A method, device, electronic device and medium for trend prediction of termite prevention

Abstract

The present application relates to the field of termite prevention, and more particularly to a termite prevention trend prediction method, device, electronic device, and medium. The method comprises: obtaining termite damage information within a preset time period in the past, determining the time series length corresponding to each building combination in the termite damage information, and performing unsupervised time series data sorting on the termite damage information to obtain first termite matrix data; then inputting the first termite matrix data into an insect pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions; then combining the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data; then performing data processing on the second termite matrix data to obtain termite damage data; then inputting the termite damage data into a preset algorithm model to generate the termite quantity for each building combination in the future preset time period. The present application has the effect of improving the accuracy of termite prediction.

Description

Technical Field

[0001] The present application relates to the field of termite prevention, and in particular to a termite prevention trend prediction method, device, electronic equipment and medium. Background Art

[0002] Termites are a destructive pest worldwide, affecting nearly every sector of the national economy. The subtropical regions of our province, where temperature and humidity are adapted, make it easier for termites to survive, causing serious damage to our national economy.

[0003] Termite prevention refers to taking measures to prevent termites from causing damage before they cause harm. However, termites' activities are hidden and are usually not discovered until they have caused huge damage. Therefore, termite monitoring has become a key issue in termite prevention. Most existing termite monitoring methods are based on placing bait in fixed monitoring nodes and then inferring termite activity by monitoring changes in the bait. However, the bait's attractant effect is greatly affected by environmental factors such as temperature and humidity. The process of termites eroding the bait is relatively slow, and termite activity information cannot be updated in a timely manner. In addition, these methods require regular inspections of the bait, which consumes a lot of manpower and material resources. From a macro perspective, there is a defect that reduces the accuracy of termite prediction. Summary of the Invention

[0004] In order to improve the accuracy of termite prediction, the present application provides a termite prevention trend prediction method, device, electronic equipment and medium.

[0005] In a first aspect, the present application provides a termite prevention trend prediction method, which adopts the following technical solutions:

[0006] A termite prevention trend prediction method, comprising:

[0007] Obtain termite damage information within a preset time period in the past, wherein the termite damage information is termite information that has occurred in different housing building types corresponding to different areas;

[0008] Analyzing the termite hazard information, determining the number of building combinations of different housing building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations, and performing unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain first termite matrix data;

[0009] Inputting the first termite matrix data into a trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combining the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data;

[0010] Data contained in the second termite matrix data is processed to obtain termite damage data, and the obtained termite damage data is input into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations in a future preset time period.

[0011] By adopting the above technical solution, when predicting the trend of termite prevention, the termite number information of different housing building types corresponding to different areas is collected to obtain termite damage information, and then the termite damage information is analyzed to obtain the number of building combinations of different housing building types and the time series length in the termite damage information, wherein the time series length and the number of building combinations are in a one-to-one correspondence. Then, the termite damage information is unsupervisedly sorted according to the time series length and the number of building combinations to obtain the first termite matrix data, and then the first termite matrix data is input into the trained pest prediction model for further analysis. Row vector features are extracted to obtain the number of termite feature dimensions, and the obtained number of termite feature dimensions is combined with the first termite matrix data for data processing to generate second termite matrix data. The data contained in the second termite matrix data is then processed one by one to obtain termite damage data. The termite damage data is input into the preset algorithm model for data extrapolation to generate the termite quantity of each building combination in the future preset time period. The staff sets the past preset time period and the future preset time period according to actual needs to obtain the termite quantity of different buildings in the future preset time period, thereby achieving the effect of improving the accuracy of termite prediction.

[0012] In another possible implementation, analyzing the termite hazard information to determine the number of building combinations of different housing building types in the termite hazard information and the length of a time series corresponding to each building combination in the number of building combinations includes:

[0013] determining at least one set of termite treatment data based on the termite damage information;

[0014] Obtaining labels for the at least one set of termite treatment data respectively to obtain building data and termite treatment time data in each set of termite treatment data;

[0015] determining whether the termite treatment data has been processed according to the termite treatment time data; if not, performing data collapse on the termite treatment data; and if the processing has been completed, correspondingly binding the building data with the termite treatment time data to obtain termite binding data;

[0016] The termite binding data is screened for building combination types to obtain the number of building combinations of different house building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations.

[0017] Through the above technical solution, when analyzing termite hazard information, at least one set of termite treatment data in the termite hazard information is obtained, and then labels are obtained for each set of termite treatment data to obtain building data and termite treatment time data in each set of termite treatment data. It is determined whether the termite treatment time data has been processed, that is, whether the processing status of the current termite treatment data is still being processed. If the processing has not been completed, the termite treatment data is collapsed and not included in the termite hazard information. If the processing is completed, the building data and the termite treatment time data are correspondingly bound to obtain termite binding data. Subsequently, the termite binding data is screened for building combination types to obtain the number of building combinations and the length of the time series. By collapsing the unprocessed termite treatment data, the accuracy of the termite hazard information is improved.

[0018] In another possible implementation, the first termite matrix data is input into a trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, including:

[0019] determining, based on the first termite matrix data, an event name, an event time, and an event area of each damage event in the termite damage information;

[0020] Inputting the event name, the event time, and the event area into the pest prediction model for vector extraction, respectively, to obtain a text feature vector corresponding to the event name, a time feature vector corresponding to the event time, and a region feature vector corresponding to the event area;

[0021] Quantity statistics are performed on the text feature vector, the time feature vector, and the region feature vector to obtain the number of termite feature dimensions.

[0022] Through the above technical solution, when obtaining the number of termite feature dimensions, the event name, event time and event area of each damage event in the termite damage information are determined according to the first termite matrix data, and then the event name, event time and event area are respectively input into the pest prediction model for vector extraction to obtain a text feature vector corresponding to the event name, a time feature vector corresponding to the event time and a regional feature vector corresponding to the event area, and then the number of termite feature dimensions is obtained by counting the text feature vectors, time feature vectors and regional feature vectors, and the feature dimensions of each damage event in the termite damage information are counted respectively, thereby achieving the effect of improving the accuracy of the number of termite feature dimensions.

