Jinhua pork quality improvement method and system based on data mining

Abstract

The invention provides a Jinhua pork quality improvement method and system based on data mining, and relates to the technical field of intelligent breeding, and the method comprises the steps: obtaining an original data set; performing data fusion according to the original data set, and obtaining a multi-modal association graph by constructing an entity-relation network topology structure; performing potential influence path extraction according to the multi-modal association map to obtain an influence path set; performing interaction effect modeling according to the influence path set to obtain a dynamic effect model; performing meat defect tracing according to the dynamic action model to obtain a defect driving factor set; and generating an optimization strategy according to the defect driving factor set to obtain a meat quality improvement decision scheme. According to the method, the multi-factor relation network is constructed by integrating Jinhua pig full-period multi-source data, the key conduction path and the dynamic interaction model influencing meat quality formation are accurately extracted, and the nonlinear antagonism relation between the leaf fat deposition tendency and intramuscular fat accumulation under the specific physiological mechanism of the Jinhua pigs is systematically quantified.

Description

Technical Field

[0001] The present invention relates to the field of intelligent pig farming technology, and in particular to a Jinhua pork quality improvement method and system based on data mining. Background Art

[0002] With growing demand for high-quality meat products, Jinhua pigs, with their unique meat flavor, early maturation and easy fattening, and abundant fat content with relatively low subcutaneous fat, have become a key resource for Jinhua ham production. However, this characteristic makes nutrient allocation during fattening particularly sensitive: excessive fat deposition in the viscera (fat) is common, while ideal intramuscular fat content cannot be consistently achieved. Current mainstream meat quality improvement technologies have shortcomings: First, most are based on lean pig breeds from abroad, failing to adapt to the Jinhua pig's unique fat metabolism pathways and energy allocation mechanisms, overlooking the inherent conflict between fat deposition propensity and intramuscular fat competition. Second, they often analyze the dynamic effects of genetics, nutritional timing, and environmental factors on fat deposition in a separate and distinct manner, lacking the ability to quantify the nonlinear antagonistic relationship between visceral fat accumulation and high-quality muscle fat formation. As a result, existing strategies cannot simultaneously control excessive fat development and precisely improve intramuscular fat deposition, hindering the improvement of core economic traits in Jinhua pork.

[0003] Based on the above shortcomings of the existing technology, there is an urgent need for a Jinhua pork quality improvement method and system based on data mining. Summary of the Invention

[0004] The purpose of the present invention is to provide a method for improving the quality of Jinhua pork based on data mining to address the above-mentioned problems. To achieve the above-mentioned purpose, the technical solutions adopted by the present invention are as follows: In a first aspect, the present application provides a method for improving Jinhua pork quality based on data mining, comprising: Acquire an original data set, wherein the original data set includes physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data, and meat quality evaluation sample data of Jinhua pig individual growth process; Performing data fusion based on the original data set, constructing an entity-relationship network topology structure by calculating the multi-dimensional implicit association strength between entities, and obtaining a multimodal association map; Extracting potential impact paths based on the multimodal association graph to obtain an impact path set; Modeling the interaction effects based on the set of influencing paths, quantifying the synergistic effects of latent variables by decomposing the nonlinear coupling relationships between the paths, and obtaining a dynamic action model; Tracing meat quality defects based on the dynamic action model, simulating pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtaining a set of defect driving factors; An optimization strategy is generated based on the set of defect driving factors to obtain a meat quality improvement decision plan.

[0005] In a second aspect, the present application also provides a Jinhua pork quality improvement system based on data mining, comprising: An acquisition module is used to acquire an original data set, wherein the original data set includes physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data and meat quality evaluation sample data of Jinhua pig individual growth process; A fusion module is used to perform data fusion based on the original data set, construct an entity-relationship network topology structure by calculating the multi-dimensional implicit association strength between entities, and obtain a multimodal association map; An extraction module, configured to extract potential impact paths based on the multimodal association graph to obtain an impact path set; A modeling module is used to model the interaction effects according to the set of influencing paths, quantify the synergistic effects of latent variables by decomposing the nonlinear coupling relationship between the paths, and obtain a dynamic action model; a traceability module for tracing meat quality defects based on the dynamic action model, simulating pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtaining a set of defect driving factors; A generation module is used to generate an optimization strategy based on the defect driving factor set to obtain a meat quality improvement decision plan.

[0006] The beneficial effects of the present invention are: The present invention integrates multi-source data from the entire life cycle of Jinhua pigs to construct a multi-factor relationship network, and accurately extracts the key conduction pathways and dynamic interaction models that affect meat quality formation. It systematically quantifies the nonlinear antagonistic relationship between the tendency of lard deposition and intramuscular fat accumulation under the unique physiological mechanism of Jinhua pigs, overcoming the limitation of the existing technology that is insufficient to analyze the dynamic coupling effects of multiple factors; then, based on the model-driven meat quality defect traceability, the core regulatory nodes are accurately located, and an optimization plan is generated to directly guide the feed formulation and feeding management adjustment, achieving the goal of effectively controlling the excessive accumulation of visceral fat while targetedly increasing the intramuscular fat content, significantly improving the uniformity of the core commodity traits of Jinhua pork and product competitiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0008] Figure 1A schematic diagram of a process for improving Jinhua pork quality based on data mining according to an embodiment of the present invention; Figure 2 Schematic diagram of the structure of the Jinhua pork quality improvement system based on data mining according to an embodiment of the present invention; Figure 3 This is a schematic structural diagram of the Jinhua pork quality improvement equipment based on data mining described in an embodiment of the present invention.

