New pollutant identification method, system and device and storage medium

Abstract

The invention discloses a new pollutant identification method, system and device, and a storage medium. The new pollutant identification method comprises the following steps: acquiring a multi-modal feature of a to-be-identified pollutant; wherein the multi-modal feature comprises at least one of a first molecular descriptor feature, a first molecular structure image feature and a first text feature; inputting the multi-modal features into a trained modal detection and completion model for modal completion to obtain second molecular descriptor features, second molecular structure image features and second text features; fusing the second molecular descriptor feature, the second molecular structure image feature and the second text feature to obtain a fused feature; and inputting the fused features into the trained new pollutant identification model for new pollutant identification to obtain a classification result of the to-be-identified pollutants, and improving the accuracy of new pollutant type identification by fusing the structured data, the image data and the semantic text data.

Description

Technical Field

[0001] This application relates to the field of new pollutant identification technology, and in particular to a new pollutant identification method, system, device and storage medium. Background Technology

[0002] With the acceleration of industrial activities and the upgrading of consumer products, new pollutants with complex structures and high potential toxicity are constantly entering the natural environment, including persistent organic pollutants, perfluoroalkyl substances, endocrine disruptors, drug residues, pesticide metabolites, plasticizers, and nanomaterials. These new pollutants are characterized by their wide variety, complex structures, high potential toxicity, diverse environmental behaviors, variable physicochemical behaviors, and complex exposure pathways, and have gradually become key factors affecting ecosystem security and human health.

[0003] Currently, the methods widely used in the identification and analysis of new pollutants include chemometric analysis, high-throughput screening, high-resolution mass spectrometry analysis, and computational simulation. However, these methods often rely on single-modal data, resulting in low identification accuracy and insufficient behavior prediction capabilities. Summary of the Invention

[0004] This application aims to at least address the technical problems existing in the prior art. To this end, this application proposes a new pollutant identification method, system, device, and storage medium that can integrate structured data, image data, and semantic text data to improve the accuracy of identifying new pollutant types.

[0005] A first aspect of this application provides a novel method for identifying pollutants, comprising the following steps:

[0006] Obtain multimodal features of the pollutant to be identified; wherein the multimodal features include at least one of a first molecule descriptor feature, a first molecule structure image feature, and a first text feature;

[0007] The multimodal features are input into the trained modality detection and completion model for modality completion, resulting in second molecule descriptor features, second molecule structure image features, and second text features.

[0008] The second molecule descriptor features, the second molecule structure image features, and the second text features are fused to obtain the fused features;

[0009] The fused features are input into the trained new pollutant identification model to identify the new pollutant, and the classification result of the pollutant to be identified is obtained.

[0010] The novel pollutant identification method according to the embodiments of this application has at least the following beneficial effects:

[0011] This method acquires multimodal features of the pollutant to be identified. These multimodal features include at least one of a first molecular descriptor feature, a first molecular structure image feature, and a first text feature. The multimodal features are then input into a trained modality detection and completion model for modality completion, yielding a second molecular descriptor feature, a second molecular structure image feature, and a second text feature. These features are then fused to obtain a fused feature. Finally, the fused feature is input into a trained new pollutant identification model for new pollutant identification, resulting in a classification result for the pollutant. By fusing structured data, image data, and semantic text data, the accuracy of identifying new pollutant types is improved.

[0012] According to some embodiments of this application, the novel pollutant identification method further includes:

[0013] The fused features are input into the trained behavior parameter prediction model to obtain the behavior parameter prediction values ​​output by the trained behavior parameter prediction model in different environments.

[0014] A comprehensive risk score is calculated based on the predicted values ​​of the aforementioned behavioral parameters.

