Ontology-based time series data extraction method, system and equipment and storage medium

Abstract

The invention provides an ontology-based time series data extraction method, system and device and a storage medium, and relates to the technical field of time series data processing.The method comprises the steps that an ontology model is constructed, and the ontology model comprises concepts, concept attributes and concept relations; determining a data extraction rule between the ontology model and the time series data, wherein the data extraction rule comprises a used ontology model and a concept name under the used ontology model; according to a data extraction rule, knowledge extraction is conducted on the time series data, a time series data ontology model is obtained, and the time series data ontology model represents the extracted knowledge graph. In this way, clear semantic information is provided for the time series data by constructing the ontology model, and therefore standardization and normalization processing of the time series data is achieved; moreover, knowledge extraction is carried out on the time series data according to the data extraction rule between the ontology model and the time series data, manual intervention is reduced, and the automation and intelligence level of the extraction process is improved.

Description

Technical Field

[0001] The present application relates to the technical field of time series data processing, and in particular to an ontology-based time series data extraction method, system, device and storage medium. Background Art

[0002] With the rapid development of information technology, time series data is being used more and more widely in various fields. When processing and analyzing time series data, time series data is extracted as knowledge (knowledge graph) for processing.

[0003] However, due to the lack of clear semantic information in time series data and the diversity of data representation methods, the interoperability between applications of different extraction systems is poor, and data integration and sharing become difficult. In addition, existing data extraction methods may rely on a large amount of manual intervention and sample training sets, with low automation and low efficiency. The extraction rules are often difficult to adapt to complex and changing data environments, resulting in inaccurate extraction results.

[0004] Therefore, an efficient and accurate time series data extraction method is urgently needed. Summary of the invention

[0005] In view of the above problems, the embodiments of the present application provide a method, system, device and storage medium for extracting time series data based on ontology, so as to overcome the above problems or at least partially solve the above problems.

[0006] In a first aspect of an embodiment of the present application, a method for extracting time series data based on ontology is disclosed, the method comprising: Constructing an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships; Determine a data extraction rule between the ontology model and the time series data, wherein the extraction rule includes: the ontology model used, and the concept name under the ontology model used; According to the data extraction rules, knowledge is extracted from the time series data to obtain a time series data ontology model, which represents the extracted knowledge graph.

[0007] Optionally, construct an ontology model, including: Based on business requirements and data characteristics, define the concepts involved in time series data, including: time series, data points, and time units; Defining the concept attributes of the concept, wherein the concept attributes include: the start time and end time of the time series, and the sampling frequency; A conceptual relationship between the concepts is defined, wherein the conceptual relationship includes: correlation between time series and sequential relationship between data points.

[0008] Optionally, knowledge extraction is performed on the time series data according to the data extraction rule to obtain a time series data ontology model, including: According to the concept name, instantiate the time series data into the ontology model used, wherein the instance includes an entity and entity attributes; Relationships are established between the instances to obtain a time series data ontology model, wherein the relationships include: correlation relationships and causal relationships.

[0009] Optionally, the method further comprises: Preprocessing the time series data to obtain preprocessed time series data; Performing feature extraction on the preprocessed time series data to obtain data features; According to the concept name, instantiating the time series data into the ontology model used includes: According to the concept name and the data feature, the time series data is instantiated into the ontology model used.

[0010] Optionally, feature extraction is performed on the preprocessed time series data to obtain data features, including: Extracting statistical features from the preprocessed time series data, wherein the statistical features are represented in the form of concept attributes in the ontology model; The trend characteristics of the preprocessed time series data are analyzed, and the trend characteristics are represented by concept relationships or specific concept attributes in the ontology model.

[0011] Optionally, the method further comprises: In the case where the time series data has periodicity, the period length and the variation pattern within the period of the preprocessed time series data are extracted to obtain periodic features, and the periodic features are represented in the form of concept attributes in the ontology model.

[0012] Optionally, the method further comprises: Performing data query according to the time series data ontology model; The potential rules of the time series data are analyzed according to the time series data ontology model, so as to perform trend prediction and anomaly discovery of the time series data based on the potential rules.

