Risk early warning method, system and device based on potential safety hazard big data and medium

Abstract

The invention relates to the technical field of network security, in particular to a risk early warning method, system and device based on potential safety hazard big data and a medium. According to the method, firstly, an initial vector space is constructed according to massive accident potential text reports to obtain a structured data matrix, secondly, unique key segmented words are extracted according to segmented words and set unique indexes to form a keyword vector space, and finally, a risk early warning instruction is generated according to a set risk level and an obtained low-dimensional potential data matrix. Through comprehensive analysis of accident potential report texts, data dimension reduction and key feature representation of potential accidents are realized, risk early warning is generated in time in combination with a set risk level, and stable operation of network security is guaranteed.

Description

Technical Field

[0001] The present invention relates to the field of network security technology, and in particular to a risk warning method, system, device and medium based on potential safety hazard big data. Background Art

[0002] With the emergence and development of new technologies such as big data, artificial intelligence, the Internet of Things, cloud computing, and mobile Internet, the network is facing a more complex situation in the field of information security than ever before, with new security issues constantly emerging and new security incidents emerging one after another. These include both endless intrusions and attacks from the outside, as well as massive text reports of accident hazards generated by internal operational violations and improper management.

[0003] With the widespread application of artificial intelligence technology in the past two years, network information systems are facing a trend of more intelligent, automated, and weaponized network attacks. Traditional security defense systems have serious risks and shortcomings and can no longer cope with increasingly complex and hidden advanced threats. Comprehensive analysis based on accident hazard text reports can effectively curb the occurrence of security incidents, which is of great significance for ensuring the safe operation of the network. Summary of the invention

[0004] Aiming at the serious risks and shortcomings of the existing security defense system and the problem that it is unable to solve the complex and hidden advanced threats, the present invention proposes a risk warning method, system, equipment and medium based on big data of safety hazards. The method first constructs an initial vector space according to a massive amount of accident hazard text reports to obtain a structured data matrix. Secondly, according to the word segmentation and the set uniqueness index, the unique key word segmentation is extracted to form a keyword vector space. Finally, according to the set risk level and the low-dimensional hazard data matrix, a risk warning instruction is generated. Through the comprehensive analysis of the accident hazard report text, the data dimension simplification and key feature representation of the hazard accident are realized, and the risk warning is generated in time to ensure the stable operation of network security.

[0005] The specific implementation contents of the present invention are as follows: A risk warning method based on safety hazard big data specifically includes the following steps: Step S1: calling a big data analysis method to obtain a text report of accident hazards, and preprocessing to obtain the accident hazard text; Step S2: construct the accident potential hazard text vector after word segmentation preprocessing to obtain a structured data matrix; Step S3: extract key words according to the structured data matrix to obtain keyword vectors; Step S4: Based on the keyword vector, call the Pearson correlation coefficient to construct a weighted graph, and map it to obtain a low-dimensional hidden danger data matrix; Step S5: Generate risk warning instructions based on the set risk level and the obtained low-dimensional hidden danger data matrix to complete the risk warning.

[0006] In order to better implement the present invention, further, the step S1 specifically includes the following steps: Step S11: calling a big data analysis method to obtain accident potential data; Step S12: Obtaining a text report of the potential accident according to the potential accident data; Step S13: Pre-process the accident potential text report to obtain the accident potential text.

[0007] In order to better implement the present invention, further, the step S2 specifically includes the following steps: Step S21: segmenting the accident potential hazard text by using stop words, and filtering to obtain the accident potential hazard text segmentation; Step S22: constructing an accident hazard text vector according to the frequencies corresponding to the accident hazard text word segments; Step S23: construct a structured data matrix based on the accident hazard text vector.

[0008] In order to better implement the present invention, further, the step S3 specifically includes the following steps: Step S31: constructing a relative frequency matrix according to the constructed structured data matrix and the obtained occurrence frequencies of the dominant accident potential codes; Step S32: according to the constructed relative frequency matrix, remove abnormal segmented words and construct a cleaned accident hazard dictionary; Step S33: constructing a strong unique accident hidden danger dictionary based on the cleaned hidden danger dictionary and the set uniqueness index; Step S34: extracting key participles of potential accident hazards according to the strong uniqueness potential accident hazard dictionary; Step S35: converting the extracted accident hazard key words into accident hazard keyword vectors.

[0009] In order to better implement the present invention, further, the step S33 specifically includes the following steps: Step S331: constructing a uniqueness index according to the set skewness and the set kurtosis; Step S332: Construct a strong unique accident hazard dictionary based on the uniqueness index.

