Sensitivity prediction model training method, sensitive information processing method and device

Abstract

The present disclosure provides a sensitivity prediction model training method, a sensitive information processing method and device, which can be applied to the fields of artificial intelligence and financial technology. The method comprises: obtaining a historical sensitive data set, the historical sensitive data set comprising an allergy data set and an environmental data set, the environmental data set comprising a plurality of types of atmospheric environmental parameters; quantitatively processing a plurality of allergy data in the allergy data set to obtain a plurality of target allergy data within a first numerical range; for any type of atmospheric environmental parameter, quantitatively processing the atmospheric environmental parameter to obtain a plurality of target environmental parameters within a second numerical range; inputting the plurality of target allergy data and the target environmental parameters into an initial sensitivity prediction model to output a predicted sensitivity; inputting the predicted sensitivity and a real sensitive data set corresponding to the historical sensitive data set into a loss function to obtain a model loss value; and adjusting model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

Description

Technical Field

[0001] This disclosure relates to the fields of artificial intelligence and financial technology, and in particular to a sensitivity prediction model training method, a sensitive information processing method, and an apparatus. Background Technology

[0002] Predicting environmental sensitivity based on environmental factors allows users to know the likelihood of experiencing allergies in the current environment, enabling them to take preventative measures in advance. One related technology involves detecting pollen concentration in the environment and analyzing this concentration using a sensitivity prediction model to determine the current environmental sensitivity.

[0003] In realizing the concept disclosed herein, the inventors discovered at least the following problems in the related technology: predicting sensitivity based solely on pollen concentration is too simplistic, resulting in low accuracy of the sensitivity prediction model. Summary of the Invention

[0004] In view of the above problems, this disclosure provides a method for training a sensitivity prediction model, a method for processing sensitive information, an apparatus, a device, a medium, and a program product.

[0005] According to a first aspect of this disclosure, a method for training a sensitivity prediction model is provided, comprising: acquiring a historical sensitivity dataset, wherein the historical sensitivity dataset includes an allergy dataset and an environmental dataset, wherein the allergy dataset includes a dataset corresponding to a target population, and the environmental dataset includes multiple types of atmospheric environmental parameters; quantizing multiple allergy data in the allergy dataset to obtain multiple target allergy data within a first numerical range; quantizing the atmospheric environmental parameters for any type to obtain multiple target environmental parameters within a second numerical range; inputting the multiple target allergy data and the target environmental parameters into an initial sensitivity prediction model and outputting a predicted sensitivity; inputting the predicted sensitivity and the real sensitivity dataset corresponding to the historical sensitivity dataset into a loss function to obtain a model loss value; and adjusting the model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

[0006] According to an embodiment of this disclosure, after obtaining the historical sensitive dataset, the method further includes: performing preprocessing operations on the data in the historical sensitive dataset to obtain a preprocessed dataset, and using the preprocessed dataset to train the initial sensitivity prediction model, wherein the preprocessing operations include outlier removal, missing value imputation, and noise minimization.

[0007] According to an embodiment of this disclosure, the first numerical range is greater than the second numerical range, and the input of multiple target allergy data and target environmental parameters into the initial sensitivity prediction model includes: vectorizing multiple target allergy data within the first numerical range and multiple target environmental parameters within the second numerical range to obtain a matrix of a preset dimension; and inputting the matrix of the preset dimension into the initial sensitivity prediction model.

[0008] According to an embodiment of this disclosure, the above-mentioned inputting the predicted sensitivity and the real sensitive dataset corresponding to the historical sensitive dataset into the loss function includes: quantizing the environmental dataset and the allergy dataset corresponding to the historical sensitive dataset in the real sensitive dataset to obtain a quantized dataset; and inputting the quantized dataset and the predicted sensitivity into the loss function.

[0009] According to an embodiment of this disclosure, before quantifying multiple allergy data in the allergy dataset to obtain multiple target allergy data within a first numerical range, the method further includes: removing the maximum and minimum values ​​in the allergy dataset and the environmental dataset respectively to obtain a removed dataset, and then quantifying the removed dataset.

[0010] A second aspect of this disclosure provides a method for processing sensitive information, comprising: acquiring a dataset to be tested and a historical dataset corresponding to the dataset to be tested, wherein the dataset to be tested includes environmental data to be tested and allergy data to be tested, and the historical dataset includes historical environmental data corresponding to the environmental data to be tested and historical allergy data corresponding to the allergy data to be tested; quantifying the historical dataset to obtain a target dataset; inputting the target dataset into the target sensitivity prediction model to output a sensitivity corresponding to the dataset to be tested; and visualizing the sensitivity.

[0011] According to embodiments of this disclosure, the sensitive information processing method further includes: determining multiple target sensitive information in a database based on preset keywords; determining a target topic for each of the target sensitive information using keyword extraction technology; determining the similarity between the target topic and the preset topic using a cosine similarity algorithm, wherein the preset topic is determined based on time information and an allergy information list, the allergy information list including allergens corresponding to each time information; sorting the multiple target sensitive information based on the similarity and a preset sorting rule to obtain a serial number for each of the target sensitive information; filtering the multiple target sensitive information based on the serial number using a preset filtering rule to obtain multiple filtered sensitive information; and visually displaying the information titles corresponding to the multiple filtered sensitive information so that the filtered sensitive information can be viewed based on the information titles.

[0012] According to embodiments of this disclosure, the sensitive information processing method further includes: responding to received location information sent from a target object, querying multiple target locations associated with the location information using a map plugin; and visually displaying the location names corresponding to the multiple target locations so that related services can be invoked based on the location names.

[0013] A third aspect of this disclosure provides a sensitivity prediction model training apparatus, comprising: a first acquisition module for acquiring a historical sensitivity dataset, wherein the historical sensitivity dataset includes an allergy dataset and an environmental dataset, wherein the allergy dataset includes a dataset corresponding to a target population, and the environmental dataset includes multiple types of atmospheric environmental parameters; a first quantization module for quantizing multiple allergy data in the allergy dataset to obtain multiple target allergy data within a first numerical range; a second quantization module for quantizing the atmospheric environmental parameters for any type of atmospheric environmental parameter to obtain multiple target environmental parameters within a second numerical range; a first prediction module for inputting the multiple target allergy data and the target environmental parameters into an initial sensitivity prediction model and outputting a predicted sensitivity; a loss determination module for inputting the predicted sensitivity and the real sensitivity dataset corresponding to the historical sensitivity dataset into a loss function to obtain a model loss value; and a model training module for adjusting the model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

[0014] A fourth aspect of this disclosure provides a sensitive information processing apparatus, comprising: a second acquisition module for acquiring a dataset to be tested and a historical dataset corresponding to the dataset to be tested, wherein the dataset to be tested includes environmental data to be tested and allergy data to be tested, and the historical dataset includes historical environmental data corresponding to the environmental data to be tested and historical allergy data corresponding to the allergy data to be tested; a third quantization module for quantizing the historical dataset to obtain a target dataset; a second prediction module for inputting the target dataset into a target sensitivity prediction model and outputting a sensitivity corresponding to the dataset to be tested; and a first display module for visually displaying the sensitivity.

[0015] A fifth aspect of this disclosure provides an electronic device comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors perform the method described above.

[0016] A sixth aspect of this disclosure also provides a computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the methods described above.

[0017] The seventh aspect of this disclosure also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0018] According to the sensitivity prediction model training method, sensitive information processing method, apparatus, equipment, medium, and program products provided in this disclosure, the initial sensitivity prediction model learns from the data by quantifying allergy data and atmospheric environmental parameters separately, effectively improving the model's learning efficiency. Furthermore, the initial sensitivity prediction model learns from both the allergy dataset and various types of atmospheric environmental parameters during training, at least partially solving the technical problem of limited data, thus making the trained target sensitivity prediction model more accurate in predicting sensitivity. Attached Figure Description

[0019] The foregoing contents, as well as other objects, features, and advantages of this disclosure, will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0020] Figure 1 This illustration schematically depicts an application scenario of a sensitivity prediction model training method or a sensitive information processing method according to embodiments of the present disclosure.

[0021] Figure 2 A flowchart illustrating a sensitivity prediction model training method according to an embodiment of the present disclosure is shown schematically.

[0022] Figure 3 A flowchart illustrating a sensitive information processing method according to an embodiment of the present disclosure is shown schematically.

[0023] Figure 4 A schematic diagram illustrating the structure of a sensitivity prediction model according to an embodiment of the present disclosure is shown.

[0024] Figure 5 A schematic diagram illustrating the results of sensitivity assessment according to embodiments of the present disclosure is shown.

[0025] Figure 6 This schematically illustrates a result display diagram of an information header according to an embodiment of the present disclosure;

[0026] Figure 7 This schematic diagram illustrates the result of displaying the location names according to an embodiment of the present disclosure;

[0027] Figure 8 This schematic diagram illustrates the structure of a sensitivity prediction model training apparatus according to an embodiment of the present disclosure;

[0028] Figure 9 This schematic diagram illustrates a structural block diagram of a sensitive information processing apparatus according to an embodiment of the present disclosure;

[0029] Figure 10 A block diagram of an electronic device suitable for implementing a sensitivity prediction model training method or a sensitive information processing method according to embodiments of the present disclosure is shown schematically. Detailed Implementation

[0030] The embodiments of the present disclosure will now be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the disclosure. In the following detailed description, numerous specific details are set forth to provide a thorough understanding of the embodiments of the present disclosure for ease of explanation. However, it will be apparent that one or more embodiments may be practiced without these specific details. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concepts of the present disclosure.

[0031] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit this disclosure. The terms “comprising,” “including,” etc., as used herein indicate the presence of the stated features, steps, operations, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, or components.

[0032] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art, unless otherwise defined. It should be noted that the terms used herein are to be interpreted in a manner consistent with the context of this specification, and not in an idealized or overly rigid way.

[0033] When using expressions such as "at least one of A, B, and C", they should generally be interpreted in accordance with the meaning that is commonly understood by a person skilled in the art (e.g., "a system having at least one of A, B, and C" should include, but is not limited to, a system having A alone, a system having B alone, a system having C alone, a system having A and B, a system having A and C, a system having B and C, and / or a system having A, B, and C, etc.).

[0034] In the technical solutions disclosed herein, the collection, storage, use, processing, transmission, provision, disclosure, and application of data (including but not limited to user personal information) comply with the provisions of relevant laws and regulations, necessary confidentiality measures have been taken, and they do not violate public order and good morals.

[0035] Common allergens include pollen allergy, dust mite allergy, and ultraviolet radiation allergy. They are also closely related to air quality indices such as season, temperature, humidity, air pollution levels, exhaust concentration, and animal dander. In related technologies, sensitivity prediction methods are often based on pollen concentration levels, categorizing sensitivity into five levels: "extremely unlikely / unlikely / relatively likely / likely / extremely likely."

[0036] In the process of sensitivity prediction, the accuracy was low due to the limited data used. In addition to pollen concentration, factors such as weather phenomena, temperature, humidity, wind speed, PM2.5, and vehicle exhaust should also be considered. Furthermore, the system only provides pollen concentration and allergy susceptibility levels without integrating relevant supplementary information, resulting in limited information for users and low user engagement.

