Multi-source feature fusion method and device
Through the multi-source feature fusion method implemented by the software platform, multi-source data fusion is automatically processed, which solves the problems of low manual operation efficiency and high error rate, improves the accuracy and efficiency of data processing, and enhances the flexibility of feature fusion.
Patent Information
- Application Number
- CN202411273017.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-11
- Publication Date
- 2024-12-06
AI Technical Summary
In the prior art, multi-source data fusion mainly relies on manual operations, which leads to high time consumption and easy errors, making it difficult to effectively process a large amount of data in the hyperparameter space.
Provide a multi-source feature fusion method, which realizes automated fusion through a software platform, including loading files to be fused, supplementing missing values, standardized processing, target feature filtering and data fusion, using merge and connection functions for feature fusion, supporting intersection and convergence and one-way fusion.
It realizes the automated fusion of multi-source data, saves labor costs, improves the accuracy and efficiency of feature fusion, reduces human errors, and enhances the flexibility and accuracy of feature fusion.
Smart Images

Figure CN119089387A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a multi-source feature fusion method and device. Background Art
[0002] In the process of constructing the hyperparameter space, data from different sources have attributes and characteristics of different dimensions. Before conducting multidimensional big data analysis and hyperparameter space mining, multi-source data from different sources must be fused first. Fusion of data from multiple sources can provide more comprehensive, accurate and reliable information, thereby supporting deeper multidimensional data analysis, decision-making and action.
[0003] Currently, multi-source parameter fusion usually adopts a manual method, which involves manually selecting the original file, selecting features from the file to copy, selecting the target file, and then pasting the corresponding features. This method requires manual operation, is time-consuming and prone to errors when processing a large amount of data in the hyperparameter space. Summary of the invention
[0004] The present application provides a multi-source feature fusion method, which can realize the automatic fusion of multi-source data and improve the accuracy and efficiency of data processing.
[0005] This application provides the following solutions:
[0006] According to a first aspect, a multi-source feature fusion method is provided, which is implemented based on a software platform, and the software platform includes an interactive interface, and is characterized in that the method includes: loading N files to be fused containing feature data to be fused, and displaying the contents of the N files to be fused in a first area of the interactive interface, where N is greater than 1, and the N files to be fused have different sources; reading the feature data to be fused in the N files to be fused, and screening target features from the feature data to be fused; fusing the N files to be fused according to the target features to obtain a fusion result, and displaying the fusion result in a second area of the interactive interface.
[0007] According to an achievable method in an embodiment of the present application, the source includes: data obtained by a remote monitoring device, data obtained by a ground monitoring device, and any combination of parameter data of the remote monitoring device and the ground monitoring device.
[0008] According to an achievable method in an embodiment of the present application, the loading of N files to be fused containing feature data to be fused includes: in response to an event in which a first component on the interactive interface is triggered, loading the N files to be fused input by a user; performing missing value supplementation and standardization processing on the feature data to be fused in the loaded files to be fused; wherein the missing value supplementation adopts a linear mean interpolation method; and the standardization processing adopts a standard normal standardization method.
[0009] According to an achievable method in an embodiment of the present application, the reading of the feature data to be fused in the N files to be fused and the screening of target features from the feature data to be fused include: in response to an event in which a second component on the interactive interface is triggered, reading out the features to be fused contained in the files to be fused; based on preset screening rules, selecting target features from the features to be fused, or, displaying the features to be fused in a third area of the interactive interface, and obtaining the features selected by the user in the third area as the target features.
[0010] According to an achievable method in an embodiment of the present application, the method further includes: obtaining a value range of the target feature input by the user in the third area, and filtering out data whose values are not within the value range from the data of the target feature in the N files to be fused.
[0011] According to an achievable method in an embodiment of the present application, the fusing of N files to be fused according to the target feature includes: in response to an event in which a third component on the interactive interface is triggered, fusing the contents of the N files to be fused containing the target feature into the same file, and the same file is the fusion result.