[0010] In order to better implement the present invention, further, the step S4 specifically includes the following steps: Step S41: Based on the keyword vector, the Pearson correlation coefficient is called to construct a weighted graph of potential accidents; Step S42: Obtain accident potential danger vertex samples from the accident potential danger weighted graph and construct an accident potential danger vertex degree matrix; Step S43: constructing an objective function according to the accident potential risk weighted graph and the accident potential risk vertex degree matrix; Step S44: According to the objective function, mapping to a low-dimensional space is performed to obtain a low-dimensional accident hazard data matrix.

[0011] In order to better implement the present invention, further, step S5 specifically includes the following steps: Step S51: Generate accident potential information according to the low-dimensional accident potential data matrix; Step S52: Generate risk warning instructions based on the set risk level and accident potential information to complete risk warning.

[0012] Based on the above-mentioned risk warning method based on safety hazard big data, in order to better realize the present invention, further, a risk warning system based on safety hazard big data is proposed, which is used to execute the above-mentioned risk warning method based on safety hazard big data; it includes a preprocessing unit, a word segmentation unit, a keyword unit, a mapping unit, and an early warning unit; The preprocessing unit is used to call the big data analysis method to obtain the accident potential hazard text report, and preprocess to obtain the accident potential hazard text; The word segmentation unit is used to segment the accident potential hazard text after word segmentation preprocessing, construct the accident potential hazard text vector, and obtain a structured data matrix; The keyword unit is used to extract the key words of the potential accident hazards according to the structured data matrix to obtain the potential accident hazard keyword vector; The mapping unit is used to construct a weighted graph of potential accident hazards by calling the Pearson correlation coefficient according to the potential accident hazard keyword vector, and map to obtain a low-dimensional potential accident hazard data matrix; The early warning unit is used to generate risk early warning instructions and complete risk early warning according to the set risk level and the obtained low-dimensional accident hazard data matrix.

[0013] Based on the above-mentioned risk warning method based on safety hazard big data, in order to better realize the present invention, further, an electronic device is proposed, including a memory and a processor; a computer program is stored on the memory; when the computer program is executed on the processor, the above-mentioned risk warning method based on safety hazard big data is implemented.

[0014] Based on the above-mentioned risk warning method based on safety hazard big data, in order to better realize the present invention, further, a computer-readable storage medium is proposed, on which computer instructions are stored; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned risk warning method based on safety hazard big data is implemented.

[0015] The present invention has the following beneficial effects: The present invention first constructs an initial vector space based on massive accident hazard text reports to obtain a structured data matrix. Secondly, based on word segmentation and set uniqueness indicators, unique key word segmentations are extracted to form a keyword vector space. Finally, based on the set risk level and the obtained low-dimensional hazard data matrix, risk warning instructions are generated. Through comprehensive analysis of accident hazard report texts, data dimension simplification and key feature representation of hazard accidents are achieved, and risk warnings are generated in a timely manner to ensure the stable operation of network security. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic flowchart of the risk warning method based on safety hazard big data provided by the present invention. DETAILED DESCRIPTION

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. It should be understood that the described embodiments are only part of the embodiments of the present invention, not all of the embodiments, and therefore should not be regarded as limiting the scope of protection. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technical personnel in this field without making creative work are within the scope of protection of the present invention.

[0018] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "disposed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0019] Embodiment 1: This embodiment proposes a risk warning method based on potential safety hazard big data, which specifically includes the following steps: Step S1: calling a big data analysis method to obtain a text report of accident hazards, and preprocessing to obtain the accident hazard text; The step S1 specifically includes the following steps: Step S11: calling a big data analysis method to obtain accident potential data; Step S12: Obtaining a text report of the potential accident according to the potential accident data; Step S13: Pre-process the accident potential text report to obtain the accident potential text.

[0020] Step S2: construct the accident potential hazard text vector after word segmentation preprocessing to obtain a structured data matrix.

[0021] The step S2 specifically includes the following steps: Step S21: segmenting the accident potential hazard text by using stop words, and filtering to obtain the accident potential hazard text segmentation; Step S22: constructing an accident hazard text vector according to the frequencies corresponding to the accident hazard text word segments; Step S23: construct a structured data matrix based on the accident hazard text vector.

[0022] Step S3: extract key words according to the structured data matrix to obtain keyword vectors; The step S3 specifically comprises the following steps: Step S31: constructing a relative frequency matrix according to the constructed structured data matrix and the obtained occurrence frequencies of the dominant accident potential codes; Step S32: According to the constructed relative frequency matrix, abnormal segmentation words are eliminated to construct a cleaned accident hazard dictionary; Step S33: construct a highly unique accident hazard dictionary based on the cleaned hazard dictionary and the set uniqueness index.

