A data processing method, device and equipment for rail transportation

Abstract

The present invention discloses a method, device, and equipment for processing rail transit data. The method comprises: obtaining data to be optimized; processing the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operational characteristics of rail transit in different dimensions; and outputting optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized. Through the above-described methods, the present invention optimizes the storage and processing methods for rail transit data processing and improves the utilization rate of rail transit data.

Description

Technical Field

[0001] The present invention relates to the field of rail transit technology, and in particular to a data processing method, device and equipment for rail transit. Background Art

[0002] Urban rail transit plays an important role in promoting urban development. Based on this, the operation and maintenance of urban rail transit is particularly important.

[0003] Due to the vast variety of rail transit equipment, rail transit-related data is equally complex and diverse. For example, the technical documentation, operating manuals, maintenance records, and other related text materials for a single vehicle model can exceed one million words. This makes it difficult to store rail transit data using a unified data format, which in turn makes data querying more difficult and reduces the utilization rate of this massive amount of rail transit-related data. Furthermore, the complexity and diversity of rail transit data make it difficult to deploy a single data processing model to uniformly handle all rail transit data. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a rail transit data processing method, device, and apparatus that overcome the above problems or at least partially solve the above problems.

[0005] According to one aspect of an embodiment of the present invention, a method for processing rail transit data is provided, comprising:

[0006] Get the data to be optimized;

[0007] Processing the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions;

[0008] Based on the obtained features to be optimized, optimization information matching the scenario corresponding to the data to be optimized is output.

[0009] Optionally, before obtaining the data to be optimized, the following steps are also included:

[0010] Get the word vector of the original data;

[0011] Obtaining a first semantic feature of the word vector according to the word vector of the original data;

[0012] Obtaining a second semantic feature of the word vector based on the first semantic feature, wherein the second semantic feature and the first semantic feature respectively represent the operation characteristics of the rail transit from different dimensions;

[0013] The rail transit knowledge graph is generated according to at least one of the first semantic feature and the second semantic feature.

[0014] Optionally, the original data includes: cloud data, basic data, historical operation and maintenance data, real-time operation and maintenance data and external environment data.

[0015] Optionally, before obtaining the data to be optimized, the following steps are also included:

[0016] If the original data includes target original data corresponding to the data to be optimized, the network parameters are optimized by at least one of the first semantic feature and the second semantic feature, and the target original data, where the target original data refers to a training label corresponding to the data to be optimized.

[0017] Optionally, obtaining a first semantic feature of the word vector based on the word vector of the original data, and obtaining a second semantic feature of the word vector based on the first semantic feature includes:

[0018] Calling the first sub-model to process the word vector of the original data to obtain the first semantic feature;

[0019] calling a second sub-model to process the first semantic feature to obtain the second semantic feature;

[0020] The first sub-model is a bidirectional GRU network model, and the second sub-model is a bidirectional LSTM model.

[0021] Optionally, the data to be optimized is processed according to a preset rail transit knowledge graph to obtain features to be optimized, including:

[0022] According to the algorithm t =f(Vs t +V′s′ t ) Get the feature to be optimized o t , where f is the activation function, V is the output weight coefficient of the forward hidden layer, and s t is the output of the forward hidden layer at the second moment, t is the second moment, V′ is the output weight coefficient of the reverse hidden layer, s′ t is the reverse hidden layer output at the second moment.

[0023] Optionally, the s t Satisfaction: t =f(Ux t +Ws t-1 );

[0024] The s′ t Satisfaction: s′ t =f(U′x t +W′s′t+1 ); where U is the input weight coefficient of forward propagation, x t is the data to be optimized at the second moment, W is the weight coefficient of the forward hidden layer, s t-1 is the forward hidden layer output at the first moment, t-1 is the first moment, the first moment is earlier than the second moment, U′ is the input weight coefficient of back propagation, W′ is the weight coefficient of the reverse hidden layer, s′ t+1 is the reverse hidden layer output at the third moment, t+1 is the third moment, and the third moment is later than the second moment.

[0025] According to another aspect of an embodiment of the present invention, a data processing device for rail transit is provided, comprising:

[0026] Acquisition module, used to obtain data to be optimized;

[0027] a processing module, configured to process the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions;

[0028] The output module is used to output optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized.

