Information processing method, storage medium, and electronic device

Abstract

The application discloses an information processing method, a storage medium and an electronic device. The method comprises the following steps: obtaining first knowledge and second knowledge of a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structure mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode, and constructing the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph. The application solves the technical problem of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, in particular, to an information processing method, a storage medium and an electronic device. BACKGROUND

[0002] In the production of agriculture and forestry, production workers often need to obtain and use relevant professional knowledge to optimize production. In the early stage, production workers obtain relevant professional knowledge by consulting scientific researchers or technical experts, however, the expert resources are limited, and this method is inefficient and most production workers are difficult to get help. In recent years, with the development of agricultural and forestry technology, especially the application of computer network, technical personnel began to build knowledge graph of agricultural and forestry production to provide relevant professional knowledge to production workers.

[0003] Therefore, how to build a professional and comprehensive knowledge graph at low cost to serve the production of agriculture and forestry has become an important problem in the related art. In view of the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The embodiments of the present application provide an information processing method, a storage medium and an electronic device to at least solve the technical problem of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

[0005] According to an aspect of the embodiments of the present application, an information processing method is provided, comprising: obtaining first knowledge and second knowledge of a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structure mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode to construct the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0006] According to another aspect of the embodiments of the present application, an information processing method is also provided, comprising: obtaining first kiwifruit knowledge and second kiwifruit knowledge of a kiwifruit to be monitored, wherein the first kiwifruit knowledge is objective kiwifruit knowledge associated with the kiwifruit to be monitored, and the second kiwifruit knowledge is subjective kiwifruit knowledge associated with the kiwifruit to be monitored; constructing a kiwifruit knowledge graph corresponding to the kiwifruit to be monitored by using the first kiwifruit knowledge and the second kiwifruit knowledge; and providing a target kiwifruit service function associated with the kiwifruit to be monitored based on the kiwifruit knowledge graph.

[0007] According to another aspect of the embodiments of the present application, there is also provided an information processing method, comprising: obtaining first crop knowledge and second crop knowledge associated with a crop to be monitored, wherein the first crop knowledge is objective crop knowledge associated with the crop to be monitored, and the second crop knowledge is subjective crop knowledge associated with the crop to be monitored; constructing a crop knowledge graph corresponding to the crop to be monitored by using the first crop knowledge and the second crop knowledge; and providing a target crop service function associated with the crop to be monitored based on the crop knowledge graph.

[0008] According to another aspect of the embodiments of the present application, there is also provided an information processing device, comprising: an obtaining module configured to obtain first knowledge and second knowledge associated with a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; a determining module configured to determine a structure mode of a knowledge graph corresponding to the crop to be monitored; a constructing module configured to perform knowledge extraction on the first knowledge and the second knowledge according to the structure mode, and construct the knowledge graph; and a service module configured to provide a target service function associated with the crop to be monitored based on the knowledge graph.

[0009] According to another aspect of the embodiments of the present application, there is also provided a computer readable storage medium comprising a stored program, wherein the program, when executed, controls a device in which the computer readable storage medium is located to perform any of the information processing methods described above.

[0010] According to another aspect of the embodiments of the present application, there is also provided an information processing system, comprising: a processor; and a memory connected to the processor, configured to provide the processor with instructions to process the following processing steps: obtaining first knowledge and second knowledge associated with a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structure mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode, and constructing the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0011] In the embodiments of the present application, first knowledge and second knowledge associated with a crop to be monitored are obtained first, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored, a structure mode of a knowledge graph corresponding to the crop to be monitored is determined, knowledge extraction is performed on the first knowledge and the second knowledge according to the structure mode, and the knowledge graph is constructed, and a target service function associated with the crop to be monitored is provided based on the knowledge graph.

[0012] It is easy to note that, through the embodiment of the present application, based on the objective knowledge and subjective knowledge associated with the crop to be monitored, the corresponding knowledge graph is constructed and the related service function is provided, the purpose of constructing the knowledge graph based on the related professional knowledge to serve the production is achieved, thereby realizing the technical effect of reducing the construction cost of the knowledge graph and the difficulty of obtaining professional knowledge, and further solving the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art. BRIEF DESCRIPTION OF DRAWINGS

[0013] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application. In the drawings:

[0014] Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing an information processing method is shown;

[0015] Figure 2 A flowchart of an information processing method according to an embodiment of the present application is shown;

[0016] Figure 3 A schematic diagram of an optional knowledge graph construction process according to an embodiment of the present application is shown;

[0017] Figure 4 A schematic diagram of an optional knowledge extraction model training process according to an embodiment of the present application is shown;

[0018] Figure 5 A schematic diagram of an optional natural language pre-training model training method according to an embodiment of the present application is shown;

[0019] Figure 6 A schematic diagram of an optional label refinement process according to an embodiment of the present application is shown;

[0020] Figure 7 A flowchart of another information processing method according to an embodiment of the present application is shown;

[0021] Figure 8 A flowchart of another information processing method according to an embodiment of the present application is shown;

[0022] Figure 9 A structural schematic diagram of an information processing device according to an embodiment of the present application is shown;

[0023] Figure 10 A structural schematic diagram of another information processing device according to an embodiment of the present application is shown;

[0024] Figure 11is a structural block diagram of another computer terminal according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to make the personnel in the technical field better understand the present application scheme, the technical scheme in the embodiment of the present application will be described clearly and completely in the following with reference to the drawings in the embodiment of the present application. Obviously, the described embodiment is only a part of the embodiment of the present application, not all. Based on the embodiment in the present application, all other embodiments obtained by the person skilled in the art without creative labor should belong to the scope of protection of the present application.

[0026] It should be noted that the terms "first", "second" and the like in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily indicate a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not necessarily limit to those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0027] First, some nouns or terms appearing in the description of the embodiments of the present application are applicable to the following explanations:

[0028] Knowledge graph: Knowledge graph is a structured semantic knowledge base, which is used to quickly describe the concepts and their mutual relationships in the physical world. Through effective processing, processing and integration of complex data documents, the knowledge graph obtains simple and clear "entity, relationship, entity" triples, and finally realizes the rapid response and reasoning of knowledge through the aggregation of a large amount of knowledge. Knowledge graph is widely used in intelligent search, intelligent question and answer, personalized recommendation, intelligence analysis and anti-fraud fields due to its powerful semantic processing ability and open interconnection ability.

[0029] Named entity recognition (NER): It refers to the task of identifying entities in natural language processing, and in the present application, it specifically refers to the task of identifying entities in knowledge graph. The essence of NER is a sequence labeling task. NER is an important basic tool in application fields such as information extraction, question and answer system, syntax analysis and machine translation.

[0030] Relation Extraction (RE): refers to the task of identifying the relationship between two entities. The essence of RE is a classification task.

[0031] Attribute Extraction (AE): refers to the task of extracting the attributes of entities or relationships from text.

[0032] Natural Language Pre-training Model: refers to a model that converts natural language into a vector containing semantic information through large-scale corpus training.

[0033] Masked Language Model (MLM) task: a training task commonly used when training a natural language pre-training model. The MLM task covers the expected words in a certain proportion, and uses the model to predict these covered words.

[0034] Natural Language Prompt: a method for converting a training task into a task consistent with the pre-training task of a pre-training model.

[0035] Embodiment 1

[0036] According to the embodiments of the present application, an information processing method embodiment is also provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0037] The method embodiment provided by the embodiment of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 A hardware structure block diagram of a computer terminal (or mobile device) for implementing an information processing method is shown. As shown in Figure 1 The computer terminal 10 (or mobile device 10) can include one or more processors 102 (the processor 102 can include but is not limited to a microprocessor MCU or a programmable logic device FPGA processing device), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it can also include a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which can be included as one of the ports in the BUS bus), a network interface, a power supply and / or a camera. Those skilled in the art can understand that Figure 1 The structure shown is only schematic, and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 can also include moreFigure 1 more or less components than those shown, or configured differently. Figure 1

[0038] It should be noted that the one or more processors 102 and / or other data processing circuitry described above can be generally referred to herein as "data processing circuitry". The data processing circuitry can be embodied in whole or in part as software, hardware, firmware, or any combination thereof. In addition, the data processing circuitry can be a single standalone processing module, or incorporated in whole or in part within any of the other elements of the computer terminal 10 (or mobile device). As referred to in embodiments of the present application, the data processing circuitry functions as a processor to control, for example, the selection of variable resistance terminal paths in connection with the interface.

[0039] The memory 104 can be used to store software programs and modules of application software, such as program instructions / data storage means corresponding to the information processing method of embodiments of the present application. The processor 102 can execute various functional applications and data processing, i.e. implement the information processing method described above, by running the software programs and modules stored in the memory 104. The memory 104 can include a high-speed random access memory, and can further include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 can further include a memory disposed remotely with respect to the processor 102, which can be connected to the computer terminal 10 through a network. Examples of the network can include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0040] The transmission device 106 is configured to receive or send data via a network. Examples of the network can include a wireless network provided by a communication provider of the computer terminal 10. In one example, the transmission device 106 includes a network interface controller (NIC) that can be connected to other network devices through a base station to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module configured to communicate with the Internet in a wireless manner.