[0023] In another possible implementation, combining the obtained termite feature dimension quantity with the first termite matrix data to generate second termite matrix data includes:

[0024] Integrating the termite feature dimension quantity with the first termite matrix data to generate termite dimension matrix data;

[0025] Performing basic data distribution exploration on the termite dimension matrix data to obtain a relative periodicity pattern of occurrence of damage events in the termite damage information, and determining a time period length based on the relative periodicity pattern;

[0026] Performing supervised time series data sorting on the termite dimension matrix data based on the time period length to obtain termite prediction matrix data;

[0027] The termite quantity trend within a future preset time period is predicted based on the termite prediction matrix data to generate second termite matrix data.

[0028] Through the above technical solution, when generating the second termite matrix data, the number of termite feature dimensions is integrated with the first termite matrix data to obtain termite dimension matrix data, and then the basic data distribution of the termite dimension matrix data is explored to obtain the relative periodicity of the occurrence of harmful events in the termite damage information, and the length of the time period is determined according to the relative periodicity. Then, based on the length of the time period, the termite dimension matrix data is supervised time series data sorted to obtain termite prediction matrix data, and then the termite prediction matrix data and the termite quantity trend in the future preset time period are predicted to generate the second termite matrix data, thereby achieving the effect of time series supervision of the first termite matrix data.

[0029] In another possible implementation, the processing of the data contained in the second termite matrix data to obtain the termite damage data includes:

[0030] Calculating a normal distribution mean and a normal distribution variance of data included in the second termite matrix data, and determining a 3σ range of the second termite matrix data based on the normal distribution mean and the normal distribution variance;

[0031] determining whether the data is outside the 3σ range; if the data is outside the 3σ range, determining a first matrix sequence of the second termite matrix data in which the data is located, calculating a sequence average based on the first matrix sequence, replacing the data with the sequence average to obtain a replaced second matrix sequence, and performing missing value processing on the second matrix sequence;

[0032] A sequence normalization process is performed on the second matrix sequence in the second termite matrix data to obtain termite damage data.

[0033] Through the above technical solution, when obtaining termite damage data, the 3σ range of the second termite matrix data is determined by calculating the normal distribution mean and normal distribution variance of the data contained in the second termite matrix data, and determining whether the current data is outside the 3σ range. If so, the data is eliminated, and the sequence average value is added to the position of the first matrix sequence where the data is located to obtain a second matrix sequence. Then, the second matrix sequence is processed for missing values to ensure the integrity of the matrix sequence. Then, the second matrix sequence in the second termite matrix data is normalized to obtain termite damage data, so as to facilitate subsequent data processing of the termite damage data.

[0034] In another possible implementation, the obtained termite damage data is input into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations within a future preset time period, and then further includes:

[0035] Obtaining the actual number of termites in each building combination in the number of building combinations within a future preset time period;

[0036] The termite quantity is denormalized based on the actual termite quantity to restore the termite quantity to the actual termite quantity.

[0037] Through the above technical solution, after the prediction of the termite quantity in the future preset time period is completed, the actual termite quantity occurring in the future preset time period is determined, and then the termite quantity is denormalized by the actual termite quantity, and the actual termite quantity is overwritten with the termite quantity, thereby achieving the effect of updating the termite quantity data.

[0038] In another possible implementation, the first termite matrix data is input into a trained pest prediction model for vector feature extraction, and then the method further includes:

[0039] determining a root mean square error of termite population based on the actual termite population and the termite population;

[0040] Setting parameters in an epoch training model in the pest prediction model according to the root mean square error of the termite quantity, and reversely iterating the set epoch training model to obtain a validation set for each round of the pest prediction model;

[0041] The validation set is evaluated and calculated to generate a loss value and an evaluation index for the validation set.

[0042] Through the above technical solution, after extracting the vector features of the first termite matrix data, the root mean square error of the termite quantity is determined based on the actual termite quantity and the termite quantity, and then the parameters of the epoch training model in the pest prediction model are set according to the root mean square error of the termite quantity. Then, the epoch training model is reversely iterated to obtain the validation set of each round in the pest prediction model, and then the validation set is calculated and evaluated to generate the loss value and evaluation index of the validation set. The staff monitors the current vector feature extraction effect through the intuitive data of the loss value and evaluation index.

[0043] In a second aspect, the present application provides a termite prevention trend prediction device, which adopts the following technical solution:

[0044] A termite prevention trend prediction device, comprising:

[0045] An information acquisition module is used to acquire termite damage information within a preset time period in the past, wherein the termite damage information is information about termites that have occurred in different housing building types corresponding to different areas;

[0046] a first matrix generation module, configured to analyze the termite hazard information, determine the number of building combinations of different housing building types in the termite hazard information and the length of a time series corresponding to each building combination in the number of building combinations, and perform unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain first termite matrix data;

[0047] A second matrix generation module is configured to input the first termite matrix data into a trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combine the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data;

[0048] The quantity prediction module is used to process the data contained in the second termite matrix data to obtain termite damage data, and input the obtained termite damage data into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations within a future preset time period.

[0049] By adopting the above technical solution, when predicting the trend of termite prevention, the termite number information of different housing building types corresponding to different areas is collected to obtain termite damage information, and then the termite damage information is analyzed to obtain the number of building combinations of different housing building types and the time series length in the termite damage information, wherein the time series length and the number of building combinations are in a one-to-one correspondence. Then, the termite damage information is unsupervisedly sorted according to the time series length and the number of building combinations to obtain the first termite matrix data, and then the first termite matrix data is input into the trained pest prediction model for further analysis. Row vector features are extracted to obtain the number of termite feature dimensions, and the obtained number of termite feature dimensions is combined with the first termite matrix data for data processing to generate second termite matrix data. The data contained in the second termite matrix data is then processed one by one to obtain termite damage data. The termite damage data is input into the preset algorithm model for data extrapolation to generate the termite quantity of each building combination in the future preset time period. The staff sets the past preset time period and the future preset time period according to actual needs to obtain the termite quantity of different buildings in the future preset time period, thereby achieving the effect of improving the accuracy of termite prediction.