[0015] According to some embodiments of this application, the predicted behavioral parameters include, but are not limited to, migration values, environmental persistence values, and potential toxicity values. The calculation of a comprehensive risk score based on the predicted behavioral parameters includes:

[0016] Multiply the migration value by the first preset weight value to obtain the first score value;

[0017] Multiply the environmental persistence value and the second preset weight value to obtain the second score value;

[0018] The potential toxicity value is multiplied by the third preset weight value to obtain the third score;

[0019] The first score, the second score, and the third score are added together to obtain the comprehensive risk score.

[0020] According to some embodiments of this application, the novel pollutant identification method further includes:

[0021] The risk level of the pollutant to be identified is obtained by classifying the risk level based on the comprehensive risk score and the preset level threshold.

[0022] The pollutants to be identified are treated based on the risk level and the preset optimization scheme.

[0023] According to some embodiments of this application, the step of inputting the multimodal features into a trained modality detection and completion model for modality completion to obtain second molecule descriptor features, second molecule structure image features, and second text features includes:

[0024] The multimodal features are input into the trained modality detection and completion model so that the trained modality detection and completion model can identify missing or incomplete features of the multimodal features;

[0025] Based on the missing or incomplete features, feature reconstruction is performed using the trained modality detection and completion model to obtain the second molecular descriptor features, the second molecular structure image features, and the second text features.

[0026] According to some embodiments of this application, obtaining the multimodal features of the pollutant to be identified includes:

[0027] Obtain the raw data of the pollutant to be identified, wherein the raw data includes at least one of structured descriptor data, structured image data, and text description data;

[0028] When the original data includes the structured descriptor data, the first molecular descriptor features are extracted based on the structured descriptor data, wherein the first molecular descriptor features include at least molecular topology, electronic structure, polarity and molecular mass;

[0029] When the original data includes the structural image data, the first molecular structure image features are extracted based on the structural image data. The first molecular structure image features include at least shape, bond connection mode and functional group spatial layout.

[0030] When the original data includes the text description data, the first text feature is extracted based on the text description data, wherein the first text feature includes at least the naming system of the pollutants to be identified, text labels, and characters.

[0031] According to some embodiments of this application, the step of extracting the features of the first molecule descriptor based on the structured descriptor data includes:

[0032] Obtain the SMILES molecular structure code of the pollutant to be identified;

[0033] The molecular structure SMILES encoding of the molecular structure is analyzed to obtain the analysis results;

[0034] Based on the analysis results, the first molecular descriptor features of the pollutant to be identified are extracted.

[0035] A second aspect of this application provides a novel pollutant identification system, the novel pollutant identification system comprising:

[0036] The data acquisition module is used to acquire multimodal features of the pollutant to be identified; wherein the multimodal features include at least one of a first molecule descriptor feature, a first molecule structure image feature, and a first text feature;

[0037] The modality completion module is used to input the multimodal features into the trained modality detection and completion model for modality completion, thereby obtaining second molecule descriptor features, second molecule structure image features, and second text features;

[0038] The feature fusion module is used to fuse the second molecule descriptor features, the second molecule structure image features, and the second text features to obtain fused features;

[0039] The new pollutant identification module is used to input the fused features into the trained new pollutant identification model to identify new pollutants and obtain the classification result of the pollutant to be identified.

[0040] This system acquires multimodal features of the pollutant to be identified. These multimodal features include at least one of a first molecular descriptor feature, a first molecular structure image feature, and a first text feature. The multimodal features are then input into a trained modality detection and completion model for modality completion, resulting in a second molecular descriptor feature, a second molecular structure image feature, and a second text feature. The second molecular descriptor feature, the second molecular structure image feature, and the second text feature are then fused to obtain a fused feature. This fused feature is then input into a trained new pollutant identification model for new pollutant identification, resulting in a classification result for the pollutant to be identified. By fusing structured data, image data, and semantic text data, the accuracy of identifying new pollutant types is improved.

[0041] A third aspect of this application provides a novel pollutant identification electronic device, including at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor, which, when executed by the at least one control processor, enables the at least one control processor to perform the novel pollutant identification method described above.