[0037] In view of the above, embodiments of this disclosure provide a sensitivity prediction model training method, a sensitive information processing method, a sensitivity prediction model training device, a sensitive information processing device, an electronic device, a readable storage medium, and a computer program product. The sensitivity prediction model training method includes: acquiring a historical sensitive dataset, wherein the historical sensitive dataset includes an allergy dataset and an environmental dataset, wherein the allergy dataset includes a dataset corresponding to a target population, and the environmental dataset includes various types of atmospheric environmental parameters; quantifying multiple allergy data in the allergy dataset to obtain multiple target allergy data within a first numerical range; quantifying atmospheric environmental parameters for any type to obtain multiple target environmental parameters within a second numerical range; inputting the multiple target allergy data and target environmental parameters into an initial sensitivity prediction model and outputting a predicted sensitivity; inputting the predicted sensitivity and the real sensitive dataset corresponding to the historical sensitive dataset into a loss function to obtain a model loss value; and adjusting the model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

[0038] Figure 1 The illustration shows an application scenario of the sensitivity prediction model training method or the sensitive information processing method according to embodiments of the present disclosure.

[0039] like Figure 1 As shown, application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, a network 104, and a server 105. The network 104 serves as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0040] Users can interact with server 105 via network 104 using at least one of the first terminal device 101, second terminal device 102, and third terminal device 103 to receive or send messages, etc. Various communication client applications can be installed on the first terminal device 101, second terminal device 102, and third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social media platform software, etc. (for example only).

[0041] The first terminal device 101, the second terminal device 102, and the third terminal device 103 can be various electronic devices with displays and support web browsing, including but not limited to smartphones, tablets, laptops, and desktop computers.

[0042] Server 105 can be a server that provides various services, such as a backend management server that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103 (this is just an example). The backend management server can analyze and process data such as received user requests, and feed back the processing results (such as web pages, information, or data obtained or generated according to user requests) to the terminal devices.

[0043] It should be noted that the sensitivity prediction model training method or sensitive information processing method provided in this disclosure embodiment can generally be executed by server 105. Correspondingly, the sensitivity prediction model training device or sensitive information processing device provided in this disclosure embodiment can generally be located in server 105. The sensitivity prediction model training method or sensitive information processing method provided in this disclosure embodiment can also be executed by a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105. Correspondingly, the sensitivity prediction model training device or sensitive information processing device provided in this disclosure embodiment can also be located in a server or server cluster that is different from server 105 and capable of communicating with the first terminal device 101, the second terminal device 102, the third terminal device 103, and / or server 105.

[0044] It should be understood that Figure 1 The number of terminal devices, networks, and servers shown is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be included.

[0045] The following will be based on Figure 1 The described scene, through Figure 2 The sensitivity prediction model training method of the disclosed embodiments is described in detail.

[0046] Figure 2 A flowchart illustrating a sensitivity prediction model training method according to an embodiment of the present disclosure is shown schematically.

[0047] like Figure 2 As shown, the sensitivity prediction model training method in this embodiment includes operations S210 to S260.

[0048] In operation S210, historical sensitive datasets are obtained, which include allergy datasets and environmental datasets. The allergy datasets include datasets corresponding to the target population, and the environmental datasets include various types of atmospheric environmental parameters.

[0049] In operation S220, multiple allergy data in the allergy dataset are quantified to obtain multiple target allergy data within a first numerical range.

[0050] In operation S230, for any type of atmospheric environmental parameter, the atmospheric environmental parameter is quantified to obtain multiple target environmental parameters within a second numerical range.

[0051] In operation S240, multiple target allergy data and target environmental parameters are input into the initial sensitivity prediction model, and the predicted sensitivity is output.

[0052] In operation S250, the predicted sensitivity and the real sensitive dataset corresponding to the historical sensitive dataset are input into the loss function to obtain the model loss value.

[0053] In operation S260, the model parameters of the initial sensitivity prediction model are adjusted based on the model loss value to obtain the target sensitivity prediction model.

[0054] According to embodiments of this disclosure, in the process of acquiring allergy datasets, since the allergy dataset includes a dataset corresponding to a target population, where the target population can be people experiencing allergy symptoms, it can be obtained from databases of dermatology hospitals or hospitals with high patient traffic. Specifically, the allergy dataset includes the number of allergy patients per day for the past year. For example, the number of allergy patients in Y1-M1-D1 is 200. It should be noted that the acquisition of allergy patient data complies with relevant laws and regulations, and necessary confidentiality measures have been taken. Furthermore, the authorization or consent of the patients was obtained before acquisition.

[0055] According to embodiments of this disclosure, during the acquisition of environmental datasets, atmospheric environmental parameters from the past year can be obtained from a database or cloud server. These atmospheric environmental parameters include various types, such as pollen concentration, temperature, humidity, wind speed, PM2.5, and vehicle exhaust concentration.

[0056] According to embodiments of this disclosure, multiple allergy data points in an allergy dataset are quantized. Based on the number of allergy sufferers per day over a historical year, all individuals in the allergy dataset are quantized into values ​​between 0 and 100. Before quantization, the maximum and minimum values ​​can be removed to make the quantization results more accurate.

[0057] According to embodiments of this disclosure, atmospheric environmental parameters are quantified for each type. For example, the daily temperature values ​​over a historical year are quantized into values ​​between 0 and 10; the daily wind speed values ​​over a historical year are quantized into values ​​between 0 and 10. Before quantization, the maximum and minimum values ​​can be removed to make the quantization results more accurate.

[0058] According to embodiments of this disclosure, multiple target allergy data and target environmental parameters are input into an initial sensitivity prediction model. The initial sensitivity prediction model learns and analyzes these data, outputting a predicted sensitivity. Here, the predicted sensitivity represents the daily predicted sensitivity over a historical year. During the training of the initial sensitivity prediction model, to train a model with more accurate prediction capabilities, it is necessary to ensure that the predicted values ​​and actual values ​​are as consistent as possible. Therefore, the predicted sensitivity and the actual sensitivity dataset corresponding to the historical sensitivity dataset are input into a loss function to obtain the model loss value. Based on the model loss value, the model parameters of the initial sensitivity prediction model are adjusted to obtain the target sensitivity prediction model.

[0059] According to embodiments of this disclosure, by quantifying allergy data and atmospheric environmental parameters separately, the initial sensitivity prediction model can learn from the data, effectively improving the model's learning efficiency. Furthermore, during the training process, the initial sensitivity prediction model involves learning from the allergy dataset and various types of atmospheric environmental parameters, at least partially solving the technical problem of limited data, thus making the trained target sensitivity prediction model more accurate in predicting sensitivity.