[0009] Markings in the figure: 800, a Jinhua pork quality improvement device based on data mining; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component; 901, acquisition module; 902, fusion module; 903, extraction module; 904, modeling module; 905, traceability module; 906, generation module. DETAILED DESCRIPTION

[0010] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0011] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are used only to distinguish the description and should not be understood as indicating or implying relative importance.

[0012] Example 1:

[0013] This embodiment provides a method for improving Jinhua pork quality based on data mining.

[0014] See also Figure 1 , the figure shows that the method includes steps S100 to S600.

[0015] Step S100, obtaining an original data set, the original data set including physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data and meat quality evaluation sample data of Jinhua pig individual growth process; Understandably, meat quality is the product of a long-term interaction of multiple factors, including genetics, environment, feeding, physiological dynamics, and post-slaughter outcomes. This step also collects key data that characterizes the pig's unique growth patterns: physiological change curves (reflecting growth and development stages), genetic markers (affecting fat metabolism), pig house environmental fluctuations (inducing stress), management events such as feeding / immunization (directly affecting nutrition), and final meat sample assessment (indicators of defects). Data acquisition in this step relies on seamless integration of automated equipment and production processes: wearable sensors attached to the pigs automatically generate time series of physiological parameters. Ambient temperature, humidity, and ammonia concentration data are collected in real time via the Internet of Things (IoT) sensor network within the pig house. Feeding records are recorded by stockmen in a structured log, either by scanning a QR code or manually entering it into a digital management system. Genetic marker loci are collected through laboratory analysis of pig ear tissue samples using gene chips or sequencing. Meat quality evaluation sample data comes from post-slaughter laboratory physical and chemical testing of specific parts and manual sensory evaluation.

[0016] Step S200: performing data fusion based on the original data set, constructing an entity-relationship network topology by calculating the multi-dimensional implicit association strength between entities, and obtaining a multimodal association graph; It should be noted that the breakthrough in data fusion and graph construction lies in the use of "entity-relationship networks" to integrate previously fragmented data silos. Traditional analysis often processes genetic data, environmental records, production data, etc. separately, ignoring the complex collaborative or antagonistic relationships between them. This step constructs a topological structure by deeply calculating the multi-dimensional implicit correlation strength between gene points and growth indicators, environmental sensor readings and feeding record events, and management actions and meat quality results. It accurately restores real complex interactive scenarios such as "gene expression is affected by temperature" and "stress events interfere with nutrient absorption efficiency" in the Jinhua pig farming environment, forming a series of dynamic correlation maps to provide structural support for subsequent exploration of causal relationships.

[0017] Step S300: extracting potential impact paths based on the multimodal association graph to obtain an impact path set; Understandably, this step's potential impact pathway extraction focuses on locating key transmission chains within the complex network map that influence meat quality. The logic is that not all associations are equally important, and it's necessary to identify traceable pathways that are strongly correlated with specific meat quality issues (such as insufficient intramuscular fat deposition and wide variation in tenderness). This step specifically designs analytical mechanisms to address the core contradiction of Jinhua pigs: a strong suet (visceral fat) production pathway but a weak intramuscular fat deposition pathway. For example, this involves tracking the potentially suppressed or bypassed pathway of "feed absorption efficiency-skeletal muscle development-intramuscular preadipocyte activation." This step directly addresses the specific meat quality goals that are most in need of optimization within the Jinhua pig breed, rather than simply screening pathways.

[0018] Step S400: Modeling the interaction effect based on the set of influencing paths, quantifying the synergistic effect of latent variables by decomposing the nonlinear coupling relationship between the paths, and obtaining a dynamic action model; Understandably, traditional linear models cannot explain a common phenomenon in Jinhua pig farming: increased nutrition may simultaneously stimulate fat deposition while inhibiting intramuscular fat accumulation (nonlinear coupling). This step deliberately decomposes the interactions at the intersection of multiple pathways (for example, during the critical growth spurt, nutrition may simultaneously activate visceral fat accumulation genes and inhibit signaling pathways that differentiate into muscle adipocyte precursors) and quantify the weight of this "internal friction" (hidden variable cooperation / competition). This accurately captures the unique nutrient allocation conflict dynamics of Jinhua pigs, establishing a computable model that truly reflects their biological reality.

[0019] Step S500: tracing meat quality defects based on the dynamic action model, simulating pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtaining a set of defect driving factors; It should be noted that the purpose of meat quality defect tracing is to conduct virtual experiments on a dynamic model, pinpointing the cause of disease with high precision and low cost. Traditional methods (such as actually changing feeding parameters for controlled experiments) are costly, time-consuming, and prone to interference. This step simulates a "pseudo-intervention" designed specifically for the Jinhua pig model by analyzing the "gradient sensitivity" of meat quality targets (such as intramuscular fat content) to changes in the entire system (such as genes, environment, and nutrition). This does not actually change the pig's diet or environment, but rather systematically explores changes in key nodes within the model (such as regulating the expression level of a key lipase) to observe the degree of improvement in the target defect (fat deficiency) and the transmission chain. This specifically addresses the difficulty of tracing the multi-factor coupling and difficult experimental reproduction of Jinhua pork quality issues, directly serving the subsequent precise regulation.

[0020] Step S600: Generate an optimization strategy based on the defect driving factor set to obtain a meat quality improvement decision plan.

[0021] Understandably, optimization strategy generation is the process of precisely translating traceability conclusions into actionable farming operations. Its core logic recognizes that single-point modifications (e.g., simply increasing intramuscular fat) can be counterproductive in complex systems (e.g., overstimulation leading to thicker lard). This step emphasizes the strategy's multi-objective balancing and dynamic fault tolerance. The resulting decision-making plan clearly defines a set of directly executable key operational guidelines, including feed formula (e.g., fatty acid ratio), environmental control parameters (e.g., temperature and humidity ranges), and feeding plans (e.g., feeding strategies during critical growth periods), ensuring stable and predictable improvement results.

[0022] Furthermore, step S200 includes steps S210 to S230.

[0023] Step S210: Perform cross-domain time series alignment processing on the original data set, unify the time base of the heterogeneous data through the time series compensation mechanism, and obtain a time-space calibration data set; Step S220: Perform multidimensional semantic association mining based on the spatiotemporal calibration dataset, and obtain a dynamic association matrix between entities by calculating the cross-modal coupling strength of cross-domain combinations of genetic loci-physiological parameters, environmental fluctuations-feeding events, and event sequences-meat quality indicators; Step S230: Evolve the network topology according to the dynamic association matrix, and iteratively generate a sparse connection structure of high-association entity edges to obtain a multimodal association graph.

[0024] It is understandable that the logical essence of steps S210 to S230 is to gradually extract explainable Jinhua pig quality formation rules from the chaotic raw data: first, through the cross-domain time series alignment in step S210 (such as the time calibration of manual feeding records and automatic sensor data), the causal analysis distortion problem caused by the asynchronous collection of multi-source data on the breeding site is solved; then, in step S220, on a unified time basis, the cross-modal coupling effect of genetic variation (such as specific fat metabolism sites) on physiological indicators (such as daily weight gain curve) is deeply explored, accurately capturing the nonlinear impact of environmental mutations (such as high temperature) and feeding interventions (such as feed adjustments) on meat quality; finally, step S230 constructs a lightweight topological network through dynamic pruning (such as retaining only key correlation edges with correlation coefficients > 0.8), condensing the scattered high-dimensional data into a computable knowledge graph pointing to the key quality regulation channels of Jinhua pigs (such as "temperature stress-feeding interruption-inhibition of intramuscular fat synthesis"), laying a structured foundation for subsequent traceability modeling.

[0025] Furthermore, step S300 includes steps S310 to S330.

[0026] Step S310: Identifying feedback loops in the biological system based on the multimodal association map, and obtaining a set of potential feedback loops by analyzing the positive and negative feedback directionality and strength of information flows in the ring topology structure involved in the meat tenderness regulation pathway; Step S320: Perform cross-level influence transmission analysis based on the potential feedback loop set, and obtain a potential influence transmission chain set by quantifying the indirect influence intensity and path dependency of the fluctuation of temperature and humidity in the feeding environment through the composition of the intestinal microbial community to the intramuscular fat deposition rate; Step S330 : identifying key cascade nodes based on the potential influencing conduction chain set, and obtaining the influencing path set by calculating the sensitivity convergence threshold of the simulated path interruption to the final meat juiciness score state.

[0027] It is important to note that S310, by analyzing the circular topology within the graph, specifically identifies the core positive and negative feedback mechanisms that cause fluctuations in Jinhua pig tenderness (e.g., the vicious cycle of persistently reinforcing appetite-suppressing signals under high temperatures), thereby revealing the self-reinforcing root causes of meat quality defects. S320 further traces the cross-hierarchical transmission chain (e.g., "sudden increase in pig house temperature - disrupted intestinal flora - reduced short-chain fatty acid synthesis - blocked differentiation of intramuscular adipocyte precursors"), quantifying the magnitude of the indirect biological effects of environmental fluctuations on meat quality (e.g., calculating the transmission coefficient that indicates a 13% decrease in intramuscular fat deposition efficiency for every 5°C temperature fluctuation), highlighting the specific transmission bottleneck that makes Jinhua pigs sensitive to hot and humid environments. S330, based on pathway disruption simulations (e.g., blocking the acetate synthesis node in the microbial metabolic pathway), calculates the sensitivity threshold for the final meat juiciness score, thereby identifying a minimal set of interventions that can significantly improve Jinhua pig-specific defects (e.g., uneven fat distribution among muscle fibers). These three steps refine the complex network into an actionable biological intervention roadmap, ensuring that the tracing conclusions are both mechanistic and practical. Specifically, the key cascade node judgment formula is:

[0028] in, Represents a set of nodes in the potential impact transmission chain; Represents the cascade node to be identified, that is, the specific regulatory unit in the conduction chain; Indicates the meat juiciness score; Indicates the node disturbance intensity; represents the sensitivity convergence threshold; Indicates the node sensitivity index; Represents the set of potential conduction chain nodes , screen out the sensitivity index Reaching the threshold A collection of nodes.

[0029] Furthermore, step S400 includes steps S410 to S430.

[0030] Step S410: Identify path intersection nodes based on the impact path set, and obtain an interaction hub node set by locating node entities shared by multiple paths and having the function of regulating the expression of intramuscular adipocyte differentiation genes; Step S420: Dynamic superposition effect modeling is performed based on the interaction hub node set, and a multi-path coupling dynamic equation is obtained by quantifying the temporal interaction intensity between the activation level of the skeletal muscle energy metabolism pathway and the fat deposition rate during the critical nutrient supply stage of the growth period; Step S430: Decoupling the cooperative competition relationship according to the multi-path coupling dynamics equation, and obtaining a dynamic action model by decomposing the nonlinear offset effect between the muscle collagen synthesis and degradation pathways and the weight of the implicit regulatory factors.

[0031] It is understandable that step S410 targets the problem of low efficiency of intramuscular fat deposition unique to Jinhua pigs by locating multi-path intersection nodes, such as simultaneously regulating the lipid synthesis gene PPARγ The shared node of " and "myostatin MSTN" solves the blind spot of traditional single pathway analysis that cannot capture nutrient allocation conflicts (skeletal muscle development occupies fat precursor cell resources). The core breakthrough lies in identifying hub entities (such as AMPK signaling nodes) that are involved in the activation of energy metabolism pathways and are affected by feeding events (such as high-protein diets) and environmental stress, integrating scattered causal chains into a target set that can simultaneously intervene in multiple meat quality defects. Step S420 targets the key time window of nutrient allocation during the rapid growth period of Jinhua pigs, and establishes a multi-path coupling equation by quantifying the dynamic inhibitory effect of the intensity of the skeletal muscle glycolysis pathway on the rate of intramuscular fat deposition (for example, for every 1 unit increase in the ability of muscle to consume glucose, the efficiency of fat precursor cell differentiation decreases by 0.7 units). Its innovation lies in the introduction of time delay factors (such as feed feeding). The last 2 hours are the period of maximum mutual interference), accurately depicting the energy metabolism paradox unique to Jinhua pigs, "high-energy diet-short-term ATP surplus-inhibition of adipocyte differentiation". Step S430 focuses on solving the core contradiction of Jinhua pigs' tenderness and juiciness, which is difficult to achieve at the same time, by decoupling the nonlinear offset between collagen synthesis (Col1a1 gene expression) and degradation (MMP1 enzyme activity) pathways (such as high temperatures in summer increase the degradation rate by 3 times, but at the same time trigger stress synthesis compensation). The key innovation lies in the introduction of implicit regulatory factor weight analysis (such as calculating that oxytocin receptor expression accounts for 62% of the regulation of collagen metabolism), revealing the cross-system mechanism of environmental intervention (such as spray cooling) that indirectly regulates meat structure through neuroendocrine pathways, so that the model can predict the net effect of compound regulatory strategies. Specifically, the multi-path coupling kinetic equation is:

[0032] in, represents the differential symbol; Indicates the concentration of intramuscular fat; Indicates the current time; represents the coefficient of fat differentiation efficiency; express nuclear receptor concentrations; represents the Hill coefficient; express half-maximal activation concentration; Indicates the rate of branched-chain amino acid uptake; represents the competitive constant for muscle synthesis; Indicates the maximum glycogen storage capacity of skeletal muscle; represents the environmental response parameter; represents the metabolic window closing rate; represents the basal inhibitory efficacy per unit of cortisol; represents the high temperature nonlinear gain factor; Indicates the real-time pig house temperature; represents the plasma cortisol concentration; e represents the natural constant; t feed Indicates the starting time of the day's feeding operation.

[0033] Furthermore, step S500 includes steps S510 to S530.

[0034] Step S510: screening meat quality defect-related paths based on the dynamic action model, and obtaining a defect-sensitive path set by calculating the global sensitivity topological coverage of the target defect quantization state relative to the model pathway; Step S520: performing a directional perturbation simulation based on the defect-sensitive path set, applying a preset step-size parameter displacement at the pig H-FABP gene expression level control point, quantifying the observable response amplitude of the displacement transmitted along the biological path to the meat quality index, and obtaining a node perturbation-index response mapping spectrum; Step S530 : performing dominant factor discrimination based on the node disturbance-index response mapping spectrum, and obtaining a set of defect driving factors by identifying control units whose contribution to the change of the target defect index exceeds a preset multiple of its normal fluctuation standard deviation.

[0035] In the meat defect tracing stage, steps S510 to S530 build a closed-loop decision chain from system modeling to precise intervention: first, through global sensitivity topology analysis (such as calculating the effect of reduced intramuscular fat content on the PPARγ The joint sensitivity of the metabolic pathway and the heat stress pathway was evaluated), and biological pathways strongly associated with the target defect were screened at the full network scale (such as the "H-FABP gene expression-fatty acid transport efficiency-muscle fat deposition" chain). Targeted virtual perturbations were then implemented (for example, the expression level of the H-FABP_rs329 gene locus unique to Jinhua pigs was adjusted by a ±15% gradient in the model). The nonlinear response amplitude of the change in this control point along the metabolic pathway to the final meat quality indicator (such as the coefficient of variation of the adipocyte diameter distribution) was accurately quantified, generating a multidimensional response spectrum containing a dose-effect relationship. Finally, based on a statistical contribution threshold, minor interference items (such as the influence of seasonal temperature fluctuations) were eliminated, and the core driving unit that can directly explain the main cause of meat quality defects (such as the methylation level of the H-FABP gene promoter) was identified, providing molecular targets and quantitative operation windows for the subsequent formulation of genetic selection or nutritional regulation programs.

[0036] Furthermore, step S520 includes steps S521 to S523.

[0037] Step S521: Define a biologically feasible perturbation interval based on the defect-sensitive pathway set. By matching the phenotypic variation range of the rs329 site of the Jinhua pig H-FABP gene, an equal-interval discretization strategy is used to generate a displacement vector sequence that meets the actual breeding / nutritional intervention capabilities, thereby obtaining a standardized perturbation parameter set. This step aims to establish an operational intervention range that is consistent with the genetic characteristics of Jinhua pigs. The core principle is to convert biological reality constraints into a discrete mathematical perturbation space by analyzing the phenotypic variation characteristics of the rs329 site of the H-FABP gene in the Jinhua pig population (the measured distribution is a normal distribution of μ=100 AU, σ=15 AU). The specific operation is to adopt an equally spaced discretization strategy to ensure that each perturbation step covers both the possible genetic variation range and the ability boundaries of actual breeding (±15%) and nutritional intervention (±30%). This design avoids meaningless simulations that exceed physiological limits and directly serves subsequent precise quantification. The implemented mathematical model is the generation of discrete perturbation vectors, and the discrete perturbation vector formula is:

[0038] in, represents the set of standardized perturbation parameters; represents the average expression level of H-FABP gene at rs329 site in Jinhua pigs; represents the relative disturbance intensity; sng(l) represents the disturbance direction; represents the disturbance displacement; Baseline expression level; L represents the maximum perturbation order; l represents the discrete perturbation index value.

[0039] Step S522: Simulating the nonlinear response of the biological pathway based on the perturbation parameter set. By constructing a system of coupled differential equations and numerically integrating them to a steady state using the fourth-order Runge-Kutta method, the dose-effect relationship of gene expression displacement along the biological pathway to the final meat quality index is quantified, thereby obtaining a node perturbation-dynamic response dataset. This step quantifies the chain conduction of gene perturbations on the final indicators of meat quality by constructing a set of differential equations specific to Jinhua pigs. The core innovation is the introduction of a transport efficiency defect parameter to model the reduced efficiency of muscle fat deposition caused by restricted fatty acid transport. A fourth-order Runge-Kutta method is used for 48 hours of numerical integration (covering the complete postprandial metabolic cycle) to capture the dose-effect relationship under steady-state conditions. This method overcomes the limitation of traditional correlation analysis that cannot analyze the conduction dynamics. The coupled differential equations are:

[0040] in, represents the baseline H-FABP expression level; represents the fatty acid transport rate; t0 represents the metabolic time; Indicates the maximum transfer capacity; It represents the defect parameter of Jinhua pig transport efficiency; represents the transport decay rate; Indicates the efficiency of muscle fat deposition; Indicates the fat degradation rate.

[0041] Step S523: Perform nonlinear feature analysis based on the node disturbance-dynamic response data set, generate a continuous response surface function through cubic spline interpolation and calculate the gradient distribution, extract the positive and negative disturbance asymmetry coefficients and sensitive window boundaries of Jinhua pigs, and obtain a disturbance-response mapping spectrum containing the biological path conduction characteristics.

[0042] This step aims to extract the specific response patterns of Jinhua pigs to genetic perturbations. Discrete data are converted into continuous response functions through cubic spline interpolation, and its first-order gradient ∇M is calculated to reveal the nonlinear sensitive range. Key findings reveal significant asymmetry between positive and negative perturbations in Jinhua pigs (the response intensity to downregulation is 52% higher than that to upregulation), and the sensitive window is determined to be [−10%, +11%]. The output includes a perturbation-response mapping spectrum containing function images, gradient distributions, and numerical parameters, directly supporting breeding decisions. These features provide a basis for avoiding ineffective intervention zones. The formula for continuous response function construction and feature extraction is:

[0043] in, represents a continuous response function; represents the B-spline basis function; represents the least squares fitting coefficient; represents the response gradient; represents the asymmetry coefficient; Represents the sensitive window; z represents the index value of the B-spline basis function.

[0044] Furthermore, step S600 includes steps S610 to S630.

[0045] Step S610: Define a multi-objective constraint space based on the defect driving factor set, and obtain a Pareto frontier search domain by setting a decision space with increasing intramuscular fat content as the priority goal, maintaining carcass lean meat percentage as the constraint boundary, and muscle antioxidant enzyme activity change threshold as the biochemical constraint condition; Step S620: Generate a set of candidate strategies that meet all constraints by simulating the coordinated changes of genetic and environmental control variables within their physiologically feasible ranges based on the Pareto front search domain; Step S630: Dynamic robustness verification is performed based on the candidate strategy set. By calculating the expected degree of guarantee of the candidate strategy for key meat quality indicators under the preset fattening cycle fluctuation range, a meat quality improvement decision plan is obtained. The meat quality improvement decision plan includes a feed formula adjustment table, breeding environment control standards, a phased feeding plan, and meat quality requirements.

[0046] In the meat quality improvement decision-making stage, steps S610 to S630 construct a closed-loop framework for multi-objective optimization and robustness verification: first, a multi-objective constraint space is defined based on a set of defect driving factors (such as H-FABP gene expression level and heat stress response node), and the improvement of intramuscular fat content (target>4.2mg / g) is set as the priority target, the carcass lean meat rate (constraint>58%) is set as the boundary limit, and the muscle antioxidant enzyme activity (SOD>135 U / mgprot) was set as the biochemical threshold, and a three-dimensional Pareto frontier search domain that took into account both quality and production performance was constructed. Then, within this search domain, the coordinated changes of genetic regulation (such as the intensity of selection at the rs329 locus) and environmental variables (such as the temperature control range of the pig house in summer) within their physiologically feasible range were simulated, and a non-dominated sorting genetic algorithm was used to generate a set of candidate strategies that simultaneously met the requirements of meat quality improvement, lean meat maintenance and oxidative stability. Finally, through dynamic robustness verification, the expected guarantee level of each candidate strategy for key meat quality indicators (intramuscular fat, shear force, drip loss) under fattening cycle fluctuations (such as feed composition deviation of ±5% and temperature fluctuation of ±3°C) was calculated (such as >90% probability of compliance). The output was an executable decision-making plan that included a precise feed formula adjustment table (linoleic acid ratio ≥1.8%), environmental control standards (cooling ≥4°C during high temperature periods), a phased feeding plan (night feeding between 60-90kg body weight), and meat quality compliance requirements (marbling score ≥3.5), ensuring the stable realization of the improvement effect in the actual Jinhua pig breeding scenario.

[0047] Example 2:

[0048] like Figure 2 As shown, this embodiment provides a Jinhua pork quality improvement system based on data mining, the system comprising: An acquisition module 901 is used to acquire an original data set, which includes physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data, and meat quality evaluation sample data of Jinhua pig individual growth process; Fusion module 902 is used to perform data fusion based on the original data set, construct an entity-relationship network topology structure by calculating the multi-dimensional implicit association strength between entities, and obtain a multimodal association map; Extraction module 903, configured to extract potential impact paths based on the multimodal association graph to obtain an impact path set; Modeling module 904 is used to model the interaction effect based on the set of influencing paths, quantify the synergistic effect of latent variables by decomposing the nonlinear coupling relationship between the paths, and obtain a dynamic action model; The traceability module 905 is used to trace the source of meat defects based on the dynamic action model, simulate pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtain a set of defect driving factors; The generation module 906 is used to generate an optimization strategy based on the defect driving factor set to obtain a meat quality improvement decision plan.

[0049] In a specific embodiment disclosed in this application, the fusion module 902 includes: The first fusion unit is used to perform cross-domain time series alignment processing based on the original data set, unify the time base of heterogeneous data through the time series compensation mechanism, and obtain a time-space calibration data set; The second fusion unit is used to perform multidimensional semantic association mining based on the spatiotemporal calibration dataset. By calculating the cross-modal coupling strength of the cross-domain combinations of genetic loci-physiological parameters, environmental fluctuations-feeding events, and event sequences-meat quality indicators, a dynamic association matrix between entities is obtained. The third fusion unit is used to evolve the network topology according to the dynamic correlation matrix, and obtain a multimodal correlation graph by iteratively generating a sparse connection structure of high-correlation entity edges.

[0050] In a specific embodiment disclosed in this application, the extraction module 903 includes: The first extraction unit is used to identify feedback loops in biological systems based on the multimodal association map. By analyzing the positive and negative feedback directionality and strength of information flow in the circular topology of the meat tenderness regulation pathway, a set of potential feedback loops is obtained. The second extraction unit is used to perform cross-level influence transmission analysis based on the potential feedback loop set. By quantifying the indirect influence intensity and path dependency of the fluctuation of temperature and humidity in the feeding environment through the composition of the intestinal microbial community to the intramuscular fat deposition rate, a potential influence transmission chain set is obtained. The third extraction unit is used to identify key cascade nodes based on the potential impact conduction chain set, and obtain the impact path set by calculating the sensitivity convergence threshold of the simulated path interruption to the final meat juiciness score state.

[0051] In a specific embodiment disclosed in this application, the modeling module 904 includes: The first modeling unit is used to identify path intersection nodes based on the impact path set, and obtain the interaction hub node set by locating the node entities shared by multiple paths and having the function of regulating the expression of intramuscular adipocyte differentiation genes; The second modeling unit is used to model the dynamic superposition effect based on the set of interaction hub nodes. By quantifying the temporal interaction intensity between the activation level of the skeletal muscle energy metabolism pathway and the fat deposition rate during the critical nutrient supply stage of the growth period, a multi-path coupling dynamic equation is obtained. The third modeling unit is used to decouple the cooperative and competitive relationship based on the multi-path coupling kinetic equation, and obtain the dynamic action model by decomposing the nonlinear offset effect between the muscle collagen synthesis and degradation pathways and the weight of the implicit regulatory factors.

[0052] In a specific embodiment disclosed in this application, the traceability module 905 includes: The first traceability unit is used to screen meat quality defect-related paths based on the dynamic action model and obtain the defect-sensitive path set by calculating the global sensitivity topological coverage of the target defect quantitative state relative to the model pathway; The second traceability unit is used to perform directional perturbation simulation based on the defect-sensitive path set. By applying a preset step parameter displacement at the pig H-FABP gene expression level control point, the observable response amplitude of the displacement transmitted along the biological path to the meat quality index is quantified, and the node perturbation-index response mapping spectrum is obtained; The third traceability unit is used to identify the dominant factors based on the node disturbance-indicator response mapping spectrum, and obtain the set of defect driving factors by identifying the control units whose contribution to the change of the target defect indicator exceeds the preset multiple of its normal fluctuation standard deviation.