[0042] A fourth aspect of this application provides a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the aforementioned novel contaminant identification method.

[0043] It should be noted that the beneficial effects of the second to fourth aspects of this application with respect to the prior art are the same as the beneficial effects of the novel pollutant identification system described above with respect to the prior art, and will not be elaborated here.

[0044] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0045] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0046] Figure 1 This is a flowchart of a novel pollutant identification method according to an embodiment of this application;

[0047] Figure 2 This is a schematic diagram of an embodiment of the novel pollutant identification system provided in this application;

[0048] Figure 3 This is a schematic diagram of the structure of an embodiment of the electronic device provided in this application. Detailed Implementation

[0049] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0050] In the description of this application, the use of terms such as "first," "second," etc., is for the purpose of distinguishing technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0051] In the description of this application, it should be understood that the orientation descriptions, such as up, down, etc., are based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application.

[0052] In the description of this application, it should be noted that, unless otherwise explicitly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0053] With the acceleration of industrial activities and the upgrading of consumer products, new pollutants with complex structures and high potential toxicity are constantly entering the natural environment, including endocrine disruptors, drug residues, pesticide metabolites, plasticizers, and nanomaterials. These new pollutants are characterized by their wide variety, complex structures, high potential toxicity, diverse environmental behaviors, variable physicochemical behaviors, and complex exposure pathways, and have gradually become key factors affecting ecosystem security and human health.

[0054] Currently, the methods widely used in the identification and analysis of new pollutants include chemometric analysis, high-throughput screening, high-resolution mass spectrometry analysis, and computational simulation. However, these methods often rely on single-modal data, resulting in low identification accuracy and insufficient behavior prediction capabilities.

[0055] To address the aforementioned technical deficiencies, embodiments of this application provide a novel pollutant identification method, system, device, and storage medium.

[0056] Please see Figure 1 This is a flowchart illustrating a novel pollutant identification method provided in an embodiment of this application. The method is applied to an electronic device, such as a server. Figure 1 As shown, the new pollutant identification method includes:

[0057] Step S101: Obtain the multimodal features of the pollutant to be identified; wherein, the multimodal features include at least one of the following: first molecule descriptor features, first molecule structure image features, and first text features;

[0058] Step S102: Input the multimodal features into the trained modality detection and completion model for modality completion to obtain the second molecule descriptor features, the second molecule structure image features, and the second text features;

[0059] Step S103: Fuse the second molecule descriptor features, the second molecule structure image features, and the second text features to obtain the fused features;

[0060] Step S104: Input the fused features into the trained new pollutant identification model to identify the new pollutant and obtain the classification result of the pollutant to be identified.

[0061] This method acquires multimodal features of the pollutant to be identified. These multimodal features include at least one of a first molecular descriptor feature, a first molecular structure image feature, and a first text feature. The multimodal features are then input into a trained modality detection and completion model for modality completion, yielding a second molecular descriptor feature, a second molecular structure image feature, and a second text feature. These features are then fused to obtain a fused feature. Finally, the fused feature is input into a trained new pollutant identification model for new pollutant identification, resulting in a classification result for the pollutant. By fusing structured data, image data, and semantic text data, the accuracy of identifying new pollutant types is improved.

[0062] In some embodiments, the new pollutant identification method further includes:

[0063] Step S201: Input the fused features into the trained behavior parameter prediction model to obtain the behavior parameter prediction values ​​output by the trained behavior parameter prediction model in different environments;

[0064] Step S202: Calculate the comprehensive risk score based on the predicted values ​​of behavioral parameters.

[0065] This application calculates a comprehensive risk score in different environments by using predicted values ​​of behavioral parameters in different environments, which can determine the pollutant risk score in different environments.

