A method and system for automatically building a knowledge base in an intelligent question-answering system

Abstract

The invention discloses a method and system for automatically constructing a knowledge base in an intelligent question-answering system, and relates to the technical field of knowledge base construction of satellite data. The method comprises: obtaining a satellite data matrix with marked patterns, wherein the patterns include: vegetation coverage, water bodies, land use, meteorological phenomena and / or geological activities; training a classifier according to the satellite data matrix with marked patterns, so that the classifier can learn the differences between various patterns; obtaining a new satellite data matrix and performing preprocessing to generate a new satellite data matrix after preprocessing, setting a pattern recognition model, and calculating a comprehensive pattern recognition value of the new satellite data matrix; inputting the comprehensive pattern recognition value into the classifier model, performing pattern classification, and finally completing the construction of a pattern knowledge base.

Description

Technical Field

[0001] The present invention belongs to the technical field of satellite data knowledge base construction, and more specifically, relates to a knowledge base automatic construction method and system in an intelligent question-answering system. Background Art

[0002] Satellite data refers to information about the earth and its environment obtained through satellite remote sensing technology. These data are usually represented by images, spectral data or other forms, and can cover large geographical areas. Satellite data is widely used in meteorological observation, environmental monitoring, agriculture, urban planning, military and other fields.

[0003] The main satellite data types include:

[0004] Optical data: image data obtained through visible light and near-infrared bands. Mainly used for research on land cover, vegetation index, land use, etc.

[0005] Radar data: Image data obtained through microwave remote sensing, which can penetrate clouds and some vegetation, and is suitable for terrain mapping, disaster monitoring, etc.

[0006] Hyperspectral data: Acquisition of data in multiple spectral bands can be used for detailed material composition analysis and classification.

[0007] LiDAR data: High-precision terrain data obtained using laser ranging technology, commonly used in 3D terrain modeling, forest measurement, etc.

[0008] Currently, there is no technical solution that can form a satellite data knowledge base with pattern recognition. Summary of the invention

[0009] In order to solve the above technical problems, the present invention proposes a method for automatically constructing a knowledge base in an intelligent question-answering system, which involves constructing a knowledge base of satellite data, including:

[0010] Obtaining a satellite data matrix with annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities;

[0011] Based on the satellite data matrix with labeled patterns, a classifier is trained to learn the differences between various patterns;

[0012] Acquire a new satellite data matrix and preprocess it, generate a preprocessed new satellite data matrix, set a pattern recognition model, and calculate a comprehensive pattern recognition value of the new satellite data matrix;

[0013] The comprehensive pattern recognition value is input into the classifier model to perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0014] Furthermore, the pattern recognition model includes:

[0015] ,

[0016] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. A second pattern recognition value.

[0017] Furthermore, the composite eigenvalues ​​of the new satellite data matrix are calculated No. First pattern recognition value include:

[0018] ,

[0019] in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the A sixth adjustment factor for the first pattern recognition value.

[0020] Furthermore, the composite eigenvalues ​​of the new satellite data matrix are calculated No. Second pattern recognition value include:

[0021] ,

[0022] in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the A sixth adjustment factor for the second pattern recognition value.

[0023] Furthermore, the composite eigenvalues ​​of the new satellite data matrix are calculated include:

[0024] ,

[0025] in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

[0026] Further, calculate the preprocessed new satellite data matrix No. The first eigenvalue include:

[0027] ,

[0028] in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

[0029] Further, calculate the preprocessed new satellite data matrix No. The second eigenvalue include:

[0030] ,

[0031] in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

[0032] Furthermore, the classifier model is: support vector machine, decision tree, random forest or k-nearest neighbor.