[0041] The display can be, for example, a touch screen liquid crystal display (LCD) that can enable a user to interact with the user interface of the computer terminal 10 (or mobile device).

[0042] It should be noted that in some alternative embodiments, the above-described Figure 1 ​The illustrated computer device (or mobile device) can include hardware elements (including circuitry), software elements (including computer code stored on a computer readable medium), or combinations of both hardware and software elements. It should be noted that Figure 1 is merely one instance of a particular concrete example and is intended to show the types of components that can be present in the above-described computer device (or mobile device).

[0043] In the above-described operating environment, the present application provides an information processing method as shown in Figure 2 Figure 2 is a flowchart of an information processing method according to an embodiment of the present application, as shown in Figure 2 The information processing method comprises the following steps:

[0044] In step S21, first knowledge and second knowledge associated with the crop to be monitored are obtained, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored.

[0045] In step S22, the structure mode of the knowledge graph corresponding to the crop to be monitored is determined.

[0046] In step S23, the first knowledge and the second knowledge are extracted according to the structure mode, and the knowledge graph is constructed.

[0047] In step S24, the target service function associated with the crop to be monitored is provided based on the knowledge graph.

[0048] Optionally, the crop to be monitored can be an agricultural crop or a forestry crop. The agricultural crop can include food crops (such as rice, corn, beans, potatoes, wheat, etc.) and economic crops (such as oil crops, vegetable crops, medicinal crops, fruit crops, etc.) and the like. Forestry crops can include industrial raw material forest crops (such as oil tea, sapis, tung tree, lacquer tree, etc.), medicinal forest crops, spice trees, etc.

[0049] Optionally, the first knowledge can be objective knowledge associated with the crop to be monitored. For example, the objective knowledge can include scientific research data, agricultural and forestry encyclopedic knowledge, structured data in agricultural and forestry related websites, Internet of Things data and statistical analysis data thereof, etc. The objective knowledge can be obtained by manual acquisition from scientific researchers, collection from the Internet, acquisition from the storage device of the Internet of Things device, acquisition from related databases, etc.

[0050] ​Optionally, the second knowledge can be subjective knowledge associated with the crop to be monitored. For example, the subjective knowledge can include expert experience knowledge, production worker feedback data, etc. The acquisition method of the subjective knowledge can include exchanging and discussing with relevant technical experts (including online exchange and offline exchange) and recording, collecting from the history of questions and answers in the agricultural and forestry related Q&A community, etc.

[0051] In the optional embodiment described above, the structural mode of the knowledge graph corresponding to the crop to be monitored can be used to determine the entity categories contained in the knowledge graph and the relationships between the entities. The knowledge graph can be constructed by knowledge extraction of the first knowledge and the second knowledge according to the structural mode.

[0052] Optionally, the target service function can be a service function provided based on the professional knowledge associated with the crop to be monitored. For example, the target service function can include knowledge query, intelligent question and answer, planting strategy recommendation, hazard identification, hazard early warning, hazard prevention strategy recommendation, etc.

[0053] It is easy to note that the method provided by the embodiment of the present application can optimize the agricultural production process using the target service function. The agricultural production process can include processes such as planting, cultivation, irrigation, fertilization, medication, harvesting, storage, and primary processing.

[0054] Specifically, other method steps included in the information processing method can also be referred to the further introduction of the embodiments of the present application in the following, which will not be described here.

[0055] In the embodiment of the present application, first, the first knowledge and the second knowledge of the crop to be monitored are acquired, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored. The structural mode of the knowledge graph corresponding to the crop to be monitored is determined. The first knowledge and the second knowledge are extracted according to the structural mode to construct the knowledge graph, and then the target service function associated with the crop to be monitored is provided based on the knowledge graph.

[0056] It is easy to note that through the embodiment of the present application, based on the objective knowledge and the subjective knowledge associated with the crop to be monitored, the corresponding knowledge graph is constructed and the related service function is provided, which achieves the purpose of constructing the knowledge graph based on the related professional knowledge to serve the production, thereby realizing the technical effect of reducing the construction cost of the knowledge graph and the difficulty of obtaining professional knowledge, and further solving the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

[0057] It should be noted that the above information processing method provided by the present application can be applied in any application scenario related to Internet of Things monitoring, knowledge graph construction, intelligent question answering, hazard identification, hazard early warning, planting strategy recommendation, hazard prevention strategy recommendation in the fields of agriculture, forestry, natural resources and ecological environment, water conservancy, meteorology, secondary disasters, etc., but is not limited thereto.

[0058] Figure 3 is a schematic diagram of an optional knowledge graph construction process according to an embodiment of the present application, as shown in the construction of an agricultural knowledge graph, agricultural encyclopedic knowledge (equivalent to the above objective knowledge) and agricultural expert knowledge (equivalent to the above subjective knowledge) need to be obtained for the knowledge extraction process. Figure 3

[0059] Still as Figure 3 shown, the construction of an agricultural knowledge graph also includes a graph concept mode design, knowledge extraction, knowledge fusion and knowledge storage process. The specific method steps of the above process can be referred to the further introduction of Figure 3 hereinbelow, which will not be described herein again.

[0060] The information processing method of the above embodiment will be further introduced hereinbelow.

[0061] In an optional embodiment, in step S23, the first knowledge and the second knowledge are subjected to knowledge extraction according to a structure mode, and a knowledge graph is constructed, including the following method steps:

[0062] Step S231, based on the structure mode, the first knowledge and the second knowledge are subjected to knowledge extraction to obtain to-be-fused entities, to-be-fused relationships and to-be-fused attributes, wherein the structure mode is used to determine the entity categories contained in the knowledge graph and the relationships possessed between entities;

[0063] Step S232, the to-be-fused entities, to-be-fused relationships and to-be-fused attributes are subjected to knowledge fusion to obtain a fusion result;

[0064] Step S233, the fusion result is stored to obtain a knowledge graph.

[0065] In the above optional embodiment, the knowledge graph can be a knowledge graph associated with the to-be-monitored crop. The structure mode of the knowledge graph can be used to determine the entity categories contained in the knowledge graph and the relationships possessed between entities.

[0066] In the above optional embodiment, the first knowledge is objective knowledge associated with the to-be-monitored crop, and the second knowledge is subjective knowledge associated with the to-be-monitored crop. Based on the structure mode, the first knowledge and the second knowledge can be subjected to knowledge extraction to further obtain the to-be-fused entities, the to-be-fused relationships and the to-be-fused attributes.​

[0067] In the above optional embodiments, knowledge fusion is performed on the entities to be fused, the relationships to be fused, and the attributes to be fused to obtain the fusion result. This knowledge fusion process can merge identical nodes, achieving the effect of removing duplication and redundancy.

[0068] In the above optional embodiments, storing the fusion results yields the aforementioned knowledge graph. This knowledge graph can be used to record the first knowledge and the second knowledge.

[0069] Still as Figure 3 As shown, constructing an agricultural knowledge graph also includes designing a conceptual schema. Specifically, this involves designing a conceptual schema for the agricultural knowledge graph, in which entities can include: phenological periods, pests and diseases, meteorological disasters, tree management, water, fertilizer and soil management, agricultural inputs, agricultural machinery, and meteorological conditions.

[0070] Optionally, the above conceptual model may also include the relationships between the entities. For example, the entity "meteorological disaster" can be determined by the entity "meteorological conditions" and preset disaster rules, and the entities "water, fertilizer and soil management" and "agricultural inputs" can have a user-user relationship (such as the fertilization behavior in water, fertilizer and soil management can use organic fertilizer in agricultural inputs), etc.

[0071] In an optional embodiment, in step S231, knowledge extraction is performed on the first knowledge and the second knowledge to obtain the entity to be fused, the relationship to be fused, and the attribute to be fused, including the following method steps:

[0072] Step S2311: Use the knowledge extraction model to extract knowledge from the first knowledge and the second knowledge to obtain the entity to be fused, the relationship to be fused, and the attribute to be fused. The knowledge extraction model is trained by machine learning using a training dataset, which is determined by the first knowledge and the second knowledge.

[0073] In the above optional embodiments, the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored. Using the above knowledge extraction model, knowledge can be extracted from the first and second knowledge to obtain the entity to be fused, the relationship to be fused, and the attribute to be fused.

[0074] In the optional embodiments described above, the knowledge extraction model can be a model trained using a training dataset through machine learning. This training dataset can be determined by the first knowledge and the second knowledge described above. For example, the training dataset can be a combination of an objective knowledge dataset and a subjective knowledge dataset of the crop to be monitored.

[0075] Optionally, the knowledge extraction process can include entity recognition (NER), relation extraction (RE) and attribute extraction (AE).

[0076] As shown in Figure 3 , the construction of the agricultural knowledge graph further includes a knowledge extraction process. According to the graph concept pattern designed in the foregoing steps, the knowledge extraction model is used to extract knowledge from the agricultural expert knowledge (equivalent to the second knowledge described above) and the agricultural encyclopedic knowledge (equivalent to the first knowledge described above). Specifically, the knowledge extraction process can include obtaining the to-be-fused entities through entity recognition. The to-be-fused relations are obtained through relation extraction, and the to-be-fused attributes are obtained through attribute extraction. The knowledge extraction model can be obtained through machine learning training using a training data set. The machine learning training process can refer to the further description of training the knowledge extraction model below, and will not be described here.

[0077] As shown in Figure 3 , the construction of the agricultural knowledge graph further includes a knowledge extraction process. According to the graph concept pattern designed in the foregoing steps, the knowledge extraction model is used to extract knowledge from the agricultural expert knowledge (equivalent to the second knowledge described above) and the agricultural encyclopedic knowledge (equivalent to the first knowledge described above). Specifically, the knowledge extraction process can include obtaining the to-be-fused entities through entity recognition. The to-be-fused relations are obtained through relation extraction, and the to-be-fused attributes are obtained through attribute extraction. The knowledge extraction model can be obtained through machine learning training using a training data set. The machine learning training process can refer to the further description of training the knowledge extraction model below, and will not be described here.

[0078] In an optional embodiment, the information processing method further includes the following method steps:

[0079] Step S24, extracting sample annotation data from the first knowledge and the second knowledge;

[0080] Step S25, obtaining a training data set based on cross-domain data, sample annotation data and a preset template, wherein the cross-domain data is knowledge graph annotation data in the remaining fields other than the application field corresponding to the to-be-monitored crop, and the preset template is used to fill in the sample annotation data and the cross-domain data;

[0081] Step S26, refining an initial label set corresponding to the sample annotation data to obtain a target label set;

[0082] Step S27, training the natural language pre-training model using the training data set and the target label set to obtain a knowledge extraction model.

[0083] In the above optional embodiment, the first knowledge is objective knowledge associated with the to-be-monitored crop, and the second knowledge is subjective knowledge associated with the to-be-monitored crop. The sample annotation data can be extracted from the first knowledge and the second knowledge.

[0084] In the optional embodiment, the cross-domain data can be knowledge graph annotation data in other fields than the application field corresponding to the crop to be monitored. For example, when the crop to be monitored is a food crop (such as wheat), the cross-domain data can be knowledge graph annotation data related to economic crops (such as vegetables, Chinese medicine, cotton, etc.) or forestry crops (such as tea-oil camellia, Chinese sycamore, tung tree, and lacquer tree, etc.). For another example, when establishing an agricultural knowledge graph, the cross-domain data can be knowledge graph annotation data in the fields of finance, medicine, and art, etc.

[0085] In the optional embodiment, the preset template can be used to fill the sample annotation data and the cross-domain data. The preset template can be preset by a technician according to an actual application scenario.

[0086] In the optional embodiment, the sample annotation data can be data enhanced by using the cross-domain data and the preset template, and then a training data set is obtained. The preset template can be a general template preset or a specified template preset.

[0087] In the optional embodiment, the initial label set corresponding to the sample annotation data can be refined to obtain the target label set. For example, in the initial label set corresponding to the sample annotation data, the label “agricultural materials” can be refined into labels such as “fertilizer” and “pesticide”. Further, the label “fertilizer” can be refined into labels such as “phosphorus fertilizer”, “potassium fertilizer”, and “boron fertilizer”. Finally, the target label set can include the initial label set and all refined labels.

[0088] In the optional embodiment, the natural language pre-training model can be trained by using the training data set and the target label set, and then the knowledge extraction model is obtained. The knowledge extraction model can be used for knowledge extraction of the first knowledge and the second knowledge of the crop to be monitored, to obtain the to-be-fused entity, the to-be-fused relationship, and the to-be-fused attribute.

[0089] Figure 4 is a schematic diagram of an optional knowledge extraction model training process according to an embodiment of the present application, as shown in Figure 4 The training process of the knowledge extraction model can include agricultural small sample annotation. A small amount of knowledge is extracted from agricultural expert knowledge (equivalent to the second knowledge) and agricultural encyclopedic knowledge (equivalent to the first knowledge) for knowledge annotation. Specifically, the knowledge annotation can include entity annotation, relationship annotation, and attribute annotation.

[0090] Still as shown in Figure 4 The training process of the knowledge extraction model can also include data enhancement. Cross-domain data and a general template (equivalent to the preset template) are used for data enhancement. Specifically, the data enhancement process can include the following method steps:

[0091] Step S2501, obtaining annotated data in fields where knowledge graphs of finance, medicine, etc. are widely applied as cross-field data;

[0092] Step S2502, adding the cross-field data and sample annotated data to the training data set;

[0093] Step S2503, constructing a data filling universal template;

[0094] Step S2504, filling the annotated entities, relationships, and attributes in the cross-field data and sample annotated data into the universal template to automatically construct a batch of data, and adding the automatically constructed batch of data to the training data set.

[0095] Optionally, the process of automatically constructing a batch of data can be: migrating the knowledge extraction strategy learned by the model from the cross-field data to the agricultural field. For example: based on the template "XXX how to do", the medical field can automatically construct "alcohol disinfection how to do", and the agricultural field can automatically construct "clear garden how to do".

[0096] It should be noted that, in the knowledge extraction model training process, due to the lack of annotated agricultural field data, relatively mature cross-field data is used for data enhancement. Through the data enhancement process in the knowledge extraction model training process, the knowledge extraction effect of the model trained can be enhanced.

[0097] Still as Figure 4 shown, the training process of the knowledge extraction model can also include the Prompt method, label refinement, and model training process. These processes can refer to the further introduction of training the knowledge extraction model below, and will not be repeated here.

[0098] In an optional embodiment, in step S27, the training data set and the target label set are used to train the natural language pre-training model to obtain the knowledge extraction model, including the following method steps:

[0099] Step S271, determining the knowledge extraction task by using the training data set;

[0100] Step S272, converting the knowledge extraction task into a pre-training task of the natural language pre-training model;

[0101] Step S273, training the natural language pre-training model by using the pre-training task and the target label set to obtain the knowledge extraction model.

[0102] In the above optional embodiments, the training dataset can be a dataset obtained by data augmentation of the sample labeled data using the aforementioned cross-domain data and the aforementioned preset template. The sample labeled data can be data extracted from the first knowledge and the second knowledge. Using this training dataset, the aforementioned knowledge extraction task can be determined.

[0103] In the above optional embodiments, the knowledge extraction task is transformed into a pre-training task for a natural language pre-training model. This pre-training task can be a masked language model (MLM) task.

[0104] In the above optional embodiments, the natural language pre-training model can be trained using the pre-training task of the natural language pre-training model and the target label set to obtain the knowledge extraction model. The target label set can be a label set obtained by refining the initial label set corresponding to the sample labeled data.

[0105] Still as Figure 4 As shown, the training process of the knowledge extraction model can also include a Prompt method. This Prompt method can be used to address the inconsistency between the pre-training tasks of the natural language pre-training model and the training tasks of the knowledge extraction model. The natural language pre-training model can use the MLM task as one of its pre-training tasks. In this case, a fixed pattern can be added to the data in the knowledge extraction model, overwriting the words indicating categories in the data, thereby training the knowledge extraction model to continue performing the MLM task.

[0106] For example, the data "January is the dormant period for kiwifruit trees." can be changed to "January is the dormant period for kiwifruit trees. Dormancy is a phenological period." Then, the word "phenological period" is overwritten, and a knowledge extraction model is used to predict the overwritten word. Through this process, the entity recognition task can be transformed into an MLM task.

[0107] It should be noted that by using the Prompt method, the semantic information learned by the natural language pre-trained model can be utilized, thereby improving the learning effect of the knowledge extraction model when learning with less sample data.

[0108] Figure 5 This is a schematic diagram of an optional training method for a natural language pre-training model according to an embodiment of the present invention. Unlike the training methods provided by the prior art, by using the Prompt method, the difference between the pre-training task of the natural language pre-training model and the training task of the knowledge extraction model is avoided, and there is no need to add a knowledge extraction model after the pre-training model.

[0109] Still as Figure 5As shown, according to the Prompt method, a mapping relationship is constructed between the knowledge extraction task and the pre-training task of the natural language pre-training model, and then the pre-training model is further trained to complete the knowledge extraction task. Through this method, the semantic information learned by the natural language pre-training model can be fully utilized, thereby solving the problem of insufficient training caused by using less sample data (equivalent to small sample labeled data in Figure 5 ).

[0110] In an optional embodiment, in step S26, the initial label set corresponding to the sample labeled data is refined to obtain the target label set, including the following method steps:

[0111] Step S261, the initial label set is decomposed to obtain an intermediate label set, wherein the initial label set is a label set corresponding to a knowledge graph, and the intermediate label set is a label set corresponding to a knowledge extraction model;

[0112] Step S262, the intermediate label set is extracted by the knowledge extraction model to obtain an extraction result;

[0113] Step S263, the extraction result is merged to obtain the target label set.

[0114] In the above optional embodiment, the initial label set is a label set corresponding to a knowledge graph, which can be a label set corresponding to sample labeled data extracted from the first knowledge and the second knowledge of the crop to be monitored. The initial label set can be decomposed to obtain the above-mentioned intermediate label set. The intermediate label set can be a label set corresponding to a knowledge extraction model.

[0115] In the above optional embodiment, the intermediate label set is extracted by the above-mentioned knowledge extraction model to obtain the above-mentioned extraction result. The knowledge extraction model can be a model obtained by training a natural language pre-training model with a training data set and a target label set. The extraction result can be merged to obtain the target label set.

[0116] Figure 6 is a schematic diagram of an optional label refinement process according to an embodiment of the present application, as shown in Figure 6 The graph label set can include labels such as "agricultural materials" and "agricultural machinery", and the graph label set can be refined into a knowledge extraction label set when the graph label set is further designed. Specifically, each label of the multiple labels in the graph label set is refined, and the refined label is added to the knowledge extraction label set. For example, the label "agricultural materials" is refined into "fertilizer", "organic fertilizer", "pesticide", etc.

[0117] Still as Figure 6As shown, knowledge extraction can be performed on the knowledge extraction label set to obtain an extraction result. Specifically, for each label of the plurality of labels in the knowledge extraction label set, a corresponding entity can be extracted, and the extracted entity can be added to the extraction result. For example, the label "fertilizer" can extract the entity "boron fertilizer", the label "organic fertilizer" can extract the entity "cow dung", and the label "pesticide" can extract the entity "stone sulfur mixture", and the like.

[0118] Still as shown Figure 6 As shown, the extraction result can be merged to obtain all entities corresponding to each label of the plurality of labels in the graph label set. For example, the entities "boron fertilizer", "cow dung", "stone sulfur mixture", and the like in the extraction result can be merged to obtain all entities corresponding to the label "agricultural materials".

[0119] It should be noted that through the above label refinement process, a better semantic connection between the labels in the knowledge graph and the data can be established, and thus the semantic similarity can be more fully utilized to help the knowledge extraction model to perform entity mining. For example, through label refinement, the entity "boron fertilizer" can be associated with the label "fertilizer" in semantics, so that the entity "boron fertilizer" is more easily extracted, and thus the problem of insufficient training samples caused by training using less sample data can be solved.

[0120] It should be noted that the focus of the present application is to extract professional knowledge in the case of less sample data, and to use the Prompt method to improve the knowledge extraction effect in the knowledge extraction model training process. However, the method for constructing a knowledge graph based on professional knowledge can be any related and implementable construction method.

[0121] It is easy to note that the knowledge graph constructed using the prior art generally stores general knowledge, which is not fine enough and has little guiding effect on agricultural and forestry workers, and a large amount of manual marking is required in the construction process of the knowledge graph, which is too high in cost. However, compared with the method provided in the related art, the method provided in the embodiment of the present application can be trained using less sample data, solving the problem of high labor cost in constructing a knowledge graph, and in addition, the knowledge stored in the knowledge graph is high in precision and strong in comprehensiveness, and can accurately guide agricultural and forestry production work.

[0122] It is easy to note that one of the beneficial effects of the method provided in the embodiment of the present application can be that a framework for constructing a knowledge graph based on less sample data is constructed, and the cost of constructing a knowledge graph is reduced. One of the beneficial effects of the method provided in the embodiment of the present application can be that the knowledge stored in the knowledge graph is high in precision, and can provide fine agricultural and forestry guidance.

[0123] In an optional embodiment, a graphical user interface is provided by the terminal device, and content displayed by the graphical user interface at least partially contains a knowledge graph service scenario. The information processing method further includes the following method steps:

[0124] Step S31, a plurality of types of candidate crops and a plurality of candidate service functions corresponding to each type of candidate crop in the plurality of types of candidate crops are displayed in the graphical user interface;

[0125] Step S32, in response to a first touch operation on the graphical user interface, a crop to be monitored is determined from the plurality of types of candidate crops;

[0126] Step S33, in response to a second touch operation on the graphical user interface, a target service function is determined for the crop to be monitored from the plurality of candidate service functions;

[0127] Step S34, in response to a third touch operation on the graphical user interface, service information corresponding to the target service function is displayed in the graphical user interface.

[0128] In the above optional embodiment, the user can at least partially obtain the above-mentioned knowledge graph service scenario through the content of the graphical user interface displayed by the terminal device. The graphical user interface can display a plurality of types of candidate crops, and can also display a plurality of candidate service functions corresponding to each type of candidate crop in the plurality of types of candidate crops. The plurality of candidate service functions can be knowledge query, intelligent question and answer, planting strategy recommendation, hazard identification, hazard warning, hazard prevention strategy recommendation, etc.

[0129] Optionally, the user can perform a first touch operation on the graphical user interface. The user can determine the crop to be monitored by touching one of the plurality of types of candidate crops in the graphical user interface.

[0130] Optionally, the user can also perform a second touch operation on the graphical user interface. The user can determine the target service function corresponding to the crop to be monitored by touching at least one of the plurality of candidate service functions displayed in the graphical user interface.

[0131] Optionally, the user can also perform a third touch operation on the graphical user interface. The user can touch a "display" button, a "recommend" button, a "determine" button, or a "show" button, etc. in the graphical user interface, so as to display the service information corresponding to the target service function in the graphical user interface. The service information can be knowledge obtained by query, answer to intelligent question and answer, recommended planting strategy, identified hazard category, hazard warning message, recommended hazard prevention strategy, etc.

[0132] In particular, the first touch operation, the second touch operation and the third touch operation can all be operations of a user touching a display screen of the terminal device with a finger and touching the terminal device. The touch operation can include single-point touch and multi-point touch, wherein the touch operation of each touch point can include clicking, long pressing, double-clicking, swiping, etc. The first touch operation, the second touch operation and the third touch operation can also be touch operations realized by a mouse, a keyboard or the like input device.

[0133] In an optional embodiment, the information processing method further comprises the following method steps:

[0134] In step S35, an operation feedback result corresponding to the service information is obtained.

[0135] In step S36, in response to an editing operation on the service information, the service information is optimized based on the operation feedback result.

[0136] In the above optional embodiment, the service information can be information corresponding to the target service function. The operation feedback result corresponding to the service information is obtained. The user can perform an editing operation on the service information (such as modifying part of the data in the service information, or redefining all the information in the service information).

[0137] For example, the service information can be knowledge obtained through query, an answer to an intelligent question and answer, a recommended planting strategy, a recognized hazard category, a hazard warning message, a recommended hazard prevention strategy, etc. The operation feedback result corresponding to the service information can be used to feed back the operation of the user on the service information, which can include acceptance, likes (such as clicking the “useful” button), opposition (such as clicking the “useless” button), editing, etc.

[0138] For example, when the service information is a recommended planting strategy, the editing operation of the user on the recommended planting strategy can be changing part of the recommended planting strategy, adding custom recommendation content to the planting strategy, or deleting the recommended planting strategy locally.

[0139] Optionally, when the user edits the service information, the service information can be optimized based on the operation feedback result corresponding to the service information.

[0140] It is easy to note that, by the method provided in this embodiment, the user can edit the service information pushed by the system, which is conducive to the user to use the method more flexibly and conveniently, and is also conducive to adjusting the knowledge extraction model training method or the graph concept mode in the knowledge graph construction process of the method according to the editing operation and the operation feedback result of the user.

[0141] In the above running environment, the present application provides a method for constructing a knowledge graph, comprising the following steps: Figure 7An information processing method is shown. Figure 7 is a flowchart of another information processing method according to an embodiment of the application, as shown in Figure 7 The information processing method comprises:

[0142] In step S71, first kiwi knowledge and second kiwi knowledge of the to-be-monitored kiwi are acquired, wherein the first kiwi knowledge is objective kiwi knowledge associated with the to-be-monitored kiwi, and the second kiwi knowledge is subjective kiwi knowledge associated with the to-be-monitored kiwi.

[0143] In step S72, a kiwi knowledge graph corresponding to the to-be-monitored kiwi is constructed using the first kiwi knowledge and the second kiwi knowledge.

[0144] In step S73, a target kiwi service function associated with the to-be-monitored kiwi is provided based on the kiwi knowledge graph.

[0145] Optionally, the first kiwi knowledge can be objective kiwi knowledge associated with the to-be-monitored kiwi. For example, the objective kiwi knowledge can include scientific research data, kiwi encyclopedic knowledge, structured data in kiwi-related websites, Internet of Things data, and statistical analysis data thereof, etc. The acquisition method of the objective kiwi knowledge can include manual acquisition from scientific researchers, collection from the Internet, acquisition from storage devices of Internet of Things devices, acquisition from related databases, etc.

[0146] Optionally, the second kiwi knowledge can be subjective kiwi knowledge associated with the to-be-monitored kiwi. For example, the subjective kiwi knowledge can include kiwi expert experience knowledge, kiwi production worker feedback data, etc. The acquisition method of the subjective kiwi knowledge can include recording through technical personnel and related technical experts exchanging and discussing (including online exchange and discussion and offline exchange and discussion), and collecting from the history of questions and answers in agriculture-related question and answer communities, etc.

[0147] Optionally, the target kiwi service function can be a kiwi service function provided based on professional kiwi knowledge associated with the to-be-monitored kiwi. For example, the target kiwi service function can include kiwi knowledge query, kiwi intelligent question and answer, kiwi planting strategy recommendation, kiwi hazard identification, kiwi hazard early warning, kiwi hazard prevention and control strategy recommendation, etc.

[0148] In the embodiment of the present application, first kiwi fruit knowledge and second kiwi fruit knowledge of the to-be-monitored kiwi fruit are acquired first, wherein the first kiwi fruit knowledge is objective kiwi fruit knowledge associated with the to-be-monitored kiwi fruit, and the second kiwi fruit knowledge is subjective kiwi fruit knowledge associated with the to-be-monitored kiwi fruit, a method of constructing a kiwi fruit knowledge graph corresponding to the to-be-monitored kiwi fruit by using the first kiwi fruit knowledge and the second kiwi fruit knowledge is adopted, and then a target kiwi fruit service function associated with the to-be-monitored kiwi fruit is provided based on the kiwi fruit knowledge graph.

[0149] It is easy to note that, by the embodiment of the present application, the corresponding kiwi fruit knowledge graph is constructed based on the objective kiwi fruit knowledge and the subjective kiwi fruit knowledge associated with the to-be-monitored kiwi fruit, and the related kiwi fruit service function is provided, so as to achieve the purpose of constructing the kiwi fruit knowledge graph based on the related professional kiwi fruit knowledge to serve the production of kiwi fruit, thereby realizing the technical effect of reducing the construction cost of the kiwi fruit knowledge graph and the difficulty of obtaining professional kiwi fruit knowledge, and further solving the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

[0150] It should be noted that the above information processing method provided by the present application can be applied in any application scenario related to kiwi fruit in the field of kiwi fruit, such as Internet of Things monitoring, knowledge graph construction, intelligent question answering, hazard identification, hazard early warning, planting strategy recommendation, and hazard prevention strategy recommendation.

[0151] In an optional embodiment, a graphical user interface is provided by the terminal device, and the content displayed by the graphical user interface at least partially contains a kiwi fruit knowledge graph service scene, and the information processing method further includes the following method steps:

[0152] Step S74, a plurality of types of candidate agricultural crops and a plurality of candidate agricultural service functions corresponding to each type of candidate agricultural crop in the plurality of types of candidate agricultural crops are displayed in the graphical user interface;

[0153] Step S75, in response to a first touch operation acting on the graphical user interface, a to-be-monitored kiwi fruit is determined from the plurality of types of candidate kiwi fruit;

[0154] Step S76, in response to a second touch operation acting on the graphical user interface, a target kiwi fruit service function is determined for the to-be-monitored kiwi fruit from the plurality of candidate kiwi fruit service functions;

[0155] Step S77, in response to a third touch operation acting on the graphical user interface, kiwi fruit service information corresponding to the target kiwi fruit service function is displayed in the graphical user interface.

[0156] In the optional embodiment described above, the user can obtain the kiwi knowledge graph service scenario described above at least partially through the graphical user interface content displayed by the terminal device. The graphical user interface can display multiple types of candidate kiwis and can also display multiple candidate kiwi service functions corresponding to each type of candidate kiwi in the multiple types of candidate kiwis. The multiple candidate agricultural service functions can be agricultural knowledge query, agricultural intelligent question answering, agricultural planting strategy recommendation, agricultural hazard identification, agricultural hazard early warning, agricultural hazard prevention and control strategy recommendation, etc.

[0157] Optionally, the user can perform a first touch operation on the graphical user interface. The user can determine the to-be-monitored kiwi by touching one of the multiple types of candidate kiwis in the graphical user interface.

[0158] Optionally, the user can also perform a second touch operation on the graphical user interface. The user can determine the target kiwi service function corresponding to the to-be-monitored kiwi by touching at least one of the multiple candidate kiwi service functions displayed in the graphical user interface. The target kiwi service function can be kiwi knowledge query, kiwi intelligent question answering, kiwi planting strategy recommendation, kiwi hazard identification, kiwi hazard early warning, kiwi hazard prevention and control strategy recommendation, etc.

[0159] Optionally, the user can also perform a third touch operation on the graphical user interface. The user can touch the "display" button, "recommend" button, "determine" button, or "show" button, etc. in the graphical user interface to cause the graphical user interface to display the kiwi service information corresponding to the target kiwi service function. The kiwi service information can be the queried kiwi knowledge, the answer to the kiwi intelligent question answering, the recommended kiwi planting strategy, the identified kiwi hazard category, the kiwi hazard early warning message, the recommended kiwi hazard prevention and control strategy, etc.

[0160] In particular, the first touch operation, the second touch operation, and the third touch operation described above can all be operations of the user touching the display screen of the terminal device with a finger. The touch operation can include single-point touch, multi-point touch, wherein the touch operation of each touch point can include clicking, long pressing, double-clicking, swiping, etc. The first touch operation, the second touch operation, and the third touch operation described above can also be touch operations realized through a mouse, a keyboard, etc.

[0161] In the running environment described above, the present application provides an information processing method as shown in Figure 8 . Figure 8 is a flowchart of another information processing method according to an embodiment of the present application, as shown in Figure 8 , the information processing method comprises:

[0162] Step S81, obtaining first fruit tree knowledge and second fruit tree knowledge of the to-be-monitored fruit tree, wherein the first fruit tree knowledge is objective fruit tree knowledge associated with the to-be-monitored fruit tree, and the second fruit tree knowledge is subjective fruit tree knowledge associated with the to-be-monitored fruit tree;

[0163] Step S82, constructing a fruit tree knowledge graph corresponding to the to-be-monitored fruit tree by using the first fruit tree knowledge and the second fruit tree knowledge;

[0164] Step S83, providing a target fruit tree service function associated with the to-be-monitored fruit tree based on the fruit tree knowledge graph.

[0165] Optionally, the to-be-monitored fruit tree can include a pome fruit tree, a stone fruit tree, a berry fruit tree, a nut fruit tree, a persimmon and jujube fruit tree, etc.

[0166] Optionally, the first fruit tree knowledge can be objective fruit tree knowledge associated with the to-be-monitored fruit tree. For example, the objective fruit tree knowledge can include scientific research data, fruit tree encyclopedic knowledge, structured data in a fruit tree related website, Internet of Things data and statistical analysis data thereof, etc. The objective fruit tree knowledge can be obtained in the following ways: manually obtained from a scientific researcher, collected from the Internet, obtained from a storage device of an Internet of Things device, obtained from a related database, etc.

[0167] Optionally, the second fruit tree knowledge can be subjective fruit tree knowledge associated with the to-be-monitored fruit tree. For example, the subjective fruit tree knowledge can include fruit tree expert experience knowledge, fruit tree producer feedback data, etc. The subjective fruit tree knowledge can be obtained in the following ways: through technical personnel and related technical experts exchange and discussion (including online exchange and discussion and offline exchange and discussion) and recording, collected from a fruit tree related question and answer community, etc.

[0168] Optionally, the target fruit tree service function can be a fruit tree service function provided based on professional fruit tree knowledge associated with the to-be-monitored fruit tree. For example, the target fruit tree service function can include fruit tree knowledge query, fruit tree intelligent question and answer, fruit tree planting strategy recommendation, fruit tree hazard identification, fruit tree hazard early warning, fruit tree hazard prevention and control strategy recommendation, etc.

[0169] In the embodiment of the application, first fruit tree knowledge and second fruit tree knowledge of a to-be-monitored fruit tree are first obtained, wherein the first fruit tree knowledge is objective fruit tree knowledge associated with the to-be-monitored fruit tree, and the second fruit tree knowledge is subjective fruit tree knowledge associated with the to-be-monitored fruit tree. A fruit tree knowledge graph corresponding to the to-be-monitored fruit tree is constructed by using the first fruit tree knowledge and the second fruit tree knowledge, and then a target fruit tree service function associated with the to-be-monitored fruit tree is provided based on the fruit tree knowledge graph.

[0170] It is easy to note that, through the embodiment of the present application, based on the objective fruit tree knowledge and the subjective fruit tree knowledge associated with the to-be-monitored fruit tree, the corresponding fruit tree knowledge graph is constructed and the related fruit tree service function is provided, the purpose of constructing the fruit tree knowledge graph based on the related professional fruit tree knowledge to serve the fruit tree production is achieved, so that the technical effect of reducing the construction cost of the fruit tree knowledge graph and the difficulty of obtaining professional fruit tree knowledge is realized, and then the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art are solved.

[0171] It should be noted that the above information processing method provided by the present application can be applied in any application scenario related to Internet of Things monitoring, knowledge graph construction, intelligent question answering, hazard identification, hazard early warning, planting strategy recommendation, hazard prevention strategy recommendation in the forestry field, but is not limited to.

[0172] It should be noted that, for the foregoing method embodiments, in order to simply describe, they are all expressed as a series of action combinations, but those skilled in the art should know that the present application is not limited by the action sequence described, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should know that the embodiments described in the specification all belong to preferred embodiments, and the actions and modules involved are not necessarily essential to the present application.

[0173] From the above description of the embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and the necessary general hardware platform, of course, it can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server, or network device, etc.) execute the method described in each embodiment of the present application.

[0174] Embodiment 2

[0175] According to the embodiment of the present application, a device embodiment for implementing the above information processing method is also provided, Figure 9 is a structural schematic diagram of an information processing device according to the embodiment of the present application, as Figure 9 shown, the device comprises: an acquisition module 901, a determination module 902, a construction module 903, a service module 904, wherein,

[0176] The acquisition module 901 is used to acquire first knowledge and second knowledge of the crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; the determination module 902 is used to determine the structural pattern of the knowledge graph corresponding to the crop to be monitored; the construction module 903 is used to extract knowledge from the first knowledge and the second knowledge according to the structural pattern and construct the knowledge graph; and the service module 904 is used to provide target service functions associated with the crop to be monitored based on the knowledge graph.

[0177] Optionally, the aforementioned construction module 903 is further configured to: extract knowledge from the first knowledge and the second knowledge based on the structural pattern to obtain entities to be fused, relationships to be fused, and attributes to be fused, wherein the structural pattern is used to determine the entity categories contained in the knowledge graph and the relationships between entities; perform knowledge fusion on the entities to be fused, relationships to be fused, and attributes to be fused to obtain the fusion result; and store the fusion result to obtain the knowledge graph.

[0178] Optionally, the aforementioned building module 903 is further configured to: extract knowledge from the first knowledge and the second knowledge using a knowledge extraction model to obtain the entity to be fused, the relationship to be fused, and the attribute to be fused, wherein the knowledge extraction model is trained using a training dataset through machine learning, and the training dataset is determined by the first knowledge and the second knowledge.

[0179] Optionally, Figure 10 This is a schematic diagram of the structure of another information processing device according to an embodiment of the present invention, such as... Figure 10 As shown, the device includes, in addition to Figure 9 In addition to all the modules shown, it also includes: a training module 905, used to extract sample labeled data from the first knowledge and the second knowledge; to obtain a training dataset based on cross-domain data, sample labeled data and preset templates, wherein the cross-domain data is knowledge graph labeled data in other domains besides the application domain corresponding to the crop to be monitored, and the preset templates are used to fill the sample labeled data and cross-domain data; to refine the initial label set corresponding to the sample labeled data to obtain the target label set; and to train the natural language pre-trained model using the training dataset and the target label set to obtain the knowledge extraction model.

[0180] Optionally, the training module 905 is further configured to: determine the knowledge extraction task using the training dataset; transform the knowledge extraction task into a pre-training task for a natural language pre-training model; and train the natural language pre-training model using the pre-training task and the target label set to obtain the knowledge extraction model.

[0181] Optionally, the training module 905 is further configured to: decompose an initial label set to obtain an intermediate label set, wherein the initial label set is a label set corresponding to the knowledge graph, and the intermediate label set is a label set corresponding to the knowledge extraction model; perform knowledge extraction on the intermediate label set by using the knowledge extraction model to obtain an extraction result; and merge the extraction result to obtain the target label set.

[0182] It should be noted that the obtaining module 901, the determining module 902, the constructing module 903, and the service module 904 correspond to steps S21 to S24 in Embodiment 1, and the four modules have the same instances and application scenarios as the corresponding steps, but are not limited to the content disclosed in Embodiment 1. It should be noted that the modules as part of the apparatus can run in the computer terminal 10 provided in Embodiment 1.

[0183] In the embodiment of the application, first, the obtaining module is used to obtain first knowledge and second knowledge associated with the crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; the determining module is used to determine a structure mode of a knowledge graph corresponding to the crop to be monitored; the constructing module is used to perform knowledge extraction on the first knowledge and the second knowledge according to the structure mode, and construct the knowledge graph; and the service module is used to provide a target service function associated with the crop to be monitored based on the knowledge graph.

[0184] It can be easily noted that, by the embodiment of the application, the objective knowledge and the subjective knowledge associated with the crop to be monitored are used to construct the corresponding knowledge graph and provide the related service function, so that the purpose of constructing the knowledge graph based on the related professional knowledge to serve the production is achieved, thereby achieving the technical effect of reducing the construction cost of the knowledge graph and the difficulty of obtaining the professional knowledge, and further solving the technical problems of the processing method based on the artificial acquisition of the professional knowledge in the related art, i.e., high cost, great difficulty, and low efficiency.

[0185] It should be noted that the preferred embodiments of the present embodiment can refer to the related description in Embodiment 1, which will not be repeated here.

[0186] Embodiment 3

[0187] According to the embodiments of the present application, an embodiment of an electronic device is also provided, which can be any one of the computing devices in the computing device group. The electronic device comprises a processor and a memory, wherein:

[0188] The memory is connected with the processor and is used to provide the processor with instructions for processing the following steps: obtaining first knowledge and second knowledge of a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structure mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode to construct the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0189] In the embodiment of the application, first knowledge and second knowledge of a crop to be monitored are obtained, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored, a structure mode of a knowledge graph corresponding to the crop to be monitored is determined, knowledge extraction is performed on the first knowledge and the second knowledge according to the structure mode to construct the knowledge graph, and a target service function associated with the crop to be monitored is provided based on the knowledge graph.

[0190] It is easy to note that, by the embodiment of the application, the corresponding knowledge graph is constructed based on the objective knowledge and the subjective knowledge associated with the crop to be monitored, and the related service function is provided, so that the purpose of constructing the knowledge graph based on the related professional knowledge to serve the production is achieved, thereby realizing the technical effect of reducing the construction cost of the knowledge graph and the difficulty of obtaining the professional knowledge, and further solving the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

[0191] It should be noted that the preferred embodiments of the present embodiment can refer to the related description in Embodiment 1, which will not be repeated here.

[0192] Embodiment 4

[0193] The embodiment of the application can provide a computer terminal, which can be any computer terminal device in a computer terminal group. Alternatively, in the present embodiment, the computer terminal can be replaced by a mobile terminal or other terminal device.

[0194] Alternatively, in the present embodiment, the computer terminal can be located in at least one network device of a plurality of network devices of a computer network.

[0195] In the present embodiment, the computer terminal can execute program codes of the following steps in the information processing method: obtaining first knowledge and second knowledge of a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structure mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode to construct the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0196] Optionally, Figure 11 is another structural block diagram of a computer terminal according to an embodiment of the present application, as shown in the figure, the computer terminal can include one or more (only one is shown in the figure) processors 122, memories 124, and peripheral interfaces 126. Figure 11

[0197] The memory can be used to store software programs and modules, such as program instructions / modules corresponding to the information processing method and device in the embodiments of the present application. The processor executes various functions and data processing by running the software programs and modules stored in the memory, that is, implements the above-mentioned information processing method. The memory can include a high-speed random access memory, and can also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some examples, the memory can further include a memory remotely arranged with respect to the processor, which can be connected to the computer terminal through a network. Examples of the above-mentioned network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof.

[0198] The processor can call the information and application programs stored in the memory through the transmission device to perform the following steps: obtaining first knowledge and second knowledge of the crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structural mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structural mode to construct the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0199] Optionally, the above-mentioned processor can further execute program codes of the following steps: based on the structural mode, performing knowledge extraction on the first knowledge and the second knowledge to obtain a to-be-fused entity, a to-be-fused relationship, and a to-be-fused attribute, wherein the structural mode is used to determine the entity categories contained in the knowledge graph and the relationships between the entities; performing knowledge fusion on the to-be-fused entity, the to-be-fused relationship, and the to-be-fused attribute to obtain a fusion result; and storing the fusion result to obtain the knowledge graph.

[0200] Optionally, the above-mentioned processor can further execute program codes of the following steps: using a knowledge extraction model to perform knowledge extraction on the first knowledge and the second knowledge to obtain a to-be-fused entity, a to-be-fused relationship, and a to-be-fused attribute, wherein the knowledge extraction model is obtained by machine learning training using a training data set, and the training data set is determined by the first knowledge and the second knowledge.

[0201] ​Optionally, the processor can further execute program codes of the following steps: extracting sample annotation data from the first knowledge and the second knowledge; obtaining a training data set based on cross-domain data, the sample annotation data, and a preset template, wherein the cross-domain data is knowledge graph annotation data in remaining domains other than an application domain corresponding to the crop to be monitored, and the preset template is used to fill in the sample annotation data and the cross-domain data; refining an initial label set corresponding to the sample annotation data to obtain a target label set; and training the natural language pre-training model using the training data set and the target label set to obtain the knowledge extraction model.

[0202] Optionally, the processor can further execute program codes of the following steps: determining a knowledge extraction task by using the training data set; converting the knowledge extraction task into a pre-training task of the natural language pre-training model; and training the natural language pre-training model using the pre-training task and the target label set to obtain the knowledge extraction model.

[0203] Optionally, the processor can further execute program codes of the following steps: decomposing an initial label set to obtain an intermediate label set, wherein the initial label set is a label set corresponding to the knowledge graph, and the intermediate label set is a label set corresponding to the knowledge extraction model; performing knowledge extraction on the intermediate label set by using the knowledge extraction model to obtain an extraction result; and merging the extraction result to obtain the target label set.

[0204] Optionally, the processor can further execute program codes of the following steps: displaying a plurality of types of candidate crops and a plurality of candidate service functions corresponding to each type of candidate crop in the plurality of types of candidate crops in the graphical user interface; determining the crop to be monitored from the plurality of types of candidate crops in response to a first touch operation acting on the graphical user interface; determining a target service function for the crop to be monitored from the plurality of candidate service functions in response to a second touch operation acting on the graphical user interface; and displaying service information corresponding to the target service function in the graphical user interface in response to a third touch operation acting on the graphical user interface.

[0205] Optionally, the processor can further execute program codes of the following steps: obtaining an operation feedback result corresponding to the service information; and optimizing the service information based on the operation feedback result in response to an editing operation acting on the service information.

[0206] The processor can call information and application programs stored in the memory through the transmission device to execute the following steps: obtaining first kiwi fruit knowledge and second kiwi fruit knowledge of the to-be-monitored kiwi fruit, wherein the first kiwi fruit knowledge is objective kiwi fruit knowledge associated with the to-be-monitored kiwi fruit, and the second kiwi fruit knowledge is subjective kiwi fruit knowledge associated with the to-be-monitored kiwi fruit; constructing a kiwi fruit knowledge graph corresponding to the to-be-monitored kiwi fruit by using the first kiwi fruit knowledge and the second kiwi fruit knowledge; and providing a target kiwi fruit service function associated with the to-be-monitored kiwi fruit based on the kiwi fruit knowledge graph.

[0207] Optionally, the processor can further execute program codes of the following steps: displaying a plurality of types of candidate agricultural crops and a plurality of candidate agricultural service functions corresponding to each type of candidate agricultural crop in the graphical user interface; determining the to-be-monitored kiwi fruit from the plurality of types of candidate kiwi fruit in response to a first touch operation on the graphical user interface; determining the target kiwi fruit service function for the to-be-monitored kiwi fruit from the plurality of candidate kiwi fruit service functions in response to a second touch operation on the graphical user interface; and displaying kiwi fruit service information corresponding to the target kiwi fruit service function in the graphical user interface in response to a third touch operation on the graphical user interface.

[0208] The processor can call information and application programs stored in the memory through the transmission device to execute the following steps: obtaining first fruit tree knowledge and second fruit tree knowledge of the to-be-monitored fruit tree, wherein the first fruit tree knowledge is objective fruit tree knowledge associated with the to-be-monitored fruit tree, and the second fruit tree knowledge is subjective fruit tree knowledge associated with the to-be-monitored fruit tree; constructing a fruit tree knowledge graph corresponding to the to-be-monitored fruit tree by using the first fruit tree knowledge and the second fruit tree knowledge; and providing a target fruit tree service function associated with the to-be-monitored fruit tree based on the fruit tree knowledge graph.

[0209] In the embodiment of the present application, first knowledge and second knowledge of the to-be-monitored crop are first obtained, wherein the first knowledge is objective knowledge associated with the to-be-monitored crop, and the second knowledge is subjective knowledge associated with the to-be-monitored crop, a structure mode of a knowledge graph corresponding to the to-be-monitored crop is determined, the first knowledge and the second knowledge are knowledge extracted according to the structure mode, and the method for constructing the knowledge graph, and then a target service function associated with the to-be-monitored crop is provided based on the knowledge graph.

[0210] It is easy to note that, through the embodiment of the present application, the corresponding knowledge graph is constructed based on the objective knowledge and the subjective knowledge associated with the to-be-monitored crop, and the related service function is provided, so as to achieve the purpose of constructing the knowledge graph based on the related professional knowledge to serve the production, thereby realizing the technical effect of reducing the construction cost of the knowledge graph and the difficulty of obtaining the professional knowledge, and further solving the technical problems of high cost, great difficulty and low efficiency of the processing method based on manual acquisition of professional knowledge in the related art.

[0211] Those skilled in the art can understand that Figure 11 The structure shown is only schematic, and the computer terminal can also be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a palm computer, a Mobile Internet Device (MID), a PAD, or other terminal devices. Figure 11 It does not limit the structure of the electronic device described above. For example, the computer terminal can further include more or fewer components (such as a network interface, a display device, etc.) than those shown in the figure, or have a different configuration from that shown in the figure. Figure 11 Figure 11

[0212] Those skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device by a program, which can be stored in a computer readable storage medium, and the storage medium can include a flash disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, etc.

[0213] According to the embodiments of the present application, an embodiment of a computer readable storage medium is also provided. Optionally, in the present embodiment, the computer readable storage medium described above can be used to save the program code executed by the information processing method provided in Embodiment 1.

[0214] Optionally, in the present embodiment, the computer readable storage medium described above can be located in any one of the computer terminals in the computer terminal group in the computer network, or in any one of the mobile terminals in the mobile terminal group.

[0215] Optionally, in the present embodiment, the computer readable storage medium is configured to store program code for performing the following steps: obtaining first knowledge and second knowledge associated with the crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structural mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structural mode, and constructing the knowledge graph; and providing a target service function associated with the crop to be monitored based on the knowledge graph.

[0216] ​​Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: performing knowledge extraction on the first knowledge and the second knowledge based on a structure mode, to obtain to-be-fused entities, to-be-fused relationships and to-be-fused attributes, wherein the structure mode is used to determine entity categories contained in the knowledge graph and relationships between entities; performing knowledge fusion on the to-be-fused entities, the to-be-fused relationships and the to-be-fused attributes, to obtain a fusion result; and storing the fusion result, to obtain the knowledge graph.

[0217] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: performing knowledge extraction on the first knowledge and the second knowledge by using a knowledge extraction model, to obtain to-be-fused entities, to-be-fused relationships and to-be-fused attributes, wherein the knowledge extraction model is obtained by machine learning training using a training data set, and the training data set is determined by the first knowledge and the second knowledge.

[0218] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: extracting sample annotation data from the first knowledge and the second knowledge; obtaining a training data set based on cross-domain data, sample annotation data and a preset template, wherein the cross-domain data is knowledge graph annotation data in a remaining field other than an application field corresponding to the to-be-monitored crop, and the preset template is used to fill the sample annotation data and the cross-domain data; refining an initial label set corresponding to the sample annotation data, to obtain a target label set; and training a natural language pre-training model by using the training data set and the target label set, to obtain the knowledge extraction model.

[0219] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: determining a knowledge extraction task by using the training data set; converting the knowledge extraction task into a pre-training task of the natural language pre-training model; and training the natural language pre-training model by using the pre-training task and the target label set, to obtain the knowledge extraction model.

[0220] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: decomposing an initial label set, to obtain an intermediate label set, wherein the initial label set is a label set corresponding to the knowledge graph, and the intermediate label set is a label set corresponding to the knowledge extraction model; performing knowledge extraction on the intermediate label set by using the knowledge extraction model, to obtain an extraction result; and merging the extraction result, to obtain the target label set.

[0221] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: displaying a plurality of types of candidate crops and a plurality of candidate service functions corresponding to each type of candidate crop in the graphical user interface; determining a crop to be monitored from the plurality of types of candidate crops in response to a first touch operation on the graphical user interface; determining a target service function for the crop to be monitored from the plurality of candidate service functions in response to a second touch operation on the graphical user interface; and displaying service information corresponding to the target service function in the graphical user interface in response to a third touch operation on the graphical user interface.

[0222] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: obtaining an operation feedback result corresponding to the service information; and optimizing the service information based on the operation feedback result in response to an editing operation on the service information.

[0223] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: obtaining first kiwi fruit knowledge and second kiwi fruit knowledge associated with the kiwi fruit to be monitored, wherein the first kiwi fruit knowledge is objective kiwi fruit knowledge associated with the kiwi fruit to be monitored, and the second kiwi fruit knowledge is subjective kiwi fruit knowledge associated with the kiwi fruit to be monitored; constructing a kiwi fruit knowledge graph corresponding to the kiwi fruit to be monitored by using the first kiwi fruit knowledge and the second kiwi fruit knowledge; and providing a target kiwi fruit service function associated with the kiwi fruit to be monitored based on the kiwi fruit knowledge graph.

[0224] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: displaying a plurality of types of candidate agricultural crops and a plurality of candidate agricultural service functions corresponding to each type of candidate agricultural crop in the graphical user interface; determining a kiwi fruit to be monitored from the plurality of types of candidate kiwi fruits in response to a first touch operation on the graphical user interface; determining a target kiwi fruit service function for the kiwi fruit to be monitored from the plurality of candidate kiwi fruit service functions in response to a second touch operation on the graphical user interface; and displaying kiwi fruit service information corresponding to the target kiwi fruit service function in the graphical user interface in response to a third touch operation on the graphical user interface.

[0225] Optionally, in the embodiment, the computer readable storage medium is configured to store program code for performing the following steps: obtaining first fruit tree knowledge and second fruit tree knowledge associated with the fruit tree to be monitored, wherein the first fruit tree knowledge is objective fruit tree knowledge associated with the fruit tree to be monitored, and the second fruit tree knowledge is subjective fruit tree knowledge associated with the fruit tree to be monitored; constructing a fruit tree knowledge graph corresponding to the fruit tree to be monitored by using the first fruit tree knowledge and the second fruit tree knowledge; and providing a target fruit tree service function associated with the fruit tree to be monitored based on the fruit tree knowledge graph.

[0226] The above-mentioned sequence numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0227] In the above-mentioned embodiments of the present application, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the relevant description of other embodiments.

[0228] In several embodiments provided in the present application, it should be understood that the disclosed technical contents can be implemented by other ways. Among them, the above-mentioned device embodiments are only schematic, for example, the division of the units is only a logical function division, and actual implementation can have another division way, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units or modules shown or discussed can be indirect coupling or communication connection through some interfaces, units or modules, which can be electrical or other forms.

[0229] The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place or distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0230] In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be realized in the form of hardware or in the form of software functional unit.

[0231] The integrated unit, if realized in the form of software functional unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application or the essential part or all or part of the technical solutions that make contributions to the prior art can be embodied in the form of software product, which is stored in a storage medium and includes a plurality of instructions for making a computer device (which can be a personal computer, a server or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application. The above-mentioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and various program code storage media.

[0232] The above merely describes the preferred embodiments of the present application, and it should be pointed out that, for those skilled in the art, several improvements and refinements can be made without departing from the principles of the present application, and these improvements and refinements should also be considered as falling within the protection scope of the present application.

Claims

1. An information processing method characterized by comprising: The method comprises the following steps: acquiring first knowledge and second knowledge of a crop to be monitored, wherein the first knowledge is objective knowledge associated with the crop to be monitored, and the second knowledge is subjective knowledge associated with the crop to be monitored; determining a structural mode of a knowledge graph corresponding to the crop to be monitored; performing knowledge extraction on the first knowledge and the second knowledge according to the structural mode by using a knowledge extraction model to construct the knowledge graph, wherein the knowledge extraction model is obtained by machine learning using a training data set, and the training data set is determined based on the first knowledge, the second knowledge, cross-domain data, and a preset template, wherein the cross-domain data is knowledge graph annotation data in fields other than an application field of the crop to be monitored; providing a target service function associated with the crop to be monitored based on the knowledge graph.

2. The information processing method according to claim 1, characterized by, The method comprises the following steps: performing knowledge extraction on the first knowledge and the second knowledge according to the structural mode by using the knowledge extraction model to construct the knowledge graph, wherein the structural mode is used to determine entity categories included in the knowledge graph and relationships between entities; performing knowledge fusion on the to-be-fused entities, the to-be-fused relationships, and the to-be-fused attributes to obtain a fusion result; storing the fusion result to obtain the knowledge graph.

3. The information processing method according to claim 1, characterized by, The method further comprises the following steps: extracting sample annotation data from the first knowledge and the second knowledge; obtaining the training data set based on the cross-domain data, the sample annotation data, and the preset template, wherein the preset template is used to fill in the sample annotation data and the cross-domain data; refining an initial label set corresponding to the sample annotation data to obtain a target label set; training a natural language pre-training model using the training data set and the target label set to obtain the knowledge extraction model.

4. The information processing method according to claim 3, characterized by, The method comprises the following steps: determining a knowledge extraction task by using the training data set; transforming the knowledge extraction task into a pre-training task of the natural language pre-training model; training a natural language pre-training model using the pre-training task and the target label set to obtain the knowledge extraction model.

5. The information processing method according to claim 3, characterized by, The method comprises the following steps: decomposing the initial label set to obtain an intermediate label set, wherein the initial label set is a label set corresponding to the knowledge graph, and the intermediate label set is a label set corresponding to the knowledge extraction model; performing knowledge extraction on the intermediate label set by using the knowledge extraction model to obtain an extraction result; merging the extraction result to obtain the target label set.

6. The information processing method according to claim 1, characterized by, The information processing method comprises the following steps: displaying a plurality of types of candidate crops and a plurality of candidate service functions corresponding to each type of candidate crop in the graphical user interface; determining the target crop to be monitored from the plurality of types of candidate crops in response to a first touch operation on the graphical user interface; determining the target service function for the target crop to be monitored from the plurality of candidate service functions in response to a second touch operation on the graphical user interface; displaying service information corresponding to the target service function in the graphical user interface in response to a third touch operation on the graphical user interface.

7. The information processing method according to claim 6, characterized by, The information processing method further comprises: obtaining an operation feedback result corresponding to the service information; optimizing the service information based on the operation feedback result in response to an editing operation on the service information.

8. An information processing method characterized by comprising: The information processing method further comprises: obtaining first kiwi fruit knowledge and second kiwi fruit knowledge of a target kiwi fruit, wherein the first kiwi fruit knowledge is objective kiwi fruit knowledge associated with the target kiwi fruit, and the second kiwi fruit knowledge is subjective kiwi fruit knowledge associated with the target kiwi fruit; determining a structure mode of a kiwi fruit knowledge graph corresponding to the target kiwi fruit; performing knowledge extraction on the first kiwi fruit knowledge and the second kiwi fruit knowledge according to the structure mode by using a knowledge extraction model to construct the kiwi fruit knowledge graph, wherein the knowledge extraction model is trained by machine learning using a training data set, and the training data set is determined based on the first kiwi fruit knowledge, the second kiwi fruit knowledge, cross-domain data, and a preset template, wherein the cross-domain data is knowledge graph annotation data in fields other than the application field of the target kiwi fruit; providing a target kiwi fruit service function associated with the target kiwi fruit based on the kiwi fruit knowledge graph.

9. The information processing method according to claim 8, characterized by, The information processing method further comprises: displaying a plurality of types of candidate agricultural crops and a plurality of candidate agricultural service functions corresponding to each type of candidate agricultural crop in the graphical user interface; determining the target kiwi fruit to be monitored from the plurality of types of candidate agricultural crops in response to a first touch operation on the graphical user interface; determining the target kiwi fruit service function for the target kiwi fruit to be monitored from the plurality of candidate agricultural service functions in response to a second touch operation on the graphical user interface; displaying kiwi fruit service information corresponding to the target kiwi fruit service function in the graphical user interface in response to a third touch operation on the graphical user interface.

10. An information processing method characterized by comprising: The information processing method further comprises: obtaining an operation feedback result corresponding to the service information; optimizing the service information based on the operation feedback result in response to an editing operation on the service information. obtaining first fruit tree knowledge and second fruit tree knowledge of a to-be-monitored fruit tree, wherein the first fruit tree knowledge is objective fruit tree knowledge associated with the to-be-monitored fruit tree, and the second fruit tree knowledge is subjective fruit tree knowledge associated with the to-be-monitored fruit tree; determining a structure mode of a fruit tree knowledge graph corresponding to the to-be-monitored fruit tree; performing knowledge extraction on the first fruit tree knowledge and the second fruit tree knowledge according to the structure mode by using a knowledge extraction model to construct the fruit tree knowledge graph, wherein the knowledge extraction model is obtained by machine learning training using a training data set, the training data set is determined based on the first fruit tree knowledge, the second fruit tree knowledge, cross-domain data, and a preset template, and the cross-domain data is knowledge graph annotation data in fields other than an application field of the to-be-monitored fruit tree; providing a target fruit tree service function associated with the to-be-monitored fruit tree based on the fruit tree knowledge graph.

11. A computer readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program controls a device in which the computer-readable storage medium is located to perform the information processing method of any one of claims 1 to 10 when the program is running.

12. An electronic device, comprising: comprise: a processor; and a memory connected to the processor, configured to provide the processor with instructions for processing the following processing steps: Step 1, obtaining first knowledge and second knowledge of a to-be-monitored crop, wherein the first knowledge is objective knowledge associated with the to-be-monitored crop, and the second knowledge is subjective knowledge associated with the to-be-monitored crop; Step 2, determining a structure mode of a knowledge graph corresponding to the to-be-monitored crop; Step 3, performing knowledge extraction on the first knowledge and the second knowledge according to the structure mode by using a knowledge extraction model to construct the knowledge graph, wherein the knowledge extraction model is obtained by machine learning training using a training data set, the training data set is determined based on the first knowledge, the second knowledge, cross-domain data, and a preset template, and the cross-domain data is knowledge graph annotation data in fields other than an application field of the to-be-monitored crop; Step 4, providing a target service function associated with the to-be-monitored crop based on the knowledge graph.

Images