[0029] According to another aspect of an embodiment of the present invention, there is provided a computing device, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus;

[0030] The memory is used to store at least one executable instruction, and the executable instruction enables the processor to execute operations corresponding to the above-mentioned rail transit data processing method.

[0031] According to another aspect of an embodiment of the present invention, a computer storage medium is provided, wherein the storage medium stores at least one executable instruction, and the executable instruction enables a processor to execute operations corresponding to the above-mentioned rail transit data processing method.

[0032] According to the solution provided by the above-mentioned embodiment of the present invention, by acquiring data to be optimized; processing the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, the knowledge graph represents the operating characteristics of the rail transit in different dimensions; based on the obtained features to be optimized, outputting optimization information that matches the scenario corresponding to the data to be optimized, a knowledge graph of rail transit can be constructed, which solves the problem of complex or unprocessable data processing caused by the huge amount of rail transit data, optimizes the storage and processing methods of rail transit data processing, and improves the utilization rate of rail transit data.

[0033] The above description is only an overview of the technical solutions of the embodiments of the present invention. In order to more clearly understand the technical means of the embodiments of the present invention, they can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiments of the present invention more obvious and easy to understand, the specific implementation methods of the embodiments of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the embodiments of the present invention. The same reference numerals are used throughout the accompanying drawings to denote the same components. In the accompanying drawings:

[0035] Figure 1 A flow chart of a method for processing rail transit data provided by an embodiment of the present invention is shown;

[0036] Figure 2 A flow chart of constructing a rail transit knowledge graph provided by an embodiment of the present invention is shown;

[0037] Figure 3 A schematic diagram of a named entity recognition model of a specific word vector stacking model provided by an embodiment of the present invention is shown;

[0038] Figure 4 A flow chart of a specific rail transit data processing method provided by an embodiment of the present invention is shown;

[0039] Figure 5 A schematic structural diagram of a rail transit data processing device provided by an embodiment of the present invention is shown;

[0040] Figure 6 A schematic structural diagram of a computing device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0041] Exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0042] Figure 1 FIG. 1 is a flow chart showing a method for processing rail transit data according to an embodiment of the present invention. Figure 1 As shown, the method includes the following steps:

[0043] Step 11: Obtain the data to be optimized;

[0044] Step 12: Processing the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions;

[0045] Step 13: output optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized.

[0046] In this embodiment, data to be optimized is obtained; the data to be optimized is processed according to a preset rail transit knowledge graph to obtain features to be optimized, and the knowledge graph represents the operating characteristics of the rail transit in different dimensions; based on the obtained features to be optimized, optimization information matching the scenario corresponding to the data to be optimized is output, thereby optimizing the storage and processing methods of rail transit data processing and improving the utilization rate of rail transit data.

[0047] Figure 2 The flow chart of constructing rail transit knowledge graph provided by the embodiment of the present invention is shown as follows: Figure 2 As shown, in an optional embodiment of the present invention, before obtaining the data to be optimized, the method further includes:

[0048] Step 21, obtain the word vector of the original data;

[0049] Step 22: obtaining a first semantic feature of the word vector based on the word vector of the original data;

[0050] Step 23: obtaining a second semantic feature of the word vector based on the first semantic feature, wherein the second semantic feature and the first semantic feature respectively represent the operation characteristics of the rail transit from different dimensions;

[0051] Step 24: Generate the rail transit knowledge graph based on at least one of the first semantic feature and the second semantic feature.

[0052] Specifically, such as Figure 3As shown, taking the problem of "light strips flashing frequently" in rail transit as an example, first, word vectors are obtained for the original data, such as named entities such as "light strips" and "light strips often"; second, the word vectors of the original data are input into the first sub-model of the named entity recognition model of the word vector stacking model. After preliminary extraction by the GRU network, a first semantic feature is obtained. The first semantic feature is then input into the second sub-model of the named entity recognition model of the word vector stacking model for further extraction, and a second semantic feature of the word vector is output. Finally, a learning algorithm is used to learn at least one of the first semantic feature and the second semantic feature to generate a rail transit knowledge graph. In this case, the word vectors are pre-trained models that can be dynamically encoded according to contextual semantics, and the stacking model is a superposition of the first and second sub-models. This multi-layer hierarchical text processing method can improve the accuracy of named entity recognition by addressing the problems of nested and excessively long named entities in the rail transit field. At the same time, when generating the rail transit knowledge graph, the learning algorithm also includes active learning. The active learning algorithm can actively infer some annotation requests and submit some filtered data to experts for annotation, effectively reducing the time and complexity of knowledge graph production.