[0013] In a second aspect of the embodiment of the present application, a time series data extraction system based on ontology is disclosed, the system comprising: An ontology construction module is used to construct an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships; A rule determination module is used to determine the data extraction rules between the ontology model and the time series data, wherein the extraction rules include: the ontology model used, and the concept name under the ontology model used The knowledge extraction module is used to extract knowledge from the time series data according to the data extraction rules to obtain a time series data ontology model, wherein the time series data ontology model represents the extracted knowledge graph.

[0014] The third aspect of the embodiments of the present application discloses an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the ontology-based time series data extraction method described in the first aspect of the embodiments of the present application when executing the computer program.

[0015] The fourth aspect of the embodiments of the present application discloses a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of the ontology-based time series data extraction method described in the first aspect of the embodiments of the present application are implemented.

[0016] The embodiments of the present application include the following advantages: In the embodiments of the present application, the ontology model includes concepts, concept attributes, and concept relationships. By constructing the ontology model, clear semantic information is provided for the time series data, thereby achieving standardization and normalization of the time series data, thereby providing interoperability of the time series data between different systems and applications, and promoting data integration and sharing. In addition, according to the data extraction rules between the ontology model and the time series data, knowledge is extracted from the time series data to reduce manual intervention and improve the automation and intelligence level of the extraction process. Moreover, the data extraction rules are determined based on the ontology model, which can better adapt to complex and changeable data environments and improve the accuracy and reliability of the extraction results. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the description of the embodiments of the present application will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0018] Figure 1 It is a flowchart of the steps of a method for extracting time series data based on ontology provided in an embodiment of the present application; Figure 2 is a schematic diagram of an ontology model provided in an embodiment of the present application; Figure 3 is a flowchart of another ontology-based time series data extraction method provided in an embodiment of the present application; Figure 4 It is a structural diagram of an ontology-based time series data extraction system provided in an embodiment of the present application; Figure 5It is a structural diagram of another ontology-based time series data extraction system provided in an embodiment of the present application; Figure 6 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] In order to better understand the technical solution of the present application, a brief introduction to the technical concepts involved in the present application is given.

[0021] Knowledge graph: Knowledge graph is a knowledge storage and retrieval method that uses a graphical structure to represent the relationship and attribute information between entities. It can integrate knowledge from different fields into a unified knowledge base, facilitating the organization, analysis and mining of information.

[0022] Knowledge extraction: Knowledge extraction is the process of automatically extracting structured knowledge from text or data. In structured data, knowledge can be extracted by parsing and extracting information such as fields, attributes, and relationships. Semi-structured data refers to data that has a certain structure but does not strictly follow a predetermined pattern. Common forms include XML (Extensible Markup Language), HTML (Hyper Text Markup Language), JSON (JavaScript Object Notation, a lightweight data exchange format), etc. Time series data is a type of semi-structured data.

[0023] In the related art, time series data extraction is mainly achieved through the following four methods.

[0024] (1) Time series data extraction based on machine learning. Time series data is processed and analyzed using machine learning algorithms (such as classification, clustering, regression, etc.). The patterns and features in the data are identified by training models, thereby achieving automatic extraction and analysis of time series data. Although this method can automatically learn the inherent laws and features of the data without manually defining complex ontology structures, it has high efficiency and scalability for large-scale data processing. However, model training requires a large amount of data and computing resources; the accuracy and effect of the model depend on the quality and diversity of the training data.

[0025] (2) Time series data extraction based on deep learning. Time series data is processed using deep learning models (such as long short-term memory networks (LSTM), recurrent neural networks (GRU), and attention-based networks (Transformer). These models can capture long-term dependencies and complex patterns in the data, thereby achieving accurate extraction and analysis of time series data. This method has powerful nonlinear modeling and feature learning capabilities; it can process high-dimensional, nonlinear, and complex time series data. However, model training requires more data and computing resources, the model's interpretability is relatively poor, and it is prone to problems such as overfitting.

