Abnormal event detection method and device, equipment, medium and product

Abstract

The invention discloses an abnormal event detection method and device, equipment, a medium and a product. The method comprises the following steps: acquiring multi-source sensing data of a plurality of fields; clustering the multi-source sensing data belonging to the same field in the plurality of fields to obtain a plurality of clustered data sets; performing feature extraction on the multi-source sensing data in each clustering data set to obtain multiple pieces of feature data; and inputting the feature data corresponding to each cluster data set into a preset abnormal event detection model to obtain a plurality of detection results. According to the embodiment of the invention, the abnormal event can be detected based on massive multi-source sensing data, and the detection efficiency of the abnormal event is improved.

Description

Technical Field

[0001] The present application belongs to the field of data processing, and in particular, relates to a method, device, equipment, medium and product for detecting abnormal events. Background Art

[0002] With the continuous development of information technology, the amount of data generated on various information channels such as the Internet and the Internet of Things has grown exponentially. The rapid increase in the share of these data has brought abundant business opportunities, such as finance, agriculture, water conservancy, urban development and other fields, but the massive data also brings great challenges to the detection of abnormal events. As a result, the current efficiency of extracting useful information from a large amount of data to detect abnormal events is low. Summary of the invention

[0003] The embodiments of the present application provide a method, device, equipment, medium and product for detecting abnormal events, which can detect abnormal events in massive data and improve the detection efficiency of abnormal events.

[0004] In a first aspect, an embodiment of the present application provides a method for detecting an abnormal event, comprising:

[0005] Acquire multi-source perception data in multiple fields;

[0006] Clustering multi-source perception data belonging to the same field in multiple fields to obtain multiple clustering data sets;

[0007] Feature extraction is performed on the multi-source perception data in each clustering data set to obtain multiple feature data;

[0008] The feature data corresponding to each clustering data set is input into a preset abnormal event detection model to obtain multiple detection results.

[0009] In a second aspect, an embodiment of the present application provides a device for detecting an abnormal event, comprising:

[0010] An acquisition module, used to acquire multi-source perception data in multiple fields;

[0011] A clustering module is used to cluster multi-source perception data belonging to the same field in multiple fields to obtain multiple clustering data sets;

[0012] An extraction module is used to extract features from the multi-source perception data in each clustering data set to obtain multiple feature data;

[0013] The detection module is used to input the feature data corresponding to each cluster data set into a preset abnormal event detection model to obtain multiple detection results.

[0014] In a third aspect, an embodiment of the present application provides an electronic device, the device comprising:

[0015] a processor and a memory storing computer program instructions;

[0016] When the processor executes the computer program instructions, it is used to execute the abnormal event detection method of the first aspect.

[0017] In a fourth aspect, an embodiment of the present application provides a computer storage medium having computer program instructions stored thereon, and when the computer program instructions are executed by a processor, the abnormal event detection method of the first aspect described above is implemented.

[0018] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program, which, when processed by a processor, implements the abnormal event detection method of the above-mentioned first aspect.

[0019] The abnormal event detection method, device, equipment, medium and product provided in the embodiments of the present application can cluster the multi-source perception data in the same field based on the multi-source perception data in multiple fields obtained, and extract features from the multi-source perception data of each clustered data set to obtain feature data, and use the abnormal event detection model to perform detection based on the feature data to obtain detection results, thereby realizing automatic detection of abnormal events in massive multi-source perception data and improving the detection efficiency of abnormal events. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solution of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0021] Figure 1 A flowchart of a method for detecting abnormal events provided in accordance with an embodiment of the present application.

[0022] Figure 2 A schematic diagram of the hardware structure of an abnormal event detection device provided in one embodiment of the present application.

[0023] Figure 3 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0024] The features and exemplary embodiments of various aspects of the present application will be described in detail below. In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only intended to explain the present application, rather than to limit the present application. For those skilled in the art, the present application can be implemented without the need for some of these specific details. The following description of the embodiments is only to provide a better understanding of the present application by illustrating the examples of the present application.

[0025] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "include..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0026] With the development of information technology, it is necessary to detect abnormal events in massive data. Timely detection of abnormal events can provide early warning of abnormal events, thereby reducing the losses caused by abnormal events. However, due to the large number of data sources and the large amount of data, the detection efficiency of abnormal events in massive data is low.

