CN114328944A - Construction method of agricultural information visualization interactive system based on knowledge graph - Google Patents
Construction method of agricultural information visualization interactive system based on knowledge graph Download PDFInfo
- Publication number
- CN114328944A CN114328944A CN202111322203.9A CN202111322203A CN114328944A CN 114328944 A CN114328944 A CN 114328944A CN 202111322203 A CN202111322203 A CN 202111322203A CN 114328944 A CN114328944 A CN 114328944A
- Authority
- CN
- China
- Prior art keywords
- data
- knowledge graph
- information
- knowledge
- constructing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a construction method of an agricultural information visualization interactive system based on a knowledge graph, which comprises the steps of obtaining crop disease and insect pest data information of a network; carrying out data preprocessing according to the obtained crop disease and insect pest data information to obtain multi-modal structured data, and carrying out training of a disease and insect pest recognition model; constructing a knowledge graph according to the preprocessed structured data and storing the knowledge graph; displaying and rendering the nodes and the relations of the knowledge graph by using a data visualization tool library and a programming technology; the preset function module is used for carrying out real-time manual interaction on the knowledge graph. The invention takes the knowledge map as a data storage mode, and carries out data visualization in a cross-media mode to visually display the data information of the crop diseases and insect pests to users. Meanwhile, disease identification and knowledge map dynamic combination are achieved, a user can quickly inquire related knowledge and update related information, and time cost for searching information from network information is saved.
Description
Technical Field
The invention relates to the technical field of agricultural information systems based on knowledge graphs, in particular to a construction method of an agricultural information visualization interactive system based on knowledge graphs.
Background
With the development of the current internet technology, the amount of information data is increasing explosively. The internet, a tool that people often seek solutions in daily study, life and work, also needs to be updated continuously. Thus, simple results based on traditional search engine searches have been increasingly inadequate for many times. The concept of the knowledge graph is provided as a product generated by the rapid development of the current internet, and is used for describing various entities and concepts existing in the real world and the connection of the entities and the entities, the entities and the concepts, so that various objects existing in the real society and various relationships among the objects can be relatively reasonably and formally described.
Agriculture is the foundation of all production, and is the foundation that human beings can survive in nature, and there is not agricultural development and the development of civilization now, and china is as traditional agricultural kingdom, and from ancient times to present, agricultural development plays indispensable role in national economy. Meanwhile, nowadays of rapid development of scientific and technological economy, the proportion of the development of the scientific and technological technology in China is higher and higher, and the development of the scientific and technological technology plays a greater and greater role in agricultural modernization construction in China. How to provide relevant reasonable and timely information for practitioners who enter related industries at first is extremely important for reducing professional thresholds of information use.
The prior art has the defects that the feedback result obtained based on the traditional search engine search is often a series of web pages or document information lists, especially for the professional field such as agriculture, a user still needs to spend a great deal of time and energy to find effective information from the information, the related crop pest description information and other related information can not be clearly obtained, the crop cultivation management information can not be effectively mastered, and agricultural work can be effectively carried out. And the direct result of the knowledge map needs to be understood by professional technicians, and ordinary people cannot directly understand and use the knowledge map.
Disclosure of Invention
The invention aims to overcome the defects in the prior art, and in order to realize the purpose, a construction method of an agricultural information visualization interactive system based on a knowledge graph is adopted to solve the problems in the background technology.
A construction method of an agricultural information visualization interactive system based on a knowledge graph comprises the following steps:
acquiring crop disease and pest data information of a network;
carrying out data preprocessing according to the obtained crop disease and insect pest data information to obtain multi-modal structured data, and carrying out training of a disease and insect pest recognition model;
constructing a knowledge graph according to the preprocessed structured data and storing the knowledge graph;
displaying and rendering the nodes and the relations of the knowledge graph by using a data visualization tool library and a programming technology;
the preset function module is used for carrying out real-time manual interaction on the knowledge graph.
As a further aspect of the invention: the specific method for acquiring the crop disease and pest data information of the network comprises the following steps: and acquiring multi-mode semi-structured crop disease and pest data information by a network data mining technology.