[0026] (3) Template-based time series data extraction. Define a set of templates or rules according to the format and structure of the time series data, and extract key information from the data by matching the templates; this method is similar to regular expression matching or string parsing. This method is simple to implement and does not require complex model training. It has a high extraction accuracy for time series data with a fixed format. However, the writing and maintenance of templates requires more manual intervention, and has poor adaptability to time series data with variable formats.

[0027] (4) Time series data extraction based on graph database. The time series data is constructed as a graph structure, and a graph database (such as Neo4j, JanusGraph, etc.) is used for storage and query. The time series data is extracted and analyzed through graph traversal and query algorithms. Although this method can intuitively represent the complex relationships between data and supports efficient graph traversal and query operations, the construction and maintenance of graph databases require a high technical threshold, and the processing performance of large-scale data may be limited.

[0028] In summary, existing time series data extraction methods have problems such as inaccurate data extraction, low efficiency, and poor data comprehensibility. In order to overcome the limitations of related technologies, the embodiment of the present application provides an ontology-based time series data extraction method, which realizes the standardized and systematic time series data extraction by constructing clear probability, concept attribute and concept relationship definitions.

[0029] Reference Figure 1 As shown, Figure 1 is a flowchart of a method for extracting time series data based on ontology provided in an embodiment of the present application. Figure 1 As shown, a method for extracting time series data based on ontology provided in an embodiment of the present application may include steps S110 to S130: Step S110: constructing an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships.

[0030] In the embodiments of the present application, a concept refers to a concept of a table in time series data; a concept attribute refers to an attribute related to a concept, and a concept attribute can be each field of a table in time series data; a concept relationship refers to a relationship between different concepts. For example, a time series in time series data is a concept, the start time of the time series is a concept attribute, and the correlation between time series is a concept relationship.

[0031] For example, Figure 2 It is a schematic diagram of an ontology model provided in an embodiment of the present application. The concepts in the ontology model may include multiple concepts, each concept has one or more conceptual attributes, and a concept may have a conceptual relationship with one or more concepts.

[0032] In a specific implementation method, constructing an ontology model includes: defining concepts involved in time series data according to business needs and data characteristics, wherein the concepts include: time series, data points and time units; defining conceptual attributes of the concepts, wherein the conceptual attributes include: the start time and end time of the time series, and the sampling frequency; defining conceptual relationships between the concepts, wherein the conceptual relationships include: the correlation between time series and the sequential relationship between data points.

[0033] In this way, by defining concepts, concept attributes and concept relationships, clear semantic information is provided for time series data, so that the standardization and normalization of time series data can be realized based on the ontology model in the future, for example, knowledge extraction and data query.

[0034] Step S120: Determine data extraction rules between the ontology model and time series data, wherein the data extraction rules include: the ontology model used, and the concept name under the ontology model used.

[0035] In an embodiment of the present application, the time series data may be data obtained from a time series database; for example, the time series data may be data in an iotdb database (wherein the iotdb database is a database that stores time series data). When extracting knowledge from time series data, it is necessary to use the ontology model and concepts to find the corresponding instances and relationships in the time series data in order to achieve extraction. The tables and fields contained in different time series data may be different, and the corresponding data extraction rules are also different. Specifically, when determining the data extraction rules between the ontology model and the time series data, you can first select an ontology model to be used, and select a concept name based on the table in the time series data to obtain the data extraction rules.

[0036] In this way, data extraction rules are automatically generated through ontology models and time series data, reducing human intervention and improving the automation and intelligence level of the extraction process. The data extraction rules can better adapt to complex and changing data environments and improve the accuracy and reliability of the extraction results.

[0037] Step S130: According to the data extraction rules, knowledge is extracted from the time series data to obtain a time series data ontology model, wherein the time series data ontology model represents the extracted knowledge graph.

[0038] In the embodiment of the present application, according to the ontology model and concept name determined in the data extraction rule, the corresponding instances and relationships are found from the time series data to perform multiple instance extractions and relationship extractions to obtain the time series data ontology model. Specifically, the instances are used as knowledge points in the knowledge graph, and the relationships between the instances are used as the edges of the knowledge graph, so that the knowledge graph is obtained through multiple instance extractions and relationship extractions, that is, the time series data ontology model is obtained.

