network or combination of networks known to those skilled in the art, including the Internet. The data processing system 1000 may communicate with one or more other data processing systems (such as server 1034, which is also not part of the data processing system 1000) via network 1014. However, alternative data processing systems may correspond to multiple data processing systems implemented as part of a distributed system, in which processors associated with several data processing systems may communicate via one or more network connections and may collectively perform tasks described as being performed by a single data processing system. Therefore, it should be understood that, when referring to a data processing system, such a system may be implemented across several data processing systems organized within a distributed system that communicate with each other via a network.

[0154] Furthermore, the term "controller" refers to any device, system, or part thereof that controls at least one operation, whether such a device is implemented in hardware, firmware, software, or some combination of at least two of these. It should be noted that the functionality associated with any particular controller can be centralized or distributed, whether local or remote.

[0155] Furthermore, it should be understood that the data processing system can be implemented as a virtual machine architecture or a virtual machine in a cloud environment. For example, processor 1002 and related components may correspond to a virtual machine running in a virtual machine environment on one or more servers. Examples of virtual machine architectures include VMware ESCi, Microsoft Hyper-V, Xen, and KVM.

[0156] Those skilled in the art will understand that the hardware shown for a data processing system can vary depending on the specific implementation. For example, the data processing system 1000 in this example may correspond to a computer, workstation, server, PC, laptop, tablet, mobile phone, and / or any other type of device / system capable of processing data and performing the functions and features associated with data processing systems, computers, processors, and / or controller operations discussed herein. The examples shown are provided for illustrative purposes only and are not intended to imply any architectural limitations on this application.

[0157] Furthermore, it should be noted that the processor described herein may reside in a server remote from the display and input devices described herein. In such an example, the display and input devices may be included in a client device that communicates with the server (and / or a virtual machine running on the server) via a wired or wireless network (which may include the Internet). In some embodiments, such a client device may, for example, execute a remote desktop application, or may correspond to a portal device that executes a remote desktop protocol with the server to send input from the input device to the server and receive visual information from the server for display via the display device. Examples of such remote desktop protocols include Teradici's PCoIP, Microsoft's RDP and RFB protocols. In such an example, the processor described herein may correspond to a virtual processor of a virtual machine running in the physical processor of the server.

[0158] As used herein, the terms “component” and “system” are intended to encompass hardware, software, or a combination of hardware and software. Thus, for example, a system or component can be a process, a process executing on a processor, or a processor itself. Furthermore, a component or system can reside on a single device or be distributed across several devices.

[0159] Furthermore, as used herein, a processor corresponds to any electronic device configured to process data via hardware circuitry, software, and / or firmware. For example, a processor as described herein may correspond to one or more (or combinations thereof) of a microprocessor, CPU, FPGA, ASIC, or any other integrated circuit (IC) or other type of circuit capable of processing data in a data processing system, which may take the form of a control board, computer, server, mobile phone, and / or any other type of electronic device.

[0160] Those skilled in the art will recognize that, for the sake of brevity and clarity, the complete structure and operation of all data processing systems applicable to this application are not shown or described herein. Instead, only the parts of the data processing system that are unique to this application or necessary for understanding this application are shown and described. The remaining construction and operation of the data processing system 1000 may conform to any of the various current implementations and practices known in the art.

[0161] Furthermore, it should be understood that the words or phrases used herein should be interpreted broadly, unless explicitly limited in some examples. For example, the terms “including” and “comprising” and their derivatives mean unlimited inclusion. Unless the context clearly indicates the opposite, the singular forms “a,” “an,” and “the” are intended to also include the plural forms. Additionally, as used herein, the term “and / or” refers to and covers any and all possible combinations of one or more associated listed items. Unless the context clearly indicates the opposite, the term “or” is inclusive, meaning and / or. The phrases “associated with” and “associated with” and their derivatives can mean including, being included, interconnected with, contained, contained within, connected to or connected to, coupled to or coupled to, able to communicate with, cooperate with, interleave, juxtapose, proximate to, bound to or bound to, have, have the nature of, etc.