[0027] Therefore, the embodiments of the present application propose a method, device, equipment, medium and product for detecting abnormal events, which can extract features from multi-source perception data in massive multi-source perception data according to different fields to obtain feature data, and use an abnormal event detection model to detect based on the feature data to obtain detection results, thereby realizing the detection of abnormal events in massive multi-source perception data and improving the detection efficiency of abnormal events. Below, the detection method of abnormal events provided by the embodiments of the present application is specifically described.

[0028] like Figure 1 As shown, the embodiment of the present application provides a method for detecting abnormal events, including:

[0029] S110: Acquire multi-source perception data in multiple fields.

[0030] Perception data is data measured and transmitted by perception devices. Perception devices are usually sensor devices. The characteristic of sensor devices is that they can collect a large amount of data in real time and dynamically. Perception data obtained from different sensors is multi-source perception data. The fields involved in multi-source perception data may include finance, agriculture, water conservancy, urban development, etc. At the same time, there are also many channels for obtaining multi-source remote sensing data, such as obtaining it from the Internet or the Internet of Things.

[0031] S120: Clustering multi-source perception data belonging to the same field in multiple fields to obtain multiple clustering data sets.

[0032] In this embodiment, a clustering algorithm can be used to cluster the acquired multi-source perception data in multiple fields, and the multi-source perception data belonging to the same field can be clustered to form a cluster data set, thereby obtaining multiple cluster data sets. For example, after clustering, the multi-source perception data acquired in the financial field, the agricultural field, the water conservancy field, and the urban development field will obtain 4 cluster data sets, and the 4 cluster data sets correspond to the above four fields one by one.

[0033] Here, the clustering algorithm is an unsupervised learning algorithm. The same group of data obtained by clustering by the clustering algorithm has similar attributes or characteristics. Commonly used clustering algorithms may include K-MEANS clustering algorithm, mean shift clustering algorithm and DBSCAN clustering algorithm.

[0034] S130: Extract features from the multi-source perception data in each cluster data set to obtain multiple feature data.

[0035] By extracting features from the clustered data sets of each field, we can obtain the feature data corresponding to each field. Here, feature data refers to data that can reflect the characteristics of different fields, and feature data in different fields may be different. For example, in the financial field, feature data includes macroeconomic data, public data of listed companies, and stock market data; in the agricultural field, feature data includes meteorological data, crop growth data, and water resources data.

[0036] S140: Inputting feature data corresponding to each cluster data set into a preset abnormal event detection model to obtain multiple detection results.

[0037] After the feature data corresponding to each cluster data set is input into the preset abnormal event detection model, multiple detection results can be obtained. Here, the preset abnormal event detection model can be a detection model integrated with an artificial intelligence decision-making system. It is conceivable that the abnormal event detection model integrated with the artificial intelligence decision-making system can improve the detection accuracy and robustness of the model, so that the abnormal event detection model can detect possible abnormal events based on the feature data of each cluster data set, and can detect the impact level of the abnormal event. For example, in the agricultural field, using the above-mentioned abnormal event detection model, it can be obtained whether agriculture is affected by disasters and the type of disasters. The type of disaster is specifically drought or waterlogging, and the impact level of drought or waterlogging can be detected, such as crop failure disasters, extremely serious disasters, very serious disasters, serious disasters, moderate disasters or mild disasters.

[0038] In some examples, the above-mentioned abnormal event detection model can be trained based on a machine learning algorithm. During training, different training samples can be set according to different fields. The training samples can include multiple historical abnormal event data. At the same time, auxiliary tools such as generative adversarial networks and autoencoders can be used to assist the model in training, thereby improving the detection efficiency and accuracy of the abnormal event detection model.

[0039] This embodiment can cluster the multi-source perception data in the same field based on the multi-source perception data in multiple fields obtained, and perform feature extraction on the multi-source perception data of each clustered data set to obtain feature data, and use the abnormal event detection model to perform detection based on the feature data to obtain detection results, thereby realizing automatic detection of abnormal events in massive multi-source perception data and improving the detection efficiency of abnormal events.