[0053] In this embodiment, the rail transit knowledge graph is constructed to describe rail transit domain knowledge. It typically extracts rail transit entities and rail transit relationships within the rail transit domain, and on this basis, constructs a semantic network. The knowledge contained in this semantic network is highly targeted and specialized. Rail transit entities are typically proper nouns within the rail transit domain, and the rail transit knowledge graph is deeply applied in practice in the rail transit field, with extremely high accuracy.

[0054] In another optional embodiment of the present invention, the original data includes: cloud data, basic data, historical operation and maintenance data, real-time operation and maintenance data and external environment data, but is not limited to the above.

[0055] In this embodiment, the cloud data generally refers to data stored on multiple virtual servers hosted by a third party, such as: health management analysis data of each device and real-time detection and operation data of the device, but not limited to the above; the basic data generally refers to basic information data, such as: device attribute data, device coding data and device maintenance manual data, but not limited to the above; the real-time operation and maintenance data generally refers to data generated by real-time maintenance and ensuring high availability of the entire service, such as: data for continuously optimizing the system architecture to improve deployment efficiency and optimizing resource utilization to improve overall ROI data, but not limited to the above; the external environment data generally refers to data located outside and has an indirect impact on all data other than external environment data, such as: weather data, but not limited to the above;

[0056] In another optional embodiment of the present invention, at least one of step 22 and step 23 further includes:

[0057] If the original data includes target original data corresponding to the data to be optimized, the network parameters are optimized by at least one of the first semantic feature and the second semantic feature, and the target original data, where the target original data refers to a training label corresponding to the data to be optimized.

[0058] In this embodiment, by performing a loss function operation on at least one of the first semantic feature and the second semantic feature and the target original data, the loss function can well reflect the gap between the model and the actual data, which can better facilitate subsequent analysis and understanding, thereby optimizing network parameters.

[0059] In another optional embodiment of the present invention, step 22 and step 23 may include:

[0060] Step 221: Call the first sub-model to process the word vector of the original data to obtain the first semantic feature;

[0061] Step 231: Call the second sub-model to process the first semantic feature to obtain the second semantic feature; the first sub-model is a bidirectional GRU network model, and the second sub-model is a bidirectional LSTM model.

[0062] Specifically, the first sub-model reads the word vector of the encoded original data and outputs a description of the phenomenon to be optimized, namely the first semantic feature. Secondly, the description of the phenomenon to be optimized is used as the input of the second sub-model, and finally a label sequence is output, namely the second semantic feature.

[0063] In this embodiment, two sub-models are used for training and recognition, so that potential named entity information is not lost.

[0064] In another optional embodiment of the present invention, the first sub-model and the second sub-model in steps 221 and 231 are loosely coupled, allowing the two-layer network to be trained independently. This loose coupling reduces training complexity. Furthermore, the first sub-model can perform preliminary recognition of named entities in a sentence, and the second sub-model can adjust and filter any incorrectly recognized entities, ultimately improving recognition accuracy.

[0065] In another optional embodiment of the present invention, step 12 may include:

[0066] Step 121, according to algorithm o t =f(Vs t +V′s′ t ) Get the feature to be optimized ot , where f is the activation function, V is the output weight coefficient of the forward hidden layer, and s t is the output of the forward hidden layer at the second moment, t is the second moment, V′ is the output weight coefficient of the reverse hidden layer, s′ t is the reverse hidden layer output at the second moment.

[0067] In this embodiment, the s t Satisfaction: t =f(Ux t +Ws t-1 );

[0068] The s′ t Satisfaction: s′ t =f(U′x t +W′s′ t+1 ); where U is the input weight coefficient of forward propagation, x t is the data to be optimized at the second moment, W is the weight coefficient of the forward hidden layer, s t-1 is the forward hidden layer output at the first moment, t-1 is the first moment, the first moment is earlier than the second moment, U′ is the input weight coefficient of back propagation, W′ is the weight coefficient of the reverse hidden layer, s′ t+1 is the reverse hidden layer output at the third moment, t+1 is the third moment, and the third moment is later than the second moment.