[0162] Furthermore, although the terms “first,” “second,” “third,” etc., may be used herein to describe various elements, functions, or actions, these elements, functions, or actions should not be limited by these terms. Rather, these numerical adjectives are used to distinguish different elements, functions, or actions. For example, a first element, function, or action may be referred to as a second element, function, or action, and similarly, a second element, function, or action may be referred to as a first element, function, or action, without departing from the scope of this application.

[0163] Furthermore, phrases such as "the processor is configured to" perform one or more functions or processes can mean that the processor is operatively configured or operablely configured to perform the function or process via software, firmware, and / or wired circuitry. For example, a processor configured to perform a function / process can correspond to a processor executing software / firmware programmed to perform the function / process, and / or can correspond to a processor having software / firmware in a memory or storage device available for execution to perform the function / process. It should also be noted that a processor "configured to" perform one or more functions or processes can also correspond to processor circuitry (e.g., an ASIC or FPGA design) specifically manufactured or "wired" to perform the function or process. Additionally, the phrase "at least one" preceding an element (e.g., a processor) configured to perform more than one function can correspond to one or more elements (e.g., processors) each performing the function, and can also correspond to two or more elements (e.g., processors) performing different functions of the one or more different functions.

[0164] Furthermore, the term "proximate to" can mean: an element is relatively close to but does not contact another element; or an element is in contact with the other part, unless the context clearly indicates the opposite.

Claims

1. A computer-implemented method, comprising: ● Receive instructions to retrieve natural language input (120A) from datasets (122A, 122B) stored in at least two independent repositories (124A, 124B); ● Extract at least two acquisition parameters (126A, 126B) from the received input (120A) based on the configurable parameter types (128A, 128B); ● Obtain the dataset (122A, 122B) corresponding to the corresponding, extracted acquisition parameters (126A, 126B) from the repository (124A, 124B); as well as ● The corresponding, acquired datasets (122A, 122B) are stored in the data storage (130).

2. The computer-implemented method according to claim 1, further comprising: ●If the corresponding extracted acquisition parameter (126) does not conform to the configurable parameter type (128), then at least one aspect (126-I) of the corresponding extracted acquisition parameter (126) is determined to be invalid; ● Prompt the user to provide user input in natural language related to the corresponding invalid aspect (126-I); ● Receive natural language response input (120B) related to the corresponding invalid aspect (126-I); and ● Extract from the received response input (120B) the corresponding acquisition parameters (126) of the corresponding valid aspect (126-V) related to the previously invalid aspect (126-I).

3. The computer-implemented method according to claim 1 or 2, in, The configurable parameter type (128) represents the name, data type, and optionally the description of the corresponding acquisition parameter (126).

4. The computer-implemented method according to any one of the preceding claims further comprises: ● Receive a list of available, standalone repositories (124).

5. The computer-implemented method according to claim 4, further comprising: ● Determine the appropriate configurable parameter type (128) for each repository (124) and optionally for each dataset (122) stored in the respective repository (124).

6. The computer-implemented method according to any one of the preceding claims further comprises: ● Prompt the user to provide user input in natural language related to the connection (132) to the corresponding repository (124) and / or the dataset (122) to be retrieved; ● Receive natural language response input (120C) related to the connection (132) to the corresponding repository (124) and / or the dataset (122) to be acquired; and ● Extract the corresponding connection (132) from the received reply input (120C).

7. The computer-implemented method according to any one of the preceding claims further comprises: ● Receive natural language input (120D) that indicates whether to activate or deactivate the corresponding stored or acquired dataset (122); ● Receive natural language input (120E) from one or more datasets in the corresponding stored or acquired datasets (122) to process; as well as ● One or more datasets in the corresponding stored or retrieved dataset (122) are processed only if the corresponding stored or retrieved dataset (122) is activated or not deactivated.

8. The computer-implemented method according to claim 7, wherein, Processing one or more of the corresponding stored or retrieved datasets further includes: ●Draw the corresponding stored, retrieved dataset (122), optionally as a function of the same corresponding stored, retrieved dataset (122) and / or as a function of one or more other corresponding stored, retrieved datasets (122); and ● Create one or more illustrations (134) or one or more video sequences of the illustrations (134) from the function described in the drawing.