As a further aspect of the invention: the specific steps of carrying out data preprocessing according to the obtained crop disease and insect pest data information to obtain multi-modal structured data and carrying out training of a disease and insect pest recognition model comprise:
performing data division according to the obtained multi-mode semi-structured crop disease and pest data information to obtain pure text data and image data;
reading plain text data, and establishing a triple relationship according to the entity, the attribute and the relationship according to the entity, the text and the text form of the entity for storage;
classifying and storing the image data, and simultaneously carrying out data preprocessing to obtain structured data, wherein the data format of the structured data comprises crops, names of diseases and insect pests corresponding to the crops, corresponding classifications and various attribute categories;
entity and relation attribute data needing visualization are separately stored in a CSV form and are used for calling a knowledge graph;
and training a recognition model for crop diseases and insect pests according to the obtained structured data.
As a further aspect of the invention: the specific method for constructing and storing the knowledge graph according to the preprocessed structured data comprises the following steps: and constructing and traversing the triples of the entities, the relations and the attributes, and storing the established knowledge graph of the entities and the relation model in a database.
As a further aspect of the invention: the specific steps of displaying and rendering the nodes and the relations of the knowledge graph by using the data visualization tool library and the programming technology comprise:
constructing a system frame interface of an agricultural information visualization interactive system, wherein the system frame interface comprises a knowledge map display module, a disease identification uploading module, a text box module and a knowledge map operation module;
constructing nodes and relations according to the obtained structured data, and generating and displaying a knowledge graph;
meanwhile, the data is imported through a knowledge graph operation module to generate a new knowledge graph and switch between different knowledge graphs.
As a further aspect of the invention: the specific steps of the system framework interface construction comprise:
establishing an agricultural information visualization interactive system by calling a data visualization tool library of a knowledge graph and combining a hypertext markup language of a front end and an element rendering language of a front end page;
importing structured data to generate a quartic media knowledge graph;
constructing a functional framework on the rear end, and combining a front end interface with the rear end;
storing other attribute information related to the nodes through the text box module and feeding back the attribute information in real time;
local pest and disease information is uploaded through the disease identification uploading module to be compared and identified.
Compared with the prior art, the invention has the following technical effects:
by adopting the technical scheme, a large amount of multi-mode semi-structured pest and disease damage data information related to crops is acquired from a network, and then the data information is preprocessed and classified and stored to obtain cross-media structured data. And constructing a cross-media knowledge graph according to the structured data, and performing information display of the knowledge graph on a system interface by constructing an agricultural information visual interactive system and combining. Meanwhile, through the arrangement of the functional module, the stored data information can be updated through the trained disease identification uploading module, and the functions of increasing, deleting, modifying and checking the data information are realized. By establishing the system, the time cost for inquiring information from massive network information can be saved.
Drawings
The following detailed description of embodiments of the invention refers to the accompanying drawings in which:
fig. 1 is a schematic diagram of steps of a construction method of an interactive agricultural information visualization system according to some embodiments disclosed in the present application;
fig. 2 is a block flow diagram of an agricultural information visualization interactive system of some embodiments disclosed herein;
fig. 3 is a data processing flow diagram of some embodiments disclosed herein.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1 and fig. 2, in an embodiment of the present invention, a method for constructing an interactive agricultural information visualization system based on a knowledge graph includes:
s1, acquiring crop disease and pest data information of the network;
in the specific implementation steps, the specific method for acquiring the crop disease and pest data information of the network comprises the following steps: and acquiring multi-mode semi-structured crop disease and pest data information by a network data mining technology. Specifically, it is necessary to construct triple structure data for the knowledge graph. Therefore, based on the mass public data of the Internet, the multi-mode semi-structured data information related to crop diseases and insect pests is acquired through a network data mining technology. The multimodal, semi-structured data information includes both plain text data and image data. Thus, the multi-modal data selection can describe the same semantic information in a more detailed way from the perspective of cross-media.
S2, preprocessing data according to the obtained crop disease and pest data information to obtain multi-modal structured data, and training a recognition model of disease and pest, wherein the method specifically comprises the following steps:
s21, dividing data according to the obtained multi-mode semi-structured crop disease and pest data information to obtain pure text data and image data;
specifically, a network data mining technology is utilized to obtain semi-structured data comprising two parts of plain text data and image data. For text files text file storage is used.