[0040] In some embodiments, after extracting features from the multi-source perception data in each cluster data set to obtain a plurality of feature data, the method further includes:

[0041] Abnormal data satisfying preset conditions are extracted from the multiple feature data corresponding to each cluster data set to obtain multiple abnormal data.

[0042] Among the multiple feature data corresponding to each cluster data set, abnormal data that meets the preset conditions can be extracted. Abnormal data is data that may cause abnormal events. Multiple feature data can be compared with historical abnormal event data to extract abnormal data that matches the historical abnormal event data, thereby obtaining multiple abnormal data. For example, taking the characteristic data in the agricultural field including meteorological data as an example, the meteorological data can be compared with the historical meteorological abnormal event data to extract abnormal meteorological data such as heavy rain and typhoon.

[0043] The feature data corresponding to each clustering data set is input into the preset abnormal event detection model to obtain multiple detection results, including:

[0044] The multiple abnormal data corresponding to each cluster data set are respectively input into the preset abnormal event detection model to obtain multiple detection results.

[0045] After extracting each of the multiple abnormal data, the multiple abnormal data can be input into a preset abnormal event detection model. Here, multiple abnormal event detection models can be set, and each abnormal event detection model corresponds to a cluster data set one by one. When the abnormal data corresponding to each cluster data set is input into the abnormal event detection model, the abnormal event detection model corresponding to the cluster data set can be input to obtain a detection result. For example, taking the extraction of abnormal data of heavy rain and typhoon in meteorological data as an example, the abnormal data of heavy rain and typhoon can be input into the abnormal event detection model corresponding to the cluster data set in the agricultural field to obtain a detection result.

[0046] In some embodiments, when the abnormal data is input into the abnormal event detection model, the feature vector of the abnormal event can be extracted. The feature vector can be used to describe the feature value of the abnormal data. Taking the above abnormal data of rainstorm and typhoon as an example, a two-dimensional feature vector (a, b) can be set based on the abnormal data. Vector a can represent the precipitation of the rainstorm, and vector b can represent the wind force of the typhoon. The purpose of extracting the feature vector is to convert the feature data into a data type that can be processed by the machine learning algorithm, thereby improving the detection accuracy of abnormal events.

[0047] In the embodiment of the present application, by extracting multiple abnormal data that meet preset conditions from the feature data corresponding to each cluster data, and inputting the multiple abnormal data into a preset detection model to obtain multiple detection results, it is possible to detect abnormal events using an abnormal event detection model and improve the detection accuracy of abnormal events.

[0048] In some embodiments, the detection result includes the probability of occurrence of the abnormal event.

[0049] When using the abnormal event detection model to detect multiple abnormal data, the abnormal event detection model can detect the type of abnormal event, and then detect the probability of abnormal event occurrence. When calculating the probability of occurrence, the abnormal data can be compared with the historical abnormal event data. When the abnormal data matches the historical abnormal event data, the number of abnormal events that occurred and the number of times the abnormal event did not occur are queried from the historical data, and then the probability of abnormal event occurrence is calculated. Here, the probability of abnormal event occurrence = the number of abnormal events that occurred / (the number of abnormal events that occurred + the number of times the abnormal event did not occur).

[0050] Taking the above-mentioned example of inputting the abnormal data of heavy rain and typhoon into the abnormal event detection model, the abnormal data can be the precipitation of heavy rain and the wind force of typhoon. The abnormal event detection model compares the precipitation and wind force with the historical precipitation and historical wind force respectively, and matches the number of times agricultural waterlogging occurred and the number of times no waterlogging occurred in the historical data of precipitation and wind force, and then derives the probability of possible occurrence of waterlogging in agriculture.

[0051] In some examples, based on different abnormal events, thresholds corresponding to different abnormal events can be set, and then the probability of occurrence can be compared with the threshold. When the probability of occurrence is greater than the threshold, warning information of the abnormal event can be generated. It can be imagined that when the probability of occurrence is greater than the threshold, it means that the possibility of an abnormal event is high, so warning information can be generated so that the public can take effective measures against possible abnormal events in a timely manner based on the warning information to reduce losses.