[0050] In one possible implementation, when analyzing the termite hazard information to determine the number of building combinations of different housing building types in the termite hazard information and the length of a time series corresponding to each building combination in the number of building combinations, the first matrix generation module is specifically configured to:

[0051] determining at least one set of termite treatment data based on the termite damage information;

[0052] Obtaining labels for the at least one set of termite treatment data respectively to obtain building data and termite treatment time data in each set of termite treatment data;

[0053] determining whether the termite treatment data has been processed according to the termite treatment time data; if not, performing data collapse on the termite treatment data; and if the processing has been completed, correspondingly binding the building data with the termite treatment time data to obtain termite binding data;

[0054] The termite binding data is screened for building combination types to obtain the number of building combinations of different house building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations.

[0055] In another possible implementation, when the second matrix generation module inputs the first termite matrix data into a trained pest prediction model to extract vector features and obtain the number of termite feature dimensions, it is specifically configured to:

[0056] determining, based on the first termite matrix data, an event name, an event time, and an event area of each damage event in the termite damage information;

[0057] Inputting the event name, the event time, and the event area into the pest prediction model for vector extraction, respectively, to obtain a text feature vector corresponding to the event name, a time feature vector corresponding to the event time, and a region feature vector corresponding to the event area;

[0058] Quantity statistics are performed on the text feature vector, the time feature vector, and the region feature vector to obtain the number of termite feature dimensions.

[0059] In another possible implementation, when the second matrix generation module combines the obtained termite feature dimension quantity with the first termite matrix data to generate the second termite matrix data, the second matrix generation module is specifically configured to: integrate the termite feature dimension quantity with the first termite matrix data to generate termite dimension matrix data;

[0060] Performing basic data distribution exploration on the termite dimension matrix data to obtain a relative periodicity pattern of occurrence of damage events in the termite damage information, and determining a time period length based on the relative periodicity pattern;

[0061] Performing supervised time series data sorting on the termite dimension matrix data based on the time period length to obtain termite prediction matrix data;

[0062] The termite quantity trend within a future preset time period is predicted based on the termite prediction matrix data to generate second termite matrix data.

[0063] In another possible implementation, when the quantity prediction module processes the data contained in the second termite matrix data to obtain the termite damage data, it is specifically configured to:

[0064] Calculating a normal distribution mean and a normal distribution variance of data included in the second termite matrix data, and determining a 3σ range of the second termite matrix data based on the normal distribution mean and the normal distribution variance;

[0065] determining whether the data is outside the 3σ range; if the data is outside the 3σ range, determining a first matrix sequence of the second termite matrix data in which the data is located, calculating a sequence average based on the first matrix sequence, replacing the data with the sequence average to obtain a replaced second matrix sequence, and performing missing value processing on the second matrix sequence;

[0066] A sequence normalization process is performed on the second matrix sequence in the second termite matrix data to obtain termite damage data.

[0067] In another possible implementation, the device further includes: a quantity acquisition module and a quantity calibration module, wherein:

[0068] The quantity acquisition module is used to obtain the actual termite quantity of each building combination in the number of building combinations within a future preset time period;

[0069] The quantity calibration module is configured to perform a denormalization process on the termite quantity based on the actual termite quantity, so as to restore the termite quantity to the actual termite quantity.

[0070] In another possible implementation, the device further includes: an error determination module, a reverse iteration module, and a calculation and evaluation module, wherein:

[0071] The error determination module is configured to determine a root mean square error of termite population based on the actual termite population and the termite population;

[0072] The reverse iteration module is used to set the parameters of the epoch training model in the pest prediction model according to the root mean square error of the termite quantity, and reversely iterate the set epoch training model to obtain a verification set for each round of the pest prediction model;

[0073] The computational evaluation module is used to perform computational evaluation on the validation set and generate a loss value and evaluation index for the validation set.

[0074] In a third aspect, the present application provides an electronic device, which adopts the following technical solution:

[0075] An electronic device, comprising:

[0076] at least one processor;

[0077] Memory;

[0078] At least one application, wherein the at least one application is stored in the memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the above-mentioned termite prevention trend prediction method.

[0079] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0080] A computer-readable storage medium includes: a computer program stored therein that can be loaded by a processor and execute the termite prevention trend prediction method.

[0081] In summary, this application has the following beneficial technical effects:

[0082] 1. When predicting the trend of termite prevention, the termite number information of different housing building types in different areas is collected to obtain termite damage information, and then the termite damage information is analyzed to obtain the number of building combinations of different housing building types and the length of the time series. The time series length and the number of building combinations are in a one-to-one correspondence. Then, the termite damage information is sorted out in an unsupervised time series according to the time series length and the number of building combinations to obtain the first termite matrix data. The first termite matrix data is then input into the trained pest prediction model for vector feature extraction. The feature extraction is performed to obtain the number of termite feature dimensions, and the obtained number of termite feature dimensions is combined with the first termite matrix data to generate the second termite matrix data. The data contained in the second termite matrix data is then processed one by one to obtain termite damage data. The termite damage data is input into the preset algorithm model for data inference to generate the number of termites for each building combination in the future preset time period. The staff sets the past preset time period and the future preset time period according to actual needs to obtain the termite quantity of different buildings in the future preset time period, thereby achieving the effect of improving the accuracy of termite prediction;

[0083] 2. When obtaining termite damage data, the 3σ range of the second termite matrix data is determined by calculating the normal distribution mean and normal distribution variance of the data contained in the second termite matrix data, and determining whether the current data is outside the 3σ range. If so, the data is eliminated, and the sequence average value is added to the position of the first matrix sequence where the data is located to obtain the second matrix sequence. The second matrix sequence is then processed for missing values to ensure the integrity of the matrix sequence. The second matrix sequence in the second termite matrix data is then normalized to obtain termite damage data for subsequent data processing of the termite damage data. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 This is a flow chart of a termite prevention trend prediction method according to an embodiment of the present application;

[0085] Figure 2 This is a block diagram of a termite prevention trend prediction method according to an embodiment of the present application;

[0086] Figure 3 It is a schematic diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0087] The following is combined with Figure 1-3 This application is described in further detail.