[0033] The present invention also proposes a knowledge base automatic construction system in an intelligent question-answering system, which involves the construction of a knowledge base of satellite data, including:

[0034] A training set acquisition module is used to acquire a satellite data matrix of annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities;

[0035] A training classifier module is used to train a classifier based on a satellite data matrix with labeled modes, so that the classifier can learn the differences between various modes;

[0036] Setting a model module to obtain and preprocess a new satellite data matrix, generate a preprocessed new satellite data matrix, set a pattern recognition model, and calculate a comprehensive pattern recognition value of the new satellite data matrix;

[0037] The pattern recognition module is used to input the comprehensive pattern recognition value into the classifier model, perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0038] Furthermore, the pattern recognition model includes:

[0039] ,

[0040] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. A second pattern recognition value.

[0041] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art:

[0042] The present invention obtains a satellite data matrix with marked patterns, wherein the patterns include: vegetation coverage, water bodies, land use, meteorological phenomena and / or geological activities; trains a classifier based on the satellite data matrix with marked patterns so that it can learn the differences between various patterns; obtains a new satellite data matrix and performs preprocessing to generate a new satellite data matrix after preprocessing, sets a pattern recognition model, and calculates a comprehensive pattern recognition value of the new satellite data matrix; inputs the comprehensive pattern recognition value into the classifier model, performs pattern classification, and finally completes the construction of a pattern knowledge base. Through the above technical solution, the present invention can automatically identify patterns in satellite data, thereby making the knowledge base more refined. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0044] Figure 2It is a structural diagram of the system of embodiment 2 of the present invention. DETAILED DESCRIPTION

[0045] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0046] The method provided by the present invention can be implemented in the following terminal environment, and the terminal may include one or more of the following components: a processor, a storage medium, and a display screen. The storage medium stores at least one instruction, and the instruction is loaded and executed by the processor to implement the method described in the following embodiment.

[0047] The processor may include one or more processing cores. The processor uses various interfaces and lines to connect various parts in the entire terminal, and executes various functions of the terminal and processes data by running or executing instructions, programs, code sets or instruction sets stored in the storage medium, and calling data stored in the storage medium.

[0048] The storage medium may include a random access memory (RAM) or a read-only memory (ROM). The storage medium may be used to store instructions, programs, codes, code sets or instructions.

[0049] The display screen is used to display the user interface of each application.

[0050] In addition, those skilled in the art can understand that the structure of the above terminal does not constitute a limitation on the terminal, and the terminal may include more or fewer components, or combine certain components, or arrange the components differently. For example, the terminal also includes components such as a radio frequency circuit, an input unit, a sensor, an audio circuit, and a power supply, which will not be described in detail here.

[0051] Example 1

[0052] like Figure 1 As shown, an embodiment of the present invention provides a method for automatically constructing a knowledge base in an intelligent question-answering system, which involves constructing a knowledge base of satellite data, including:

[0053] Step 101, obtaining a satellite data matrix with marked patterns (obtaining original satellite data matrix from various satellite sensors, including different types of remote sensing data such as optical, radar, infrared, etc.), wherein the patterns include: vegetation cover, water body, land use, meteorological phenomenon and / or geological activity;

[0054] Specifically, for example, the mode can be:

[0055] Vegetation Cover:

[0056] Normal vegetation: healthy forests, grasslands, etc.

[0057] Abnormal vegetation: Vegetation affected by disease, drought or fire.

[0058] Water bodies:

[0059] Normal bodies of water such as rivers, lakes and oceans.

[0060] Abnormal water bodies: pollution, algal blooms, and drying up.

[0061] Land Use:

[0062] Farmland: Contains different types of crops.

[0063] Urban areas: buildings, roads, etc.

[0064] Desertification: Soil degradation and desert expansion.

[0065] Meteorological phenomena:

[0066] Cloud Cover: Normal cloud distribution.

[0067] Extreme weather: hurricanes, tornadoes, blizzards, etc.

[0068] Geological activity:

[0069] Normal terrain: mountains, plains, river valleys, etc.

[0070] Abnormal terrain: landslides, earthquake-affected areas, etc.