S22, reading plain text data stored in a text file shape, and establishing a triple relation according to the entity, the text and the text form of the entity according to the entity, the attribute and the relation for storage;
the image data is stored in a classified manner, as shown in fig. 3, and data preprocessing is performed by using a manual inspection and normalization manner to obtain structured data, wherein the data format of the structured data includes crops, names of corresponding crop diseases and insect pests, corresponding classifications, and attribute categories. Specifically, for example, the tomato disease category is taken as an example, the entity of tomato early blight comprises: english name, synonym, brief introduction, harm symptom, pathogen, infection circulation, occurrence rule, prevention and treatment method and other information.
S23, independently storing the entity needing visualization and the relationship attribute data in a CSV form for calling the knowledge graph;
in a specific embodiment, disease entities such as tomato early blight, tomato late blight and the like are stored in a tomato disease entity file and stored in a CSV form, and similarly, tomato disease relationships are also stored in a CSV file, so that visual part calling of subsequent data is facilitated.
And S24, training a recognition model for crop diseases and insect pests according to the obtained structured data.
In the specific embodiment, according to the existing different disease data sets, such as data sets of tomato early blight, tomato late blight and the like, Resnet50, vgg16 or other deep learning models are used for training and screening out the model with the best disease identification effect, so that the identification result can be obtained by classifying the diseases after subsequent local uploading.
Specifically, the data set is divided into 9: 1 into a training set and a test set.
In the embodiment, a deep learning network is adopted to construct an image identification model, a neural network model obtains the difference between a predicted image tag and a true value through forward propagation, a gradient is obtained through a loss function, and then parameters of the model are updated through reverse propagation of the gradient. As ResNet has a good generalization effect on most image data sets and a good recognition effect on a target data set, a ResNet model is proposed to be adopted for model training of crop diseases and insect pests, and disease classification is conveniently carried out after subsequent local uploading to obtain a recognition result.
S3, constructing a knowledge graph according to the preprocessed structured data and storing the knowledge graph;
the specific method comprises the following steps: and constructing and traversing the triples of the entities, the relations and the attributes, and storing the established knowledge graph of the entities and the relation model in a database.
And S4, displaying and rendering the nodes and the relations of the knowledge graph by using a data visualization tool library and a programming technology. Specifically, on the basis of a traditional method for constructing a text knowledge graph by independently utilizing a Neo4j tool, the method utilizes a data visualization tool library and a programming technology to display and render nodes and relations of the knowledge graph, and displays multi-modal information in a cross-media knowledge graph mode in front of a user in a more direct and attractive mode through the visualization technology, and the method specifically comprises the following steps:
s41, constructing a system frame interface of the agricultural information visualization interactive system, wherein the system frame interface comprises a knowledge map display module, a disease identification uploading module, a text box module and a knowledge map operation module;
specifically, a required system framework interface is designed and constructed by calling a data visualization tool library and adopting a hypertext markup language (HTML) of a front end, an element rendering language (CSS) of a front end page and JavaScript. And simultaneously setting corresponding interface operation buttons and a search box and the like. And the display and rendering of information such as nodes, relations and the like of the cross-media knowledge graph facing the agriculture are realized.
S42, constructing nodes and relations according to the obtained structured data, and generating and displaying a knowledge graph;
meanwhile, the data is imported through a knowledge graph operation module to generate a new knowledge graph and switch between different knowledge graphs.
Specifically, the constructed structured data is used for constructing nodes and relations, the cross-media knowledge graph is generated and displayed in the corresponding area of the interface, and other attributes of the entity are stored in the corresponding area. The map can also be generated by setting a CSV file uploading button beside the title, other attribute information is imported into the MySQL database to generate a new knowledge map, and a switching button is set, so that switching can be performed between maps of different crops.
And S5, carrying out real-time manual interaction on the knowledge graph through a preset function module.
Specifically, the front-end interface display can be associated with the database through a preset function module.
The specific steps of the system framework interface construction comprise:
establishing an agricultural information visualization interactive system by calling a data visualization tool library of a knowledge graph and combining a hypertext markup language of a front end and an element rendering language of a front end page;
importing structured data to generate a quartic media knowledge graph;
constructing a functional framework on the rear end, and combining a front end interface with the rear end;
storing other attribute information related to the nodes through the text box module and feeding back the attribute information in real time;
local pest and disease information is uploaded through the disease identification uploading module to be compared and identified.