[0088] After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed, but as long as they are within the scope of the claims of this application, they are protected by patent law.

[0089] To make the purpose, technical solutions, and advantages of the embodiments of this application more clear, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0090] In addition, the term "and / or" as used herein simply describes an association between related objects, indicating that three possible relationships exist. For example, "a termite prevention trend prediction method, apparatus, electronic device, and medium and / or B" can represent: the existence of a termite prevention trend prediction method, apparatus, electronic device, and medium alone; the existence of a termite prevention trend prediction method, apparatus, electronic device, and medium and B together; and the existence of B alone. Furthermore, the character " / " as used herein, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0091] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0092] The embodiment of the present application provides a termite prevention trend prediction method, which is executed by an electronic device, which can be a server or a terminal device, wherein the server can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smart phone, a tablet computer, a laptop computer, a desktop computer, etc., but is not limited to this. The terminal device and the server can be directly or indirectly connected through wired or wireless communication, and the embodiment of the present application does not limit this. Figure 1 As shown, the method includes:

[0093] Step S10: Acquire termite damage information within a preset time period in the past.

[0094] The termite damage information is the termite information that has occurred in different housing building types corresponding to different areas.

[0095] For the embodiment of the present application, the preset time period in the past was input by the staff through a designated terminal device, and the designated terminal device includes a tablet, a mobile phone, and a computer.

[0096] Specifically, the staff enters a past preset time period (for example: January 1, 2020 - October 3, 2021) in the designated terminal device, and then the designated terminal device sends the past preset time period to the electronic device for processing. After the electronic device receives the past preset time period, it obtains information on the number of termites that have occurred in different types of house buildings corresponding to different areas in the past preset time period.

[0097] Specifically, one possible implementation method for obtaining termite damage information is a big data acquisition method, which uses big data technology to obtain all termite damage information, and then filters all termite damage information according to a preset time period in the past to obtain the termite damage information within the preset time period in the past.

[0098] Step S11: Analyze the termite hazard information, determine the number of building combinations of different house building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations, and perform unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain the first termite matrix data.

[0099] Specifically, termite damage information refers to information about termites that occurred in different building types over a preset time period. This information includes the specific location and start time of the termite damage. For example, a termite damage incident occurred on *th day of *th month of **** in a brick-and-wood building in ** village, ** district, ** city, ** province. The estimated number of termites in this incident was 18,000, and the incident has been resolved. After obtaining the termite damage information, the electronic device extracts the specific location from the information to determine the building combinations with different termite damage information (one of the building combinations in the number of building combinations is a brick-and-wood building in ** village, ** district, ** city, ** province). The number of building combinations is then counted to obtain the number of building combinations.

[0100] Specifically, a time series is a set of random variables ordered by time. It is typically the result of observing an underlying process at a given sampling rate over equally spaced time periods. Time series data essentially reflects the changing trends of one or more random variables over time. The core of time series forecasting methods is to discover these patterns from the data and use them to estimate future data.

[0101] In the embodiment of the present application, the time series length represents the length of the termite damage information that changes over time.

[0102] The unsupervised time series data of termite damage information was sorted according to the length of the time series and the number of building combinations, and the following first termite matrix data was obtained:

[0103]

[0104] Where m is the number of building combinations and n is the length of the time series.

[0105] Step S12: Input the first termite matrix data into the trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combine the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data.

[0106] Specifically, when inputting the first termite matrix data into the pest prediction model for vector feature extraction, it is necessary to obtain matrix data samples in advance. The matrix data samples include the first termite matrix data formed by all past termite damage information and the vector features in the first termite matrix data. Then, a pest prediction model is created, and the pest prediction model is trained based on the matrix data samples to obtain a trained pest prediction model.

[0107] Specifically, the pre-research model is a pre-trained neural network model. Neural networks (NNs) are complex network systems formed by a large number of simple processing units (called neurons) that are widely interconnected. They reflect many basic characteristics of human brain function and are highly complex nonlinear dynamic learning systems. Neural networks have large-scale parallel, distributed storage and processing, self-organization, self-adaptation, and self-learning capabilities, making them particularly suitable for handling imprecise and ambiguous information processing problems that require simultaneous consideration of many factors and conditions. The development of neural networks is related to neuroscience, mathematical science, cognitive science, computer science, artificial intelligence, information science, cybernetics, robotics, microelectronics, psychology, optical computing, molecular biology, and other fields.

[0108] Specifically, the first pest matrix data is input into the pre-research model for vector feature extraction, and the extracted vector features are counted to obtain the number of feature dimensions, where the vector features include event text vector features, time vector features, and building combination vector features in termite damage information. Then, the number of feature dimensions is combined with the first termite matrix data to obtain the second termite matrix data.

[0109] Step S13, processing the data contained in the second termite matrix data to obtain termite damage data, and inputting the obtained termite damage data into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations in a future preset time period.

[0110] For the embodiments of the present application, a bidirectional LSTM model is used as an example to illustrate the preset algorithm model, including but not limited to the bidirectional LSTM model.

[0111] Specifically, the preset algorithm model is constructed, and the main body of the model adopts bidirectional LSTM as the trend prediction model. LSTM mainly consists of a forget gate, an input gate, and an output gate;

[0112] Forget gate:f t =σ(W f [h t-1 ,x t ]+b f );

[0113] Input Gate:

[0114] After filtering the information through the forget gate and input gate, the historical memory and the memory content of the current stage are merged, and the generated value is:

[0115]

[0116] Output Gate: After the LSTM described above, a layer of LSTM network is connected in the reverse direction. Through this process, a BI-LSTM layer can be obtained. Since several groups of building combinations are trained together, a joint learning layer of architectural spatial features is added, and the associated vector matrix size is initialized to M*V*K. The output vector of the last LSTM layer is taken, transposed and multiplied by the associated vector parameter matrix, and finally connected to the regression loss function to complete the construction of the preset algorithm model.

[0117] The embodiment of the present application provides a method for predicting the trend of termite prevention. When predicting the trend of termite prevention, termite damage information is obtained by collecting information on the number of termites that have occurred in different housing building types corresponding to different areas. The termite damage information is then analyzed to obtain the number of building combinations of different housing building types and the length of the time series in the termite damage information, wherein the time series length and the number of building combinations are in a one-to-one correspondence. The termite damage information is then unsupervisedly sorted into time series data according to the time series length and the number of building combinations to obtain first termite matrix data. The first termite matrix data is then input into a trained insect matrix. The damage prediction model performs vector feature extraction to obtain the number of termite feature dimensions, and combines the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data. The data contained in the second termite matrix data is then processed one by one to obtain termite damage data. The termite damage data is input into the preset algorithm model for data extrapolation to generate the number of termites for each building combination in the future preset time period. The staff sets the past preset time period and the future preset time period according to actual needs to obtain the termite quantity of different buildings in the future preset time period, thereby achieving the effect of improving the accuracy of termite prediction.

[0118] In a possible implementation of an embodiment of the present application, step S11 specifically includes step S111 (not shown in the figure), step S112 (not shown in the figure), step S113 (not shown in the figure) and step S114 (not shown in the figure), wherein step S111 determines at least one set of termite treatment data based on termite hazard information.

[0119] Specifically, the termite treatment data contained in the termite damage information is reported by enthusiastic citizens and building users to the reporting personnel, and then the information reporters record and register the reported information. The termite treatment data includes the location of the termite damage, the details of the termite damage, the degree of termite damage, the time when the termite damage occurred, and the time when the termite damage was resolved.

[0120] Step S112 , performing label acquisition on at least one set of termite treatment data to obtain building data and termite treatment time data in each set of termite treatment data.

[0121] Specifically, label acquisition is performed based on data labels in at least one set of termite treatment data to obtain specified label content, namely, by obtaining the location of termite damage, termite damage details, termite damage degree, termite damage occurrence time, and termite damage resolution time.

[0122] Step S113, determining whether the termite treatment data has been completed based on the termite treatment time data. If not, the termite treatment data is collapsed. If the treatment is completed, the building data is correspondingly bound to the termite treatment time data to obtain termite binding data.

[0123] Specifically, by obtaining the information corresponding to the termite damage resolution time tag, it is determined whether the current termite treatment data has been processed. If the termite damage resolution time tag does not have corresponding time information, it means that the termite treatment data is still in the process of processing, so the termite treatment data is collapsed. If the thermal termite damage resolution time tag has corresponding time information, the building data is bound to the termite treatment time data.

[0124] Step S114 , filtering the termite binding data by building combination type to obtain the number of building combinations of different house building types in the termite damage information and the time series length corresponding to each building combination in the number of building combinations.

[0125] In a possible implementation of the embodiment of the present application, step S12 specifically includes step S121 (not shown in the figure), step S122 (not shown in the figure), and step S123 (not shown in the figure), wherein:

[0126] Step S121 : determining the event name, event time, and event area of each damage event in the termite damage information based on the first termite matrix data.

[0127] Specifically, the event name corresponding to the hazardous event is determined according to x in the first termite matrix data, the event time corresponding to the hazardous event is determined according to n in the first termite matrix data, and the event area corresponding to the hazardous event is determined according to m in the first termite matrix data.

[0128] In step S122 , the event name, event time, and event area are input into the pest prediction model for vector extraction to obtain a text feature vector corresponding to the event name, a time feature vector corresponding to the event time, and a region feature vector corresponding to the event area.

[0129] Step S123 , performing quantitative statistics on the text feature vectors, the time feature vectors, and the region feature vectors to obtain the number of termite feature dimensions.

[0130] Specifically, the total number of eigenvectors in a matrix is calculated as: number = n - rank of the eigenvector matrix, or number = nr(λE-A), where n is the rank. Not every matrix can be diagonalized. If a matrix has distinct eigenvalues, it can definitely be diagonalized. The projections (i.e., coordinates) of the eigenvectors onto the basis vectors assume an h-dimensional vector space. Therefore, they can be directly represented as coordinate vectors. Using basis vectors, linear transformations can also be represented using a simple matrix multiplication.

[0131] In a possible implementation of an embodiment of the present application, step S12 specifically includes step S124 (not shown in the figure), step S125 (not shown in the figure), step S126 (not shown in the figure) and step S127 (not shown in the figure), wherein, in step S124, the number of termite feature dimensions is integrated with the first termite matrix data to generate termite dimension matrix data.

[0132] Specifically, the number of termite feature dimensions is used as a dimension to be integrated with the first termite matrix data. The embodiment of the present application uses pytorch technology for illustration, including but not limited to an implementable method of pytorch technology.

[0133] The number of termite feature dimensions is added to the first termite matrix data in a dimension-wise manner through the instruction “out.unsqueeze(-1)” in pytorch to achieve dimension integration.

[0134] Specifically, PyTorch is an open-source Python machine learning library based on Torch, used for applications such as natural language processing. Developed primarily by Facebook's AI team, it not only offers powerful GPU acceleration but also supports dynamic neural networks, a feature currently lacking in many mainstream frameworks like TensorFlow. PyTorch offers two advanced features: 1. Powerful GPU-accelerated tensor computations (like Numpy); 2. Deep neural networks with automatic differentiation systems. In addition to Facebook, PyTorch has also been adopted by Twitter, GMU, and Salesforce.

[0135] After integrating the termite feature dimension quantity with the first termite matrix data, the following termite dimension matrix data is obtained:

[0136]

[0137] , where v represents the number of termite feature dimensions.

[0138] Step S125 , performing basic data distribution exploration on the termite dimension matrix data, obtaining the relative periodicity of the occurrence of the damage events in the termite damage information, and determining the length of the time period based on the relative periodicity.

[0139] Specifically, the termite dimension matrix data is imported into an Excel spreadsheet, and the Python integrated Jupyter environment is configured to perform basic data distribution exploration on the n time series in the ant dimension matrix data. The main purpose is to find the relative periodicity of the corresponding sequence, and then determine the length of the time period based on the relative periodicity.

[0140] Step S126 , performing supervised time series data sorting on the termite dimension matrix data based on the time period length to obtain termite prediction matrix data.

[0141] Specifically, t is used to replace the time periodicity length, and n in the termite dimension matrix data is replaced by t to obtain the new termite prediction matrix data:

[0142]

[0143] Step S127: predicting the termite quantity trend within a preset time period in the future based on the termite prediction matrix data to generate second termite matrix data.

[0144] Specifically, assuming that the preset future time period is k, that is, the moving step length is k-step prediction, the second termite matrix data is obtained:

[0145]

[0146] In a possible implementation of the embodiment of the present application, step S13 specifically includes step S131 (not shown in the figure), step S132 (not shown in the figure), and step S133 (not shown in the figure), wherein:

[0147] Step S131 : Calculate the normal distribution mean and normal distribution variance of the data contained in the second matrix data, and determine the 3σ range of the second matrix data based on the normal distribution mean and normal distribution variance.

[0148] Specifically, step S132 determines whether the data is outside the 3σ range. If the data is outside the 3σ range, the first matrix sequence of the second termite matrix data where the data is located is determined, the sequence average is calculated according to the first matrix sequence, the data is replaced with the sequence average to obtain the replaced second matrix sequence, and the second matrix sequence is processed for missing values.

[0149] Specifically, the 3σ range is based on the interference or noise of singular data caused by repeated measurements of equal precision under normal distribution, which makes it difficult for the normal distribution to be satisfied. If the absolute value of the residual error of a measurement value in a set of measurement data is greater than 3σ, then the measurement value is a bad value and should be eliminated. The error equal to ±3σ is usually regarded as the limit error. For a random error with normal distribution, the probability of falling outside ±3σ is only 0.27%. The probability of it occurring in a finite number of measurements is very small, so the 3σ criterion exists. The 3σ criterion is the most commonly used and simplest criterion for judging gross errors. It is generally applied to situations where the number of measurements is sufficiently large (n≥30) or when n>10 is used for rough judgment.

[0150] Specifically, missing values refer to data clustering, grouping, deletion or truncation caused by lack of information in the second matrix sequence. The processing of missing values is generally divided into deleting cases with missing values and interpolating missing values. The embodiment of the present application processes the second matrix sequence by deleting cases with missing values. There are mainly simple deletion method and weighting method for deleting cases with missing values. The simple deletion method is the most primitive method for processing missing values. It deletes cases with missing values. If the data missing problem can be achieved by simply deleting a small part of the sample, then this method is the most effective. When the type of missing value is non-completely random missing, the deviation can be reduced by weighting the complete data. After marking the cases with incomplete data, different weights are assigned to the complete data cases. The weights of the cases can be obtained through logistic or probit regression.

[0151] Step S133 : performing sequence normalization processing on the second matrix sequence in the second termite matrix data to obtain termite damage data.

[0152] Normalization comes in two forms: converting numbers to decimals between 0 and 1, and converting dimensionless expressions to dimensional expressions. This approach is primarily designed to facilitate data processing, as it maps data to a range of 0 to 1, making it faster and more convenient.

[0153] The specific processing normalization method is:

[0154] In a possible implementation of the embodiment of the present application, step S13 further includes: step Sa (not shown in the figure) and step Sb (not shown in the figure), wherein:

[0155] Step Sa, obtaining the actual number of termites in each building combination in the number of building combinations within a future preset time period.

[0156] Specifically, the actual number of termites in a future preset time period is obtained. For example, if the future preset time period is one month, then after one month, the actual number of termites in each building combination is obtained.

[0157] Step Sb: performing denormalization processing on the termite quantity based on the actual termite quantity to restore the termite quantity to the actual termite quantity.

[0158] Specifically, the termite quantity is denormalized so that the termite quantity is restored to the data before the normalization process, and then the actual termite quantity is overwritten on the termite quantity.

[0159] In a possible implementation of the embodiment of the present application, step S12 further includes: step Sd (not shown in the figure), step Se (not shown in the figure), and step Sf (not shown in the figure), wherein:

[0160] Step Sd: determining a root mean square error of termite population based on the actual termite population and the termite population.

[0161] Specifically, the root mean square error is the square root of the ratio of the square of the deviation between the observed value and the true value and the number of observations n. The root mean square error calculation formula is: Re = √[∑di^2 / n], where n is the number of measurements, and di is a set of actual termite counts and the deviation of termite counts.

[0162] In step Se, the parameters of the epoch training model in the pest prediction model are set according to the root mean square error of the termite quantity, and the epoch training model after setting is reversely iterated to obtain the verification set of each round in the pest prediction model.

[0163] Specifically, when a complete dataset passes through the neural network once and returns once, this process is called an epoch of model training. An epoch refers to the process in which all data is fed into the network to complete a single forward computation and backpropagation. Because an epoch is often too large for a computer to handle, it is divided into several smaller batches. During training, iterating all the data once is insufficient; multiple iterations are required for convergence. In actual training, all data is divided into several batches, and a portion of the data is fed in at a time. Gradient descent itself is an iterative process. The parameters of the epoch-trained model are set based on the root mean squared error of the termite population. This yields a validation set for each round of the pest prediction model, namely the termite population validation set.

[0164] In step Sf, the validation set is evaluated and the loss value and evaluation index of the validation set are generated.

[0165] Specifically, the computational evaluation of the validation set includes the following steps:

[0166] Calculate the percentage of the validation set and the training set of the pest prediction model to obtain the loss value;

[0167] Compare the loss value with the standard loss value table to obtain the evaluation index.

[0168] For example: the current loss value is 50%, and the 50% corresponding to the standard loss value table is a level 2 indicator.

[0169] The above embodiment introduces a termite prevention trend prediction method from the perspective of method flow. The following embodiment introduces a termite prevention trend prediction device from the perspective of a virtual module or virtual unit. Please refer to the following embodiment for details.

[0170] The present application embodiment provides a termite prevention trend prediction device, such as Figure 2 As shown, the termite prevention trend prediction device 20 may specifically include: an information acquisition module 21, a first matrix generation module 22, a second matrix generation module 23 and a quantity prediction module 24, wherein:

[0171] The information acquisition module 21 is used to obtain termite damage information within a preset time period in the past. The termite damage information is the number of termites that have occurred in different building types corresponding to different areas;

[0172] The first matrix generation module 22 is configured to analyze the termite hazard information, determine the number of building combinations of different housing building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations, and perform unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain first termite matrix data;

[0173] The second matrix generation module 23 is used to input the first termite matrix data into the trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combine the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data;

[0174] The quantity prediction module 24 is used to process the data contained in the second termite matrix data to obtain termite damage data, and input the obtained termite damage data into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the building combination quantity within a future preset time period.

[0175] In one possible implementation of the embodiment of the present application, when the first matrix generation module 22 analyzes termite hazard information and determines the number of building combinations of different housing building types in the termite hazard information and the length of the time series corresponding to each building combination in the number of building combinations, it is specifically configured to:

[0176] determining at least one set of termite treatment data based on the termite damage information;

[0177] Obtain labels for at least one set of termite treatment data to obtain building data and termite treatment time data in each set of termite treatment data;

[0178] Determine whether the termite treatment data has been completed based on the termite treatment time data. If the treatment has not been completed, decompose the termite treatment data. If the treatment has been completed, bind the building data to the termite treatment time data to obtain termite binding data.

[0179] The building combination types are screened for termite binding data to obtain the number of building combinations of different house building types in termite damage information and the time series length corresponding to each building combination in the number of building combinations.

[0180] In another possible implementation of the embodiment of the present application, when the second matrix generation module 23 inputs the first termite matrix data into the trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, it is specifically configured to: determine the event name, event time, and event area of each damage event in the termite damage information based on the first termite matrix data;

[0181] The event name, event time, and event area are input into the pest prediction model for vector extraction, and a text feature vector corresponding to the event name, a time feature vector corresponding to the event time, and a regional feature vector corresponding to the event area are obtained.

[0182] The text feature vectors, time feature vectors and regional feature vectors are counted to obtain the number of termite feature dimensions.

[0183] In another possible implementation of the embodiment of the present application, when the second matrix generating module 23 combines the obtained termite feature dimension quantity with the first termite matrix data to generate the second termite matrix data, it is specifically configured to: integrate the termite feature dimension quantity with the first termite matrix data to generate the termite dimension matrix data;

[0184] Conduct basic data distribution exploration on termite dimension matrix data to obtain the relative periodicity of termite damage events in the information, and determine the length of the time period based on the relative periodicity.

[0185] Based on the length of the time period, supervised time series data sorting is performed on the termite dimension matrix data to obtain the termite prediction matrix data;

[0186] The termite quantity trend within a future preset time period is predicted based on the termite prediction matrix data to generate second termite matrix data.

[0187] In another possible implementation of the embodiment of the present application, when the population prediction module 24 processes the data contained in the second termite matrix data to obtain the termite damage data, it is specifically configured to:

[0188] Calculating a normal distribution mean and a normal distribution variance of the data contained in the second matrix data, and determining a 3σ range of the second matrix data based on the normal distribution mean and the normal distribution variance;

[0189] Determine whether the data is outside the 3σ range. If the data is outside the 3σ range, determine the first matrix sequence of the second termite matrix data where the data is located, calculate the sequence average value based on the first matrix sequence, replace the data with the sequence average value to obtain the replaced second matrix sequence, and perform missing value processing on the second matrix sequence;

[0190] A sequence normalization process is performed on the second matrix sequence in the second termite matrix data to obtain termite damage data.

[0191] In another possible implementation of the embodiment of the present application, the device 20 further includes: a quantity acquisition module and a quantity calibration module, wherein:

[0192] The quantity acquisition module is used to obtain the actual termite quantity of each building combination in the future preset time period; the quantity calibration module is used to perform denormalization processing on the termite quantity based on the actual termite quantity and restore the termite quantity to the actual termite quantity.

[0193] In another possible implementation of the embodiment of the present application, the apparatus 20 further includes: an error determination module, a reverse iteration module, and a calculation evaluation module, wherein:

[0194] an error determination module for determining a root mean square error of termite population based on the actual termite population and the termite population;

[0195] The reverse iteration module is used to set the parameters in the epoch training model of the pest prediction model according to the root mean square error of the termite quantity, and reversely iterate the epoch training model after setting to obtain the validation set of each round in the pest prediction model; the calculation and evaluation module is used to calculate and evaluate the validation set and generate the loss value and evaluation index of the validation set.

[0196] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0197] The embodiment of the present application also introduces an electronic device from the perspective of a physical device, such as Figure 3 As shown, Figure 3The electronic device 300 shown includes, in addition to conventional configuration devices, a processor 301 and a memory 303. The processor 301 and the memory 303 are connected, for example, via a bus 302. Optionally, the electronic device 300 may also include a transceiver 304. It should be noted that in actual applications, the number of transceivers 304 is not limited to one, and the structure of the electronic device 300 does not constitute a limitation on the embodiments of the present application.

[0198] The processor 301 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor 301 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, and the like.

[0199] The bus 302 may include a path for transmitting information between the above components. The bus 302 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 302 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0200] The memory 303 may be a ROM (Read Only Memory) or other type of static storage device that can store static information and instructions, a RAM (Random Access Memory) or other type of dynamic storage device that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto.

[0201] The memory 303 is used to store application code for executing the solution of the present application, and the execution is controlled by the processor 301. The processor 301 is used to execute the application code stored in the memory 303 to implement the content shown in the above method embodiment.