[0069] Figure 4 A flow chart of a specific rail transit data processing method provided by an embodiment of the present invention is shown. Figure 4 As shown in the figure, the method is mainly divided into three parts: input data, operation and maintenance data brain, and output service.

[0070] The first part is input data: that is, data collection, including collecting cloud data, basic data, historical operation and maintenance data, real-time operation and maintenance data, and external environment data.

[0071] The second part, the Operations and Maintenance Data Brain, involves model building. This involves using deep learning models based on deep networks, reinforcement learning, and natural language processing, along with big data-based data mining algorithms. This system autonomously learns the diverse knowledge contained in the input data and builds an unstructured rail transit knowledge graph. This knowledge graph continuously enriches the relevant knowledge base as operations and maintenance data grows.

[0072] The third part is outputting services: after the data to be optimized is input into the model described in the second part, the corresponding services are output according to the results of the model processing and combined with the actual scenarios of rail transit operation and maintenance. The output corresponding services mainly serve the decision-making of the leadership, the maintenance tasks of the operation and maintenance personnel, and the production mode optimization of the management personnel, but are not limited to the above.

[0073] In the above embodiment of the present invention, the user can establish a knowledge base in combination with the generated rail transit knowledge graph, and at the same time utilize the machine learning algorithm deployed in the background of the system. It mines any possible association relationship and automatically mines strong association relationships, such as the fault type, phenomenon, and cause of different components inside the equipment or the association between different professional equipment and the external environment. It utilizes the association rules, deep learning and other technologies deployed in the system, analyzes the fault phenomenon through natural language processing technology, extracts entities and keywords, and uses semantic similarity technology to find similar fault cases that have occurred in the past. Combined with the causes of similar fault cases, it analyzes the probability of the possible causes of the current fault case and recommends the optimal strategy.

[0074] Figure 5 FIG. 5 shows a schematic diagram of the structure of a data processing device 50 for rail transportation provided by an embodiment of the present invention. Figure 5 As shown, the device includes:

[0075] An acquisition module 51 is used to acquire data to be optimized;

[0076] A processing module 52 is configured to process the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions;

[0077] The output module 53 is configured to output optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized.

[0078] Optionally, the processing module 52 is further configured to obtain word vectors of the original data;

[0079] Obtaining a first semantic feature of the word vector according to the word vector of the original data;

[0080] Obtaining a second semantic feature of the word vector based on the first semantic feature, wherein the second semantic feature and the first semantic feature respectively represent the operation characteristics of the rail transit from different dimensions;

[0081] The rail transit knowledge graph is generated according to at least one of the first semantic feature and the second semantic feature.

[0082] Optionally, the original data includes: cloud data, basic data, historical operation and maintenance data, real-time operation and maintenance data and external environment data.

[0083] Optionally, the processing module 52 is also used to optimize network parameters through at least one of the first semantic feature and the second semantic feature, and the target original data if the original data includes target original data corresponding to the data to be optimized, and the target original data refers to the training label corresponding to the data to be optimized.

[0084] Optionally, the processing module 52 is further configured to call a first sub-model to process the word vector of the original data to obtain the first semantic feature;

[0085] calling a second sub-model to process the first semantic feature to obtain the second semantic feature;

[0086] The first sub-model is a bidirectional GRU network model, and the second sub-model is a bidirectional LSTM model.

[0087] Optionally, the processing module 52 is further configured to: t =f(Vs t +V′s′ t ) Get the feature to be optimized o t , where f is the activation function, V is the output weight coefficient of the forward hidden layer, and s t is the output of the forward hidden layer at the second moment, t is the second moment, V′ is the output weight coefficient of the reverse hidden layer, s′ t is the reverse hidden layer output at the second moment.