9. The computer-implemented method according to any one of the preceding claims, in, The corresponding acquired dataset (122) is associated with the corresponding device (138), such as the detectable observables (136) of the automated device in the automated system (140). The method further includes: ● Receive a document (142) relating to the corresponding, acquired dataset (122) and / or the corresponding device (138), wherein the document (142) represents at least one of device attributes, device components, device configuration, device performance parameters, device operating parameters, device status parameters, and device condition parameters, or any combination of device attributes, device components, device configuration, device performance parameters, device operating parameters, device status parameters, and device condition parameters; ● Identify at least one aspect (142-122) of the received corresponding document (142) in relation to the corresponding, acquired dataset (122); and ●For example, via a display device (112), the corresponding acquired dataset (122) and the relevant aspects (142-122) of the received document (142) are conveyed to the user.

10. The computer-implemented method according to claim 9, further comprising: ● Identify at least one difference (144) between the corresponding acquired dataset (122) and the corresponding received document (142) and the relevant corresponding aspects (142-122); and ● For example, information (144-I) regarding the corresponding differences (144) is conveyed to the user via a display device (112), and / or This results in a change in the operating state of the corresponding device (138) or the automation system (140) including the corresponding device (138).

11. The computer-implemented method according to claim 9 or 10, in, The corresponding acquired dataset (122) includes time-series data of the detectable observations (136) of the corresponding device (138).

12. The computer-implemented method according to any one of claims 9 to 11, in, The document (142) includes at least two distinct aspects (142-122), Wherein, the corresponding aspects (142-122) characterize the corresponding equipment attributes, corresponding equipment components, corresponding equipment configurations, corresponding equipment performance parameters, corresponding equipment operating parameters, corresponding equipment status parameters, and corresponding equipment condition parameters, or characterize any combination of the corresponding equipment attributes, corresponding equipment components, corresponding equipment configurations, corresponding equipment performance parameters, corresponding equipment operating parameters, corresponding equipment status parameters, and corresponding equipment condition parameters, and The method further includes: ●The analysis results are determined by analyzing the corresponding, acquired dataset (122) using the relevant aspects (142-122); and ●Communicate the confirmed analysis results to the user.

13. The computer-implemented method according to any one of the preceding claims, further comprising: ● The sequence of steps is stored in the data memory (130); ● Provide users with the corresponding sequence of steps to choose from; as well as ●If selected by the user, the corresponding sequence of steps is executed. The sequence of steps includes at least three of the following steps: o Retrieve the dataset (122) corresponding to the corresponding, extracted retrieval parameters (126) from the repository (124); o Determine the appropriate configurable parameter type (128) for each repository (124) and optionally for each dataset (122) stored in the respective repository (124); o One or more datasets in the corresponding stored or retrieved dataset (122) are processed only if the corresponding stored or retrieved dataset (122) is activated or not deactivated; o For example, via a display device (112), the corresponding stored or retrieved dataset (122) and / or processed, stored, retrieved dataset (122) are communicated to the user; o plots the corresponding stored, retrieved dataset (122) as a function of the same corresponding stored, retrieved dataset (122) and / or as a function of one or more other corresponding stored, retrieved datasets (122); o creates one or more illustrations (134) or one or more video sequences of the illustrations (134) from the function that draws them. o Display the corresponding illustration (134) or a video sequence of the illustration (134) created via the display device (112); o For example, via a display device (112), the relevant aspects (142-122) of the acquired dataset (122) and the received document (142) are conveyed to the user. o For example, information (144-I) about the corresponding differences (144) is conveyed to the user via a display device (112), and / or o Change the operating state of the corresponding device (138) or the automation system (140) including the corresponding device (138).

14. A computer system (100, 118) arranged and configured to perform the steps of a computer-implemented method according to any one of the preceding claims.

15. A computer program product (162) comprising computer program code that, when executed by a computer system (100, 118), causes the computer system (100, 118) to perform the method of any one of claims 1 to 13.

16. A computer-readable medium (160) comprising a computer program product (162) including computer program code that, when executed by a computer system (100, 118), causes the computer system (100, 118) to perform the method of any one of claims 1 to 13.