[0060] According to embodiments of this disclosure, after obtaining the historical sensitive dataset, the following operations may also be included:

[0061] The data in the historical sensitive dataset is preprocessed to obtain a preprocessed dataset, which is then used to train the initial sensitivity prediction model. The preprocessing operations include outlier removal, missing value imputation, and noise minimization.

[0062] According to embodiments of this disclosure, during data preprocessing, outliers can be detected and removed using a statistical rolling method. Data can be detected using a time difference method to complete the data on the time axis, thus completing the missing value imputation operation. The rolling average method can be used to minimize noise in the time series data, thereby minimizing noise. Preprocessing the data facilitates quantization and makes the quantization results more accurate.

[0063] According to embodiments of this disclosure, where the first numerical range is greater than the second numerical range, inputting multiple target allergy data and target environmental parameters into the initial sensitivity prediction model may include the following operations:

[0064] Multiple target allergy data within the first numerical range and multiple target environmental parameters within the second numerical range are vectorized to obtain a matrix of a preset dimension; the matrix of the preset dimension is then input into the initial sensitivity prediction model.

[0065] According to embodiments of this disclosure, daily target allergy data and target environmental data from the past year are vectorized. Since the target environmental data includes quantified pollen concentration, temperature, humidity, wind speed, PM2.5, and vehicle exhaust concentration, a 7-dimensional matrix is ​​obtained after vectorization. This 7-dimensional matrix is ​​used as input to an initial sensitivity prediction model, which outputs the predicted sensitivity corresponding to the 7-dimensional matrix.

[0066] According to embodiments of this disclosure, inputting the predicted sensitivity and the real sensitive dataset corresponding to the historical sensitive dataset into the loss function may include the following operations:

[0067] The environmental dataset and allergy dataset, which correspond to the historical sensitive dataset in the real sensitive dataset, are quantized to obtain quantized datasets; the quantized datasets and the predicted sensitivity are then input into the loss function.

[0068] According to embodiments of this disclosure, the real sensitive dataset corresponds to the historical dataset. For example, the historical sensitive dataset consists of daily allergy and environmental datasets from Y1-M1-D1 to Y1-M1-D2. The real sensitive dataset consists of daily allergy and environmental datasets from Y2-M1-D1 to Y2-M1-D2. The historical sensitive dataset corresponds to a year earlier than the real sensitive dataset. The environmental and allergy datasets within the real sensitive dataset are quantized separately. The number of allergy sufferers per day in the allergy dataset is quantized to a value between 0 and 100. Various types of atmospheric environmental parameters in the environmental dataset are quantized to values ​​between 0 and 10.

[0069] According to embodiments of this disclosure, the quantized dataset and the prediction sensitivity are input into a loss function to obtain the model loss value. By adjusting the initial sensitivity prediction model parameters, the model loss value is minimized, thereby making the model prediction more accurate. The calculation formula is as follows:

[0070]

[0071] Where M represents the total number of training samples, y represents the quantized dataset, y' represents the prediction sensitivity, and L represents the model loss value.

[0072] The following will be based on Figure 1 The described scene, through Figures 3-6 The sensitive information processing method of the disclosed embodiments will be described in detail.

[0073] Figure 3 A flowchart illustrating a sensitive information processing method according to an embodiment of the present disclosure is shown schematically.

[0074] like Figure 3 As shown, the sensitive information processing method of this embodiment includes operations S310 to S340.

[0075] In operation S310, the test dataset and the corresponding historical dataset are obtained. The test dataset includes test environment data and test allergy data, and the historical dataset includes historical environment data corresponding to the test environment data and historical allergy data corresponding to the test allergy data.

[0076] The S320 is used to quantize the historical dataset to obtain the target dataset.

[0077] In operation S330, the target dataset is input into the target sensitivity prediction model, and the sensitivity corresponding to the dataset to be tested is output.

[0078] The sensitivity is visualized when operating the S340.

[0079] According to embodiments of this disclosure, a test dataset and a corresponding historical dataset are obtained. For example, the test dataset consists of environmental and allergy data corresponding to Y2-M1-D1, and the historical dataset consists of environmental and allergy data corresponding to Y1-M1-D1, wherein the historical dataset corresponds to a time one year earlier than the test dataset. The environmental and allergy data in the historical dataset are then quantized. The number of allergy sufferers per day in the allergy dataset is quantized to a value between 0 and 100. Various types of atmospheric environmental parameters in the environmental dataset are quantized to values ​​between 0 and 10.

[0080] Figure 4A schematic diagram illustrating the structure of a sensitivity prediction model according to an embodiment of the present disclosure is shown.

[0081] According to embodiments of this disclosure, such as Figure 4 As shown, the quantified allergy data y back The environmental data X is input into the target sensitivity prediction model, and the output is the sensitivity corresponding to the data to be tested. (Global Forecast). The target sensitivity prediction model consists of multiple stacks. After prediction, the output of each stack is obtained and displayed as a curve. These include the output "Trend" for Stack1, the output "seasonally" for Stack2, and the output "exogenous" for StackS. By fitting multiple curves, the final output curve of the model (Forecast Period) is obtained. This curve is then compared with historical data (Backcast Period) to determine the difference between the predicted and historical data.

[0082] According to embodiments of this disclosure, each stack includes multiple blocks, and each block is the most basic structural module of the model, composed of multiple fully connected layers (FC Stack). Each block contains two main parts: the first part maps the input time series into expansion coefficients (a low-dimensional vector storing the inherent information of the time series), and the second part maps the expansion coefficients back to the time series. The specific calculation formula is as follows:

[0083]

[0084] The input to the target sensitivity prediction model is (y back ,X),y back Let represent quantified allergy data, and X represent quantified environmental data. s represents the s-th stack, and b represents the b-th block. The input to each block is the input of the previous block minus its output. In this way, the model processes the stack residuals that previous layers could not fit correctly. The final prediction is the sum of the predictions from each block. s,b This represents the output of the (b-1)th block. and X b-1 The parameters FC of the FC Stack in the fully connected neural network FCNN are then passed through two linear functions to obtain the spread coefficients. and

[0085]

[0086] in, This represents the prediction value of the b-th block in the s-th stack for the future. This represents the predicted value of the b-th block in the s-th stack for the past. V represents a linear mapping vector, which will be fitted during training.

[0087]

[0088] in, Represents the true historical value, residual The prediction for the future in the s-th stack is the sum of the predictions of all blocks.

[0089]

[0090] The sensitivity is the sum of all stack output values. This indicates the sensitivity of the data to be tested.

[0091] Figure 5 A schematic diagram illustrating the results of sensitivity assessment according to an embodiment of this disclosure is provided.

[0092] According to embodiments of this disclosure, such as Figure 5 As shown, the sensitivity is visualized, with the allergy index prediction representing the sensitivity output by the target sensitivity prediction model. Arcs from left to right indicate increasing severity, reflecting the current level of sensitivity. Below the arcs, the daily sensitivity, or allergy index, is displayed. Environmental data such as pollen concentration, temperature, humidity, and PM2.5 levels, along with the corresponding date, are also shown for sensitivity reference. This visualization interface can be displayed on relevant financial office software. This software can be used for mobile or remote work, information exchange, and training. Users can access this visualization interface by clicking the relevant icon on the financial office software and then clicking the "Allergy Index" icon to view sensitivity and environmental data related information.

[0093] According to embodiments of this disclosure, the sensitive information processing method may further include the following operations:

[0094] Multiple target sensitive information items are identified in the database based on preset keywords. Keyword extraction technology is used to determine the target topic of each target sensitive information item. A cosine similarity algorithm is used to determine the similarity between the target topic and the preset topic, which is determined based on time information and an allergy information list, including allergens corresponding to each time item. The multiple target sensitive information items are sorted based on similarity and preset sorting rules to obtain a sequence number for each item. Preset filtering rules are used to filter the multiple target sensitive information items based on their sequence numbers, resulting in multiple filtered sensitive information items. The information titles corresponding to the multiple filtered sensitive information items are visualized to allow viewing of the filtered sensitive information based on its title.

[0095] According to embodiments of this disclosure, web crawling technology is used to download target sensitive information whose titles contain the preset keyword "allergy". The target sensitive information can be related articles, which can be obtained from professional medical websites. The target topic of each piece of target sensitive information is determined using TF-IDF (term frequency–inverse document frequency, a commonly used weighting technique for information retrieval and data mining). For example, the target topic of article 1 is "Beijing, plants, pets, allergies, drugs, sneezing", and the target topic of article 2 is "pollen, allergies, symptoms, runny nose, prevention, hospital".

[0096] According to embodiments of this disclosure, the similarity between a target topic and a preset topic is determined using a cosine similarity algorithm. The preset topic is determined based on time information and an allergy information list. As the time information changes, the corresponding allergens identified in the allergy information list change, thus making the preset topic different.

[0097] According to embodiments of this disclosure, in determining the similarity between the target topic and the preset topic, assuming the target topic of article 1 is A and the preset topic is B. A and B are two n-dimensional vectors, where A is [A1, A2, ... An] and B is [B1, B2, ... Bn]. Then, the cosine of the angle θ between A and B is the cosine similarity between A and B, expressed by the formula:

[0098]

[0099] According to an embodiment of this disclosure, the keywords of a preset topic are represented by a vector [1 1 1 11 1]. If a keyword appears in the keywords of article 1, it is represented as 1; otherwise, it is represented as 0. Thus, the keywords of article A can be represented by a vector [1 0 0 0 01], and the keywords of article 2 are represented as [1 1 1 01 1]. According to formula (6), the cosine similarity between the keyword vectors of article 1 and article 2 and the preset keyword vector can be calculated.

[0100] Figure 6 The diagram illustrates the result display of the information header according to an embodiment of the present disclosure.

[0101] According to embodiments of this disclosure, sensitive information is sorted by similarity to obtain a sequence number for each piece of information. The top ten pieces of sensitive information with the highest to lowest similarity are then filtered to obtain ten filtered sensitive information articles. The titles of these ten filtered sensitive information articles are then visualized, allowing users to click and view the original text. This visualization interface can be displayed on relevant financial office software. Users can access this visualization interface by clicking the relevant icon on the financial office software and then further clicking the "Popular Science Article" icon to view the sensitive information. Figure 6 As shown, the visualization displays ten articles; for example, the first article's title is "Prevention of Pollen Allergies."

[0102] Figure 7 A schematic diagram illustrating the result of location names according to an embodiment of the present disclosure is provided.

[0103] According to embodiments of this disclosure, the sensitive information processing method may further include the following operations:

[0104] In response to the location information received from the target object, the system uses a map plugin to query multiple target locations associated with the location information; the system then visualizes the location names corresponding to the multiple target locations so that related services can be invoked based on the location names.

[0105] According to embodiments of this disclosure, such as Figure 7 As shown, the system receives location information from the target object, integrates a map plugin to present the results, and displays multiple target locations associated with the location information. For example, nearby hospitals and pharmacies. The location names are displayed, along with related services, such as a link to register for appointments at hospitals, and directions to the hospital, thus providing users with comprehensive information and increasing user engagement. This visualization interface can be displayed on relevant financial office software. Users can access the visualization interface by clicking the relevant icon on the financial office software and then clicking the "Nearby" icon to view the location name and its associated services.

[0106] Based on the above-described sensitivity prediction model training method, this disclosure also provides a sensitivity prediction model training device. The following will combine... Figure 7 The device is described in detail.

[0107] Figure 8 A schematic block diagram of a sensitivity prediction model training apparatus according to an embodiment of the present disclosure is shown.

[0108] like Figure 8 As shown, the sensitivity prediction model training device 800 of this embodiment includes a first acquisition module 810, a first quantization module 820, a second quantization module 820, a first prediction module 830, a loss determination module 840, and a model training module 850.

[0109] The first acquisition module 810 is used to acquire a historical sensitive dataset, which includes an allergy dataset and an environmental dataset. The allergy dataset includes a dataset corresponding to the target population, and the environmental dataset includes various types of atmospheric environmental parameters. In one embodiment, the first acquisition module 810 can be used to perform the operation S210 described above, which will not be repeated here.

[0110] The first quantization module 820 is used to quantize multiple allergy data in the allergy dataset to obtain multiple target allergy data within a first numerical range. In one embodiment, the first quantization module 820 can be used to perform the operation S220 described above, which will not be repeated here.

[0111] The second quantization module 830 is used to quantify atmospheric environmental parameters for any type of atmospheric environmental parameter to obtain multiple target environmental parameters within a second numerical range. In one embodiment, the second quantization module 830 can be used to perform the operation S230 described above, which will not be repeated here.

[0112] The first prediction module 840 is used to input multiple target allergy data and target environmental parameters into an initial sensitivity prediction model and output the predicted sensitivity. In one embodiment, the first prediction module 840 can be used to perform the operation S240 described above, which will not be repeated here.

[0113] The loss determination module 850 is used to input the predicted sensitivity and the real sensitive dataset corresponding to the historical sensitive dataset into the loss function to obtain the model loss value. In one embodiment, the loss determination module 850 can be used to perform the operation S250 described above, which will not be repeated here.

[0114] The model training module 860 is used to adjust the model parameters of the initial sensitivity prediction model based on the model loss value to obtain the target sensitivity prediction model. In one embodiment, the model training module 860 can be used to perform the operation S260 described above, which will not be repeated here.

[0115] According to embodiments of this disclosure, by quantifying allergy data and atmospheric environmental parameters separately, the initial sensitivity prediction model can learn from the data, effectively improving the model's learning efficiency. Furthermore, during the training process, the initial sensitivity prediction model involves learning from the allergy dataset and various types of atmospheric environmental parameters, at least partially solving the technical problem of limited data, thus making the trained target sensitivity prediction model more accurate in predicting sensitivity.

[0116] According to embodiments of this disclosure, the sensitivity prediction model training apparatus 800 further includes a preprocessing module.

[0117] The preprocessing module is used to preprocess the data in the historical sensitive dataset to obtain a preprocessed dataset, which is then used to train the initial sensitivity prediction model. The preprocessing operations include outlier removal, missing value imputation, and noise minimization.

[0118] According to embodiments of this disclosure, the first prediction module 840 includes a matrix determination submodule and a first input submodule.

[0119] The matrix determination submodule is used to vectorize multiple target allergy data within a first numerical range and multiple target environmental parameters within a second numerical range to obtain a matrix of preset dimensions.

[0120] The first input submodule is used to input a matrix of preset dimensions into the initial sensitivity prediction model.

[0121] According to embodiments of this disclosure, the loss determination module 850 includes a true quantization submodule and a second input submodule.

[0122] The Real Quantization submodule is used to quantize the environmental dataset and allergy dataset, which correspond to the historical sensitive dataset, in the real sensitive dataset to obtain the quantized dataset.

[0123] The second input submodule is used to input the quantized dataset and prediction sensitivity into the loss function.

[0124] According to embodiments of this disclosure, the sensitivity prediction model training apparatus 800 further includes a removal module.

[0125] The removal module is used to remove the maximum and minimum values ​​from the allergy dataset and the environmental dataset respectively, resulting in a removed dataset, which is then quantized.

[0126] Based on the above-described sensitive information processing method, this disclosure also provides a sensitive information processing apparatus. The following will be combined with... Figure 9 The device is described in detail.

[0127] Figure 9A schematic block diagram of a sensitivity prediction model training apparatus according to an embodiment of the present disclosure is shown.

[0128] like Figure 9 As shown, the sensitive information processing device 900 of this embodiment includes a second acquisition module 910, a third quantization module 920, a second prediction module 930, and a first display module 940.

[0129] The second acquisition module 910 is used to acquire the dataset to be tested and the historical dataset corresponding to the dataset to be tested. The dataset to be tested includes environmental data to be tested and allergy data to be tested. The historical dataset includes historical environmental data corresponding to the environmental data to be tested and historical allergy data corresponding to the allergy data to be tested. In one embodiment, the second acquisition module 910 can be used to perform the operation S310 described above, which will not be repeated here.

[0130] The third quantization module 920 is used to quantize the historical dataset to obtain the target dataset. In one embodiment, the third quantization module 920 can be used to perform the operation S320 described above, which will not be repeated here.

[0131] The second prediction module 930 is used to input the target dataset into the target sensitivity prediction model and output the sensitivity corresponding to the dataset to be tested. In one embodiment, the second prediction module 930 can be used to perform the operation S330 described above, which will not be repeated here.

[0132] The first display module 940 is used to visualize the sensitivity. In one embodiment, the first display module 940 can be used to perform the operation S340 described above, which will not be repeated here.

[0133] According to embodiments of this disclosure, the sensitive information processing device 900 further includes an information determination module, a topic determination module, a similarity determination module, a sorting module, a filtering module, and a second display module.

[0134] The information determination module is used to determine multiple target sensitive information in the database based on preset keywords.

[0135] The topic determination module is used to determine the target topic for each piece of sensitive information by using keyword extraction technology.

[0136] The similarity determination module is used to determine the similarity between the target topic and the preset topic using the cosine similarity algorithm. The preset topic is determined based on time information and an allergy information list, which includes allergens corresponding to each time information.

[0137] The sorting module is used to sort multiple target sensitive information based on similarity and preset sorting rules to obtain the sequence number of each target sensitive information.