[0066] In some embodiments, the predicted behavioral parameters include, but are not limited to, migration values, environmental persistence values, and potential toxicity values. A comprehensive risk score is calculated based on these predicted behavioral parameters, including:

[0067] Step S301: Multiply the migration value and the first preset weight value to obtain the first score value;

[0068] Step S302: Multiply the environmental persistence value and the second preset weight value to obtain the second score value;

[0069] Step S303: Multiply the potential toxicity value and the third preset weight value to obtain the third score value;

[0070] Step S304: Add the first score, the second score, and the third score to obtain the comprehensive risk score.

[0071] The first, second, and third preset weight values ​​mentioned above are constant values ​​that are pre-set to add up to 1 according to actual needs.

[0072] This application improves the accuracy of risk scoring by assigning different weights to the predicted values ​​of different behavioral parameters.

[0073] In some embodiments, the new pollutant identification method further includes:

[0074] Step S401: Based on the comprehensive risk score and the preset level threshold, the risk level of the pollutant to be identified is determined.

[0075] Step S402: Treat the pollutants to be identified based on the risk level and the preset optimization scheme.

[0076] The aforementioned preset level thresholds are constant values ​​that are pre-set according to actual needs.

[0077] Specifically, the risk level can be set as high if it is above a preset value, and low if it is below a preset value; and by taking different optimization measures for different risk levels, the efficiency of response is improved.

[0078] In some embodiments, multimodal features are input into a trained modality detection and completion model for modality completion, resulting in second molecule descriptor features, second molecule structure image features, and second text features, including:

[0079] Step S501: Input the multimodal features into the trained modality detection and completion model so that the trained modality detection and completion model can identify the missing or incomplete features of the multimodal features;

[0080] Step S502: Based on missing or incomplete features, feature reconstruction is performed using a trained modality detection and completion model to obtain second molecule descriptor features, second molecule structure image features, and second text features.

[0081] The aforementioned missing features can be determined by the modality detection and completion model trained with the aforementioned multimodal features as lacking at least one of the aforementioned second molecule descriptor features, second molecule structure image features, and second text features.

[0082] The aforementioned incomplete features can be determined as incomplete by the trained modality detection and completion model by at least one of the first molecule descriptor features, first molecule structure image features, and first text features.

[0083] This application improves the accuracy of species identification by using feature completion as the data foundation for subsequent prediction.

[0084] In some embodiments, obtaining the multimodal features of the pollutant to be identified includes:

[0085] Step S601: Obtain the raw data of the pollutant to be identified, wherein the raw data includes at least one of structured descriptor data, structured image data, and text description data;

[0086] Step S602: If the original data includes structured descriptor data, extract the first molecule descriptor features based on the structured descriptor data, wherein the first molecule descriptor features include at least molecular topology, electronic structure, polarity and molecular mass.

[0087] Step S603: If the original data includes structural image data, extract the first molecule structural image features based on the structural image data. The first molecule structural image features include at least shape, bond connection mode and functional group spatial layout.

[0088] Step S604: If the original data includes text description data, extract the first text feature based on the text description data, wherein the first text feature includes at least the naming system of the pollutant to be identified, text label and characters.

[0089] In some embodiments, the structured descriptor includes the molecular fingerprint, physicochemical parameters, and SMILES encoding of the new pollutant. The molecular structure SMILES encoding is parsed to obtain the first molecular descriptor features of the pollutant to be identified. The first molecular descriptor features include a series of descriptor features such as molecular topology, electronic structure, polarity, and molecular mass. The first molecular structure image features include image pattern features such as shape, bond connection mode, and spatial layout of functional groups. The first text features include the naming system of the pollutant to be identified, text labels, and characters.

[0090] This application improves the accuracy of species identification by extracting multimodal features, which serve as the data basis for subsequent predictions.

[0091] In some embodiments, based on structured descriptor data, the first molecule descriptor features are extracted, including:

[0092] Step S701: Obtain the SMILES molecular structure code of the pollutant to be identified;

[0093] Step S702: Perform molecular structure analysis on the SMILES encoding of the molecular structure to obtain the analysis results;

[0094] Step S703: Based on the analysis results, extract the first molecular descriptor features of the pollutant to be identified.