[0202] The electronic devices include, but are not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Servers and the like are also possible. Figure 3 The electronic device shown is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0203] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0204] The above are only some of the implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A termite prevention trend prediction method, characterized in that: include Obtain termite damage information within a preset time period in the past, wherein the termite damage information is termite information that has occurred in different housing building types corresponding to different areas; Analyzing the termite hazard information, determining the number of building combinations of different housing building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations, and performing unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain first termite matrix data; Inputting the first termite matrix data into a trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combining the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data; The step of combining the obtained termite feature dimension quantity with the first termite matrix data to generate second termite matrix data includes: Integrating the termite feature dimension quantity with the first termite matrix data to generate termite dimension matrix data; Performing basic data distribution exploration on the termite dimension matrix data to obtain a relative periodicity pattern of occurrence of damage events in the termite damage information, and determining a time period length based on the relative periodicity pattern; Performing supervised time series data sorting on the termite dimension matrix data based on the time period length to obtain termite prediction matrix data; Predicting termite quantity trends within a future preset time period based on the termite prediction matrix data to generate second termite matrix data; Data contained in the second termite matrix data is processed to obtain termite damage data, and the obtained termite damage data is input into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations in a future preset time period.

2. The method according to claim 1, characterized in that Analyzing the termite hazard information to determine the number of building combinations of different housing building types in the termite hazard information and the length of a time series corresponding to each building combination in the number of building combinations includes: determining at least one set of termite treatment data based on the termite damage information; Obtaining labels for the at least one set of termite treatment data respectively to obtain building data and termite treatment time data in each set of termite treatment data; determining whether the termite treatment data has been processed according to the termite treatment time data; if not, performing data collapse on the termite treatment data; and if the processing has been completed, correspondingly binding the building data with the termite treatment time data to obtain termite binding data; The termite binding data is screened for building combination types to obtain the number of building combinations of different house building types in the termite hazard information and the time series length corresponding to each building combination in the number of building combinations.

3. The method according to claim 1, characterized in that The first termite matrix data is input into the trained pest prediction model to extract vector features to obtain the number of termite feature dimensions, including: determining, based on the first termite matrix data, an event name, an event time, and an event area of each damage event in the termite damage information; Inputting the event name, the event time, and the event area into the pest prediction model for vector extraction, respectively, to obtain a text feature vector corresponding to the event name, a time feature vector corresponding to the event time, and a region feature vector corresponding to the event area; Quantity statistics are performed on the text feature vector, the time feature vector, and the region feature vector to obtain the number of termite feature dimensions.

4. The method according to claim 1, wherein The step of processing the data contained in the second termite matrix data to obtain termite damage data includes: Calculating a normal distribution mean and a normal distribution variance of data included in the second termite matrix data, and determining a 3σ range of the second termite matrix data based on the normal distribution mean and the normal distribution variance; determining whether the data is outside the 3σ range; if the data is outside the 3σ range, determining a first matrix sequence of the second termite matrix data in which the data is located, calculating a sequence average based on the first matrix sequence, replacing the data with the sequence average to obtain a replaced second matrix sequence, and performing missing value processing on the second matrix sequence; A sequence normalization process is performed on the second matrix sequence in the second termite matrix data to obtain termite damage data.

5. The method according to claim 1, wherein The obtained termite damage data is input into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations within a future preset time period, and then further includes: Obtaining the actual number of termites in each building combination in the number of building combinations within a future preset time period; The termite quantity is denormalized based on the actual termite quantity to restore the termite quantity to the actual termite quantity.

6. The method according to claim 5, characterized in that The step of inputting the first termite matrix data into a trained pest prediction model for vector feature extraction further includes: determining a root mean square error of termite population based on the actual termite population and the termite population; Setting parameters in an epoch training model in the pest prediction model according to the root mean square error of the termite quantity, and reversely iterating the set epoch training model to obtain a validation set for each round of the pest prediction model; The validation set is evaluated and calculated to generate a loss value and an evaluation index for the validation set.

7. A termite prevention trend prediction device, characterized in that: include: An information acquisition module is used to acquire termite damage information within a preset time period in the past, wherein the termite damage information is information about termites that have occurred in different housing building types corresponding to different areas; a first matrix generation module, configured to analyze the termite hazard information, determine the number of building combinations of different housing building types in the termite hazard information and the length of a time series corresponding to each building combination in the number of building combinations, and perform unsupervised time series data sorting on the termite hazard information based on the time series length and the number of building combinations to obtain first termite matrix data; A second matrix generation module is configured to input the first termite matrix data into a trained pest prediction model to perform vector feature extraction to obtain the number of termite feature dimensions, and combine the obtained number of termite feature dimensions with the first termite matrix data to generate second termite matrix data; When the second matrix generation module combines the obtained termite feature dimension quantity with the first termite matrix data to generate the second termite matrix data, it is specifically used to: Integrating the termite feature dimension quantity with the first termite matrix data to generate termite dimension matrix data; Performing basic data distribution exploration on the termite dimension matrix data to obtain a relative periodicity pattern of occurrence of damage events in the termite damage information, and determining a time period length based on the relative periodicity pattern; Performing supervised time series data sorting on the termite dimension matrix data based on the time period length to obtain termite prediction matrix data; Predicting termite quantity trends within a future preset time period based on the termite prediction matrix data to generate second termite matrix data; The quantity prediction module is used to process the data contained in the second termite matrix data to obtain termite damage data, and input the obtained termite damage data into a preset algorithm model for data calculation to generate the termite quantity of each building combination in the number of building combinations within a preset time period in the future.

8. An electronic device, characterized in that: The electronic device includes: at least one processor; Memory; At least one application, wherein the at least one application is stored in a memory and configured to be executed by at least one processor, and the at least one application is configured to: execute the termite prevention trend prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed in a computer, the computer is caused to execute the termite prevention trend prediction method according to any one of claims 1 to 6.

Images