S51, the knowledge graph display module completes the functions of free node dragging and interface amplification:
by calling the existing knowledge map visualization tool and combining with the use of HTML + CSS + D3.js or other modes to import structured data, the required cross-media knowledge map of related crop diseases is generated, the problem that the object is insufficiently described by a single text of a traditional knowledge map database Neo4j can be well solved, and the related object can be better described from different dimensions. Meanwhile, the interface zooming-in and zooming-out functions are realized through the front-end interface language.
S52, establishing a knowledge graph operation module for the user to operate the knowledge graph in real time:
the use of a back-end writing frame, a database and the connection with the front end adopt SpringBoot + Neo4j + MySQL + Nginx, and the back-end function is combined with the front-end interface to achieve the related operation function. The functions of adding, deleting, modifying and inquiring maps can be realized by setting buttons and a search box through the MySQL language of the traditional database, and related functions can also be realized through professional Cypher sentences, so that professional and non-professional people can use the system conveniently.
Meanwhile, the instant feedback effect which can be displayed on the graph during operation by a user needs to be highlighted, so that the user can sense related information more visually and concisely, the relevant nodes need to be subjected to highlighted effect operation when being operated, the relevant nodes are difficult to directly find in the whole graph, the nodes need to be displayed and amplified independently after being searched through a search box, the graph is hidden temporarily, and subsequent related operation on the nodes is facilitated.
S53, establishing a special text box module to store other attribute information related to the node, and feeding back other related information in the text box in real time when the target is placed on a specific node;
s54, establishing a special crop disease identification uploading module, setting an interface to associate the locally uploaded image with a rear-end disease identification module, comparing the locally uploaded image with the locally uploaded relevant disease and insect pests through a trained model to give a relevant reference result, highlighting relevant nodes on the map by the method after the disease identification result appears, and displaying relevant information and cultivation management suggestions in real time in a text box.
Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that changes, modifications, substitutions and alterations can be made in these embodiments without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents, which should be construed as being within the scope of the invention.
Claims (6)
1. A construction method of an agricultural information visualization interactive system based on a knowledge graph is characterized by comprising the following steps:
acquiring crop disease and pest data information of a network;
carrying out data preprocessing according to the obtained crop disease and insect pest data information to obtain multi-modal structured data, and carrying out training of a disease and insect pest recognition model;
constructing a knowledge graph according to the preprocessed structured data and storing the knowledge graph;
displaying and rendering the nodes and the relations of the knowledge graph by using a data visualization tool library and a programming technology;
the preset function module is used for carrying out real-time manual interaction on the knowledge graph.
2. The method for constructing the knowledge-graph-based agricultural information visualization interactive system according to claim 1, wherein the specific method for acquiring the crop pest data information of the network is as follows: and acquiring multi-mode semi-structured crop disease and pest data information by a network data mining technology.
3. The method for constructing the knowledge-graph-based agricultural information visualization interactive system according to claim 1 or 2, wherein the specific steps of performing data preprocessing according to the obtained crop pest data information to obtain multi-modal structured data and performing training of a pest identification model include:
performing data division according to the obtained multi-mode semi-structured crop disease and pest data information to obtain pure text data and image data;
reading plain text data, and establishing a triple relationship according to the entity, the attribute and the relationship according to the entity, the text and the text form of the entity for storage;
classifying and storing the image data, and simultaneously carrying out data preprocessing to obtain structured data, wherein the data format of the structured data comprises crops, names of diseases and insect pests corresponding to the crops, corresponding classifications and various attribute categories;
entity and relation attribute data needing visualization are separately stored in a CSV form and are used for calling a knowledge graph;
and training a recognition model for crop diseases and insect pests according to the obtained structured data.
4. The method for constructing the interactive system for agricultural information visualization based on the knowledge graph as claimed in claim 1, wherein the specific method for constructing and storing the knowledge graph according to the preprocessed structured data is as follows: and constructing and traversing the triples of the entities, the relations and the attributes, and storing the established knowledge graph of the entities and the relation model in a database.