[0138] The filtering module is used to filter multiple sensitive information targets based on their serial numbers using preset filtering rules, thereby obtaining multiple filtered sensitive information.

[0139] The second display module is used to visually display the information titles corresponding to multiple filtered sensitive information items, so that the filtered sensitive information can be viewed based on the information titles.

[0140] According to embodiments of this disclosure, the sensitive information processing apparatus 900 further includes a location determination module and a third display module.

[0141] The location determination module is used to query multiple target locations associated with the received location information from the target object using a map plugin.

[0142] The third display module is used to visualize the location names corresponding to multiple target locations, so that relevant services can be called based on the location names.

[0143] According to embodiments of this disclosure, any multiple modules selected from the first acquisition module 810, the first quantization module 820, the second quantization module 820, the first prediction module 830, the loss determination module 840, and the model training module 850, or the second acquisition module 910, the third quantization module 920, the second prediction module 930, and the first display module 940, can be combined into one module, or any one of these modules can be split into multiple modules. Alternatively, at least some of the functions of one or more of these modules can be combined with at least some of the functions of other modules and implemented in one module. According to embodiments of this disclosure, at least one of the first acquisition module 810, the first quantization module 820, the second quantization module 820, the first prediction module 830, the loss determination module 840, and the model training module 850, or the second acquisition module 910, the third quantization module 920, the second prediction module 930, and the first display module 940, can be at least partially implemented as hardware circuits, such as field-programmable gate arrays (FPGAs), programmable logic arrays (PLAs), systems-on-a-chip, systems-on-a-substrate, systems-on-package, application-specific integrated circuits (ASICs), or implemented in hardware or firmware by any other reasonable means of integrating or packaging circuits, or implemented in any one of the three methods of software, hardware, and firmware, or in a suitable combination of any of them. Alternatively, at least one of the first acquisition module 810, the first quantization module 820, the second quantization module 820, the first prediction module 830, the loss determination module 840, and the model training module 850, or the second acquisition module 910, the third quantization module 920, the second prediction module 930, and the first display module 940, can be at least partially implemented as a computer program module, which can perform corresponding functions when the computer program module is run.

[0144] Figure 10 A block diagram of an electronic device suitable for implementing a sensitivity prediction model training method or a sensitive information processing method according to embodiments of the present disclosure is shown schematically.

[0145] like Figure 10 As shown, an electronic device 1000 according to an embodiment of the present disclosure includes a processor 1001, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage portion 1008 into a random access memory (RAM) 1003. The processor 1001 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or an associated chipset and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 1001 may also include onboard memory for caching purposes. The processor 1001 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present disclosure.

[0146] RAM 1003 stores various programs and data required for the operation of electronic device 1000. Processor 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Processor 1001 performs various operations of the method flow according to embodiments of the present disclosure by executing programs in ROM 1002 and / or RAM 1003. It should be noted that the programs may also be stored in one or more memories other than ROM 1002 and RAM 1003. Processor 1001 may also perform various operations of the method flow according to embodiments of the present disclosure by executing programs stored in said one or more memories.

[0147] According to embodiments of this disclosure, the electronic device 1000 may further include an input / output (I / O) interface 1005, which is also connected to a bus 1004. The electronic device 1000 may also include one or more of the following components connected to the input / output (I / O) interface 1005: an input section 1006 including a keyboard, mouse, etc.; an output section 1007 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a LAN card, modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the input / output (I / O) interface 1005 as needed. A removable medium 1011, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on the drive 1010 as needed so that computer programs read from it can be installed into the storage section 1008 as needed.

[0148] This disclosure also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments; or it may exist independently and not assembled into the device / apparatus / system. The computer-readable storage medium carries one or more programs that, when executed, implement the method according to the embodiments of this disclosure.

[0149] According to embodiments of this disclosure, the computer-readable storage medium may be a non-volatile computer-readable storage medium, such as including, but not limited to: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this disclosure, the computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to embodiments of this disclosure, the computer-readable storage medium may include ROM 1002 and / or RAM 1003 and / or one or more memories other than ROM 1002 and RAM 1003 described above.

[0150] Embodiments of this disclosure also include a computer program product comprising a computer program containing program code for performing the methods shown in the flowchart. When the computer program product is run on a computer system, the program code enables the computer system to implement the sensitivity prediction model training method or sensitive information processing method provided in the embodiments of this disclosure.

[0151] When the computer program is executed by the processor 1001, it performs the functions defined in the system / apparatus of this disclosure embodiments. According to embodiments of this disclosure, the systems, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0152] In one embodiment, the computer program may rely on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may also be transmitted and distributed in the form of signals over a network medium, and downloaded and installed via the communication section 1009, and / or installed from a removable medium 1011. The program code contained in the computer program can be transmitted using any suitable network medium, including but not limited to: wireless, wired, etc., or any suitable combination thereof.

[0153] In such an embodiment, the computer program can be downloaded and installed from a network via communication section 1009, and / or installed from removable medium 1011. When the computer program is executed by processor 1001, it performs the functions defined in the system of this disclosure embodiment. According to embodiments of this disclosure, the systems, devices, apparatuses, modules, units, etc., described above can be implemented by computer program modules.

[0154] According to embodiments of this disclosure, program code for executing the computer programs provided in embodiments of this disclosure can be written in any combination of one or more programming languages. Specifically, these computational programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C", or similar programming languages. The program code can execute entirely on the user's computing device, partially on the user's device, partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0155] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0156] Those skilled in the art will understand that the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways, even if such combinations or combinations are not explicitly described in this disclosure. In particular, the features described in the various embodiments and / or claims of this disclosure can be combined or combined in various ways without departing from the spirit and teachings of this disclosure. All such combinations and / or combinations fall within the scope of this disclosure.

[0157] The embodiments of this disclosure have been described above. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of this disclosure. Although various embodiments have been described above, this does not mean that the measures in the various embodiments cannot be used advantageously in combination. The scope of this disclosure is defined by the appended claims and their equivalents. Various substitutions and modifications can be made by those skilled in the art without departing from the scope of this disclosure, and all such substitutions and modifications should fall within the scope of this disclosure.