[0095] This application extracts molecular descriptor features through molecular structure SMILES encoding, which serves as the data basis for subsequent predictions, thereby improving the accuracy of species identification.

[0096] Additionally, refer to Figure 2 One embodiment of this application provides a novel pollutant identification system, including a data acquisition module 1100, a modal completion module 1200, a feature fusion module 1300, and a novel pollutant identification module 1400, wherein:

[0097] The data acquisition module 1100 is used to acquire the multimodal features of the pollutant to be identified; wherein, the multimodal features include at least one of the following: first molecule descriptor features, first molecule structure image features, and first text features;

[0098] The modality completion module 1200 is used to input multimodal features into the trained modality detection and completion model for modality completion, and obtain second molecule descriptor features, second molecule structure image features and second text features;

[0099] The feature fusion module 1300 is used to fuse the second molecule descriptor features, the second molecule structure image features, and the second text features to obtain the fused features;

[0100] The new pollutant identification module 1400 is used to input the fused features into the trained new pollutant identification model to identify new pollutants and obtain the classification results of the pollutants to be identified.

[0101] This system acquires multimodal features of the pollutant to be identified. These multimodal features include at least one of a first molecular descriptor feature, a first molecular structure image feature, and a first text feature. The multimodal features are then input into a trained modality detection and completion model for modality completion, resulting in a second molecular descriptor feature, a second molecular structure image feature, and a second text feature. The second molecular descriptor feature, the second molecular structure image feature, and the second text feature are then fused to obtain a fused feature. This fused feature is then input into a trained new pollutant identification model for new pollutant identification, resulting in a classification result for the pollutant to be identified. By fusing structured data, image data, and semantic text data, the accuracy of identifying new pollutant types is improved.

[0102] It should be noted that the system embodiments described above are based on the same inventive concept as the method embodiments described above. Therefore, the relevant content of the method embodiments described above is also applicable to the system embodiments described above, and will not be repeated here.

[0103] Figure 3 A schematic diagram of the rule mining hardware structure provided in an embodiment of this application is shown.

[0104] The new pollutant identification device may include a processor 301 and a memory 302 storing computer program instructions.

[0105] Specifically, the processor 301 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0106] Memory 302 may include mass storage for data or instructions. For example, and not limitingly, memory 302 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 302 may include removable or non-removable (or fixed) media. Where appropriate, memory 302 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 302 is non-volatile solid-state memory.

[0107] In some embodiments, memory 302 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Thus, generally, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this disclosure.

[0108] The processor 301 implements any of the novel pollutant identification methods in the above embodiments by reading and executing computer program instructions stored in the memory 302.

[0109] In one example, the new contaminant identification device may also include a communication interface 303 and a bus 310. For example, Figure 3 As shown, the processor 301, memory 302, and communication interface 303 are connected through bus 310 and complete communication with each other.

[0110] The communication interface 303 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0111] Bus 310 includes hardware, software, or both, that couples components of the new contaminant identification device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 310 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0112] This new pollutant identification device can execute the new pollutant identification method described in this application based on a three-dimensional design model, thereby achieving a combination of... Figure 1 and Figure 2 A novel method and system for identifying pollutants are described.

[0113] Furthermore, in conjunction with the novel pollutant identification methods described in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the novel pollutant identification methods described in the above embodiments.

[0114] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0115] The functional blocks shown in the above block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0116] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0117] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0118] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A novel pollutant identification method, characterized in that, The new pollutant identification method includes: Obtain multimodal features of the pollutant to be identified; wherein the multimodal features include at least one of a first molecule descriptor feature, a first molecule structure image feature, and a first text feature; The multimodal features are input into the trained modality detection and completion model for modality completion, resulting in second molecule descriptor features, second molecule structure image features, and second text features. The second molecule descriptor features, the second molecule structure image features, and the second text features are fused to obtain the fused features; The fused features are input into the trained new pollutant identification model to identify the new pollutant, and the classification result of the pollutant to be identified is obtained.

2. The novel pollutant identification method according to claim 1, characterized in that, The new pollutant identification method also includes: The fused features are input into the trained behavior parameter prediction model to obtain the behavior parameter prediction values ​​output by the trained behavior parameter prediction model in different environments. A comprehensive risk score is calculated based on the predicted values ​​of the aforementioned behavioral parameters.

3. The novel pollutant identification method according to claim 2, characterized in that, The predicted behavioral parameters include, but are not limited to, migration values, environmental persistence values, and potential toxicity values. The calculation of a comprehensive risk score based on the predicted behavioral parameters includes: Multiply the migration value by the first preset weight value to obtain the first score value; Multiply the environmental persistence value and the second preset weight value to obtain the second score value; The potential toxicity value is multiplied by the third preset weight value to obtain the third score; The first score, the second score, and the third score are added together to obtain the comprehensive risk score.

4. The novel pollutant identification method according to claim 2, characterized in that, The new pollutant identification method also includes: The risk level of the pollutant to be identified is obtained by classifying the risk level based on the comprehensive risk score and the preset level threshold. The pollutants to be identified are treated based on the risk level and the preset optimization scheme.

5. The novel pollutant identification method according to claim 1, characterized in that, The process of inputting the multimodal features into a trained modality detection and completion model for modality completion, resulting in second molecule descriptor features, second molecule structure image features, and second text features, includes: The multimodal features are input into the trained modality detection and completion model so that the trained modality detection and completion model can identify missing or incomplete features of the multimodal features; Based on the missing or incomplete features, feature reconstruction is performed using the trained modality detection and completion model to obtain the second molecular descriptor features, the second molecular structure image features, and the second text features.

6. The novel pollutant identification method according to claim 1, characterized in that, The acquisition of multimodal features of the pollutant to be identified includes: Obtain the raw data of the pollutant to be identified, wherein the raw data includes at least one of structured descriptor data, structured image data, and text description data; When the original data includes the structured descriptor data, the first molecular descriptor features are extracted based on the structured descriptor data, wherein the first molecular descriptor features include at least molecular topology, electronic structure, polarity and molecular mass; When the original data includes the structural image data, the first molecular structure image features are extracted based on the structural image data. The first molecular structure image features include at least shape, bond connection mode and functional group spatial layout. When the original data includes the text description data, the first text feature is extracted based on the text description data, wherein the first text feature includes at least the naming system of the pollutants to be identified, text labels, and characters.

7. The novel pollutant identification method according to claim 6, characterized in that, The step of extracting the features of the first molecule descriptor based on the structured descriptor data includes: Obtain the SMILES molecular structure code of the pollutant to be identified; The molecular structure SMILES encoding is used to perform molecular structure parsing to obtain the first molecular descriptor features of the pollutant to be identified.

8. A novel pollutant identification system, characterized in that, The new pollutant identification system includes: The data acquisition module is used to acquire multimodal features of the pollutant to be identified; wherein the multimodal features include at least one of a first molecule descriptor feature, a first molecule structure image feature, and a first text feature; The modality completion module is used to input the multimodal features into the trained modality detection and completion model for modality completion, thereby obtaining second molecule descriptor features, second molecule structure image features, and second text features; The feature fusion module is used to fuse the second molecule descriptor features, the second molecule structure image features, and the second text features to obtain fused features; The new pollutant identification module is used to input the fused features into the trained new pollutant identification model to identify new pollutants and obtain the classification result of the pollutant to be identified.

9. A novel pollutant identification device, characterized in that, It includes at least one control processor and a memory for communicatively connecting to the at least one control processor; the memory stores instructions executable by the at least one control processor to enable the at least one control processor to perform a novel pollutant identification method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions for causing a computer to perform a novel pollutant identification method as described in any one of claims 1 to 7.

Images