[0088] Optionally, the s t Satisfaction: t =f(Ux t +Ws t-1 );

[0089] The s′ t Satisfaction: s′ t =f(U′x t +W′s′ t+1 ); where U is the input weight coefficient of forward propagation, x t is the data to be optimized at the second moment, W is the weight coefficient of the forward hidden layer, s t-1 is the forward hidden layer output at the first moment, t-1 is the first moment, the first moment is earlier than the second moment, U′ is the input weight coefficient of back propagation, W′ is the weight coefficient of the reverse hidden layer, s′ t+1 is the reverse hidden layer output at the third moment, t+1 is the third moment, and the third moment is later than the second moment.

[0090] It should be noted that this embodiment is an apparatus embodiment corresponding to the above method embodiment, and all implementation methods in the above method embodiment are applicable to the embodiment of this apparatus and can achieve the same technical effects.

[0091] An embodiment of the present invention provides a non-volatile computer storage medium, wherein the computer storage medium stores at least one executable instruction, and the computer executable instruction can execute the rail transit data processing method in any of the above method embodiments.

[0092] Figure 6 The schematic diagram of the structure of the computing device provided by the embodiment of the present invention is shown. The specific embodiment of the present invention does not limit the specific implementation of the computing device.

[0093] like Figure 6 As shown, the computing device may include: a processor, a communication interface, a memory, and a communication bus.

[0094] The processor, communication interface, and memory communicate with each other via a communication bus. The communication interface is used to communicate with other devices, such as client devices or other server network elements. The processor is used to execute programs, specifically, the steps described in the aforementioned embodiment of the rail transit data processing method for a computing device.

[0095] Specifically, the program may include program codes including computer operation instructions.

[0096] The processor may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present invention. The one or more processors included in the computing device may be processors of the same type, such as one or more CPUs, or processors of different types, such as one or more CPUs and one or more ASICs.

[0097] The memory is used to store programs. The memory may include high-speed RAM memory and may also include non-volatile memory (non-volatile memory), such as at least one disk storage.

[0098] The program can be specifically configured to cause a processor to execute the rail transit data processing method described in any of the aforementioned method embodiments. The specific implementation of each step in the program can be found in the corresponding descriptions of the corresponding steps and units in the aforementioned rail transit data processing method embodiments, and will not be repeated here. Those skilled in the art will clearly understand that, for ease and brevity of description, the specific operating processes of the devices and modules described above can refer to the corresponding process descriptions in the aforementioned method embodiments, and will not be repeated here.

[0099] The algorithm or display provided herein is not inherently related to any particular computer, virtual system or other device. Various general-purpose systems can also be used together with the teachings based on this. According to the above description, it is obvious that the structure required for constructing such systems. In addition, the embodiment of the present invention is not directed to any specific programming language. It should be understood that various programming languages ​​can be utilized to implement the content of the embodiment of the present invention described herein, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the embodiment of the present invention.

[0100] In the description provided herein, numerous specific details are described. However, it is understood that embodiments of the present invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques are not shown in detail so as not to obscure the understanding of this description.

[0101] Similarly, it should be understood that in order to streamline the embodiments of the invention and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the embodiments of the invention are sometimes grouped together into a single embodiment, figure, or description thereof. However, this disclosed approach should not be interpreted as reflecting an intention that the claimed embodiments of the invention require more features than are expressly recited in each claim. Rather, as reflected in the claims below, inventive aspects lie in less than all of the features of the individual embodiments disclosed above. Accordingly, the claims that follow the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of the invention.

[0102] Those skilled in the art will appreciate that the modules in the devices in the embodiments may be adaptively changed and arranged in one or more devices different from the embodiments. The modules or units or components in the embodiments may be combined into one module or unit or component, and in addition may be divided into multiple submodules or subunits or subcomponents. All features disclosed in this specification (including the accompanying claims, abstracts and drawings) and all processes or units of any method or device disclosed herein may be combined in any combination, except that at least some of such features and / or processes or units are mutually exclusive. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstracts and drawings) may be replaced by an alternative feature providing the same, equivalent or similar purpose.

[0103] Furthermore, those skilled in the art will appreciate that although some embodiments herein include certain features included in other embodiments but not other features, combinations of features from different embodiments are intended to be within the scope of the present invention and to form different embodiments. For example, in the claims below, any of the claimed embodiments may be used in any combination.