[0053] In a specific embodiment disclosed in this application, the generation module 906 includes The first generation unit is used to define a multi-objective constraint space based on the defect driving factor set. By setting the decision space with increasing intramuscular fat content as the priority goal, maintaining carcass lean meat percentage as the constraint boundary, and the threshold of muscle antioxidant enzyme activity change as the biochemical constraint condition, the Pareto frontier search domain is obtained. The second generation unit is used to generate a candidate strategy set that meets all constraints by simulating the coordinated changes of genetic and environmental control variables within their physiological feasible ranges based on the Pareto frontier search domain; The third generation unit is used to perform dynamic robustness verification based on the candidate strategy set. By calculating the expected guarantee of the candidate strategy for key meat quality indicators under the preset fattening cycle fluctuation range, the meat quality improvement decision plan is obtained. The meat quality improvement decision plan includes a feed formula adjustment table, breeding environment control standards, a phased feeding plan and meat quality requirements.

[0054] Example 3:

[0055] Corresponding to the above method embodiment, this embodiment also provides a Jinhua pork quality improvement device based on data mining. The Jinhua pork quality improvement device based on data mining described below and the Jinhua pork quality improvement method based on data mining described above can be referenced to each other.

[0056] Figure 3 FIG is a block diagram of a Jinhua pork quality improvement device 800 based on data mining according to an exemplary embodiment. Figure 3 As shown, the Jinhua pork quality improvement device 800 based on data mining may include: a processor 801, a memory 802. The Jinhua pork quality improvement device 800 based on data mining may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0057] The processor 801 is used to control the overall operation of the Jinhua pork quality improvement device 800 based on data mining to complete all or part of the steps of the aforementioned Jinhua pork quality improvement method based on data mining. The memory 802 is used to store various types of data to support the operation of the Jinhua pork quality improvement device 800 based on data mining. This data may include, for example, instructions for any application or method operating on the Jinhua pork quality improvement device 800 based on data mining, as well as application-related data such as contact information, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signal may be further stored in the memory 802 or transmitted via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be a keyboard, a mouse, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the Jinhua pork quality improvement device 800 based on data mining and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0058] In an exemplary embodiment, a Jinhua pork quality improvement device 800 based on data mining can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned Jinhua pork quality improvement method based on data mining.

[0059] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided. When executed by a processor, the program instructions implement the steps of the aforementioned method for improving Jinhua pork quality based on data mining. For example, the computer-readable storage medium may be the aforementioned memory 802 including the program instructions. The program instructions may be executed by the processor 801 of the device 800 for improving Jinhua pork quality based on data mining to implement the aforementioned method for improving Jinhua pork quality based on data mining.

[0060] 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. A method for improving Jinhua pork quality based on data mining, characterized in that: include: Acquire an original data set, wherein the original data set includes physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data, and meat quality evaluation sample data of Jinhua pig individual growth process; Performing data fusion based on the original data set, constructing an entity-relationship network topology structure by calculating the multi-dimensional implicit association strength between entities, and obtaining a multimodal association map; Extracting potential impact paths based on the multimodal association graph to obtain an impact path set; Modeling the interaction effects based on the set of influencing paths, quantifying the synergistic effects of latent variables by decomposing the nonlinear coupling relationships between the paths, and obtaining a dynamic action model; Tracing meat quality defects based on the dynamic action model, simulating pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtaining a set of defect driving factors; An optimization strategy is generated based on the defect driving factor set to obtain a meat quality improvement decision plan.

2. The Jinhua pork quality improvement method based on data mining according to claim 1, characterized in that: Performing data fusion according to the original data set includes: Performing cross-domain time series alignment processing on the original data set, unifying the time base of the heterogeneous data through a time series compensation mechanism, and obtaining a spatiotemporal calibration data set; Perform multidimensional semantic association mining based on the spatiotemporal calibration dataset, and obtain a dynamic association matrix between entities by calculating the cross-modal coupling strength of cross-domain combinations of genetic loci-physiological parameters, environmental fluctuations-feeding events, and event sequences-meat quality indicators; The network topology is evolved according to the dynamic association matrix, and a sparse connection structure of high-association entity edges is iteratively generated to obtain a multimodal association graph.

3. The Jinhua pork quality improvement method based on data mining according to claim 1, characterized in that: Extracting potential impact paths based on the multimodal association graph includes: Identifying feedback loops in biological systems based on the multimodal association map, and obtaining a set of potential feedback loops by analyzing the positive and negative feedback directionality and strength of information flows in the circular topology of the meat tenderness regulatory pathway; Based on the potential feedback loop set, a cross-level impact transmission analysis was performed to obtain a potential impact transmission chain set by quantifying the indirect impact intensity and path dependency of the fluctuations in temperature and humidity of the feeding environment through the composition of the intestinal microbial community to the intramuscular fat deposition rate; The key cascade nodes are identified based on the potential impact conduction chain set, and the impact path set is obtained by calculating the sensitivity convergence threshold of the simulated path interruption to the final meat juiciness score state.

4. The Jinhua pork quality improvement method based on data mining according to claim 1, characterized in that: Interaction effect modeling is performed based on the set of influence paths, including: Identifying path intersection nodes based on the impact path set, and obtaining an interaction hub node set by locating node entities shared by multiple paths and having the function of regulating the expression of intramuscular adipocyte differentiation genes; Dynamic superposition effect modeling is performed based on the interaction hub node set, and the multi-path coupling dynamic equation is obtained by quantifying the temporal interaction intensity between the activation level of the skeletal muscle energy metabolism pathway and the fat deposition rate during the critical nutrient supply stage of the growth period; The cooperative competition relationship is decoupled according to the multi-path coupling kinetic equation, and a dynamic action model is obtained by decomposing the nonlinear offset effect between the muscle collagen synthesis and degradation pathways and the weight of the implicit regulatory factors.

5. The Jinhua pork quality improvement method based on data mining according to claim 1, characterized in that: The meat quality defect traceability is performed based on the dynamic action model, including: Screening meat quality defect-related pathways based on the dynamic action model, and obtaining a defect-sensitive pathway set by calculating the global sensitivity topological coverage of the target defect quantization state relative to the model pathway; Directed perturbation simulation is performed based on the defect-sensitive path set. By applying a preset step parameter displacement at the pig H-FABP gene expression level control point, the observable response amplitude of the displacement transmitted along the biological path to the meat quality index is quantified to obtain a node perturbation-indicator response mapping spectrum; The dominant factor is identified according to the node disturbance-index response mapping spectrum, and the defect driving factor set is obtained by identifying the control units whose contribution to the change of the target defect index exceeds a preset multiple of its normal fluctuation standard deviation.

6. A Jinhua pork quality improvement system based on data mining, characterized in that: include: An acquisition module is used to acquire an original data set, wherein the original data set includes physiological parameter time series data, genetic marker site data, environmental sensor data, feeding record event data and meat quality evaluation sample data of Jinhua pig individual growth process; A fusion module is used to perform data fusion based on the original data set, construct an entity-relationship network topology structure by calculating the multi-dimensional implicit association strength between entities, and obtain a multimodal association map; An extraction module, configured to extract potential impact paths based on the multimodal association graph to obtain an impact path set; A modeling module is used to model the interaction effects according to the set of influencing paths, quantify the synergistic effects of latent variables by decomposing the nonlinear coupling relationship between the paths, and obtain a dynamic action model; a traceability module for tracing meat quality defects based on the dynamic action model, simulating pseudo-intervention responses by analyzing the gradient sensitivity of target parameters to system variables, and obtaining a set of defect driving factors; A generation module is used to generate an optimization strategy based on the defect driving factor set to obtain a meat quality improvement decision plan.

7. The Jinhua pork quality improvement system based on data mining according to claim 6 is characterized in that: The fusion module includes: A first fusion unit is configured to perform cross-domain time series alignment processing on the original data set, unify the time base of the heterogeneous data through a time series compensation mechanism, and obtain a spatiotemporal calibration data set; The second fusion unit is used to perform multidimensional semantic association mining based on the spatiotemporal calibration data set, and obtain a dynamic association matrix between entities by calculating the cross-modal coupling strength of cross-domain combinations of genetic loci-physiological parameters, environmental fluctuations-feeding events, and event sequences-meat quality indicators; The third fusion unit is used to perform network topology evolution according to the dynamic association matrix, and obtain a multimodal association graph by iteratively generating a sparse connection structure of high-association entity edges.

8. The Jinhua pork quality improvement system based on data mining according to claim 6 is characterized in that: The extraction module includes: a first extraction unit for identifying feedback loops in the biological system based on the multimodal association map, and obtaining a set of potential feedback loops by analyzing the positive and negative feedback directionality and strength of information flows in the cyclic topological structure of the meat tenderness regulation pathway; A second extraction unit is configured to perform a cross-level influence transmission analysis based on the potential feedback loop set, and obtain a potential influence transmission chain set by quantifying the indirect influence intensity and path dependency of the fluctuation of temperature and humidity in the feeding environment via the composition of the intestinal microbial community to the intramuscular fat deposition rate; The third extraction unit is used to identify key cascade nodes according to the potential influence conduction chain set, and obtain the influence path set by calculating the sensitivity convergence threshold of the simulated path interruption to the final meat juiciness score state.

9. The Jinhua pork quality improvement system based on data mining according to claim 6 is characterized in that: The modeling module includes: A first modeling unit is configured to identify path intersection nodes according to the set of influencing paths, and obtain an interaction hub node set by locating node entities shared by multiple paths and having the function of regulating the expression of intramuscular adipocyte differentiation genes; A second modeling unit is configured to perform dynamic superposition effect modeling based on the interaction hub node set, and obtain a multi-path coupling kinetic equation by quantifying the temporal interaction intensity between the activation level of the skeletal muscle energy metabolism pathway and the fat deposition rate during the critical nutrient supply stage of the growth period; The third modeling unit is used to decouple the cooperative competition relationship according to the multi-path coupling kinetic equation, and obtain a dynamic action model by decomposing the nonlinear offset effect between the muscle collagen synthesis and degradation pathways and the weight of the implicit regulatory factors.

10. The Jinhua pork quality improvement system based on data mining according to claim 6 is characterized in that: The traceability module includes: A first traceability unit is configured to screen meat quality defect-related paths according to the dynamic action model, and obtain a defect-sensitive path set by calculating a global sensitivity topological coverage of a target defect quantization state relative to a model path; The second traceability unit is used to perform a directional perturbation simulation based on the defect-sensitive path set, by applying a preset step parameter displacement at the pig H-FABP gene expression level control point, quantifying the observable response amplitude of the displacement transmitted along the biological path to the meat quality index, and obtaining a node perturbation-indicator response mapping spectrum; The third traceability unit is used to identify the dominant factors based on the node disturbance-indicator response mapping spectrum, and obtain the set of defect driving factors by identifying the control units whose contribution to the change of the target defect indicator exceeds a preset multiple of its normal fluctuation standard deviation.

Images