Claims

1. A sensitivity prediction model training method, comprising: obtaining a historical sensitive data set, wherein the historical sensitive data set comprises an allergy data set and an environmental data set, wherein the allergy data set comprises a data set corresponding to a target population, and the environmental data set comprises a plurality of types of atmospheric environmental parameters; quantizing a plurality of allergy data in the allergy data set to obtain a plurality of target allergy data within a first numerical range; quantizing any type of the atmospheric environmental parameters to obtain a plurality of target environmental parameters within a second numerical range; inputting the plurality of target allergy data and the target environmental parameters into an initial sensitivity prediction model to output a predicted sensitivity; inputting the predicted sensitivity and a true sensitive data set corresponding to the historical sensitive data set into a loss function to obtain a model loss value; adjusting model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

2. The method of claim 1, wherein, After obtaining the historical sensitive data set, further comprising: performing a preprocessing operation on the data in the historical sensitive data set to obtain a preprocessed data set, so as to train the initial sensitivity prediction model using the preprocessed data set, wherein the preprocessing operation comprises outlier removal, missing value supplementation, and noise minimization.

3. The method of claim 1 or 2, wherein, The first numerical range is greater than the second numerical range, and the inputting of the plurality of target allergy data within the first numerical range and the plurality of target environmental parameters within the second numerical range into the initial sensitivity prediction model comprises: vectorizing the plurality of target allergy data within the first numerical range and the plurality of target environmental parameters within the second numerical range to obtain a matrix of a preset dimension; inputting the matrix of the preset dimension into the initial sensitivity prediction model.

4. The method of claim 1 or 2, wherein, The inputting of the predicted sensitivity and the true sensitive data set corresponding to the historical sensitive data set into the loss function comprises: quantizing the environmental data set and the allergy data set corresponding to the historical sensitive data set in the true sensitive data set respectively to obtain a quantized data set; inputting the quantized data set and the predicted sensitivity into the loss function.

5. The method of claim 1 or 2, wherein, Before the quantizing of the plurality of allergy data in the allergy data set to obtain the plurality of target allergy data within the first numerical range, further comprising: removing the maximum and minimum values in the allergy data set and the environmental data set respectively to obtain a removed data set, so as to quantize the removed data set.

6. A sensitive information processing method, comprising: obtaining a to-be-tested data set and a historical data set corresponding to the to-be-tested data set, wherein the to-be-tested data set comprises to-be-tested environmental data and to-be-tested allergy data, and the historical data set comprises historical environmental data corresponding to the to-be-tested environmental data and historical allergy data corresponding to the to-be-tested allergy data; quantizing the historical data set to obtain a target data set; inputting the target data set into the target sensitivity prediction model of any one of claims 1-5 to output a sensitivity corresponding to the to-be-tested data set. visualize the sensitivity.

7. The method of claim 6, further comprising: determining a plurality of target sensitive information in a database based on a preset keyword; determining a target topic of each of the target sensitive information by using a keyword extraction technique; determining a similarity between the target topic and a preset topic by using a cosine similarity algorithm, wherein the preset topic is determined based on time information and a list of allergens, and the list of allergens includes an allergen corresponding to each of the time information; sorting a plurality of the target sensitive information based on the similarity and a preset arrangement rule to obtain a serial number of each of the target sensitive information; screening a plurality of the target sensitive information based on the serial number by using a preset screening rule to obtain a plurality of screened sensitive information; visualizing information titles corresponding to the plurality of screened sensitive information to enable viewing of the screened sensitive information based on the information titles.

8. The method of claim 6, further comprising: in response to receiving location information sent by a target object, querying a plurality of target locations associated with the location information by using a map plug-in; visualizing location names corresponding to the plurality of target locations to enable calling of related services based on the location names.

9. A sensitivity prediction model training apparatus, comprising: a first acquisition module configured to acquire a historical sensitive data set, wherein the historical sensitive data set includes an allergen data set and an environmental data set, the allergen data set includes a data set corresponding to a target population, and the environmental data set includes a plurality of types of atmospheric environmental parameters; a first quantization module configured to quantize a plurality of allergen data in the allergen data set to obtain a plurality of target allergen data within a first numerical range; a second quantization module configured to quantize any type of the atmospheric environmental parameters to obtain a plurality of target environmental parameters within a second numerical range; a first prediction module configured to input the plurality of target allergen data and the target environmental parameters into an initial sensitivity prediction model to output a predicted sensitivity; a loss determination module configured to input the predicted sensitivity and a true sensitive data set corresponding to the historical sensitive data set into a loss function to obtain a model loss value; a model training module configured to adjust model parameters of the initial sensitivity prediction model based on the model loss value to obtain a target sensitivity prediction model.

10. A sensitive information processing apparatus, comprising: a second acquisition module configured to acquire a to-be-tested data set and a historical data set corresponding to the to-be-tested data set, wherein the to-be-tested data set includes to-be-tested environmental data and to-be-tested allergen data, and the historical data set includes historical environmental data corresponding to the to-be-tested environmental data and historical allergen data corresponding to the to-be-tested allergen data; a third quantization module configured to quantize the historical data set to obtain a target data set; a second prediction module, configured to input the target data set into the target sensitivity prediction model in any one of claims 1-5, and output a sensitivity corresponding to the target data set; a first display module, configured to visually display the sensitivity. 11.An electronic device, comprising: one or more processors; a storage device configured to store one or more programs, wherein the one or more programs, when executed by the one or more processors, enable the one or more processors to perform the method according to any one of claims 1-5 or 6-8. 12.A computer-readable storage medium having stored thereon executable instructions that, when executed by a processor, cause the processor to perform the method according to any one of claims 1-5 or 6-8. 13.A computer program product, comprising a computer program that, when executed by a processor, implements the method according to any one of claims 1-5 or 6-8.

Images