Intelligent comprehensive management method and system for germplasm resources of Chinese orchid

Abstract

The invention discloses an intelligent comprehensive management method and system for Chinese orchid germplasm resources, and the method comprises the steps: collecting multi-source heterogeneous data of the Chinese orchid germplasm resources, carrying out the data standardization modeling and multi-dimensional label classification, and constructing a Chinese orchid knowledge graph fusing the genetic characteristics, character indexes and environmental factors of varieties; performing dynamic semantic association from the knowledge graph according to the user interaction behaviors, and constructing a user feature association table; and related entities and relational data in the knowledge graph are extracted according to the association table, germplasm resource recommendation information is generated in combination with weights such as the flowering period, the fragrance index and the market popularity, and a weighted CNN prediction model is introduced to perform accurate planting prediction on recommended varieties. According to the method, the multi-omics analysis, the knowledge graph and the deep learning technology are fused, accurate recommendation, safety control and intelligent cultivation decision making of Chinese orchid germplasm resources are achieved, and the resource management efficiency and the industrial application value are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of horticultural plant genetic resource management, and in particular to an intelligent comprehensive management method and system for Chinese orchid germplasm resources that integrates multi-omics data analysis, knowledge graph construction, and artificial intelligence algorithms. Background Art

[0002] my country has a long history of collecting, preserving, and utilizing orchid germplasm resources, but numerous challenges persist in the systematic management of these resources. For example, germplasm information is fragmented and lacks a unified nomenclature system; genetic information and multi-omics data are not systematically recorded; cultivation management relies on experience, lacking data-driven scientific decision-making; and data standards are inconsistent across institutions, resulting in low levels of resource sharing. These factors hinder the efficient conservation, accurate evaluation, and widespread utilization of China's orchid germplasm resources, hindering the development and upgrading of the industry.

[0003] To address these challenges, this paper proposes an intelligent, integrated management method and system for cymbidium germplasm resources that integrates multi-omics data (including genomic, transcriptomic, metabolomic, proteomic, and phenotypic data) with practical production and management information. This method integrates resource integration, feature extraction, intelligent recommendation, and growth prediction. Leveraging artificial intelligence and knowledge graph technology, it enables precise recommendation and scientific management of germplasm resources, significantly improving evaluation accuracy, recommendation efficiency, and the scientific nature of promotion and utilization. This provides technical support for the digital and intelligent transformation of the cymbidium industry. Summary of the Invention

[0004] This invention addresses the current challenges of fragmented information, inconsistent standards, and a lack of intelligent analysis and prediction in the management of Chinese cymbidium germplasm resources. It proposes a comprehensive intelligent management method and system for Chinese cymbidium germplasm resources. This solution integrates multiple technologies, including multi-omics data processing, knowledge graph construction, convolutional neural network (CNN) semantic analysis, graph neural network (GNN) semantic representation, data encryption, and access control. This approach enables comprehensive intelligent management of Chinese cymbidium germplasm resources, from data collection and knowledge graph construction to user feature extraction, resource recommendations, and planting predictions. This invention significantly improves germplasm resource management efficiency, recommendation accuracy, and scientific decision-making, providing technical support for the digitalization and intelligentization of the Chinese cymbidium industry.

[0005] The first aspect of the present invention provides an intelligent integrated management method for cymbidium germplasm resources, comprising the following steps: S1: Collect multi-source heterogeneous data of Cymbidium germplasm resources, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, and cultivation environment monitoring data; perform standardized modeling and multi-dimensional label classification on the multi-source heterogeneous data, and generate triple data based on variety genetic similarity, trait co-occurrence, and environmental adaptability. Use the triple data to construct a knowledge graph that integrates Cymbidium characteristic weights; S2: Real-time collection and analysis of user interaction behavior data at multiple time points. Contextual semantic analysis and feature extraction are performed on the user behavior data based on the CNN semantic model to obtain user behavior characteristics including flowering period preference, fragrance preference, and market popularity. Based on these user behavior characteristics, dynamic association analysis is performed on relevant entities and relationship data in the knowledge graph. Combined with the temporal correlation of each user behavior characteristic, a feature association table is generated for each user. S3: Based on the feature association table, extract entity and relationship data that match the user's features from the knowledge graph, perform semantic conversion based on the multi-dimensional feature parameters of the Chinese orchid varieties, such as flowering period, leaf type, and fragrance index, to generate germplasm resource recommendation information; S4: Select a matching encryption algorithm based on the germplasm resource recommendation information and its data type, configure access permissions according to user roles and assign independent keys, record access logs and recommendation data call records, and perform encrypted backup storage; S5: Based on the germplasm resource recommendation information, a variety of recommended Cymbidium varieties are obtained, and the entity nodes of the recommended varieties are semantically represented in the knowledge graph through a preset graph neural network, and the semantic weights of the recommended Cymbidium varieties are set. Combined with the historical and real-time growth sequences and semantic weights, a weighted CNN prediction model is introduced to predict the growth status and flowering period of Cymbidium.

[0006] In this solution, S1 specifically includes: The multi-source heterogeneous data includes: Basic classification information, including variety category, flower type category, and foliage category; Geographic distribution information, including geographic coordinates of origin and cultivation areas, altitude, climate zones, and seasonal temperature and humidity changes; Genetic and molecular information, including genome sequence, transcriptome expression profile, and metabolome aroma compound profile; Phenotypic and trait information: leaf shape, flower color, fragrance index, flowering period, and stress resistance index; Perform data formatting, data cleaning, missing value filling and outlier processing on multi-source heterogeneous data, and perform unit unification and threshold normalization preprocessing according to national orchid industry standards; After preprocessing, data modeling is performed based on a multidimensional tag system, which includes genetic tags, trait tags, environmental adaptation tags, and market popularity tags; Based on the labeled data, the entity extraction, attribute feature extraction and entity relationship construction of orchids were performed to obtain triple data. The entity relationship includes the genetic similarity relationship between varieties, the co-occurrence relationship of traits, the environmental adaptability relationship and the fragrance correlation relationship; The triplet data is introduced into the weight factor of the Chinese orchid characteristics for weighted processing to generate a knowledge graph integrating the weight of the Chinese orchid characteristics; Graphical visualization is performed based on the knowledge graph structure, and each entity node and its relationship edges are interactively displayed through the user terminal to support variety comparison, feature retrieval and germplasm structure analysis.

[0007] In this solution, the user terminal includes: Mobile terminal: used to install a dedicated application for managing Chinese orchid germplasm resources, supports taking images of flowers, leaves, and roots and uploading them for identification, and supports uploading fragrance sensor data and geographic positioning information; Computer terminal: with high-resolution map visualization and batch data analysis functions, supporting multi-window comparison of variety characteristics, export of variety identification reports and cultivation plans; Dedicated collection equipment terminal: Integrates aroma component sensing module, environmental parameter collection module and RFID / QR code scanning module, used to quickly collect aroma spectrum of Chinese orchid varieties, real-time climate parameters and germplasm identity information on site, and synchronizes updates with the knowledge graph system; Cloud interactive interface: supports remote calling of knowledge graph data, online subscription of recommended variety updates, and receiving cultivation warning information pushed by prediction models.

[0008] In this solution, S2 specifically includes: Based on the interaction between users and the system, the system collects user behavior data at multiple time points in real time. The behavior data includes variety search records and click frequency, flowering period selection preferences and seasonal trends, fragrance category preferences and historical evaluation data, browsing time of leaf art type and flower color combination, market transaction and collection operation records; Constructing a CNN semantic model that combines the multimodal features of orchids, wherein the input of the semantic model includes a text description vector, a fragrance spectrum feature vector, and a variety image feature vector; Through the semantic model, contextual semantic information is extracted from the entity data, relationship data, and attribute data in the knowledge graph to obtain graph document data containing three characteristics: flowering period, fragrance type, and market popularity, and multi-dimensional semantic representation is performed based on the document vector; Using the semantic model to perform semantic analysis on user behavior data, generating user behavior features including time-dependent features, and performing associated entity retrieval and relationship data analysis in the knowledge graph based on the user behavior features to obtain retrieval entity data that meets the user's interests; In the knowledge graph, the retrieval entity data is marked and an associated knowledge dataset is generated. In the associated knowledge dataset, the similarity based on the variety characteristics is calculated for each associated knowledge data, and the relationship strength is numerically represented and serialized to obtain a first sequence. Combined with the time series information of user behavior, the time correlation between each associated knowledge data and other associated knowledge data is calculated, and the time correlation is serialized to obtain a second sequence. The correlation coefficient between the first sequence and the second sequence is calculated based on the Pearson correlation coefficient, and the absolute value of the correlation coefficient is used as the interest level of the associated knowledge data; All associated knowledge data are sorted according to interest, and user behavior features are stored corresponding to associated knowledge data to form a feature association table that is independently maintained for each user.