[0071] Step 102, training a classifier based on the satellite data matrix of the marked modes so that it can learn the differences between various modes;

[0072] Specifically, the classifier model is: support vector machine, decision tree, random forest or k-nearest neighbor.

[0073] Step 103, obtaining a new satellite data matrix and preprocessing it, generating a preprocessed new satellite data matrix, setting a pattern recognition model, and calculating a comprehensive pattern recognition value of the new satellite data matrix;

[0074] Specifically, the pattern recognition model includes:

[0075] ,

[0076] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The ReLU (Rectified Linear Unit) function is a commonly used activation function, which is widely used in deep learning and neural networks. The mathematical expression is:

[0077] f(x)=max(0,x), when the input x is greater than zero, the output is f(x)=x, when the input x is less than or equal to zero, the output is f(x)=0.

[0078] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include:

[0079] ,

[0080] in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the A sixth adjustment factor for the first pattern recognition value.

[0081] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include:

[0082] ,

[0083] in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the A sixth adjustment factor for the second pattern recognition value.

[0084] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix include:

[0085] ,

[0086] in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

[0087] Specifically, calculate the new satellite data matrix after preprocessing No. The first eigenvalue include:

[0088] ,

[0089] in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

[0090] Specifically, calculate the new satellite data matrix after preprocessing No. The second eigenvalue include:

[0091] ,

[0092] in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

[0093] Step 104, input the comprehensive pattern recognition value into the classifier model to perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0094] Example 2

[0095] like Figure 2 As shown, an embodiment of the present invention further provides a knowledge base automatic construction system in an intelligent question-answering system, which involves the construction of a knowledge base of satellite data, including:

[0096] A training set acquisition module is used to acquire a satellite data matrix of annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities;

[0097] A training classifier module is used to train a classifier based on a satellite data matrix with labeled modes, so that the classifier can learn the differences between various modes;

[0098] Specifically, the classifier model is: support vector machine, decision tree, random forest or k-nearest neighbor.

[0099] Setting a model module to obtain and preprocess a new satellite data matrix, generate a preprocessed new satellite data matrix, set a pattern recognition model, and calculate a comprehensive pattern recognition value of the new satellite data matrix;

[0100] Specifically, the pattern recognition model includes:

[0101] ,

[0102] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. A second pattern recognition value.

[0103] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include:

[0104] ,

[0105] in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the A sixth adjustment factor for the first pattern recognition value.

[0106] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include:

[0107] ,

[0108] in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the A sixth adjustment factor for the second pattern recognition value.

[0109] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix include:

[0110] ,

[0111] in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

[0112] Specifically, calculate the new satellite data matrix after preprocessing No. The first eigenvalue include:

[0113] ,

[0114] in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

[0115] Specifically, calculate the new satellite data matrix after preprocessing No. The second eigenvalue include:

[0116] ,

[0117] in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

[0118] The pattern recognition module is used to input the comprehensive pattern recognition value into the classifier model, perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0119] Example 3

[0120] An embodiment of the present invention further proposes a storage medium storing a plurality of instructions, wherein the instructions are used to implement the method for automatically constructing a knowledge base in the intelligent question-answering system.

[0121] Optionally, in this embodiment, the above storage medium may be located in any computer terminal in a computer terminal group in a computer network, or in any mobile terminal in a mobile terminal group.

[0122] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps: Step 101, obtaining a satellite data matrix of annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities;

[0123] Step 102, training a classifier based on the satellite data matrix of the marked modes so that it can learn the differences between various modes;

[0124] Specifically, the classifier model is: support vector machine, decision tree, random forest or k-nearest neighbor.

[0125] Step 103, obtaining a new satellite data matrix and preprocessing it, generating a preprocessed new satellite data matrix, setting a pattern recognition model, and calculating a comprehensive pattern recognition value of the new satellite data matrix;

[0126] Specifically, the pattern recognition model includes:

[0127] ,

[0128] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. A second pattern recognition value.

[0129] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include:

[0130] ,

[0131] in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the A sixth adjustment factor for the first pattern recognition value.

[0132] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include:

[0133] ,

[0134] in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the A sixth adjustment factor for the second pattern recognition value.

[0135] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix include:

[0136] ,

[0137] in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

[0138] Specifically, calculate the new satellite data matrix after preprocessing No. The first eigenvalue include:

[0139] ,

[0140] in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

[0141] Specifically, calculate the new satellite data matrix after preprocessing No. The second eigenvalue include:

[0142] ,

[0143] in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

[0144] Step 104, input the comprehensive pattern recognition value into the classifier model to perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0145] Example 4

[0146] An embodiment of the present invention also proposes an electronic device, including a processor and a storage medium connected to the processor, wherein the storage medium stores multiple instructions, and the instructions can be loaded and executed by the processor so that the processor can execute the method for automatically constructing a knowledge base in an intelligent question and answer system.

[0147] Specifically, the electronic device of this embodiment may be a computer terminal, and the computer terminal may include: one or more processors, and a storage medium.

[0148] Among them, the storage medium can be used to store software programs and modules, such as the automatic construction method of the knowledge base in an intelligent question-and-answer system in an embodiment of the present invention, the corresponding program instructions / modules, and the processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium, that is, realizing the automatic construction method of the knowledge base in the above-mentioned intelligent question-and-answer system. The storage medium may include a high-speed random storage medium, and may also include a non-volatile storage medium, such as one or more magnetic storage systems, flash memory, or other non-volatile solid-state storage media. In some instances, the storage medium may further include a storage medium remotely arranged relative to the processor, and these remote storage media may be connected to the terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0149] The processor may call the information and application program stored in the storage medium through the transmission system to perform the following steps: Step 101, obtaining a satellite data matrix of annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities;

[0150] Step 102, training a classifier based on the satellite data matrix of the marked modes so that it can learn the differences between various modes;

[0151] Specifically, the classifier model is: support vector machine, decision tree, random forest or k-nearest neighbor.

[0152] Step 103, obtaining a new satellite data matrix and preprocessing it, generating a preprocessed new satellite data matrix, setting a pattern recognition model, and calculating a comprehensive pattern recognition value of the new satellite data matrix;

[0153] Specifically, the pattern recognition model includes:

[0154] ,

[0155] in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. A second pattern recognition value.

[0156] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include:

[0157] ,

[0158] in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the A sixth adjustment factor for the first pattern recognition value.

[0159] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include:

[0160] ,

[0161] in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the A sixth adjustment factor for the second pattern recognition value.

[0162] Specifically, calculate the composite eigenvalues ​​of the new satellite data matrix include:

[0163] ,

[0164] in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

[0165] Specifically, calculate the new satellite data matrix after preprocessing No. The first eigenvalue include:

[0166] ,

[0167] in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

[0168] Specifically, calculate the new satellite data matrix after preprocessing No. The second eigenvalue include:

[0169] ,

[0170] in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

[0171] Step 104, input the comprehensive pattern recognition value into the classifier model to perform pattern classification, and finally complete the construction of the pattern knowledge base.

[0172] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0173] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0174] In the several embodiments provided by the present invention, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the system embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0175] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0176] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0177] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only storage medium (ROM, Read-Only Memory), random access storage medium (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0178] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. A method for automatically constructing a knowledge base in an intelligent question-answering system, involving the construction of a knowledge base for satellite data, characterized in that: include: Obtaining a satellite data matrix with annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities; Based on the satellite data matrix with labeled patterns, a classifier is trained to learn the differences between various patterns; Acquire a new satellite data matrix and preprocess it, generate a preprocessed new satellite data matrix, set a pattern recognition model, and calculate a comprehensive pattern recognition value of the new satellite data matrix; Inputting the comprehensive pattern recognition value into the classifier model to perform pattern classification, and finally completing the construction of the pattern knowledge base; The pattern recognition model includes: ， in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. a second pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include: ， in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the a sixth adjustment factor for the first pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include: ， in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the a sixth adjustment factor for the second pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix include: ， in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

2. The method for automatically constructing a knowledge base in an intelligent question-answering system according to claim 1, characterized in that: Calculate the new satellite data matrix after preprocessing No. The first eigenvalue include: ， in, For the The first adjustment factor for the first eigenvalue, For the The second adjustment factor for the first eigenvalue, For the The third adjustment factor for the first eigenvalue, For the The fourth adjustment factor for the first eigenvalue, For the The fifth adjustment factor for the first eigenvalue, For the The sixth adjustment factor for the first eigenvalue, For the The seventh adjustment factor for the first eigenvalue, For the The eighth adjustment factor for the first eigenvalue.

3. The method for automatically constructing a knowledge base in an intelligent question-answering system according to claim 1, characterized in that: Calculate the new satellite data matrix after preprocessing No. The second eigenvalue include: ， in, For the The first adjustment factor for the second eigenvalue, For the The second adjustment factor for the second eigenvalue, For the The third adjustment factor for the second eigenvalue, For the The fourth adjustment factor for the second eigenvalue, For the The fifth adjustment factor for the second eigenvalue, For the The sixth adjustment factor for the second eigenvalue.

4. The method for automatically constructing a knowledge base in an intelligent question-answering system according to claim 1, characterized in that: The classifier model is: support vector machine, decision tree, random forest or k nearest neighbor.

5. A knowledge base automatic construction system in an intelligent question-answering system, involving the construction of a knowledge base of satellite data, characterized in that: include: A training set acquisition module is used to acquire a satellite data matrix of annotated patterns, wherein the patterns include: vegetation cover, water bodies, land use, meteorological phenomena and / or geological activities; A training classifier module is used to train a classifier based on a satellite data matrix with labeled modes, so that the classifier can learn the differences between various modes; Setting a model module to obtain and preprocess a new satellite data matrix, generate a preprocessed new satellite data matrix, set a pattern recognition model, and calculate a comprehensive pattern recognition value of the new satellite data matrix; A pattern recognition module, used to input the comprehensive pattern recognition value into the classifier model, perform pattern classification, and finally complete the construction of a pattern knowledge base; The pattern recognition model includes: ， in, is the composite eigenvalue of the new satellite data matrix The comprehensive pattern recognition value of is the number of modes, For the An adjustment factor for the overall pattern recognition value, For the The weight of the first pattern recognition value, For the The weight of the second pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. The first pattern recognition value, is the composite eigenvalue of the new satellite data matrix No. a second pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix No. First pattern recognition value include: , in, For the a first adjustment factor for the first pattern recognition value, For the a second adjustment factor for the first pattern recognition value, For the a third adjustment factor for the first pattern recognition value, For the a fourth adjustment factor of the first pattern recognition value, For the a fifth adjustment factor of the first pattern recognition value, For the a sixth adjustment factor for the first pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix No. Second pattern recognition value include: , in, For the a first adjustment factor for a second pattern recognition value, For the a second adjustment factor for a second pattern recognition value, For the a third adjustment factor for the second pattern recognition value, For the a fourth adjustment factor for the second pattern recognition value, For the a fifth adjustment factor for the second pattern recognition value, For the a sixth adjustment factor for the second pattern recognition value; Calculate the composite eigenvalues ​​of the new satellite data matrix include: , in, is the number of eigenvalues, For the The adjustment factor for the composite eigenvalues, For the The adjustment factor for the first eigenvalue, is the new satellite data matrix after preprocessing No. The first eigenvalue, For the The weight of the second eigenvalue, For the The adjustment factor for the second eigenvalue, is the new satellite data matrix after preprocessing No. The second eigenvalue.

Images