[0039] In a specific implementation method, knowledge is extracted from the time series data according to the data extraction rule to obtain a time series data ontology model, including step D1 and step D2: Step D1: According to the concept name, instantiate the time series data into the ontology model used, wherein the instance includes entities and entity attributes.

[0040] Step D2: Establish relationships between the instances to obtain a time series data ontology model, wherein the relationships include: correlation relationships and causal relationships.

[0041] In an embodiment of the present application, entities include time series, data points, and time units; when instantiating the time series data into the ontology model used, the time series, data points, or time units are used as entities according to the concept name, and the attributes of the entity are instantiated together into the ontology model used, that is, as a knowledge point in the knowledge graph. For example, taking the time series as an entity, its corresponding entity attributes include: attribute information such as statistical characteristics (characterizing the statistical characteristics of the time series), trend characteristics (characterizing the trend changes of the time series), and periodic characteristics (characterizing the periodicity of the time series). Next, a relationship is established between the instances, that is, the edges of the knowledge graph are formed, thereby obtaining a complete time series data ontology model. In addition, after obtaining the time series data ontology model, the time series data ontology model can be stored in a graph database.

[0042] Through the above implementation process, a unified ontology model is constructed to provide clear semantic information for time series data, thereby realizing the standardization and normalization of time series data, thereby providing interoperability between different systems and applications, and promoting data integration and sharing. In addition, according to the data extraction rules between the ontology model and time series data, knowledge is extracted from time series data to reduce manual intervention and improve the automation and intelligence level of the extraction process. Moreover, the data extraction rules are determined based on the ontology model, which can better adapt to the complex and changing data environment and improve the accuracy and reliability of the extraction results. In this way, a more accurate, efficient, easy to understand and reuse time series data extraction method is realized.

[0043] In combination with the above embodiments, in one implementation, the present application embodiment further provides a method for extracting time series data based on ontology. In this method, in addition to the above steps, the following steps may also be included: Step A1: preprocess the time series data to obtain preprocessed time series data.

[0044] Specifically, the time series data is cleaned and converted. Data cleaning refers to cleaning the time series data, removing noise, outliers, and duplicate data; data conversion refers to converting the time series data into a form suitable for ontology model representation as needed, for example, discretizing or standardizing continuous data.

[0045] Step A2: extracting features from the preprocessed time series data to obtain data features.

[0046] In an embodiment of the present application, key data features are extracted from preprocessed time series data, and the extracted data features are represented in the ontology model in the form of concept attributes or concept relationships.

[0047] Specifically, feature extraction is performed on the preprocessed time series data to obtain data features, including sub-steps A11 to A12: Step A11: extracting statistical features from the preprocessed time series data, wherein the statistical features are represented in the form of concept attributes in the ontology model.

[0048] Step A12: Analyze the trend characteristics of the pre-processed time series data, where the trend characteristics are represented by concept relationships or specific concept attributes in the ontology model.

[0049] In the embodiment of the present application, statistical features refer to statistical characteristics that can reflect time series data, for example, statistical features can be features such as mean, variance, standard deviation, etc. Trend features can reflect the changing trend of time series data, for example, trend features can be changing trends such as linear trend and nonlinear trend.

[0050] Optionally, feature extraction is performed on the preprocessed time series data to obtain data features, further comprising sub-step A13: Step A13: When the time series data is periodic, extract the period length and the change pattern within the period of the preprocessed time series data to obtain periodic features, and the periodic features are represented in the form of concept attributes in the ontology model.

[0051] For example, for a time series in the preprocessed time series data, the period length (e.g., 10 seconds is a period length) and the change pattern within the period (e.g., the data changes in the form of a sine function within a period) can be determined according to the change of the time series.

[0052] Furthermore, in the above step D1, "according to the concept name, instantiating the time series data into the ontology model used" specifically includes: according to the concept name and the data feature, instantiating the time series data into the ontology model used.

[0053] In an embodiment of the present application, when instantiating the time series data into the ontology model used, the data features obtained in the above step A3 are instantiated into the ontology model as entity attributes, thereby obtaining knowledge points in the knowledge graph; then, by establishing relationships between instances, a complete time series data ontology model is obtained.

[0054] In one embodiment, the time series data comes from a time series database (for example, an iotdb database). Before knowledge extraction is performed on the time series data, it is also necessary to dynamically connect to the time series database and map the ontology model in use with the time series database. Specifically, dynamically connecting to the time series database includes: configuring the service IP, port, user name, password, root path, group, device, and physical quantity of the time series database to query a single or multiple data in the time series database, thereby dynamically connecting to the time series database. In this way, based on the mapping between the ontology model and the time series database, and according to the data extraction rules, knowledge extraction is performed on the time series data to obtain a time series data ontology model.

[0055] In combination with the above embodiments, in one implementation, the present application embodiment further provides a method for extracting time series data based on ontology. In this method, in addition to the above steps, the following steps may also be included: Step B1: perform data query according to the time series data ontology model.

[0056] Step B2: Analyze the potential rules of the time series data according to the time series data ontology model, so as to perform trend prediction and anomaly discovery of the time series data based on the potential rules.

[0057] In an embodiment of the present application, the time series data is extracted into a time series data ontology model (i.e., a knowledge graph). The time series data ontology model can better reflect the potential information and rules of the time series data through the relationship between instances. Therefore, data query based on the time series data ontology model is used to improve the efficiency of data query; and query and analysis operations are performed based on the time series data ontology model to deeply explore the potential information and rules of the time series data center, and provide strong support for predicting data trends, discovering anomalies, etc. through potential information and rules. In some embodiments, the correlation analysis between time series, attribute-based data filtering, relational reasoning and other operations can also be performed based on the time series data ontology model. In this way, complex data query and data analysis are realized through the time series data ontology model, which enhances the query and analysis capabilities of time series data.

[0058] The following is a specific example to illustrate the ontology-based time series data extraction method provided in the embodiment of the present application. Figure 3 As shown, Figure 3 It is a flowchart of another ontology-based time series data extraction method provided in an embodiment of the present application. Specifically, the method includes steps S310 to S370: Step S310: construct an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships.

[0059] Specifically, according to business needs and data characteristics, the concepts involved in time series data are defined, and the concepts include: time series, data points and time units; the conceptual attributes of the concepts are defined, and the conceptual attributes include: the start time and end time of the time series, and the sampling frequency; the conceptual relationships between the concepts are defined, and the conceptual relationships include: the correlation between time series and the sequential relationship between data points.

[0060] Step S320: Determine data extraction rules between the ontology model and time series data, wherein the data extraction rules include: the ontology model used, and the concept name under the ontology model used.

[0061] Step S330: preprocess the time series data to obtain preprocessed time series data.

[0062] Specifically, the time series data is cleaned and converted. Data cleaning refers to cleaning the time series data, removing noise, outliers, and duplicate data; data conversion refers to converting the time series data into a form suitable for ontology model representation as needed, for example, discretizing or standardizing continuous data.

[0063] Step S340: extracting features from the preprocessed time series data to obtain data features.

[0064] Specifically, statistical features are extracted from the pre-processed time series data, and the statistical features are represented in the form of concept attributes in the ontology model; trend features of the pre-processed time series data are analyzed, and the trend features are represented by concept relationships or specific concept attributes in the ontology model. In the case where the time series data has periodicity, the period length and the change pattern within the period of the pre-processed time series data are extracted to obtain periodic features, and the periodic features are represented in the form of concept attributes in the ontology model.

[0065] Step S350: instantiate the time series data into the ontology model used according to the concept name and the data feature.

[0066] Step S360: Establish relationships between the instances to obtain a time series data ontology model, wherein the relationships include: correlation relationships and causal relationships.

[0067] Step S370: performing data query according to the time series data ontology model; and analyzing the potential rules of the time series data according to the time series data ontology model to perform trend prediction and anomaly discovery of the time series data based on the potential rules.

[0068] Through the above implementation process, the standardization and normalization of time series data are realized through the processes of ontology model construction, data preprocessing, data feature extraction, time series data instantiation and instance relationship establishment, so as to provide interoperability of time series data between different systems and applications and promote data integration and sharing. According to the data extraction rules between the ontology model and time series data, knowledge is extracted from time series data to reduce manual intervention and improve the automation and intelligence level of the extraction process. Moreover, the data extraction rules are determined according to the ontology model, which can better adapt to the complex and changing data environment and improve the accuracy and reliability of the extraction results. In addition, the fast and accurate extraction and in-depth analysis of time series data are realized through the time series data ontology model, which enhances the query and analysis capabilities of time series data.

[0069] In this way, the technical problems existing in the representation, extraction, query, integration, sharing and deep mining of time series data are solved, the efficiency and quality of data processing are improved, and strong support is provided for data analysis and decision-making.

[0070] The present application also provides a time series data extraction system based on ontology, referring to Figure 4 As shown, Figure 4 : is a structural diagram of a time series data extraction system based on ontology provided in an embodiment of the present application, the system comprising: An ontology construction module 410 is used to construct an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships; The rule determination module 420 is used to determine the data extraction rules between the ontology model and the time series data. The data extraction rules include: the ontology model used, and the concept name under the ontology model used. The knowledge extraction module 430 is used to extract knowledge from the time series data according to the data extraction rules to obtain a time series data ontology model, and the time series data ontology model represents the extracted knowledge graph.

[0071] Based on this system, a unified ontology model is constructed to provide clear semantic information for time series data, thereby realizing the standardization and normalization of time series data, thereby providing interoperability between different systems and applications, and promoting data integration and sharing. In addition, according to the data extraction rules between the ontology model and time series data, knowledge is extracted from time series data, reducing manual intervention and improving the automation and intelligence level of the extraction process. Moreover, the data extraction rules are determined based on the ontology model, which can better adapt to complex and changing data environments and improve the accuracy and reliability of the extraction results.

[0072] In some optional embodiments, referring to Figure 5 As shown, Figure 5 It is a structural diagram of another ontology-based time series data extraction system provided in an embodiment of the present application. Specifically, the system includes: an ontology construction module 410, a rule determination module 420 and a knowledge extraction module 430.

[0073] Specifically, the ontology building module 410 includes: The concept definition module 4101 is used to define the concepts involved in the time series data according to business requirements and data characteristics, and the concepts include: time series, data points and time units; Attribute definition module 4102, used to define the concept attributes of the concept, the concept attributes include: the start time and end time of the time series, and the sampling frequency; The relationship definition module 4103 is used to define the conceptual relationship between the concepts, and the conceptual relationship includes: the correlation between time series and the sequential relationship between data points.

[0074] In an optional embodiment, the knowledge extraction module 430 includes: An instantiation module 4301 is used to instantiate the time series data to the ontology model used according to the concept name, wherein the instance includes an entity and entity attributes; The relationship establishment module 4302 is used to establish relationships between the instances to obtain a time series data ontology model, wherein the relationships include: correlation relationships and causal relationships.

[0075] In an optional embodiment, the system further includes: A data preprocessing module 440 is used to preprocess the time series data to obtain preprocessed time series data; A feature extraction module 450 is used to extract features from the preprocessed time series data to obtain data features; The instantiation module 4301 is also used to instantiate the time series data into the ontology model used according to the concept name and the data characteristics.

[0076] In an optional embodiment, the feature extraction module 450 includes: A statistical feature extraction module 4501 is used to extract statistical features from the preprocessed time series data, wherein the statistical features are represented in the form of concept attributes in the ontology model; The trend feature extraction module 4502 is used to analyze the trend features of the pre-processed time series data, and the trend features are represented by the concept relationship or specific concept attributes in the ontology model.

[0077] In an optional embodiment, the feature extraction module 450 further includes: The periodic feature extraction module 4503 is used to extract the period length and the change pattern within the period of the preprocessed time series data when the time series data is periodic, and obtain the periodic feature. The periodic feature is represented in the form of concept attributes in the ontology model.

[0078] In an optional embodiment, the system includes: A data query module 460, configured to perform data query according to the time series data ontology model; The data analysis module 470 is used to analyze the potential rules of the time series data according to the time series data ontology model, so as to perform trend prediction and anomaly discovery of the time series data based on the potential rules.

[0079] The present application also provides an electronic device, referring to Figure 6 , Figure 6 Schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 6 As shown, the electronic device 600 includes: a memory 610 and a processor 620. The memory 610 and the processor 620 are connected via a bus communication. A computer program is stored in the memory 610. The computer program can be run on the processor 620 to implement the steps of the ontology-based time series data extraction method described in the embodiment of the present application.

[0080] The embodiment of the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the ontology-based time series data extraction method described in the embodiment of the present application are implemented.

[0081] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0082] The embodiments of the present application are described with reference to the flowcharts and / or block diagrams of the methods, systems and devices according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0083] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0084] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable terminal device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0085] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present application.

[0086] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0087] The above is a detailed introduction to the ontology-based time series data extraction method, system, device and storage medium provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A time series data extraction method based on ontology, characterized in that: The method comprises: Constructing an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships; Determine a data extraction rule between the ontology model and the time series data, wherein the data extraction rule includes: the ontology model used, and the concept name under the ontology model used; According to the data extraction rules, knowledge is extracted from the time series data to obtain a time series data ontology model, which represents the extracted knowledge graph.

2. The method according to claim 1, characterized in that Construct an ontology model, including: Based on business requirements and data characteristics, define the concepts involved in time series data, including: time series, data points, and time units; Defining the concept attributes of the concept, wherein the concept attributes include: the start time and end time of the time series, and the sampling frequency; A conceptual relationship between the concepts is defined, wherein the conceptual relationship includes: correlation between time series and sequential relationship between data points.

3. The method according to claim 1, characterized in that According to the data extraction rules, knowledge is extracted from the time series data to obtain a time series data ontology model, including: According to the concept name, instantiate the time series data into the ontology model used, wherein the instance includes an entity and entity attributes; Relationships are established between the instances to obtain a time series data ontology model, wherein the relationships include: correlation relationships and causal relationships.

4. The method according to claim 3, characterized in that The method further comprises: Preprocessing the time series data to obtain preprocessed time series data; Performing feature extraction on the preprocessed time series data to obtain data features; According to the concept name, instantiating the time series data into the ontology model used includes: According to the concept name and the data feature, the time series data is instantiated into the ontology model used.

5. The method according to claim 4, characterized in that Feature extraction is performed on the preprocessed time series data to obtain data features, including: Extracting statistical features from the preprocessed time series data, wherein the statistical features are represented in the form of concept attributes in the ontology model; The trend characteristics of the preprocessed time series data are analyzed, and the trend characteristics are represented by concept relationships or specific concept attributes in the ontology model.

6. The method according to claim 5, characterized in that The method further comprises: In the case where the time series data has periodicity, the period length and the variation pattern within the period of the preprocessed time series data are extracted to obtain periodic features, and the periodic features are represented in the form of concept attributes in the ontology model.

7. The method according to claim 1, characterized in that The method further comprises: Performing data query according to the time series data ontology model; The potential rules of the time series data are analyzed according to the time series data ontology model, so as to perform trend prediction and anomaly discovery of the time series data based on the potential rules.

8. A time series data extraction system based on ontology, characterized in that: The system comprises: An ontology construction module is used to construct an ontology model, wherein the ontology model includes: concepts, concept attributes, and concept relationships; A rule determination module is used to determine the data extraction rules between the ontology model and the time series data, wherein the data extraction rules include: the ontology model used, and the concept name under the ontology model used The knowledge extraction module is used to extract knowledge from the time series data according to the data extraction rules to obtain a time series data ontology model, wherein the time series data ontology model represents the extracted knowledge graph.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the ontology-based time series data extraction method described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the ontology-based temporal data extraction method described in any one of claims 1 to 7 are implemented.

Images