[0009] In this solution, S3 specifically includes: Based on the interest level in the feature association table and the multidimensional feature weights of the orchid varieties, multiple associated knowledge data that meet the user's preferences are screened out. The multidimensional feature weights include: Flowering period matching: calculated based on the user's regional climate conditions and the historical flowering period prediction results of Chinese orchid varieties; Fragrance compatibility: calculated based on the similarity of fragrance spectra and user fragrance preference labels; Preference for foliage and flower color combinations: calculated based on visual feature scores of the foliage and flower color categories of the variety; Market circulation popularity: a comprehensive score based on transaction records, number of collections and online discussion popularity; Environmental adaptability index: calculated based on the matching degree between the variety cultivation environment requirements and the environmental parameters of the user's location; Extract the entity and relationship data corresponding to the above-mentioned associated knowledge data from the knowledge graph, perform semantic conversion, and generate multimodal germplasm resource recommendation information including variety name, trait characteristics, flowering period prediction, fragrance index, cultivation suggestions and market reference price; Prioritizing the recommended information based on weighted interest, wherein the weighted interest is a weighted sum of flowering period matching, fragrance compatibility, market circulation popularity, and environmental adaptability index parameters; The highest priority recommendation information will be displayed first on the user terminal interface, and secondary screening and sorting will be supported based on flowering period, fragrance type, and market popularity.

[0010] In this solution, S4 specifically includes: Based on the germplasm resource recommendation information generated by S3, determine the business scenario and recommended data type of the current recommendation process; The business scenarios include variety query scenarios, cultivation management scenarios, variety trading and circulation scenarios, variety copyright confirmation and infringement monitoring scenarios; Analyze the security level of recommended data based on business scenarios and recommended data types, and match the corresponding encryption algorithm. Use asymmetric encryption algorithms for sensitive data and symmetric encryption algorithms for non-sensitive data. Use end-to-end encryption protocols for cross-platform data transmission; Configure access rights and assign independent keys based on user roles, including ordinary users, registered merchants, breeding institutions, and platform administrators. Different roles have different restrictions on access scope, data export, and secondary distribution permissions; Combined with the blockchain traceability mechanism, the unique identity identification, transaction records, copyright information and recommendation data of national orchid varieties are stored on the chain; During the user access process, access logs, variety transaction records, copyright certificate call records and germplasm resource recommendation information access records are extracted through permission verification, and the data of the above access process are encrypted and stored in the disaster recovery backup system.

[0011] In this solution, S5 specifically includes: Based on the recommended information of germplasm resources, a variety of recommended Chinese orchid varieties are obtained, and entity and attribute data related to the variety are extracted from the knowledge graph to construct a graph structure containing variety characteristics, genetic information, trait information, geographic information and cultivation parameters; In the graph structure, the relationship information of the five dimensions of variety, trait, geography, genetics and quantity indicators is analyzed, and the following indicators specific to the orchid industry are added to the relationship information: Climate adaptability coefficient: calculated based on the similarity between the meteorological data of the target cultivation area and the historical cultivation environment of the variety; Flowering time window prediction value: Based on historical flowering period data and real-time environmental monitoring data, the time series analysis model is used to predict the most probable flowering start and end dates; Fragrance stability index: Based on the aroma component test results of previous years, it evaluates the degree of fluctuation of the aroma component ratio of the variety in different environments; The entity nodes of the recommended Chinese orchid varieties are semantically represented through a preset graph neural network, and the comprehensive semantic weight of each recommended Chinese orchid variety is calculated in combination with the above-mentioned industry-specific indicators; The historical growth sequences and current real-time growth sequences of various recommended Chinese orchid varieties are input into the weighted CNN prediction model, and the comprehensive semantic weight is introduced as the authenticity of the growth sequence for sequence learning and prediction training. During the prediction process, the output includes growth status level, flowering time window, and fragrance stability rating. Recommended cultivation measures are set through the prediction output, and the prediction results are generated and automatically pushed to the user terminal.

[0012] In S5 of this solution, the prediction results obtained by predicting the growth status and flowering period of Cymbidium orchid are displayed in a multimodal visualization manner through the user terminal, and the multimodal visualization includes: Flowering period forecast calendar view: Mark the expected flowering start and end dates in the form of a timeline, and dynamically display the overlap of flowering periods of different varieties; Aroma component radar chart: displays the relative content and stability index of major aroma compounds; Climate suitability heat map: displays suitable cultivation areas on the map based on the climate suitability coefficient, and supports zooming and area filtering; Growth status dynamic graph: Generate curves or animations based on historical and real-time growth sequences to show the changing trends of plant height, leaf number, and flower bud differentiation indicators; Cultivation suggestion panel: automatically generates corresponding fertilization, irrigation, temperature and humidity control, and shading suggestions based on the prediction results, and exports them as a management plan; The visual interface supports interactive operations, including screening, comparison and collection by flowering period, fragrance or climatic conditions, and superimposed analysis of the prediction information of the selected variety and historical cultivation records.

[0013] The second aspect of the present invention also provides an intelligent integrated management system for Chinese orchid germplasm resources, the system comprising: Memory: used to store the intelligent integrated management program for Chinese orchid germplasm resources and multi-source heterogeneous data sets, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, cultivation environment monitoring data, user behavior data, and market transaction records; Processor: used to execute the management program to implement the steps of the above-mentioned intelligent comprehensive management method for Chinese orchid germplasm resources; Aroma sensor interface module: used to connect to gas chromatography-mass spectrometry or electronic nose aroma component collection equipment, and directly transmit the collected aroma spectrum data to the system database; Climate and environmental data acquisition module: used to collect light intensity, air temperature and humidity, soil temperature and humidity, and carbon dioxide concentration in real time, and dynamically linked with the weighted CNN prediction model; Blockchain node module: used to store unique product identification, transaction contracts, copyright certificates, and recommended data call records; User interaction terminal interface: used to connect mobile terminals, computer terminals and special collection equipment, and used for real-time display and interactive operation of flowering calendars, fragrance radar maps, climate heat maps and multimodal prediction results.

[0014] A third aspect of the present invention provides a computer-readable storage medium, which stores program instructions for executing the above-mentioned intelligent integrated management method for Chinese orchid germplasm resources.

[0015] The present invention discloses a method and system for intelligent, integrated management of Chinese orchid germplasm resources. The method comprises: collecting multi-source, heterogeneous data on Chinese orchid germplasm resources and performing data standardization modeling and multidimensional label classification to construct a Chinese orchid knowledge graph that integrates variety genetic characteristics, trait indicators, and environmental factors; dynamically performing semantic associations within the knowledge graph based on user interaction behaviors to construct a user feature association table; extracting relevant entities and relationship data from the knowledge graph based on the association table, generating germplasm resource recommendation information based on weights such as flowering period, fragrance index, and market popularity, and introducing a weighted CNN prediction model to accurately predict the planting of recommended varieties. The present invention integrates multi-omics analysis, knowledge graphs, and deep learning technologies to achieve accurate recommendation, safe management, and intelligent cultivation decision-making for Chinese orchid germplasm resources, significantly improving resource management efficiency and industrial application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart showing an intelligent integrated management method for Chinese cymbidium germplasm resources according to the present invention is shown; Figure 2 Shows a schematic diagram of the module of the present invention; Figure 3 A schematic diagram of the containerized deployment module and blockchain traceability module of the present invention is shown; Figure 4 It shows a schematic diagram of the operation of the prediction credibility verification and scheduling module of the present invention; Figure 5 A block diagram of an intelligent integrated management system for Chinese orchid germplasm resources according to the present invention is shown; DETAILED DESCRIPTION

[0017] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0018] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0019] Figure 1 The flowchart of the intelligent comprehensive management method of cymbidium germplasm resources of the present invention is shown.

[0020] like Figure 1 As shown, the first aspect of the present invention provides an intelligent integrated management method for Chinese orchid germplasm resources, comprising: S1: Collect multi-source heterogeneous data of Cymbidium germplasm resources, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, and cultivation environment monitoring data; perform standardized modeling and multi-dimensional label classification on the multi-source heterogeneous data, and generate triple data based on variety genetic similarity, trait co-occurrence, and environmental adaptability. Use the triple data to construct a knowledge graph that integrates Cymbidium characteristic weights; S2: Real-time collection and analysis of user interaction behavior data at multiple time points. Contextual semantic analysis and feature extraction are performed on the user behavior data based on the CNN semantic model to obtain user behavior characteristics including flowering period preference, fragrance preference, and market popularity. Based on these user behavior characteristics, dynamic association analysis is performed on relevant entities and relationship data in the knowledge graph. Combined with the temporal correlation of each user behavior characteristic, a feature association table is generated for each user. S3: Based on the feature association table, extract entity and relationship data that match the user's features from the knowledge graph, perform semantic conversion based on the multi-dimensional feature parameters of the Chinese orchid varieties, such as flowering period, leaf type, and fragrance index, to generate germplasm resource recommendation information; S4: Select a matching encryption algorithm based on the germplasm resource recommendation information and its data type, configure access permissions according to user roles and assign independent keys, record access logs and recommendation data call records, and perform encrypted backup storage; S5: Based on the germplasm resource recommendation information, a variety of recommended Cymbidium varieties are obtained, and the entity nodes of the recommended varieties are semantically represented by a preset graph neural network (GNN) in the knowledge graph. The semantic weights of the recommended Cymbidium varieties are set, and the weighted CNN prediction model is introduced to predict the growth status and flowering period of Cymbidium by combining historical and real-time growth sequences and semantic weights.

[0021] It should be noted that in the present invention, the platform (or system) also includes a germplasm resource database, that is, a system database, which is a digital management platform dedicated to the management, retrieval, archiving and sharing of plant and animal germplasm information. Through the unified processing and structured classification (graph storage) of the present invention, the visualization efficiency and user understanding ability of Chinese orchid germplasm resource data and multi-source heterogeneous data are significantly improved, while also having a higher level of reusability and standardization in subsequent system calls and data interface development. Chinese orchid characteristic weights include the characteristic weights represented by each attribute and relationship data in the knowledge graph, such as geographic location, seasonal temperature and humidity, phenotypic and trait indicators, genetic similarity, trait co-occurrence, environmental adaptability characteristics and other related factors and characteristic weights. Genetic similarity, trait co-occurrence and environmental adaptability are stored as attribute and relationship data.

[0022] Figure 2 Shows a schematic diagram of the module of the present invention; It should be noted that the present invention includes a generation module (implementation process corresponds to S1), an acquisition module (implementation process corresponds to S2, S3), an encryption module (implementation process corresponds to S4), and an analysis and prediction module (implementation process corresponds to S4). Each module is connected at the software and hardware levels through system integration to form a complete platform for comprehensive management, analysis and prediction of Chinese orchid germplasm resource data, such as Figure 2 shown.

[0023] According to an embodiment of the present invention, S1 specifically includes: The multi-source heterogeneous data includes: Basic classification information, including variety category, flower type category, and foliage category; Geographic distribution information, including geographic coordinates of origin and cultivation areas, altitude, climate zones, and seasonal temperature and humidity changes; Genetic and molecular information, including genome sequence, transcriptome expression profile, and metabolome aroma compound profile; Phenotypic and trait information: leaf shape, flower color, fragrance index, flowering period, and stress resistance index; Perform data formatting, data cleaning, missing value filling and outlier processing on multi-source heterogeneous data, and perform unit unification and threshold normalization preprocessing according to national orchid industry standards; After preprocessing, data modeling is performed based on a multidimensional tag system, which includes genetic tags, trait tags, environmental adaptation tags, and market popularity tags; Based on the labeled data, the entity extraction, attribute feature extraction and entity relationship construction of orchids were performed to obtain triple data. The entity relationship includes the genetic similarity relationship between varieties, the co-occurrence relationship of traits, the environmental adaptability relationship and the fragrance correlation relationship; The triplet data is introduced into the weight factor of the Chinese orchid characteristics for weighted processing to generate a knowledge graph integrating the weight of the Chinese orchid characteristics; Graphical visualization is performed based on the knowledge graph structure, and each entity node and its relationship edges are interactively displayed through the user terminal to support variety comparison, feature retrieval and germplasm structure analysis.

[0024] It should be noted that entities are nodes in the graph, each node representing a Chinese orchid variety or sample, and each edge represents relational data, which represents similarities, genetic connections, or common growth conditions between different varieties. A knowledge graph is constructed based on Chinese orchid germplasm resources and their associated characteristics (such as variety, geographical environment, growth cycle, and climatic factors). Attribute data includes information such as the geographic distribution and planting characteristics of entity nodes. A knowledge graph is a type of graph-structured data that can establish graph-based relationships between different Chinese orchid varieties. It visualizes abstract germplasm data through graphics and images, enhancing users' understanding of resource structure and characteristics, improving interactive experience and decision-making efficiency. Furthermore, it can efficiently analyze the associated information of different Chinese orchid varieties, providing a data foundation for subsequent germplasm resource recommendations.

[0025] The basic classification, geographical distribution, germplasm characteristics and labeling classification methods in multi-source heterogeneous data are as follows: Basic classification: According to growth habits and adaptability, Chinese orchid germplasm resources are divided into multiple varieties, such as Chunlan, Huilan, Jianlan, Molan, Hanlan, Lianbanlan, Chunjian and Doubanlan.

[0026] Geographical distribution: Cymbidium resources are particularly abundant in Asia and Oceania, with major production areas including China, Japan, South Korea, Vietnam, Myanmar and Australia.

[0027] Germplasm characteristics: Different varieties have different genetic characteristics such as growth cycle, leaf shape, flower color, fragrance type and stress resistance.

[0028] Classification method: supports multi-dimensional classification, including labeling management of germplasm resources by variety, flower shape, region, genotype, growth environment and other dimensions.

[0029] During user interaction and data storage, the system (i.e., platform) of the present invention can implement the following data processing: The system automatically records user query behavior on the platform, including keywords, access history, and interaction behavior, for use in user interest modeling and behavioral profile analysis. Based on user query frequency and click behavior, it extracts user preference trends for specific flower types or categories, such as a tendency toward unusual flower varieties. The system uses encryption algorithms to protect key processes and data, ensuring user privacy and platform data security. Unauthorized users cannot read raw data. The system implements user role permission management, distinguishing between ordinary users, researchers, and platform administrators, with different roles having differentiated data access and operation permissions. The system defines access permissions and defines access levels for different roles based on security policies, achieving effective isolation and permission control of core data resources.

[0030] According to an embodiment of the present invention, the user terminal includes: Mobile terminal: used to install a dedicated application for managing Chinese orchid germplasm resources, supports taking images of flowers, leaves, and roots and uploading them for identification, and supports uploading fragrance sensor data and geographic positioning information; Computer terminal: with high-resolution map visualization and batch data analysis functions, supporting multi-window comparison of variety characteristics, export of variety identification reports and cultivation plans; Dedicated collection equipment terminal: Integrates aroma component sensing module, environmental parameter collection module and RFID / QR code scanning module, used to quickly collect aroma spectrum of Chinese orchid varieties, real-time climate parameters and germplasm identity information on site, and synchronizes updates with the knowledge graph system; Cloud interactive interface: supports remote calling of knowledge graph data, online subscription of recommended variety updates, and receiving cultivation warning information pushed by prediction models.