[0104] The various component embodiments of the present invention can be implemented in hardware, or in software modules running on one or more processors, or in a combination thereof. It should be understood by those skilled in the art that a microprocessor or digital signal processor (DSP) can be used in practice to implement some or all of the functions of some or all of the components according to the embodiments of the present invention. The embodiments of the present invention can also be implemented as a device or apparatus program (e.g., a computer program and a computer program product) for executing part or all of the methods described herein. Such a program implementing an embodiment of the present invention can be stored on a computer-readable medium, or can have the form of one or more signals. Such a signal can be downloaded from an Internet website, or provided on a carrier signal, or provided in any other form.

[0105] It should be noted that the above embodiments illustrate rather than limit the embodiments of the invention, and that a person skilled in the art may devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between brackets should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The embodiments of the invention may be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names. The steps in the above embodiments should not be understood as limiting the order of execution unless otherwise specified.

Claims

1. A rail transit data processing method, characterized in that: The method consists solely of the following steps: Get the data to be optimized; Processing the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions; Outputting optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized; Before obtaining the data to be optimized, it also includes: Get the word vector of the original data; Obtaining a first semantic feature of the word vector according to the word vector of the original data; Obtaining a second semantic feature of the word vector based on the first semantic feature, where the second semantic feature and the first semantic feature respectively represent the operating characteristics of the rail transit from different dimensions; if the original data includes target original data corresponding to the data to be optimized, optimizing network parameters using at least one of the first semantic feature and the second semantic feature and the target original data, where the target original data refers to a training label corresponding to the data to be optimized; Generating the rail transit knowledge graph according to the first semantic feature and the second semantic feature; Obtaining a first semantic feature of the word vector according to the word vector of the original data, and obtaining a second semantic feature of the word vector according to the first semantic feature, including: Calling a bidirectional GRU network model to process the word vectors of the original data to obtain the first semantic feature; Calling a bidirectional LSTM model to process the first semantic feature to obtain the second semantic feature; The data to be optimized is processed according to a preset rail transit knowledge graph to obtain features to be optimized, including: According to the algorithm t =f(Vs t +V′s t ′) Get the feature to be optimized o t , where f is the activation function, V is the output weight coefficient of the forward hidden layer, and s t is the output of the forward hidden layer at the second moment, t is the second moment, V′ is the output weight coefficient of the reverse hidden layer, s t ′ is the reverse hidden layer output at the second moment; The rail transit data processing method also includes an active learning algorithm that can actively infer some labeling requests and submit some filtered data to experts for labeling.

2. The rail transit data processing method according to claim 1, characterized in that: The raw data includes: cloud data, basic data, historical operation and maintenance data, real-time operation and maintenance data and external environment data.

3. The rail transit data processing method according to claim 2, characterized in that: The s t Satisfaction: t =f(Ux t +Ws t-1 ); The s′ t Satisfaction: s′ t =f(U′x t +W′s′ t+1 ); where U is the input weight coefficient of forward propagation, x t is the data to be optimized at the second moment, W is the weight coefficient of the forward hidden layer, s t-1 is the forward hidden layer output at the first moment, t-1 is the first moment, the first moment is earlier than the second moment, U′ is the input weight coefficient of back propagation, W′ is the weight coefficient of the reverse hidden layer, s′ t+1 is the reverse hidden layer output at the third moment, t+1 is the third moment, and the third moment is later than the second moment.

4. A data processing device for rail transit, characterized in that: The device is used to implement the data processing method according to any one of claims 1 to 3, and the device includes: Acquisition module, used to obtain data to be optimized; a processing module, configured to process the data to be optimized according to a preset rail transit knowledge graph to obtain features to be optimized, wherein the knowledge graph represents the operating features of the rail transit in different dimensions; The output module is used to output optimization information that matches the scenario corresponding to the data to be optimized based on the obtained features to be optimized.

5. A computing device comprising: A processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; The memory is used to store at least one executable instruction, and when the at least one executable instruction is executed, the processor executes the rail transit data processing method according to any one of claims 1 to 3.

6. A computer storage medium, wherein at least one executable instruction is stored in the storage medium, and when the executable instruction is executed, a computing device executes the rail transit data processing method according to any one of claims 1 to 3.

Images