[0052] In some examples, the warning information includes prompt information of the warning level. Specifically, for a certain abnormal event, multiple thresholds can be set, such as setting a first threshold, a second threshold, and a third threshold sorted from small to large, and comparing the probability of occurrence with the above multiple thresholds. Based on the comparison results, the warning level prompt information can be generated. For example, when the probability of occurrence is greater than or equal to the third threshold, the warning level is special; when the probability of occurrence is greater than or equal to the second threshold and less than the third threshold, the warning level is high; when the probability of occurrence is greater than or equal to the first threshold and less than the second threshold, the warning level is medium; when the probability of occurrence is less than the first threshold, the warning level is low.

[0053] In some examples, when the warning level is high or above, warning information can be released through the integrated media platform to prompt the public to take effective measures in a timely manner and prepare for a rainy day.

[0054] In some examples, the warning information also includes prompt information of the warning range. Specifically, the multi-source perception data of different fields obtained above may include remote sensing data of different regions. When the detection result that an abnormal event may occur in the target area is detected, prompt information can be generated for the target area as the warning range. For example, the multi-source perception data obtained includes remote sensing data of area A, and features are extracted from the remote sensing data including area A. After the extracted feature data is input into a preset abnormal event detection model, it is obtained that the agricultural coverage area S in area A may be flooded, so prompt information that flooding occurs in the agricultural coverage area S in area A can be generated.

[0055] By including the probability of abnormal events in the detection results, setting a threshold, comparing the probability of occurrence with the threshold, and generating warning information when the probability of occurrence is greater than the threshold, the public can be prompted to take effective measures in time for possible abnormal events to reduce losses. Including warning level and / or warning range in the warning information can make the warning information clearer and further reduce the adverse effects of abnormal events.

[0056] In some embodiments, after clustering the multi-source perception data belonging to the same field in multiple fields to obtain multiple clustered data sets, the method further includes:

[0057] The multiple clustering data sets are cleaned according to the cleaning methods corresponding to different fields to obtain the multiple cleaned clustering data sets.

[0058] Data cleaning is to filter and remove duplicate and redundant data, complete the missing parts, and correct or delete incorrect data. Common data cleaning methods include binning, regression and clustering.

[0059] After obtaining multiple clustering data sets, the corresponding clustering data sets can be cleaned according to different cleaning methods according to different fields. For example, in the agricultural field, the clustering method can be used to clean the clustering data sets corresponding to the agricultural field, so as to find the noise and remove it. In the financial field, the binning method can be used to clean the clustering data sets corresponding to the financial field to achieve the purpose of eliminating noise.

[0060] Feature extraction is performed on the multi-source perception data in each clustering data set to obtain multiple feature data, including:

[0061] Feature extraction is performed on each cleaned clustering data set to obtain multiple feature data.

[0062] When performing feature extraction on the cleaned cluster data, not only can valid data be extracted, but the cluster data set can also be further refined and classified. For example, in the cluster data set in the agricultural field, the multiple feature data extracted include meteorological data, crop growth data, and water resources data.

[0063] The cleaned clustering data set is subjected to feature extraction to obtain multiple feature data, which can make the detection results based on the feature data more accurate and improve the detection accuracy of abnormal events.

[0064] like Figure 2 As shown, an embodiment of the present application provides a detection device for abnormal events, including:

[0065] Acquisition module 201 is used to acquire multi-source perception data in multiple fields.

[0066] The clustering module 202 is used to cluster multi-source perception data belonging to the same field in multiple fields to obtain multiple cluster data sets.

[0067] The extraction module 203 is used to extract features from the multi-source perception data in each cluster data set to obtain multiple feature data.

[0068] The detection module 204 is used to input the feature data corresponding to each cluster data set into a preset abnormal event detection model to obtain multiple detection results.

[0069] In some embodiments, the extraction module 203 may also be used to:

[0070] Extracting abnormal data satisfying preset conditions from multiple feature data corresponding to each clustering data set respectively to obtain multiple abnormal data;

[0071] The detection module 204 can be specifically used for:

[0072] The multiple abnormal data corresponding to each cluster data set are respectively input into the preset abnormal event detection model to obtain multiple detection results.

[0073] In some embodiments, the detection result includes the probability of occurrence of the abnormal event.