[0023] The step S33 specifically includes the following steps: Step S331: constructing a uniqueness index according to the set skewness and the set kurtosis; Step S332: Construct a strong unique accident hazard dictionary based on the uniqueness index.

[0024] Step S34: extracting key participles of potential accident hazards according to the strong uniqueness potential accident hazard dictionary; Step S35: converting the extracted accident hazard key words into accident hazard keyword vectors.

[0025] Step S4: Based on the keyword vector, call the Pearson correlation coefficient to construct a weighted graph and map it to obtain a low-dimensional hidden danger data matrix.

[0026] The step S4 specifically comprises the following steps: Step S41: Based on the keyword vector, the Pearson correlation coefficient is called to construct a weighted graph of potential accident hazards; Step S42: Obtain accident potential vertices samples from the accident potential weighted graph and construct an accident potential vertex degree matrix; Step S43: constructing an objective function according to the accident potential risk weighted graph and the accident potential risk vertex degree matrix; Step S44: According to the objective function, mapping to a low-dimensional space is performed to obtain a low-dimensional accident hazard data matrix.

[0027] Step S5: Generate risk warning instructions based on the set risk level and the obtained low-dimensional hidden danger data matrix to complete the risk warning.

[0028] The step S5 specifically comprises the following steps: Step S51: Generate accident potential information according to the low-dimensional accident potential data matrix; Step S52: Generate risk warning instructions based on the set risk level and accident potential information to complete risk warning.

[0029] Working principle: Embodiment 2: This embodiment is based on the above embodiment 1. Figure 1 As shown, a specific embodiment is described in detail.

[0030] Step S1: Call the big data analysis method to obtain the accident hazard text report, and pre-process the accident hazard text.

[0031] First, call big data analysis to zoom in and obtain the accident hazard text report, and perform text preprocessing to eliminate text that is not related to the accident hazard, and obtain the preprocessed accident hazard text.

[0032] Step S2: construct the accident potential hazard text vector after word segmentation preprocessing to obtain a structured data matrix.

[0033] The unstructured text formed by the preprocessed accident potential hazard text is segmented. Accurate segmentation can better describe the phenomenon caused by the accident potential hazard. Otherwise, it may lead to a deviation in the understanding of the accident potential hazard. Therefore, this embodiment filters out single characters, punctuation marks, auxiliary words and other words that are obviously irrelevant to the description of accident hazards by means of stop words to improve the effect and accuracy of word segmentation; then, a vector space is constructed based on the initially obtained m accident hazard word segmentations, and according to the frequency corresponding to the word segmentations, the frequency is filled into the corresponding position in the corresponding m-dimensional vector to obtain the accident hazard text vector.

[0034] Finally, a structured data matrix is ​​constructed based on the accident hazard text vector.

[0035] Step S3: Extract key words according to the structured data matrix to obtain keyword vectors.

[0036] Firstly, K types of dominant accident hazard codes are selected in the structured data matrix according to the frequency of occurrence of accident hazard codes to obtain k accident hazard data; the relative frequency of the accident hazard word segments in the k accident hazard data is calculated to obtain the relative frequency vector; and the various types of dominant accident hazard codes are integrated to obtain the relative frequency matrix.

[0037] Secondly, for various accident hazard codes, the more unique the segmentation words used in the accident hazard description are, the easier it is to distinguish. Before looking for unique accident hazard segmentation words, it is necessary to eliminate abnormal and unique accident hazard segmentation words. Such accident hazard segmentation words usually appear very rarely in accident hazard data, which may be caused by spelling errors, rare words or uncommon domain-specific terms, and are not widespread. Therefore, based on the relative frequency matrix, the abnormal and unique accident hazard segmentation words are eliminated, and the cleaned accident hazard dictionary is constructed based on the accident hazard data after elimination.

[0038] Then, according to the cleaned accident hazard dictionary, the skewness and kurtosis are set to construct the uniqueness index, and the uniqueness of each accident hazard segmentation is further measured by deeply analyzing the symmetry of the relative frequency data distribution of the remaining segmentations; therefore, the first a strong unique accident hazard segmentations are selected from the cleaned accident hazard dictionary to construct a strong unique accident hazard dictionary; According to the relative frequency matrix, the first b high relative frequency words in each type of accident hazard code are selected to obtain a strong applicability accident hazard dictionary, and then c key accident hazard words are obtained according to the strong uniqueness accident hazard dictionary and the strong applicability accident hazard dictionary to construct a classic accident hazard dictionary; Finally, based on the extracted c key accident hazard segmentations, an accident hazard keyword vector space is constructed, and the initial unstructured accident hazard text data in various accident hazard events are converted into accident hazard keyword vectors with a variable dimension of c, so as to realize the description of accident site operation information with low-dimensional numerical structured data.

[0039] Step S4: Based on the keyword vector, call the Pearson correlation coefficient to construct a weighted graph and map it to obtain a low-dimensional hidden danger data matrix.

[0040] In order to further explore the intrinsic correlation between accident hazard keyword vectors and realize feature dimension reduction and representation of accident hazard keyword vectors, this embodiment firstly calls the Pearson correlation coefficient based on the keyword vectors to construct an accident hazard weighted graph; Secondly, the accident potential vertices samples are obtained from the accident potential weighted graph, and the accident potential vertex degree matrix is ​​constructed; Then, the objective function is constructed according to the accident potential hazard weighted graph and the accident potential hazard vertex degree matrix; Finally, according to the objective function, the low-dimensional accident hazard data matrix is ​​obtained by mapping it to the low-dimensional space.

[0041] Step S5: Generate risk warning instructions based on the set risk level and the obtained low-dimensional hidden danger data matrix to complete the risk warning.

[0042] First, based on the low-dimensional accident hazard data matrix, the accident hazard information is generated; then, based on the set risk level and accident hazard information, a risk warning instruction is generated to complete the risk warning. The risk levels in this embodiment include A, B, C, and D, and each risk level corresponds to a different accident hazard code.

[0043] Working principle: This embodiment extracts accident hazard keyword features, and introduces skewness and kurtosis to construct a uniqueness index based on the relative frequency matrix, extracts accident hazard key participles including uniqueness and applicability, constructs a classic dictionary of accident hazards, and forms an accident hazard keyword vector space. In order to further explore the potential relationship between accident hazard keyword vectors, the Pearson correlation coefficient is called to realize the feature representation of the accident hazard keyword vector. Finally, risk warnings are generated in time according to the set risk level to ensure the stable operation of network security.

[0044] The other parts of this embodiment are the same as those of the above-mentioned embodiment 1, and thus will not be described in detail.

[0045] Embodiment 3: Based on any one of the above-mentioned embodiments 1-2, this embodiment proposes a risk warning system based on safety hazard big data, which is used to execute the above-mentioned risk warning method based on safety hazard big data; it includes a preprocessing unit, a word segmentation unit, a keyword unit, a mapping unit, and an early warning unit; The preprocessing unit is used to call the big data analysis method to obtain the accident potential hazard text report, and preprocess to obtain the accident potential hazard text; The word segmentation unit is used to segment the accident potential hazard text after word segmentation preprocessing, construct the accident potential hazard text vector, and obtain a structured data matrix; The keyword unit is used to extract the key words of the potential accident hazards according to the structured data matrix to obtain the potential accident hazard keyword vector; The mapping unit is used to construct a weighted graph of potential accident hazards by calling the Pearson correlation coefficient according to the potential accident hazard keyword vector, and map to obtain a low-dimensional potential accident hazard data matrix; The early warning unit is used to generate risk early warning instructions and complete risk early warning according to the set risk level and the obtained low-dimensional accident hazard data matrix.

[0046] This embodiment also proposes an electronic device, including a memory and a processor; a computer program is stored on the memory; when the computer program is executed on the processor, the above-mentioned risk warning method based on safety hazard big data is implemented.

[0047] This embodiment also proposes a computer-readable storage medium, on which computer instructions are stored; when the computer instructions are executed on the above-mentioned electronic device, the above-mentioned risk warning method based on safety hazard big data is implemented.

[0048] The other parts of this embodiment are the same as any one of the above-mentioned embodiments 1-2, so they will not be repeated here.

[0049] The processor involved in the embodiments of the present application may be a chip. For example, it may be a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0050] The memory involved in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0051] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0052] Those of ordinary skill in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0053] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0054] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the modules is only a logical function division. There may be other division methods in actual implementation, such as multiple modules or components can be combined or integrated into another device, 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 devices or modules, which can be electrical, mechanical or other forms.

[0055] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical modules, that is, they may be located in one device or distributed on multiple devices. Some or all of the modules may be selected according to actual needs to achieve the purpose of the present embodiment.

[0056] In addition, each functional module in each embodiment of the present application may be integrated into one device, or each module may exist physically separately, or two or more modules may be integrated into one device.

[0057] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using a software program, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium may be any available medium that a computer can access or may include one or more servers, data centers and other data storage devices that can be integrated with the medium. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a DVD), or a semiconductor medium (eg, a solid state disk (SSD)).

[0058] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A risk warning method based on big data of potential safety hazards, characterized in that: The specific steps include: Step S1: calling a big data analysis method to obtain a text report of accident hazards, and preprocessing to obtain the accident hazard text; Step S2: construct the accident potential hazard text vector after word segmentation preprocessing to obtain a structured data matrix; Step S3: extract key words according to the structured data matrix to obtain keyword vectors; Step S4: Based on the keyword vector, call the Pearson correlation coefficient to construct a weighted graph, and map it to obtain a low-dimensional hidden danger data matrix; Step S5: Generate risk warning instructions based on the set risk level and the obtained low-dimensional hidden danger data matrix to complete the risk warning.

2. The risk warning method based on big data of potential safety hazards according to claim 1 is characterized in that: The step S1 specifically includes the following steps: Step S11: calling a big data analysis method to obtain accident potential data; Step S12: Obtaining a text report of the potential accident according to the potential accident data; Step S13: Pre-process the accident potential text report to obtain the accident potential text.

3. The risk warning method based on big data of potential safety hazards according to claim 2 is characterized in that: The step S2 specifically includes the following steps: Step S21: segmenting the accident potential hazard text by using stop words, and filtering to obtain the accident potential hazard text segmentation; Step S22: constructing an accident hazard text vector according to the frequencies corresponding to the accident hazard text word segments; Step S23: construct a structured data matrix based on the accident hazard text vector.

4. The risk warning method based on big data of potential safety hazards according to claim 3 is characterized in that: The step S3 specifically comprises the following steps: Step S31: constructing a relative frequency matrix according to the constructed structured data matrix and the obtained occurrence frequencies of the dominant accident potential codes; Step S32: according to the constructed relative frequency matrix, remove abnormal segmented words and construct a cleaned accident hazard dictionary; Step S33: constructing a strong unique accident hidden danger dictionary based on the cleaned hidden danger dictionary and the set uniqueness index; Step S34: extracting key participles of potential accident hazards according to the strong uniqueness potential accident hazard dictionary; Step S35: converting the extracted accident hazard key words into accident hazard keyword vectors.

5. The risk warning method based on big data of potential safety hazards according to claim 4 is characterized in that: The step S33 specifically includes the following steps: Step S331: constructing a uniqueness index according to the set skewness and the set kurtosis; Step S332: Construct a strong unique accident hazard dictionary based on the uniqueness index.

6. The risk warning method based on big data of potential safety hazards according to claim 4 is characterized in that: The step S4 specifically comprises the following steps: Step S41: Based on the keyword vector, the Pearson correlation coefficient is called to construct a weighted graph of potential accidents; Step S42: Obtain accident potential danger vertex samples from the accident potential danger weighted graph and construct an accident potential danger vertex degree matrix; Step S43: constructing an objective function according to the accident potential risk weighted graph and the accident potential risk vertex degree matrix; Step S44: According to the objective function, mapping to a low-dimensional space is performed to obtain a low-dimensional accident hazard data matrix.

7. The risk warning method based on big data of potential safety hazards according to claim 1 is characterized in that: The step S5 specifically comprises the following steps: Step S51: Generate accident potential information according to the low-dimensional accident potential data matrix; Step S52: Generate risk warning instructions based on the set risk level and accident potential information to complete risk warning.

8. A risk warning system based on big data of potential safety hazards, used to execute the risk warning method based on big data of potential safety hazards as claimed in claim 1; characterized in that: It includes pre-processing unit, word segmentation unit, keyword unit, mapping unit and early warning unit; The preprocessing unit is used to call the big data analysis method to obtain the accident potential hazard text report, and preprocess to obtain the accident potential hazard text; The word segmentation unit is used to segment the accident potential hazard text after word segmentation preprocessing, construct the accident potential hazard text vector, and obtain a structured data matrix; The keyword unit is used to extract the key words of the potential accident hazards according to the structured data matrix to obtain the potential accident hazard keyword vector; The mapping unit is used to construct a weighted graph of potential accident hazards by calling the Pearson correlation coefficient according to the potential accident hazard keyword vector, and map to obtain a low-dimensional potential accident hazard data matrix; The early warning unit is used to generate risk early warning instructions and complete risk early warning according to the set risk level and the obtained low-dimensional accident hazard data matrix.

9. An electronic device, characterized in that: It comprises a memory and a processor; a computer program is stored in the memory; when the computer program is executed on the processor, the risk warning method based on safety hazard big data as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions; when the computer instructions are executed on the electronic device as described in claim 9, the risk warning method based on safety hazard big data as described in any one of claims 1-7 is implemented.

Images