[0012] According to an achievable method in an embodiment of the present application, the merging of the contents of the N files to be merged containing the target features into the same file includes: merging the data of file a and file b using a Merge function, and stacking the results of the merging process using a Concat function to obtain a c file; wherein, when N is equal to 2, the a file is the first file of the N files to be merged, the b file is the second file of the N files to be merged, and the c file is the fusion result; when N is greater than 2, the first file of the N files to be merged is used as the a file, and the second file is used as the b file; merging the data of the a file and the b file using the Merge function, and stacking the results of the merging process using the Concat function to obtain the c file; updating the a file to the currently obtained c file, and updating the b file to the next unmerged file among the N files to be merged, and continuing to execute the step of merging the data of the a file and the b file using the Merge function until all the files to be merged are merged into the same c file.
[0013] According to an achievable method in an embodiment of the present application, the merging of the data of file a and file b using the Merge function includes: selecting a merging method from the merging methods that can be used by the Merge function, and merging the data of file a and file b according to the selected merging method; wherein the merging methods that can be used by the Merge function include Inner method, Left method, Right method and Outer method; the Inner method is to merge the file a and the file b according to the intersection of the target features of the file a and the file b; the Left method is to merge the file a and the file b according to the target features in the file a; the Right method is to merge the file a and the file b according to the target features in the file b; the Outer method is to merge the file a and the file b according to the union of the target features of the file a and the file b.
[0014] According to the second aspect, a multi-source feature fusion device is provided, characterized in that the device includes: a file loading module, used to load N files to be fused containing feature data to be fused, wherein N is greater than 1, and the N files to be fused have different sources; a feature screening module, used to read the feature data to be fused in the N files to be fused, and screen target features from the feature data to be fused; a data fusion module, used to fuse the N files to be fused according to the target features to obtain a fusion result; and an interface display module, used to display the contents of the N files to be fused in a first area of the interface and to display the fusion result in a second area of the interface.
[0015] According to a third aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.
[0016] According to a fourth aspect, there is provided an electronic device, comprising:
[0017] one or more processors; and
[0018] A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of the method described in any one of the first aspects above.
[0019] According to the specific embodiments provided in this application, this application discloses the following technical effects:
[0020] 1) The multi-source feature fusion method provided in this application can automatically load files and perform operations such as feature screening and feature fusion. It realizes the automatic fusion of multi-source data, saves labor costs, reduces possible human errors, and improves the accuracy and efficiency of feature fusion.
[0021] 2) This application supplements missing values and standardizes the feature data to be fused in the fusion file. On the one hand, it improves the accuracy of feature fusion. On the other hand, it reduces the impact of data on downstream tasks (such as data mining and analysis) through missing value supplementation, and improves the performance of downstream tasks and algorithm stability through data standardization.
[0022] 3) This application screens the features to be fused, selects the target features to be fused, and can select the value range of the data for the target features, making the feature fusion more targeted. Users can independently select the target features according to actual needs, further improving the efficiency and accuracy of feature fusion.
[0023] 4) The present application can fuse files based on the intersection or union of target features in the files to be fused or based on the target features of a certain file to be fused, thereby improving the flexibility of feature fusion.
[0024] Of course, any product implementing the present application does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS
[0025] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments 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 creative work.
[0026] Figure 1 A flowchart of a multi-source feature fusion method provided in an embodiment of the present application.
[0027] Figure 2 A flowchart of a multi-source feature fusion method provided in an embodiment of the present application.
[0028] Figure 3 A schematic diagram of a software interaction interface on which the multi-source feature fusion method provided in an embodiment of the present application is based.
[0029] Figure 4 A schematic block diagram of a multi-source feature fusion device provided in an embodiment of the present application.
[0030] Figure 5 A schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0031] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the present application belong to the scope of protection of this application.
[0032] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings.
[0033] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.
[0034] The word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.
[0035] Currently, multi-source feature fusion is mainly carried out manually. The file to be fused is selected manually, and features are selected from the file to be fused to be copied; then the target file is selected and the copied content is pasted to the corresponding position of the target file. However, the manual method is time-consuming and has a high risk of error, and is not suitable for feature fusion of large amounts of data.
[0036] In view of this, the present application provides a new idea. Figure 1 This is a flow chart of a multi-source feature fusion method provided in an embodiment of the present application. This method can be performed by a multi-source feature fusion device. Figure 1 As shown in , the method may include the following steps:
[0037] Step 101, load N files to be fused containing feature data to be fused, and display the contents of the N files to be fused in the first area of the interactive interface, wherein N is greater than 1, and the N files to be fused are from different sources.
[0038] Step 102, reading the feature data to be fused in N files to be fused, and screening the target features from the feature data to be fused.
[0039] Step 103: fuse the N files to be fused according to the target features to obtain a fusion result, and display the fusion result in the second area of the interactive interface.
[0040] It can be seen from the above process that the multi-source feature fusion method provided by this application can automatically load files and perform operations such as feature screening and feature fusion, thereby realizing the automatic fusion of multi-source data, saving labor costs, and improving the accuracy and efficiency of feature fusion.
[0041] The following is a detailed description of each step in the above process and the effects that can be further produced in conjunction with the embodiments. It should be noted that the limitations such as "first" and "second" involved in the present disclosure do not have limitations in terms of size, order and quantity, and are only used to distinguish in name, for example, "first area" and "second area" are used to distinguish two areas.
[0042] First, the above step 101, i.e., "loading N files to be fused containing feature data to be fused, and displaying the contents of the N files to be fused in the first area of the interactive interface, wherein N is greater than or equal to 1, and the N files to be fused are from different sources" is described in detail in conjunction with the embodiment.
[0043] Figure 2 A flow chart of the multi-source feature fusion method provided in the embodiment of the present application is shown in FIG. Figure 2 , the present application is used for the fusion of multi-source features, wherein multi-source features represent features from multiple sources, wherein different sources may be different origins of data, different acquisition methods, etc. Multi-source feature data often has attributes and features of different dimensions. The features to be fused are stored in a file to be fused, and the features to be fused in a file to be fused all correspond to the same source. The file to be fused can be directly generated after measurement by certain parameter acquisition devices, or it can be obtained after some data processing. The present application fuses N files to be fused, where N is greater than or equal to 1, that is, the present application can be used for the feature fusion of files to be fused from two or more sources.
[0044] The method of the present application can be applied to the fusion of multiple data source features, and users can apply the method according to actual needs. For example, if the present application is applied to the field of spatial analysis, the sources of the files to be fused may include telemetry data, situation data, and product data. Telemetry data refers to data obtained from a long distance through remote sensors or instruments, which are usually used to monitor, measure or track the status, performance or other relevant information of remote objects or environments. Telemetry data contains features such as environmental parameters (temperature, humidity, wind speed, etc.), operating status and location information (coordinates, longitude and latitude, direction, etc.); situation data usually refers to data collected in a specific field or scenario, which is used to describe the current situation, status or trend. It mainly includes some real-time indicators (meteorological conditions and time data, etc.), geographic information (location information and geographic boundaries, etc.) and sensor data (temperature, pressure, etc.); product data refers to the parameter data of the equipment used to monitor the above parameters, usually including information such as the material and shape of the product.
[0045] Specifically, the source of the file to be fused may include: data obtained through remote monitoring equipment, data obtained through ground monitoring equipment, and any combination of parameter data of remote monitoring equipment and ground monitoring equipment. Among them, the remote monitoring equipment can obtain telemetry data, the ground monitoring equipment can obtain situation data, and the parameter data of the remote monitoring equipment and the ground monitoring equipment are the parameter information of the equipment itself. The telemetry data, situation data, and the parameter data of the equipment itself can be data obtained in the same time period.
[0046] The multi-source feature fusion method of the present application is implemented based on a software platform, which includes an interactive interface. Figure 3 A schematic diagram of a software interaction interface based on the multi-source feature fusion method provided in the embodiment of the present application, such as Figure 3 As shown, the interactive interface can be set with multiple trigger components and areas, and the file to be fused input by the user can be loaded by triggering the first component in the interactive interface. For example, a "Load Input File" button is set on the interactive page of the software platform. After the user clicks the button, the file selection and file upload can be performed to import the file to be fused. More specifically, the software platform of the present application is implemented based on Python, and the file to be fused can be read based on the Read function in Python. Batch selection can be performed, and single file selection can also be performed.
[0047] The content of the uploaded file to be merged is displayed in the first area of the interactive interface. A scroll bar may be set in the first area, and the user may browse the content of the file to be merged completely by dragging the scroll bar.
[0048] In the process of acquiring multi-source feature data, there are often problems with data reading errors or failure to obtain some data, so there are often some missing values in the input files to be fused. These missing values will affect data mining and analysis. Therefore, the present application can also process missing values before multi-source data fusion. There are many ways to process missing values, such as filling with the median or mean of the features to be fused corresponding to the missing values.
[0049] As a preferred implementation, the missing values can be filled using the linear mean interpolation method. This mean filling method makes the missing values after filling more consistent with the probability distribution of the existing data, and can well compensate for the impact of the missing values. Specifically, assuming that the missing value of the data point is to be estimated, the formula for mean interpolation is:
[0050] y i =(y i―1 +y i+1 ) / twenty one)
[0051] where is what the missing value is to be inserted into, and and are the values near the missing value.
[0052] The present application can also perform data standardization before multi-source data fusion. Data from different sources often have different units and measurement ranges. Features with larger ranges will account for larger scale changes in the final results, ultimately affecting the analysis results and efficiency of multidimensional data and hyperparameter space. Therefore, the present application can also perform data standardization before multi-source data fusion. Standardization can be done in a variety of ways, such as: scaling the data to between 0 and 1 through minimum-maximum normalization; or applying natural logarithmic transformation or other logarithmic transformation methods to the data.
[0053] In order to ensure that the processed data conforms to the probability distribution, as a preferred implementation, the data is normalized using the standard normalization method. After the original data is normalized, the mean of the data becomes 0 and the standard deviation becomes 1. The specific formula is as shown in formula (2).
[0054]
[0055] Among them, is the original data, the value after normal standardization, and is the mean and standard deviation of the original data.
[0056] The above step 102, namely "reading the feature data to be fused in N files to be fused, and selecting target features from the feature data to be fused", is described in detail below in conjunction with an embodiment.
[0057] The features contained in the files to be fused are the features to be fused. This application reads the features to be fused contained in the files from N files to be fused, and selects target features from the feature data to be fused. The target features are the features that need to be retained after the N files to be fused are fused.
[0058] Among them, the target feature can be selected based on the preset screening rules, and the preset screening rules can be pre-set according to the actual situation. For example, the preset screening rules can take all the features to be fused as the target features, or set a preset target feature list and take the features to be fused in the list as the target features. The target feature can also be selected by the user. The user can independently select the features he needs as the target feature through the features to be fused displayed on the interactive interface.
[0059] On the software platform, in response to the event that the second component on the interface is triggered, the features to be fused contained in the file to be fused are read out; based on the preset screening rules, the target features are selected from the features to be fused, or the features to be fused are displayed in the third area of the interface, and the features selected by the user in the third area are obtained as the target features. Specifically, when the user clicks the "Feature Filter" button in the interactive interface, the third area of the software interface displays the features to be fused. A selection box can be set in front of each feature to be fused, and the user can check the feature in the selection box, and all the checked features to be fused are the target features.
[0060] Furthermore, the user can select the data corresponding to the target feature for fusion. The user filters the data content in N files to be fused by limiting the value range of the target feature. Specifically, the third area is also provided with an input component for the value range corresponding to the target feature, such as an input box or a selection button. The software platform obtains the value range of the target feature entered by the user in the third area, filters out the data whose values are not within the value range from the data of the target feature, and uses the filtered data for subsequent feature fusion.
[0061] The above step 103, namely "fusing N files to be fused according to target features to obtain a fusion result, and displaying the fusion result in the second area of the interactive interface" is described in detail below in conjunction with an embodiment.
[0062] Read the target features and their corresponding data in N files to be fused, and fuse the above contents into one file, which is the fusion result.
[0063] There are many ways to fuse N files to be fused. The N files can be fused according to the intersection of the target features of the N files to be fused, that is, the fused file only contains the target features contained in each file to be fused and the data corresponding to the target features in all files to be fused; the N files can also be fused according to the union of the target features of the N files to be fused, that is, the fused file contains all the target features and the data corresponding to the target features in all files to be fused; the files can also be fused according to the target features contained in a certain file to be fused, that is, the fused file contains all the target features contained in the file to be fused and the data corresponding to the target features in all files to be fused.
[0064] This application fuses N files to be merged. When N is equal to 2, two files can be directly merged; when N is greater than 2, two files to be merged can be merged first, and the fusion result can be used as a new file to be merged and then merged with other files to be merged until all files to be merged are merged into one file.
[0065] Specifically, the fusion of the files to be fused can be realized based on the Merge and Concat functions in the Pandas data processing and analysis library of Python. The files to be fused are stored in the format of DataFrame, and each row of DataFrame is a key and its corresponding data value, where the key value can represent the target feature. The data of file a and file b are merged using the Merge function, and the result of the merging process is stacked using the Concat function to obtain file c. Among them, when N is equal to 2, file a is the first file among the N files to be fused, file b is the second file among the N files to be fused, and file c is the fusion result. When N is greater than 2, the first file among the N files to be fused is used as file a, and the second file is used as file b; the data of file a and file b are merged using the Merge function, and the result of the merging process is stacked using the Concat function to obtain file c; file a is updated to the currently obtained c file, and file b is updated to the next unfused file among the N files to be fused, and the step of merging the data of file a and file b using the Merge function is continued until all the files to be fused are fused into the same c file.
[0066] The Merge function can connect different DataFrames together based on one or more keys. There are four connection methods: Inner, Left, Right, and Outer. When the Merge function is used to merge file a and file b, the Inner method is to merge the two files based on the intersection of the target features of file a and file b; the Left method is to merge the two files based on the target features in file a; the Right method is to merge the two files based on the target features in file b; and the Outer method is to merge the two files based on the union of the target features of file a and file b. The selection of Inner, Left, Right, and Outer can be set to the default method, or the user can modify the parameters corresponding to the Merge function code according to actual needs. After using the Merge function to achieve the fusion of the target features, the Concat function is used to merge the data content corresponding to the target features.
[0067] Furthermore, in the Merge function, you can also use the on parameter to specify the connection key, or use the left_on and right_on parameters to specify the connection keys of the left and right files to be fused respectively, so as to achieve more flexible feature fusion. The Concat function can stack two or more multi-source features together along an axis (the "axis" here refers to rows or columns). You can use the axis parameter to specify the axis to be stacked. axis=0 means stacking along rows. In this case, the number of columns of the two multi-source features must be the same; axis=1 means stacking along columns. In this case, the number of rows of the two multi-source features must be the same. By limiting the parameters in the Merge and Concat functions, the flexibility of feature fusion can be further enhanced.
[0068] In the fusion process, for the file to be fused that does not have a certain target feature, the value of the target feature can be filled with the mean value first, and then the above fusion process is performed based on the filled file to be fused.
[0069] In the software platform, in response to the event that the third component on the interactive interface is triggered, the target feature and its corresponding data are fused into a file as the fusion result. Specifically, when the user clicks the "Data Analysis" button on the interactive interface, the system will analyze, filter and fuse the multi-source data according to the user-set conditions, and output the output file by default to the folder where the system's running program is located. The output file will be named after the output time. The results are displayed in the second area on the interactive interface. The fusion results can be automatically saved, and the results are output by default to the folder where the system's running program is located. The output file is named after the output time. Avoid unnecessary risks caused by misoperation, thereby improving the stability of multi-source data fusion.
[0070] The above method provided in the embodiment of the present application can be applied to a variety of application scenarios, including but not limited to: in the field of medical health, by integrating the patient's electronic health records, medical images, genomic data, wearable device monitoring data, etc., to provide a more comprehensive assessment of the patient's health status; in the field of financial services, by combining market data, transaction data, news reports, social media sentiment analysis, etc., more accurate risk assessment and investment strategy formulation. In the field of environmental monitoring, by integrating data sources such as meteorological station data, satellite remote sensing data, and geographic information systems (GIS), climate change analysis and natural disaster warnings are carried out.
[0071] The significance of fusing multi-source data is to provide more comprehensive, accurate and reliable information, thereby supporting more in-depth multi-dimensional data analysis, decision-making and action. In other words, the downstream system of the multi-source feature fusion method provided in the embodiment of the present application is a data analysis system. For example, the fusion of features of multiple data sources including telemetry data, situation data, product data, etc. facilitates the subsequent analysis and mining of multi-dimensional big data and hyperparameter space, thereby further making decisions or model training based on the results of the analysis and mining.
[0072] The fusion of multi-source data can improve data quality and integrity. Data from different sources may have different accuracy, coverage and integrity. The fusion of multi-source data can make up for the shortcomings of a single data source and improve the overall data quality and integrity. It can also increase the diversity of data. Data from different sources may cover different aspects and dimensions. By fusing multi-source data, you can obtain broader and more comprehensive information, so as to better understand the object or problem under study. The fusion of multi-source data can also discover hidden associations and patterns between and within data. By associating and integrating data from different sources, you may find associations and patterns between different data, thereby generating new insights and discoveries. The fusion of multi-source data can improve the credibility of data analysis. Cross-validation and comparison of data provided by multiple data sources can enhance the credibility and accuracy of data analysis and reduce the possibility of misjudgment. The fusion of multi-source data supports more fine-grained analysis and insights. The fusion of multi-source data can provide more fine-grained and detailed information, thereby supporting deeper data analysis and insights and helping to discover more specific problems and opportunities. In addition, it can also improve the basis and effect of decision-making. Comprehensive analysis based on multi-source data can provide more comprehensive and accurate decision-making basis, thereby supporting smarter and more effective decisions and actions. And the fusion of multi-source data can better deal with data uncertainty and risk. Fusion of multi-source data can reduce the uncertainty and risk brought by a single data source and enhance the robustness and reliability of data. In summary, the fusion of multi-source data helps to improve the quality, credibility and value of data, and provide support for deeper multi-dimensional data analysis, smarter decisions and more effective actions.
[0073] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0074] According to an embodiment of another aspect, a multi-source feature fusion device is provided. Figure 4 FIG. 4 is a schematic block diagram of a multi-source feature fusion device according to an embodiment. Figure 4 As shown, the device 400 includes:
[0075] The file loading module 401 is used to load N to-be-fused files containing the feature data to be fused, where N is greater than 1 and the N to-be-fused files are from different sources.
[0076] The feature screening module 402 is used to read the feature data to be fused in the N files to be fused, and screen the target features from the feature data to be fused.
[0077] The data fusion module 403 is used to fuse N files to be fused according to target features to obtain a fusion result.
[0078] The interface display module 404 is used to display the contents of the N files to be merged in the first area of the interface and to display the fusion result in the second area of the interface.
[0079] As one possible implementation manner, the source includes data obtained by a remote monitoring device, data obtained by a ground monitoring device, or any combination of parameter data of the remote monitoring device and the ground monitoring device.
[0080] As one of the achievable methods, when loading N files to be fused containing feature data to be fused, the file loading module 401 can be configured as follows: in response to an event in which the first component on the interactive interface is triggered, the N files to be fused input by the user are loaded; missing values are supplemented and standardized for the feature data to be fused in the loaded files to be fused; wherein the missing values are supplemented using a linear mean interpolation method; and the standardization process uses a standard normal standardization method.
[0081] As one of the achievable methods, the feature screening module 402 can be configured to read the feature data to be fused in N files to be fused and screen the target features from the feature data to be fused as follows: in response to an event in which the second component on the interactive interface is triggered, read out the features to be fused contained in the files to be fused; based on preset screening rules, select the target features from the features to be fused, or, display the features to be fused in the third area of the interactive interface, and obtain the features selected by the user in the third area as the target features.
[0082] As one achievable method, the feature screening module 402 may be configured to obtain a value range of the target feature input by the user in the third area, and filter out data whose values are not within the value range from the target feature data in the N files to be fused.
[0083] As one of the feasible ways, when fusing N files to be fused according to the target features, the data fusion module 403 can be configured as follows: in response to an event in which a third component on the interactive interface is triggered, the contents of the N files to be fused containing the target features are merged into the same file, and the same file is the fusion result.
[0084] As one of the feasible ways, the data fusion module 403 can be configured as follows when fusing the contents containing the target features in N files to be fused into the same file: using the Merge function to merge the data of file a and file b, and using the Concat function to stack the results of the merging process to obtain file c; wherein, when N is equal to 2, file a is the first file of the N files to be fused, file b is the second file of the N files to be fused, and file c is the fusion result; when N is greater than 2, the first file of the N files to be fused is used as file a, and the second file is used as file b; using the Merge function to merge the data of file a and file b, and using the Concat function to stack the results of the merging process to obtain file c; updating file a to the currently obtained file c, updating file b to the next unfused file among the N files to be fused, and continuing to execute the step of merging the data of file a and file b using the Merge function until all the files to be fused are fused into the same file c.
[0085] As one of the achievable methods, the data fusion module 403 can be configured to merge the data of file a and file b by using the merge function: select a merge method from the merge methods available for the Merge function, and merge the data of file a and file b according to the selected merge method. Among them, the merge methods available for the Merge function include Inner method, Left method, Right method and Outer method; Inner method is to merge file a and file b according to the intersection of the target features of file a and file b; Left method is to merge file a and file b according to the target features in file a; the Right method is to merge file a and file b according to the target features in file b; Outer method is to merge file a and file b according to the union of the target features of file a and file b.
[0086] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative work.
[0087] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and provide corresponding operation entrances for users to choose to authorize or refuse.
[0088] In addition, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the steps of any one of the methods in the aforementioned method embodiments are implemented.
[0089] And an electronic device, comprising:
[0090] one or more processors; and
[0091] A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of the method described in any one of the aforementioned method embodiments.
[0092] The present application also provides a computer program product, including a computer program, which implements the steps of any one of the methods in the aforementioned method embodiments when executed by a processor.
[0093] in, Figure 5The architecture of the electronic device is shown as an example, which may include a processor 510, a video display adapter 511, a disk drive 512, an input / output interface 513, a network interface 514, and a memory 520. The processor 510, the video display adapter 511, the disk drive 512, the input / output interface 513, the network interface 514, and the memory 520 may be communicatively connected via a communication bus 530.
[0094] The processor 510 may be implemented by a general-purpose CPU, a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, etc., and is used to execute relevant programs to implement the technical solutions provided in this application.
[0095] The memory 520 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 520 can store an operating system 521 for controlling the operation of the electronic device 500, and a basic input and output system (BIOS) 522 for controlling the low-level operation of the electronic device 500. In addition, a web browser 523, a data storage management system 524, and a multi-source feature fusion device 525, etc. can also be stored. The above-mentioned multi-source feature fusion device 525 can be an application program that specifically implements the operations of the aforementioned steps in the embodiment of the present application. In short, when the technical solution provided by the present application is implemented by software or firmware, the relevant program code is stored in the memory 520 and is called and executed by the processor 510.
[0096] The input / output interface 513 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.
[0097] The network interface 514 is used to connect to a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).
[0098] The bus 530 comprises a pathway for transmitting information between the various components of the device (eg, the processor 510, the video display adapter 511, the disk drive 512, the input / output interface 513, the network interface 514, and the memory 520).
[0099] It should be noted that, although the above device only shows a processor 510, a video display adapter 511, a disk drive 512, an input / output interface 513, a network interface 514, a memory 520, a bus 530, etc., in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include components necessary for implementing the solution of the present application, and does not necessarily include all the components shown in the figure.
[0100] It can be known from the description of the above implementation methods that those skilled in the art can clearly understand that the present application can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a computer program product, which can be stored in a storage medium such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.
[0101] The technical solution provided by the present application is described in detail above. The principle and implementation method of the present application are described in detail using specific examples. The description of the above embodiments is only used to help understand the method and core idea of the present application. At the same time, for those skilled in the art, 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 limiting the present application.
Claims
1. A multi-source feature fusion method, the method is implemented based on a software platform, the software platform includes an interactive interface, characterized in that: The method comprises: Loading N to-be-fused files containing to-be-fused feature data, and displaying the contents of the N to-be-fused files in the first area of the interactive interface, where N is greater than 1 and the N to-be-fused files are from different sources; Reading the feature data to be fused in the N files to be fused, and screening target features from the feature data to be fused; The N files to be fused are fused according to the target features to obtain a fusion result, and the fusion result is displayed in the second area of the interactive interface.
2. The method according to claim 1, characterized in that The sources include: data acquired through a remote monitoring device, data acquired through a ground monitoring device, and any combination of parameter data of the remote monitoring device and the ground monitoring device.
3. The method according to claim 1, characterized in that The loading of N to-be-fused files containing to-be-fused feature data comprises: In response to an event in which the first component on the interactive interface is triggered, N files to be fused input by a user are loaded; missing values are supplemented and standardized for the feature data to be fused in the loaded files to be fused; The missing values are supplemented by linear mean interpolation method; the standardization process adopts standard normal standardization method.
4. The method according to claim 1, characterized in that The reading of the feature data to be fused in the N files to be fused and screening target features from the feature data to be fused includes: In response to an event in which a second component on the interactive interface is triggered, reading out the to-be-fused features contained in the to-be-fused file; Based on a preset screening rule, a target feature is selected from the features to be fused, or the features to be fused are displayed in a third area of the interactive interface, and a feature selected by a user in the third area is obtained as the target feature.
5. The method according to claim 4, characterized in that The method further comprises: The value range of the target feature input by the user in the third area is obtained, and data whose values are not within the value range are filtered out from the data of the target feature in the N files to be fused.
6. The method according to claim 1, characterized in that The fusing of the N files to be merged according to the target feature includes: in response to an event in which a third component on the interactive interface is triggered, fusing the contents of the N files to be merged containing the target feature into a same file, wherein the same file is the fusion result.
7. The method according to claim 6, characterized in that The step of fusing the contents of the N files to be fused into one file including the target feature comprises: The data of file a and file b are merged by using the Merge function, and the result of the merging process is stacked by using the Concat function to obtain file c; wherein, When N is equal to 2, the a file is the first file of the N files to be fused, the b file is the second file of the N files to be fused, and the c file is the fusion result; When N is greater than 2, the first file among the N files to be merged is used as the a file, and the second file is used as the b file; the data of the a file and the b file are merged using the Merge function, and the results of the merging process are stacked using the Concat function to obtain the c file; the a file is updated to the currently obtained c file, and the b file is updated to the next unmerged file among the N files to be merged, and the step of merging the data of the a file and the b file using the Merge function is continued until all the files to be merged are merged into the same c file.
8. The method according to claim 7, wherein the step of merging the data of file a and file b by using the Merge function comprises: Selecting a merging method from the merging methods available for the Merge function, and merging the data of file a and file b according to the selected merging method; Among them, the merging methods that can be used in the Merge function include Inner method, Left method, Right method and Outer method; the Inner method is to merge the a file and the b file according to the intersection of the target features of the a file and the b file; the Left method is to merge the a file and the b file according to the target features in the a file; the Right method is to merge the a file and the b file according to the target features in the b file; the Outer method is to merge the a file and the b file according to the union of the target features of the a file and the b file.
9. A multi-source feature fusion device, characterized in that: The device comprises: A file loading module, used to load N to-be-fused files containing to-be-fused feature data, where N is greater than 1 and the N to-be-fused files are from different sources; A feature screening module, used for reading the feature data to be fused in the N files to be fused, and screening target features from the feature data to be fused; A data fusion module is used to fuse N files to be fused according to the target features to obtain a fusion result; The interface display module is used to display the contents of the N files to be merged in the first area of the interface and to display the fusion result in the second area of the interface.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.
11. An electronic device, characterized in that: include: one or more processors; as well as A memory associated with the one or more processors, the memory being used to store program instructions, wherein the program instructions, when read and executed by the one or more processors, execute the steps of the method described in any one of claims 1 to 8.
Citation Information
Patent Citations
Data fusion method and device
CN110119413A
Multi-event service lineage data fusion method oriented to service collaboration process
CN113268547A
Data fusion system and method based on cloud platform and edge computing
CN114912508A
Text feature analysis method and device, storage medium and electronic equipment
CN115455963A
Multi-dimensional data fusion processing method, platform and equipment and storage medium
CN115879057A