[0074] In some embodiments, the detection module 204 may also be used to:

[0075] When the probability of occurrence of an abnormal event is greater than a threshold, an early warning message of the abnormal event is generated.

[0076] In some embodiments, the warning information includes prompt information of the warning level and / or warning range.

[0077] In some embodiments, the apparatus further comprises:

[0078] A cleaning module is used to clean multiple clustering data sets according to cleaning methods corresponding to different fields to obtain multiple cleaned clustering data sets;

[0079] The extraction module 203 may also be used to:

[0080] Feature extraction is performed on each cleaned clustering data set to obtain multiple feature data.

[0081] The device of the above embodiment is used to implement the corresponding abnormal event detection method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0082] Figure 3 A schematic diagram of the hardware structure of an electronic device provided in an application embodiment.

[0083] The electronic device 300 may include a processor 301 and a memory 302 storing computer program instructions.

[0084] Specifically, the processor 301 may include a central processing unit (CPU), or an application specific integrated circuit (ASIC), or may be configured to implement one or more integrated circuits of the embodiments of the present application.

[0085] The memory 302 may include a large capacity memory for data or instructions. By way of example and not limitation, the memory 302 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a magnetic tape, or a universal serial bus (USB) drive or a combination of two or more of these. In appropriate cases, the memory 302 may include a removable or non-removable (or fixed) medium. In appropriate cases, the memory 302 may be inside or outside the integrated gateway disaster recovery device. In a specific embodiment, the memory 902 is a non-volatile solid-state memory.

[0086] In certain embodiments, memory 302 includes a read-only memory (ROM). The ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of the above, where appropriate.

[0087] The memory may include a read-only memory (ROM), a random access memory (RAM), a magnetic disk storage medium device, an optical storage medium device, a flash memory device, an electrical, optical or other physical / tangible memory storage device. Thus, typically, the memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., a memory device) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to the first aspect of the present application.

[0088] The processor 301 implements any one of the abnormal event detection methods in the above embodiments by reading and executing computer program instructions stored in the memory 302 .

[0089] In one example, the electronic device may further include a communication interface 303 and a bus 304. Figure 3The processor 301, the memory 302, and the communication interface 303 are connected via a bus 304 and communicate with each other.

[0090] The communication interface 303 is mainly used to implement communication between various modules, devices, units and / or equipment in the embodiments of the present application.

[0091] Bus 304 includes hardware, software or both, and the parts of online data flow billing equipment are coupled to each other. For example, but not limitation, bus may include accelerated graphics port (AGP) or other graphics bus, enhanced industrial standard architecture (EISA) bus, front-end bus (FSB), hypertransport (HT) interconnection, industrial standard architecture (ISA) bus, infinite bandwidth interconnection, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnection (PCI) bus, PCI-Express (PCI-X) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus or other suitable bus or two or more of these combinations. In appropriate cases, bus 304 may include one or more buses. Although the present application embodiment describes and shows a specific bus, the present application considers any suitable bus or interconnection.

[0092] The electronic device of the above embodiment is used to implement the corresponding abnormal event detection method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0093] In addition, in combination with the abnormal event detection method in the above embodiment, the embodiment of the present application can provide a computer storage medium for implementation. The computer storage medium stores computer program instructions; when the computer program instructions are executed by the processor, any abnormal event detection method in the above embodiment is implemented.

[0094] In addition, in combination with the abnormal event detection method in the above embodiment, the embodiment of the present application can provide a computer program product for implementation. When the computer program product instructions are executed by a processor of an electronic device, any abnormal event detection method in the above embodiment is implemented.

[0095] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0096] The functional blocks shown in the above-described block diagram can be implemented as hardware, software, firmware or a combination thereof. When implemented in hardware, it can be, for example, an electronic circuit, an application specific integrated circuit (ASIC), appropriate firmware, a plug-in, a function card, etc. When implemented in software, the elements of the present application are programs or code segments that are used to perform the required tasks. The program or code segment can be stored in a machine-readable medium, or transmitted on a transmission medium or a communication link by a data signal carried in a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, optical fiber media, radio frequency (RF) links, etc. The code segment can be downloaded via a computer network such as the Internet, an intranet, etc.

[0097] It should also be noted that the exemplary embodiments mentioned in this application describe some methods or devices based on a series of steps or devices. However, this application is not limited to the order of the above steps, that is, the steps can be performed in the order mentioned in the embodiment, or in a different order from the embodiment, or several steps can be performed simultaneously.

[0098] The above reference is according to the method of the embodiment of the present application, the flow chart of the device (device) and the computer program product and / or the block diagram described various aspects of the present application.It should be understood that each square box in the flow chart and / or the block diagram and the combination of each square box in the flow chart and / or the block diagram can be realized by computer program instructions.These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer or other programmable data processing device to produce a machine so that these instructions executed by the processor of the computer or other programmable data processing device enable the realization of the function / action specified in one or more square boxes of the flow chart and / or the block diagram.Such a processor can be but is not limited to a general-purpose processor, a special-purpose processor, a special application processor or a field programmable logic circuit.It can also be understood that each square box in the block diagram and / or the flow chart and the combination of the square boxes in the block diagram and / or the flow chart can also be realized by the dedicated hardware that performs the specified function or action, or can be realized by the combination of dedicated hardware and computer instructions.

[0099] The above is only a specific implementation of the present application. Those skilled in the art can clearly understand that for the convenience and simplicity of description, the specific working processes of the above-described devices, modules and units can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. It should be understood that the protection scope of the present application is not limited to this. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application.

Claims

1. A method for detecting abnormal events, characterized in that: include: Acquire multi-source perception data in multiple fields; Clustering the multi-source perception data belonging to the same field in multiple fields to obtain multiple clustered data sets; Feature extraction is performed on the multi-source perception data in each clustering data set to obtain multiple feature data; The feature data corresponding to each cluster data set is input into a preset abnormal event detection model to obtain multiple detection results.

2. The abnormal event detection method according to claim 1, characterized in that: After extracting features from the multi-source perception data in each cluster data set to obtain a plurality of feature data, the method further includes: Extracting abnormal data satisfying preset conditions from the plurality of feature data corresponding to each cluster data set respectively, to obtain a plurality of abnormal data; The feature data corresponding to each cluster data set is input into a preset abnormal event detection model to obtain multiple detection results, including: The multiple abnormal data corresponding to each cluster data set are respectively input into a preset abnormal event detection model to obtain multiple detection results.

3. The abnormal event detection method according to claim 2, characterized in that: The detection result includes the probability of occurrence of abnormal events.

4. The method for detecting abnormal events according to claim 3, characterized in that: After inputting the plurality of abnormal data corresponding to each cluster data set into a preset abnormal event detection model to obtain a plurality of detection results, the method further includes: When the probability of occurrence of the abnormal event is greater than a threshold, early warning information of the abnormal event is generated.

5. The abnormal event detection method according to claim 4, characterized in that: The warning information includes prompt information of the warning level and / or warning range.

6. The abnormal event detection method according to claim 1, characterized in that: After clustering the multi-source perception data belonging to the same field in multiple fields to obtain multiple cluster data sets, the method further includes: Cleaning the multiple clustering data sets according to cleaning methods corresponding to different fields to obtain multiple cleaned clustering data sets; The feature extraction is performed on the multi-source perception data in each cluster data set to obtain multiple feature data, including: Feature extraction is performed on each cleaned clustering data set to obtain multiple feature data.

7. A device for detecting abnormal events, characterized in that: include: An acquisition module, used to acquire multi-source perception data in multiple fields; A clustering module, used for clustering the multi-source perception data belonging to the same field in multiple fields to obtain multiple clustering data sets; An extraction module is used to extract features from the multi-source perception data in each clustering data set to obtain multiple feature data; The detection module is used to input the feature data corresponding to each cluster data set into a preset abnormal event detection model to obtain multiple detection results.

8. An electronic device, characterized in that: The device comprises: a processor, and a memory storing computer program instructions; The processor reads and executes the computer program instructions to implement the abnormal event detection method according to any one of claims 1 to 6.

9. A readable storage medium, characterized in that: The readable storage medium stores computer program instructions, and when the computer program instructions are executed by a processor, the abnormal event detection method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is processed by a processor, the abnormal event detection method as described in any one of claims 1 to 6 is implemented.

Images