[0031] According to an embodiment of the present invention, the S2 is specifically: Based on the interaction between users and the system, the system collects user behavior data at multiple time points in real time. The behavior data includes variety search records and click frequency, flowering period selection preferences and seasonal trends, fragrance category preferences and historical evaluation data, browsing time of leaf art type and flower color combination, market transaction and collection operation records; Constructing a CNN semantic model that combines the multimodal features of orchids, wherein the input of the semantic model includes a text description vector, a fragrance spectrum feature vector, and a variety image feature vector; Through the semantic model, contextual semantic information is extracted from the entity data, relationship data, and attribute data in the knowledge graph to obtain graph document data containing three characteristics: flowering period, fragrance type, and market popularity, and multi-dimensional semantic representation is performed based on the document vector; Using the semantic model to perform semantic analysis on user behavior data, generating user behavior features including time-dependent features, and performing associated entity retrieval and relationship data analysis in the knowledge graph based on the user behavior features to obtain retrieval entity data that meets the user's interests; In the knowledge graph, the retrieval entity data is marked and an associated knowledge dataset is generated. In the associated knowledge dataset, the similarity based on the variety characteristics is calculated for each associated knowledge data, and the relationship strength is numerically represented and serialized to obtain a first sequence. Combined with the time series information of user behavior, the time correlation between each associated knowledge data and other associated knowledge data is calculated, and the time correlation is serialized to obtain a second sequence. The correlation coefficient between the first sequence and the second sequence is calculated based on the Pearson correlation coefficient, and the absolute value of the correlation coefficient is used as the interest level of the associated knowledge data; All associated knowledge data are sorted according to interest, and user behavior features are stored corresponding to associated knowledge data to form a feature association table that is independently maintained for each user.

[0032] It's important to note that the CNN semantic model, a semantic analysis model based on convolutional neural networks, leverages the semantic information representation of user behavior characteristics to further retrieve corresponding associated entity data (national orchid). During associated entity retrieval and relationship data analysis within the knowledge graph, the corresponding retrieved entity data is retrieved, and dynamic association analysis of entity and relationship data within the knowledge graph is performed. The contextual representation of behavioral characteristics within the knowledge graph is evaluated, and the corresponding entity data is labeled.

[0033] In the knowledge graph, relationships between entities are generally determined by the correlations between orchids from different countries, including geographical location, growth patterns, varieties, and climate factors. User behavior data, including browsing history, search keywords, click volume, and conversion behavior, possesses time-series and context-dependent characteristics. This data can be used to characterize user profiles, including demographic information, behavioral pathways, and interests.

[0034] The CNN semantic model performs semantic analysis and feature classification on feature data through convolutional layers, pooling layers and fully connected layers.

[0035] In the user behavior data (or user behavior feature vector), each user behavior data corresponds to a time node, each associated knowledge data also corresponds to a time node, and each user corresponds to a related knowledge data set. Graph document data is specifically a type of contextual data, constructed based on the entity data of the knowledge graph. In the present invention, this contextual data is used to retrieve entities associated with user behavior features. Specifically, the retrieval entity data is labeled with the corresponding associated entities. The associated knowledge data includes the retrieval entity data, the associated entity, the corresponding graph association information of the entity, and the entity attribute information.

[0036] Relationship strength is derived from edge information in the graph. Stronger strength indicates a closer association between entities, and in the knowledge graph, closer distances between entities. Here, we use category characteristics as the association dimension for analysis. The first sequence stores multiple association strength values, representing the associations between a piece of associated knowledge data and multiple other associated knowledge data entities. The second sequence stores multiple temporal correlations, derived through temporal correlation analysis. Temporal correlation is the distance between two time nodes; the greater the distance, the greater the temporal correlation. Temporal correlation analysis can reflect the interaction characteristics and similarities of users over time, thereby evaluating the effectiveness of recommended data and uncovering potential user resource data needs.

[0037] The feature association table stores the association status between user behavior features, associated knowledge data, interest level and data.

[0038] It is worth mentioning here that in the interactive analysis and recommendation of traditional germplasm resource platforms, user-based behavior analysis is often based on a single dimension, the real-time recommendation effect is poor, and the retrieval efficiency of real-time related data between high-frequency interactive users and germplasm big data is low. It is difficult to adapt to efficient and dynamic recommendations under the condition of high-frequency interaction between big data and multiple users, and there is a lack of recommendation association pattern analysis methods, resulting in a single dimension of recommendation data and difficulty in exploring users' potential needs.

[0039] In the present invention, behavioral feature analysis is performed by collecting user behavior data at multiple time nodes, and before that, a multi-dimensional knowledge graph of Cymbidium germplasm resources with contextual semantic information is constructed. Based on the semantic analysis form, graph knowledge data retrieval analysis and entity association analysis are performed on different behavioral features, and multiple entity data are retrieved to construct associated knowledge data. Time correlation is introduced, and the relationship between the corresponding associated knowledge data of users at different time nodes is analyzed. The interest level of recommended knowledge is set, and a feature association table is maintained for each user. In addition, based on the real-time update of newly added user behavior characteristics or knowledge graphs, the associated table can be updated in real time at the same time, so as to achieve efficient and accurate generation of recommendation data for different users, and build a recommendation model that combines germplasm resources with graphs, effectively utilizing the correlation between graph knowledge and user characteristics, and combining semantic model analysis to effectively analyze user behavior characteristics and build user portraits for associated knowledge retrieval, so as to achieve dynamic data recommendation for users, and further set the update and maintenance rules of the associated table to improve the real-time and applicability of recommendations, and realize dynamic and efficient recommendation analysis for more users and big data germplasm platforms. Based on the associated knowledge of the graph, it can dynamically predict the user's future potential needs, and automatically match and push the model that best suits the content or resources, thereby improving user satisfaction and interaction efficiency.

[0040] According to an embodiment of the present invention, S3 specifically includes: Based on the interest level in the feature association table and the multidimensional feature weights of the orchid varieties, multiple associated knowledge data that meet the user's preferences are screened out. The multidimensional feature weights include: Flowering period matching: calculated based on the user's regional climate conditions and the historical flowering period prediction results of Chinese orchid varieties; Fragrance compatibility: calculated based on the similarity of fragrance spectra and user fragrance preference labels; Preference for foliage and flower color combinations: calculated based on visual feature scores of the foliage and flower color categories of the variety; Market circulation popularity: a comprehensive score based on transaction records, number of collections and online discussion popularity; Environmental adaptability index: calculated based on the matching degree between the variety cultivation environment requirements and the environmental parameters of the user's location; Extract the entity and relationship data corresponding to the above-mentioned associated knowledge data from the knowledge graph, perform semantic conversion, and generate multimodal germplasm resource recommendation information including variety name, trait characteristics, flowering period prediction, fragrance index, cultivation suggestions and market reference price; Prioritizing the recommended information based on weighted interest, wherein the weighted interest is a weighted sum of flowering period matching, fragrance compatibility, market circulation popularity, and environmental adaptability index parameters; The highest priority recommendation information will be displayed first on the user terminal interface, and secondary screening and sorting will be supported based on conditions such as flowering period, fragrance, market popularity, etc.

[0041] In the feature association table, interest degree can achieve one-time screening, and multi-dimensional feature weights can achieve two-time screening, and interest-weighted screening can be performed in combination with multi-dimensional evaluation criteria.

[0042] According to an embodiment of the present invention, the S4 is specifically: Based on the germplasm resource recommendation information generated by S3, determine the business scenario and recommended data type of the current recommendation process; The business scenarios include variety query scenarios, cultivation management scenarios, variety trading and circulation scenarios, variety copyright confirmation and infringement monitoring scenarios; Analyze the security level of recommended data based on business scenarios and recommended data types, and match the corresponding encryption algorithm. Use symmetric encryption algorithms (AES, SM4, etc.) for non-sensitive data such as general search data; Use asymmetric encryption algorithms (RSA, SM2, etc.) and digital signatures for sensitive data such as transaction contracts and copyright certificates; Use end-to-end encryption protocol for cross-platform data transmission.

[0043] Configure access rights and assign independent keys based on user roles, including ordinary users, registered merchants, breeding institutions, and platform administrators. Different roles have different restrictions on access scope, data export, and secondary distribution permissions; Combined with the blockchain traceability mechanism, the unique identity of the national orchid varieties, transaction records, copyright information and call records of recommended data are stored on the chain, making the data tamper-proof and traceable; During the user access process, access logs, variety transaction records, copyright certificate call records and germplasm resource recommendation information access records are extracted through permission verification, and the above access data are encrypted and stored in the disaster recovery backup system to prevent data loss and illegal leakage.

[0044] It should be noted that based on the business scenario and the type of recommended data, we can analyze the data permission level and security requirement level in the corresponding recommended scenario, and further match different levels of encryption algorithms to ensure data security. Market popularity refers to market circulation popularity.

[0045] S4's encryption process ensures the system's audit traceability and disaster recovery capabilities, and encryption scenarios include but are not limited to typical usage scenarios such as resource retrieval, planting service appointments, variety selection and comparison.

[0046] Recommendation data types can be further categorized into multimodal information, including structured content data (e.g., database fields), connected data (e.g., similarity associations), and graphic data (e.g., flower pattern images + text descriptions). The encryption algorithm can be symmetric or asymmetric, such as DES and AES, offering advantages such as fast encryption speed, high security, and low implementation costs, making it suitable for real-time protection of the recommendation process. User roles include, but are not limited to, administrators, reviewers, and ordinary users, with different roles possessing varying system access permissions and operational boundaries.

[0047] This method matches the appropriate encryption algorithm to the recommendation scenario and data type, and assigns access permissions based on user roles, enabling data protection and permission management throughout the recommendation process. Furthermore, user access records and resource call data are captured and encrypted for backup, further enhancing the system's overall data security, auditability, and disaster recovery capabilities.

[0048] According to an embodiment of the present invention, the S5 is specifically: Based on the recommended information of germplasm resources, a variety of recommended Chinese orchid varieties are obtained, and entity and attribute data related to the variety are extracted from the knowledge graph to construct a graph structure containing variety characteristics, genetic information, trait information, geographic information and cultivation parameters; In the graph structure, the relationship information of the five dimensions of variety, trait, geography, genetics and quantity indicators is analyzed, and the following indicators specific to the orchid industry are added to the relationship information: Climate adaptability coefficient: calculated based on the similarity between the target cultivation area's multi-year meteorological data (temperature, humidity, light, precipitation) and the variety's historical cultivation environment; Flowering time window prediction value: Based on historical flowering period data and real-time environmental monitoring data, the time series analysis model is used to predict the most probable flowering start and end dates; Fragrance stability index: Based on the aroma component test results of previous years, it evaluates the degree of fluctuation of the aroma component ratio of the variety in different environments; The entity nodes of the recommended Chinese orchid varieties are semantically represented through a preset graph neural network, and the comprehensive semantic weight of each recommended Chinese orchid variety is calculated in combination with the above-mentioned industry-specific indicators; The historical growth sequences and current real-time growth sequences of various recommended Chinese orchid varieties are input into the weighted CNN prediction model, and the comprehensive semantic weight is introduced as the authenticity of the growth sequence for sequence learning and prediction training. The output during the prediction process includes growth status level, flowering time window, and fragrance stability rating. Recommended cultivation measures are set through the prediction output, and the prediction results are generated and automatically pushed to the user terminal to assist in variety selection and cultivation management decisions.

[0049] It should be noted that in sequence prediction training, semantic weights are used as the authenticity of the historical growth sequences of different varieties. Based on the authenticity of the data, the sequence training of the CNN prediction model can be biased towards sequences with higher authenticity. Targeted sequence feature learning is performed based on the weights to construct an accurate prediction model for current germplasm resource recommendation information. In addition, during each prediction training process, semantic weights can be introduced to set the authenticity of different training sequences for loss calculation in the sequence prediction results, so that the weighted CNN prediction model can be biased towards variety sequences with higher authenticity for feature learning. The prediction results include growth status level, flowering time window, fragrance stability rating, and recommended cultivation measures.

[0050] The graph structure is a subgraph based on the S1 knowledge graph, but the relational data is different. Graph neural networks (GNNs) specifically include GCN and GAT.

[0051] The specific data characteristics of the five dimensions of variety, trait, geography, genetics, entity, and quantity indicators are as follows: Variety information: such as the name of the Chinese orchid, scientific name, variety origin, breeder information, parent materials, etc.; Trait information: such as flower color, leaf shape, fragrance, flowering period, polysaccharide or sesquiterpene content, etc. Geographic information: such as planting area, longitude and latitude, altitude, climate parameters (temperature, humidity, etc.); Genetic information: such as genotype, phenotypic differences, allele frequency, and heterozygosity; Quantitative indicators: such as sample size, number of plants, number of flowers, pollination rate, etc.

[0052] The weighted CNN prediction model uses the Conv1D one-dimensional convolutional layer and the one-dimensional pooling layer to capture the temporal features and dependencies of the sequence at different time scales. During the prediction process, the model outputs results through the fully connected layer. Furthermore, the data requires sequence normalization preprocessing before importing the historical growth sequence. In crop prediction, the output of the prediction sequence can be used to obtain flowering period classifications (such as early-flowering, medium-flowering, and late-flowering) or specific predicted date ranges and statuses (such as "expected flowering between late March and early April," growth status predictions, etc.).

[0053] A pre-set graph neural network represents the semantic information of the entity nodes of the recommended orchid varieties. The semantic weight of each recommended orchid variety is analyzed through the graph structure. Specifically, the contextual importance of each entity is evaluated based on the semantic relationship between the entities of the recommended orchid varieties in the graph structure. The context is formed by semantic analysis of all entity relationship attribute information in the graph structure. The criticality of the entity in the context is analyzed in combination with the graph structure to obtain the contextual importance. In the graph structure, the relationship data between each entity can be analyzed to obtain the association degree. The association degree is determined by the number and strength of the entity edges. The number and strength of the entity edges are determined by the relationship between five dimensions.

[0054] Both context importance and entity relevance can be analyzed as semantic weights. Furthermore, semantic weights can be obtained based on a weighted average of the two values. Optionally, on the basis of the semantic weights, specific indicators of the orchid industry can be added for secondary weight calculation, and the climate adaptability coefficient, flowering time window prediction value, fragrance stability index, etc. can be introduced for comprehensive semantic weight calculation. Based on the prediction needs and status analysis needs of orchids, different types of weight indicators can be freely matched as semantic weights for calculation to improve the platform's adaptability to big data analysis.

[0055] It's worth noting that the flowering period of Cymbidium orchids is influenced by a complex coupling of factors, including the genetic characteristics of the variety, environmental factors (temperature, light, humidity), and growth history. Existing technologies offer limited methods for predicting plantings based on recommended Cymbidium orchid planting resource data, and lack growth forecasting analysis that integrates varietal, geographic, and genetic associations. Traditional prediction methods struggle to handle inter-varietal correlations and high-dimensional spatiotemporal data (growth sequences), resulting in poor generalization.

[0056] Based on this, the present invention constructs a sub-graph for the recommended Chinese orchid varieties to form a graph structure, sets relational data based on multi-dimensional Chinese orchid characteristics to obtain a complete graph structure, uses a neural network to extract the semantic information of the recommended Chinese orchid varieties in the graph structure and calculates the semantic weight, and uses the semantic weight and growth sequence to learn the sequence characteristics of the recommended Chinese orchid varieties through the CNN prediction model. Finally, the CNN prediction model is used to perform accurate growth prediction for the recommended Chinese orchid varieties.

[0057] The present invention can realize heterogeneous data fusion prediction, combine graph structure (variety association, geographical association, etc.) with time series environmental data for fusion prediction training, realize dynamic adaptability of recommended variety prediction, and realize nonlinear response modeling of planting prediction and multi-dimensional environmental fluctuations. The present invention uses graph neural networks to realize cross-variety semantic feature propagation and fusion learning, which significantly improves the accuracy and generalization ability of national orchid flowering time prediction, especially showing good prediction effects in scenarios with a wide variety of varieties and strong data heterogeneity.

[0058] During the training process of the CNN prediction model, a supervision mechanism can be introduced or the model structure can be iteratively optimized for hyperparameters based on reinforcement learning; Furthermore, when the graph structure is sparse or sample data is insufficient, it can automatically switch to a random walk model (such as DeepWalk or node2vec) to capture semantic feature relationships. This walk strategy samples paths, learns semantic relationships between entity nodes, and then connects time series for lightweight prediction. A random walk model constructs sequences on a graph by simulating node-to-node walk behavior and uses sequence context to learn node representations. It can be used in scenarios where the graph structure is sparse or where deep learning models are difficult to train.

[0059] According to an embodiment of the present invention, in S5, the prediction results obtained by predicting the growth status and flowering period of the Chinese orchid are displayed in a multimodal visualization manner through the user terminal, and the multimodal visualization includes: Flowering period forecast calendar view: Mark the expected flowering start and end dates in the form of a timeline, and dynamically display the overlap of flowering periods of different varieties; Aroma component radar chart: displays the relative content and stability index of major aroma compounds (such as linalool, geraniol, phenylethyl alcohol, etc.); Climate suitability heat map: displays suitable cultivation areas on the map based on the climate suitability coefficient, and supports zooming and area filtering; Growth status dynamic chart: Generate curves or animations based on historical and real-time growth sequences to show the changing trends of indicators such as plant height, number of leaves, and flower bud differentiation; Cultivation suggestion panel: automatically generates corresponding fertilization, irrigation, temperature and humidity control, and shading suggestions based on the prediction results, and supports users to export them as a management plan with one click.

[0060] The visual interface supports interactive operations, including filtering, comparison and collection by flowering period, fragrance or climatic conditions, and allows superimposed analysis of the prediction information of the selected variety and historical cultivation records to assist users in making variety selection and management decisions.

[0061] According to an embodiment of the present invention, a containerized deployment module is further included, which includes: Decomposition modules: determine the system architecture of the Cymbidium germplasm resources data management system and decompose the system architecture into multiple portable modules; Select modules: Obtain requirements for portable modules and select cloud service providers and development infrastructure based on these requirements; Packaging module: writing application code according to the system architecture and packaging the portable module, cloud service provider and infrastructure; Startup Module: Create and manage containers in real time based on Docker technology, deploy packaged modules into containers, and start them in the cloud, ensuring that the containers are correctly configured and compatible with the infrastructure.

[0062] It should be noted that the Cymbidium germplasm resource data management system is a software system for efficient management of Cymbidium germplasm resources. The system architecture may include: user interface layer, business logic layer, data storage layer, etc. The Cymbidium germplasm resource data management system is the system of the present invention, and multiple portable modules include but are not limited to: Germplasm resource information management module: realizes basic functions such as adding, deleting, modifying and querying germplasm resources; Data analysis and management module: provides statistical analysis tools, supports chart visualization, and assists decision-making; Permission management and authentication module: provides functions such as user registration, login, access control, and permission classification; Data backup and recovery module: realize automatic data backup, manual recovery, and prevent data loss or damage.

[0063] Cloud service providers such as Alibaba Cloud, Huawei Cloud, Tencent Cloud, Amazon AWS, and Microsoft Azure support elastic computing, storage, and network resources. Infrastructure includes cloud resource configurations such as container runtime environments, virtual machines, network topology, object storage, and load balancers. Docker is a container technology platform that supports rapid build, delivery, and deployment. It encapsulates applications and their dependencies into images, enabling consistent operation across platforms. This invention leverages the Docker technology platform to achieve portable deployment of each module.

[0064] According to an embodiment of the present invention, a blockchain traceability module is also included, which includes: Collection and recording module: collects key data nodes and records hash values; On-chain module: uses a consortium chain structure to write verified data to the chain and bind it to the smart contract; Verification module: Set up traceability verification based on user role access policy; Evidence storage module: uses hash values ​​and timestamp information on the chain to form data evidence; It should be noted that the blockchain platform can be a consortium chain platform that supports controllable permissions, such as Hyperledger Fabric and Fisco BCOS, and the smart contract language can be Go or Solidity.

[0065] Key data nodes include resource registration, analysis records, recommendation results, and predicted tags. The verification module ensures that the data accessed by visitors is authentic, complete, and untampered. The evidence storage module meets the need for trusted traceability of Chinese orchid germplasm resources in scientific research and trading scenarios. Application scenarios for the blockchain traceability module include: variety rights protection, breeding process traceability, circulation record storage, and user behavior log tamper-proofing.

[0066] This implementation combines blockchain traceability technology with a containerized deployment mechanism. This not only ensures the security, traceability, and integrity of key data within the Cymbidium germplasm resource management system, but also leverages technologies like Docker to enable portability, ease of deployment, and high scalability of system modules. The containerized structure facilitates rapid deployment in diverse environments (on-premises, edge, and cloud), while blockchain ensures trustworthiness and verifiability throughout the entire process, providing robust digital security for research, breeding, and trading scenarios.

[0067] Figure 3 A schematic diagram of the containerized deployment module and blockchain traceability module of the present invention is shown; like Figure 3 The figure shows a schematic diagram of the combination of the containerized deployment module and the blockchain traceability module. The upper left part of the figure includes four portable modules, the upper right part shows the Docker container and blockchain traceability part, and the lower part shows that the containerized deployment module is modularly deployed based on the system architecture by decomposing, selecting and packaging modules, and performing operations such as requesting, uploading and verifying the data collected after deployment.

[0068] Figure 4 It shows a schematic diagram of the operation of the prediction credibility verification and scheduling module of the present invention; According to an embodiment of the present invention, a prediction credibility verification and scheduling module is also included, which includes: First Verification Module: When a user initiates a prediction analysis request, a unique task identifier and input data hash summary are generated for each task and recorded on the blockchain; Second verification module: Before the system executes the prediction model, it obtains the fingerprint of the current container operating environment, recalculates the summary together with the prediction parameters, and writes it to the chain; Smart contract module: The smart contract deployed on the blockchain platform automatically verifies whether the task summary matches the container environment, and triggers predictive analysis after verification; On-chain result recording module: Generates a hash summary of the prediction output result and writes it into the blockchain with the system private key signature; Heterogeneous container scheduling module: Automatically schedules the most appropriate instance to execute the task across multiple container nodes based on the predicted task's label attributes; Block query and audit module: Based on the user or administrator's ability to view the on-chain status, call history, input and output summary, and model version of the prediction task.

[0069] It should be noted that the present invention provides a prediction credibility verification and scheduling module based on the deep integration of blockchain and containerized deployment, which is used to enhance data security, execution transparency and process traceability in the prediction and analysis process of Chinese orchid germplasm resources.

[0070] The task identifier is a unique identifier generated for each prediction request, used for on-chain task indexing and subsequent tracking. The input data hash digest uses algorithms such as SHA-256 to digest key prediction inputs to ensure that input parameters have not been tampered with. The task identifier consists of a UUID and a timestamp.

[0071] The container runtime environment fingerprint includes but is not limited to: model image name and version, container host information, GPU or CPU hardware identifier, call timestamp, container unique ID, etc.

[0072] The on-chain result recording module realizes the non-repudiation and a posteriori verifiability of the results.

[0073] Smart contracts are implemented in Solidity or Fabric chaincode and are used to perform a series of trusted process controls such as parameter matching verification, task authorization verification, and scheduling logic triggering.

[0074] The heterogeneous container scheduling module reads task tags and runtime status to schedule across multiple container instances, ensuring correct model versions, satisfied permissions, and resource adaptation to support the flexible execution of complex tasks. Tag attributes include model type, permission level, node load, etc.

[0075] The blockchain is deployed using a private chain (such as Hyperledger Fabric, Ethereum private chain, etc.). All prediction-related records are written into the block in the form of transactions, forming a complete prediction process log chain for auditing and accountability tracing.

[0076] The prediction credibility verification and scheduling module realizes the trusted record of the entire process from prediction request initiation to model execution, result output, and audit backtracking, improving the transparency, security and accountability traceability of the system's germplasm resource predictions. It is suitable for key scenarios such as scientific research verification, smart agricultural management, and variety certification and filing.

[0077] The prediction credibility verification and scheduling module can be seamlessly integrated with Docker containers, Kubernetes orchestration platforms, blockchain nodes, and predictive analysis services (such as deep learning model APIs), and provide services externally through RESTful interfaces or gRPC protocols.

[0078] The blockchain trusted verification and container scheduling module proposed in this invention greatly improves the performance of the national orchid germplasm resource prediction and analysis platform in terms of security, compliance, and controllability, and provides a solid foundation for achieving large-scale multi-node collaborative prediction.

[0079] By combining blockchain with containerized deployment, this invention achieves traceability, verifiability, and tamper-proof management of the entire process of orchid growth prediction tasks, greatly improving the credibility and transparency of prediction results. The execution process of the prediction task, including key links such as input data, model version, operating environment, and output results, can be recorded on the chain to ensure that all actions are non-repudiable and facilitate subsequent traceability and auditing. At the same time, the container fingerprint mechanism ensures the consistency of the model execution environment, avoids prediction bias due to system differences, and improves the stability and reproducibility of the results. The system also supports elastic container scheduling based on heterogeneous resources, and can intelligently allocate computing tasks according to task priority and resource status, significantly improving overall response efficiency and resource utilization. In scenarios where multiple departments or multiple institutions collaborate, the blockchain traceability mechanism effectively prevents data abuse and enhances the compliance and data security of the collaborative process. Combined with on-chain encryption summaries and container-level access control, it can also resist illegal access and malicious tampering. In addition, containerized deployment supports rapid integration, flexible expansion, and version rollback, providing technical support for the engineering implementation and continuous iteration of the Guolan Intelligent Prediction Platform, and promoting its large-scale application in scientific research, agricultural management, and commercial services.

[0080] According to an embodiment of the present invention, the further embodiment includes: In one interaction cycle, the user behavior characteristics and germplasm resource recommendation information of multiple interactive users are collected; The user behavior features are vectorized to obtain the behavior feature vector. The Word2Vec model is used to construct text information from the knowledge graph, and the germplasm resource recommendation information is semantically analyzed based on the CNN semantic model, and word vectors are formed by combining the text information. The feature vector and word vector of each user are used as clustering sample data, and users are clustered based on the kmeans clustering algorithm. Multiple user groups are obtained based on the clustering results. Grouping the prediction results by multiple user groups to obtain multiple prediction data groups; Each prediction data set is encrypted and stored on the chain.

[0081] It should be noted that the present invention clusters users with similar prediction characteristics and behavioral characteristics based on two-dimensional information (behavioral characteristics and prediction characteristics) and classifies and packages the prediction data on the chain. In the subsequent tracing of prediction results and historical data queries, it can effectively improve the efficiency of tracing back queries for similar data. Moreover, through the two-dimensional cluster analysis of prediction data, it can effectively mine user prediction results with similar retrieval and prediction characteristics, and can achieve efficient and stable data chaining, encrypted storage and traceability query processes in multi-user high-frequency interaction scenarios. In the process of clustering feature vectors and word vectors as cluster sample data, the similarity between cluster sample data is measured by the mean of the Euclidean distance of the two vectors.

[0082] Figure 5 A block diagram of an intelligent integrated management system for cymbidium germplasm resources of the present invention is shown.

[0083] The second aspect of the present invention further provides an intelligent integrated management system 5 for Chinese orchid germplasm resources, the system comprising: Memory: used to store the intelligent integrated management program for Chinese orchid germplasm resources and multi-source heterogeneous data sets, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, cultivation environment monitoring data, user behavior data, and market transaction records; Processor: used to execute the management program to implement the steps of the above-mentioned intelligent comprehensive management method for Chinese orchid germplasm resources; Aroma sensor interface module: used to connect to aroma component collection equipment such as gas chromatography-mass spectrometry (GC-MS) or electronic nose, and directly transmit the collected aroma spectrum data to the system database; Climate and environmental data acquisition module: used to collect light intensity, air temperature and humidity, soil temperature and humidity, and carbon dioxide concentration in real time, and dynamically linked with the prediction model; Blockchain node module: used to store unique product identification, transaction contracts, copyright certificates, and recommended data call records, to achieve decentralized storage and traceable management of data; User interactive terminal interface: used to connect mobile terminals, computer terminals and special collection equipment, supporting real-time display and interactive operation of multimodal prediction results such as flowering calendars, fragrance radar maps, and climate heat maps.

[0084] The intelligent integrated management system for cymbidium germplasm resources can realize any step of the above-mentioned intelligent integrated management method for cymbidium germplasm resources when it is in operation.

[0085] The third aspect of the present invention also provides a computer-readable storage medium, which stores program instructions for executing any one of the intelligent integrated management methods for Chinese cymbidium germplasm resources.

[0086] This invention discloses an intelligent, integrated management method and system for Chinese orchid germplasm resources. The system includes the following steps: collecting basic information about Chinese orchid germplasm resources, generating a knowledge graph through standardized modeling and classification; analyzing user behavior data in real time, extracting features using a CNN semantic model, and dynamically associating these features with the knowledge graph to generate a user feature association table; extracting graph data from the feature association table to generate recommended germplasm resource information; matching encryption algorithms and configuring role permissions to achieve secure access to recommended information and data backup; and predicting planting results using a CNN model based on the graph neural network semantic representation and growth sequence of recommended varieties. This invention integrates graphs and deep learning technologies to achieve precise recommendations for germplasm resources and secure data management, thereby improving the efficiency of germplasm resource management and the scientific nature of decision-making.

[0087] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0088] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0089] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0090] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0091] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0092] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. An intelligent integrated management method for cymbidium germplasm resources, characterized in that: include: S1: Collect multi-source heterogeneous data of Cymbidium germplasm resources, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, and cultivation environment monitoring data; perform standardized modeling and multi-dimensional label classification on the multi-source heterogeneous data, and generate triple data based on variety genetic similarity, trait co-occurrence, and environmental adaptability. Use the triple data to construct a knowledge graph that integrates Cymbidium characteristic weights; S2: Real-time collection and analysis of user interaction behavior data at multiple time points. Contextual semantic analysis and feature extraction are performed on the user behavior data based on the CNN semantic model to obtain user behavior characteristics including flowering period preference, fragrance preference, and market popularity. Based on the user behavior characteristics, dynamic association analysis is performed on the relevant entities and relationship data in the knowledge graph, and the time correlation of each user behavior characteristic is combined to generate a feature association table for each user; S3: Based on the feature association table, extract entity and relationship data that match the user's features from the knowledge graph, perform semantic conversion based on the multi-dimensional feature parameters of the Chinese orchid varieties, such as flowering period, leaf type, and fragrance index, to generate germplasm resource recommendation information; S4: Select a matching encryption algorithm based on the germplasm resource recommendation information and its data type, configure access permissions according to user roles and assign independent keys, record access logs and recommendation data call records, and perform encrypted backup storage; S5: Based on the germplasm resource recommendation information, a variety of recommended Cymbidium varieties are obtained, and the entity nodes of the recommended varieties are semantically represented in the knowledge graph through a preset graph neural network, and the semantic weights of the recommended Cymbidium varieties are set. Combined with the historical and real-time growth sequences and semantic weights, a weighted CNN prediction model is introduced to predict the growth status and flowering period of Cymbidium.

2. The intelligent integrated management method for cymbidium germplasm resources according to claim 1, characterized in that: Said S1 specifically includes: The multi-source heterogeneous data includes: Basic classification information, including variety category, flower type category, and foliage category; Geographic distribution information, including geographic coordinates of origin and cultivation areas, altitude, climate zones, and seasonal temperature and humidity changes; Genetic and molecular information, including genome sequence, transcriptome expression profile, and metabolome aroma compound profile; Phenotypic and trait information: leaf shape, flower color, fragrance index, flowering period, and stress resistance index; Perform data formatting, data cleaning, missing value filling and outlier processing on multi-source heterogeneous data, and perform unit unification and threshold normalization preprocessing according to national orchid industry standards; After preprocessing, data modeling is performed based on a multidimensional tag system, which includes genetic tags, trait tags, environmental adaptation tags, and market popularity tags; Based on the labeled data, the entity extraction, attribute feature extraction and entity relationship construction of orchids were performed to obtain triple data. The entity relationship includes the genetic similarity relationship between varieties, the co-occurrence relationship of traits, the environmental adaptability relationship and the fragrance correlation relationship; The triplet data is introduced into the weight factor of the Chinese orchid characteristics for weighted processing to generate a knowledge graph integrating the weight of the Chinese orchid characteristics; Graphical visualization is performed based on the knowledge graph structure, and each entity node and its relationship edges are interactively displayed through the user terminal to support variety comparison, feature retrieval and germplasm structure analysis.

3. The intelligent integrated management method for cymbidium germplasm resources according to claim 2, characterized in that: The user terminal includes: Mobile terminal: used to install a dedicated application for managing Chinese orchid germplasm resources, supports taking images of flowers, leaves, and roots and uploading them for identification, and supports uploading fragrance sensor data and geographic positioning information; Computer terminal: with high-resolution map visualization and batch data analysis functions, supporting multi-window comparison of variety characteristics, export of variety identification reports and cultivation plans; Dedicated collection equipment terminal: Integrates aroma component sensing module, environmental parameter collection module and RFID / QR code scanning module, used to quickly collect aroma spectrum of Chinese orchid varieties, real-time climate parameters and germplasm identity information on site, and synchronizes updates with the knowledge graph system; Cloud interactive interface: supports remote calling of knowledge graph data, online subscription of recommended variety updates, and receiving cultivation warning information pushed by prediction models.

4. The intelligent integrated management method for cymbidium germplasm resources according to claim 1, characterized in that: The S2 specifically includes: Based on the interaction between users and the system, the system collects user behavior data at multiple time points in real time. The behavior data includes variety search records and click frequency, flowering period selection preferences and seasonal trends, fragrance category preferences and historical evaluation data, browsing time of leaf art type and flower color combination, market transaction and collection operation records; Constructing a CNN semantic model that combines the multimodal features of orchids, wherein the input of the semantic model includes a text description vector, a fragrance spectrum feature vector, and a variety image feature vector; Through the semantic model, contextual semantic information is extracted from the entity data, relationship data, and attribute data in the knowledge graph to obtain graph document data containing three characteristics: flowering period, fragrance type, and market popularity, and multi-dimensional semantic representation is performed based on the document vector; Using the semantic model to perform semantic analysis on user behavior data, generating user behavior features including time-dependent features, and performing associated entity retrieval and relationship data analysis in the knowledge graph based on the user behavior features to obtain retrieval entity data that meets the user's interests; In the knowledge graph, the retrieval entity data is marked and an associated knowledge dataset is generated. In the associated knowledge dataset, the similarity based on the variety characteristics is calculated for each associated knowledge data, and the relationship strength is numerically represented and serialized to obtain a first sequence. Combined with the time series information of user behavior, the time correlation between each associated knowledge data and other associated knowledge data is calculated, and the time correlation is serialized to obtain a second sequence. The correlation coefficient between the first sequence and the second sequence is calculated based on the Pearson correlation coefficient, and the absolute value of the correlation coefficient is used as the interest level of the associated knowledge data; All associated knowledge data are sorted according to interest, and user behavior features are stored corresponding to associated knowledge data to form a feature association table that is independently maintained for each user.

5. The intelligent integrated management method for cymbidium germplasm resources according to claim 4, characterized in that: The S3 specifically includes: Based on the interest level in the feature association table and the multidimensional feature weights of the orchid varieties, multiple associated knowledge data that meet the user's preferences are screened out. The multidimensional feature weights include: Flowering period matching: calculated based on the user's regional climate conditions and the historical flowering period prediction results of Chinese orchid varieties; Fragrance compatibility: calculated based on the similarity of fragrance spectra and user fragrance preference labels; Preference for foliage and flower color combinations: calculated based on visual feature scores of the foliage and flower color categories of the variety; Market circulation popularity: a comprehensive score based on transaction records, number of collections and online discussion popularity; Environmental adaptability index: calculated based on the matching degree between the variety cultivation environment requirements and the environmental parameters of the user's location; Extract the entity and relationship data corresponding to the above-mentioned associated knowledge data from the knowledge graph, perform semantic conversion, and generate multimodal germplasm resource recommendation information including variety name, trait characteristics, flowering period prediction, fragrance index, cultivation suggestions and market reference price; Prioritizing the recommended information based on weighted interest, wherein the weighted interest is a weighted sum of flowering period matching, fragrance compatibility, market circulation popularity, and environmental adaptability index parameters; The highest priority recommendation information will be displayed first on the user terminal interface, and secondary screening and sorting will be supported based on flowering period, fragrance type, and market popularity.

6. The intelligent integrated management method for cymbidium germplasm resources according to claim 1, characterized in that: The S4 specifically includes: Based on the germplasm resource recommendation information generated by S3, determine the business scenario and recommended data type of the current recommendation process; The business scenarios include variety query scenarios, cultivation management scenarios, variety trading and circulation scenarios, variety copyright confirmation and infringement monitoring scenarios; Analyze the security level of recommended data based on business scenarios and recommended data types, and match the corresponding encryption algorithm. Use asymmetric encryption algorithms for sensitive data and symmetric encryption algorithms for non-sensitive data. Use end-to-end encryption protocols for cross-platform data transmission; Configure access rights and assign independent keys based on user roles, including ordinary users, registered merchants, breeding institutions, and platform administrators. Different roles have different restrictions on access scope, data export, and secondary distribution permissions; Combined with the blockchain traceability mechanism, the unique identity identification, transaction records, copyright information and recommendation data of national orchid varieties are stored on the chain; During the user access process, access logs, variety transaction records, copyright certificate call records and germplasm resource recommendation information access records are extracted through permission verification, and the data of the above access process are encrypted and stored in the disaster recovery backup system.

7. The intelligent integrated management method for cymbidium germplasm resources according to claim 1, characterized in that: The S5 specifically includes: Based on the recommended information of germplasm resources, a variety of recommended Chinese orchid varieties are obtained, and entity and attribute data related to the variety are extracted from the knowledge graph to construct a graph structure containing variety characteristics, genetic information, trait information, geographic information and cultivation parameters; In the graph structure, the relationship information of the five dimensions of variety, trait, geography, genetics and quantity indicators is analyzed, and the following indicators specific to the orchid industry are added to the relationship information: Climate adaptability coefficient: calculated based on the similarity between the meteorological data of the target cultivation area and the historical cultivation environment of the variety; Flowering time window prediction value: Based on historical flowering period data and real-time environmental monitoring data, the time series analysis model is used to predict the most probable flowering start and end dates; Fragrance stability index: Based on the aroma component test results of previous years, it evaluates the degree of fluctuation of the aroma component ratio of the variety in different environments; The entity nodes of the recommended Chinese orchid varieties are semantically represented through a preset graph neural network, and the comprehensive semantic weight of each recommended Chinese orchid variety is calculated in combination with the above-mentioned industry-specific indicators; The historical growth sequences and current real-time growth sequences of various recommended Chinese orchid varieties are input into the weighted CNN prediction model, and the comprehensive semantic weight is introduced as the authenticity of the growth sequence for sequence learning and prediction training. During the prediction process, the output includes growth status level, flowering time window, and fragrance stability rating. Recommended cultivation measures are set through the prediction output, and the prediction results are generated and automatically pushed to the user terminal.

8. The intelligent integrated management method for cymbidium germplasm resources according to claim 1, characterized in that: In S5, the prediction results obtained by predicting the growth status and flowering period of the Cymbidium orchid are displayed in a multimodal visualization manner through the user terminal, and the multimodal visualization includes: Flowering period forecast calendar view: Mark the expected flowering start and end dates in the form of a timeline, and dynamically display the overlap of flowering periods of different varieties; Aroma component radar chart: displays the relative content and stability index of major aroma compounds; Climate suitability heat map: displays suitable cultivation areas on the map based on the climate suitability coefficient, and supports zooming and area filtering; Growth status dynamic graph: Generate curves or animations based on historical and real-time growth sequences to show the changing trends of plant height, leaf number, and flower bud differentiation indicators; Cultivation suggestion panel: automatically generates corresponding fertilization, irrigation, temperature and humidity control, and shading suggestions based on the prediction results, and exports them as a management plan; The visual interface supports interactive operations, including screening, comparison and collection by flowering period, fragrance or climatic conditions, and superimposed analysis of the prediction information of the selected variety and historical cultivation records.

9. An intelligent integrated management system for Chinese orchid germplasm resources, characterized in that: The system includes: Memory: used to store the intelligent integrated management program for Chinese orchid germplasm resources and multi-source heterogeneous data sets, including genome sequences, transcriptome expression profiles, metabolome aroma components, phenotypic image data, cultivation environment monitoring data, user behavior data, and market transaction records; Processor: used for executing the management program to implement the steps of the intelligent integrated management method for Chinese orchid germplasm resources as claimed in claim 1; Aroma sensor interface module: used to connect to gas chromatography-mass spectrometry or electronic nose aroma component collection equipment, and directly transmit the collected aroma spectrum data to the system database; Climate and environmental data acquisition module: used to collect light intensity, air temperature and humidity, soil temperature and humidity, and carbon dioxide concentration in real time, and dynamically linked with the weighted CNN prediction model; Blockchain node module: used to store unique product identification, transaction contracts, copyright certificates, and recommended data call records; User interaction terminal interface: used to connect mobile terminals, computer terminals and special collection equipment, and used for real-time display and interactive operation of flowering calendars, fragrance radar maps, climate heat maps and multimodal prediction results.

10. A computer-readable storage medium, characterized in that The medium stores program instructions for executing the intelligent integrated management method for Chinese orchid germplasm resources as described in any one of claims 1 to 8.

Images