5. The method for constructing an interactive system for agricultural information visualization based on a knowledge graph according to claim 1, wherein the specific steps of displaying and rendering nodes and relationships of the knowledge graph by using a data visualization tool library and a programming technology comprise:
constructing a system frame interface of an agricultural information visualization interactive system, wherein the system frame interface comprises a knowledge map display module, a disease identification uploading module, a text box module and a knowledge map operation module;
constructing nodes and relations according to the obtained structured data, and generating and displaying a knowledge graph;
meanwhile, the data is imported through a knowledge graph operation module to generate a new knowledge graph and switch between different knowledge graphs.
6. The method for constructing the interactive system for agricultural information visualization based on the knowledge-graph as claimed in claim 5, wherein the specific steps of the system framework interface construction include:
establishing an agricultural information visualization interactive system by calling a data visualization tool library of a knowledge graph and combining a hypertext markup language of a front end and an element rendering language of a front end page;
importing structured data to generate a quartic media knowledge graph;
constructing a functional framework on the rear end, and combining a front end interface with the rear end;
storing other attribute information related to the nodes through the text box module and feeding back the attribute information in real time;
local pest and disease information is uploaded through the disease identification uploading module to be compared and identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111322203.9A CN114328944A (en) | 2021-11-09 | 2021-11-09 | Construction method of agricultural information visualization interactive system based on knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111322203.9A CN114328944A (en) | 2021-11-09 | 2021-11-09 | Construction method of agricultural information visualization interactive system based on knowledge graph |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114328944A true CN114328944A (en) | 2022-04-12 |
Family
ID=81045386
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111322203.9A Pending CN114328944A (en) | 2021-11-09 | 2021-11-09 | Construction method of agricultural information visualization interactive system based on knowledge graph |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114328944A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115050014A (en) * | 2022-06-15 | 2022-09-13 | 河北农业大学 | Small sample tomato disease identification system and method based on image text learning |
-
2021
- 2021-11-09 CN CN202111322203.9A patent/CN114328944A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115050014A (en) * | 2022-06-15 | 2022-09-13 | 河北农业大学 | Small sample tomato disease identification system and method based on image text learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Lohmann et al. | Concentri cloud: Word cloud visualization for multiple text documents | |
US11093520B2 (en) | Information extraction method and system | |
Pivk et al. | Transforming arbitrary tables into logical form with TARTAR | |
CN114595344B (en) | Crop variety management-oriented knowledge graph construction method and device | |
CN111598702A (en) | Knowledge graph-based method for searching investment risk semantics | |
CN110781315A (en) | Food safety knowledge map and construction method of related intelligent question-answering system | |
CN103440287A (en) | Web question-answering retrieval system based on product information structuring | |
Ansari et al. | Intelligent question answering system based on artificial neural network | |
CN111538847A (en) | Ningxia rice knowledge graph construction method | |
CN112115252B (en) | Intelligent auxiliary writing processing method and device, electronic equipment and storage medium | |
CN111651614A (en) | Method and system for constructing medicated diet knowledge graph, electronic equipment and storage medium | |
CN104881428B (en) | A kind of hum pattern extraction, search method and the device of hum pattern webpage | |
CN115599899A (en) | Intelligent question-answering method, system, equipment and medium based on aircraft knowledge graph | |
CN114328944A (en) | Construction method of agricultural information visualization interactive system based on knowledge graph | |
Schneider et al. | Visualizing topical quotations over time to understand news discourse | |
Torget et al. | Mapping texts: Combining text-mining and geo-visualization to unlock the research potential of historical newspapers | |
Zhuge | Dimensionality on summarization | |
Zhu et al. | Question answering on agricultural knowledge graph based on multi-label text classification | |
Wang et al. | Research on the construction method of rice knowledge graph | |
CN113051402A (en) | Fishery knowledge recommendation method based on knowledge graph | |
CN110555196B (en) | Method, apparatus, device and storage medium for automatically generating article | |
Zhao | Interactive visual data exploration: A multi-focus approach | |
Graves et al. | Towards Visual Overviews for Open Government Data. | |
JP5716223B1 (en) | Thought support dictionary, thought support dictionary system, thought support system, thought support method, thought support program, thought support program storage medium, and thought support data storage medium | |
Pagliosa et al. | MIST: multiscale information and summaries of texts |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |