Livestock veterinarian breeding data analysis system

Abstract

The present application relates to the technical field of livestock data processing, in particular to a livestock and veterinary breeding data analysis system, which comprises: a state transition analysis module collects data of the breeding cycle of female livestock, integrates the data of all female livestock to construct a transition probability matrix, predicts the future state transition path of female livestock according to the transition probability matrix, and generates a female livestock state transition evaluation result. In the present application, by analyzing the data of each state of the breeding cycle of female livestock, the state change path of female livestock at different stages is determined, accurate prediction and quantitative evaluation of the state in the breeding cycle are realized, and the formulation of the breeding plan is more accurate. By evaluating the residence time and transition trend of each stage, the best breeding time is predicted, and the overall breeding efficiency is improved. Based on the similarity analysis of behavior data such as activity frequency, body temperature and number of reproductions, individual female livestock is divided into different behavior groups, and more targeted reproduction management strategies are formulated for different groups.

Description

Technical Field

[0001] This invention relates to the field of livestock data processing technology, and in particular to a livestock veterinary mating data analysis system. Background Technology

[0002] Livestock data processing improves the efficiency, economic benefits, and sustainability of livestock farming by collecting, processing, storing, and analyzing livestock production data. This includes animal health monitoring, breeding data analysis, nutritional formula optimization, production process tracking, and environmental monitoring. With the development of IoT, big data, and AI technologies, livestock data processing can automate data collection and analysis, helping livestock managers monitor production in real time.

[0003] Among them, the livestock and veterinary mating data analysis system helps managers better formulate mating plans and optimize reproductive efficiency by collecting, integrating, and analyzing mating-related data, such as the estrus cycle of female animals, mating frequency, reproductive success rate, and veterinary records. The system not only enables precise management of individual animal reproductive information but also improves the overall herd's reproductive rate and reduces resource waste, making it an important tool for achieving scientific and modern management in animal husbandry.

[0004] While existing systems can collect and integrate reproductive data during the management of female livestock reproduction, they struggle with detailed analysis and prediction of the animal's condition throughout the reproductive cycle. This can lead to inaccurate timing of mating, reducing reproductive success rates. Furthermore, current technologies fail to adequately address individual differences and behavioral characteristics among female animals, hindering the development of targeted mating plans. Although the systems record basic health information, they are not conducive to rapid, continuous monitoring of physiological parameters and real-time anomaly identification, potentially missing hidden health risks and increasing the likelihood of reproductive failure. Simultaneously, the systems' inability to promptly capture dynamic changes in data processing results in insufficient flexibility in reproductive management. This impacts the effective utilization of resources and reduces the health and reproductive efficiency of female animals. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a livestock and veterinary breeding data analysis system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a livestock and veterinary mating data analysis system comprising:

[0007] The state transition analysis module collects data on the reproductive cycle of female animals, integrates data from all female animals to construct a transition probability matrix, predicts the future state transition path of female animals based on the transition probability matrix, and generates state transition assessment results for female animals.

[0008] The behavior clustering and grouping module, based on the evaluation results of the female animal state transition and the behavioral characteristic data of the female animal, unifies the behavioral characteristic data into feature vectors, analyzes the distance between feature vectors, divides the groups according to similarity, and obtains the behavioral characteristic group division results.

[0009] The fluctuation anomaly detection module continuously records the physiological data of female animals in a specific behavioral group based on the physiological information of the female animals in the behavioral feature group segmentation results, analyzes the basic fluctuation of physiological information, identifies abnormal fluctuation physiological information, and generates abnormal physiological fluctuation analysis results.

[0010] The mating data analysis module integrates and analyzes the reproductive data of female animals by combining the assessment results of female animal state transitions, the results of behavioral characteristic group segmentation, and the results of abnormal physiological fluctuation analysis. Based on the previous mating success rate data, it selects the best mating time and generates recommended mating information.

[0011] As a further aspect of the present invention, the step of obtaining the transition probability matrix is ​​specifically as follows:

[0012] Based on the estrus, pregnancy, lactation and recovery period status data of female animals in the reproductive cycle, the time and status identifier of the status changes of each female animal are recorded, and the data are organized in chronological order to obtain the status sequence data of female animals.

[0013] Based on the female animal state sequence data, the following formula is used:

[0014]

[0015] Calculate the probability P of transitioning from state i to state j. ij This allows us to obtain the probability information of state transitions.

[0016] Where, N ij N is the total number of times the female animal transitions from state i to state j. ik T is the number of transitions the female animal makes from state i to state k. k σ is the average dwell time in state k, where k is the index variable, m is the total number of states, and σ is the average dwell time in state k. T It is the standard deviation of the time the female animal stays in state i;

[0017] Based on the transition probability information between states, the number of times the female animal transitions from each state to other states is counted, and the state transition frequencies are integrated. Based on the state transition data of the female animal during the reproductive cycle, a transition probability matrix is ​​constructed.

[0018] As a further aspect of the present invention, the step of obtaining the predicted future state transition path of the female animal specifically includes:

[0019] Based on the transition probability matrix, the dwell time and probability of state transition of female animals in each state during the reproductive cycle are analyzed. The probability value of each state in the matrix is ​​extracted in turn and the data is correlated to obtain the state transition trend evaluation result.

[0020] Based on the state transition trend assessment results, the node of the current state is taken as the starting point of the path, the probability of the next state transition is analyzed, and the nodes are gradually added to the transition path to generate the state transition assessment results of the female animal.

[0021] As a further aspect of the present invention, the step of obtaining the distance between the analytical feature vectors specifically comprises:

[0022] Based on the assessment results of the female animal's state transition and behavioral characteristic data, data on the female animal's activity frequency, body temperature, weight, and number of reproductions were extracted. The data were then standardized and integrated into feature vectors to obtain the feature vector set of the female animal.

[0023] Based on the feature vector set of the female animal, the distance between the feature vectors is analyzed using the following formula:

[0024]

[0025] Calculate the Euclidean distance d between female animals q and r across all behavioral characteristics. qr Establish groupings based on the behavioral characteristics of female livestock;

[0026] Among them, z qt′ and z rt′ represents the values ​​of female animal q and female animal r on the behavioral feature at the t′ term, respectively, and s is the feature vector dimension, representing the number of behavioral features, currently activity frequency, body temperature, weight, and number of reproductions.

[0027] As a further aspect of the present invention, the steps for obtaining the behavioral feature group segmentation results are specifically as follows:

[0028] Based on the grouping of the female livestock's behavioral characteristics, behavioral characteristic data of the female livestock group are extracted, including activity frequency, body temperature, weight, and number of reproductions. The data are statistically analyzed and processed, and the characteristic differences between groups are determined to obtain the characteristic difference analysis results.

[0029] Based on the results of the feature difference analysis, the behavioral feature range of each group is determined, and the grouping is further optimized by merging groups with similar feature ranges into type groups, adjusting the boundary features of each type group, and optimizing the consistency of female animal features within the same type group to obtain the behavioral feature group division results.

[0030] As a further aspect of the present invention, the step of obtaining the baseline fluctuation of the analyzed physiological information specifically includes:

[0031] Based on the behavioral characteristics of the group segmentation results, physiological information such as body temperature, heart rate and activity frequency of the female animals are extracted from the group. The data is organized in chronological order, the basic fluctuation of physiological information is monitored, and the physiological fluctuation record of the group is obtained.

[0032] Based on the physiological fluctuation records of the group, the following formula is used:

[0033]

[0034] Calculate the change value ΔP′ of the y-th physiological information item. y The results of the analysis of changes in physiological parameters were obtained;

[0035] Among them, P′ y,h Let P' represent the y-th physiological parameter value at the current time h. y,h-1 Let W represent the y-th physiological parameter value at the previous time h-1, D represent the difference in the number of days between the two measurements, and W represent the value of the y-th physiological parameter at the previous time h-1. y S is the weighting coefficient of the y-th physiological parameter. y It is the seasonally adjusted value of the y-th physiological parameter.

[0036] As a further aspect of the present invention, the step of acquiring physiological information to identify abnormal fluctuations specifically includes:

[0037] Based on the analysis results of the changes in the physiological parameters, the daily recorded changes in body temperature, heart rate and activity frequency are compared with the conventional range of previous statistics to determine whether each data exceeds the normal fluctuation range and generate the screening results of abnormal data.

[0038] Based on the screening results of the abnormal data, all data marked as abnormal are recorded in chronological order, the frequency and time distribution of abnormal fluctuations are statistically analyzed, the presence of periodic characteristics is checked and analyzed, and abnormal physiological fluctuation analysis results are generated.

[0039] As a further aspect of the present invention, the step of selecting the optimal mating time specifically comprises:

[0040] Based on the assessment results of the female animal state transition, the results of the behavioral characteristic group segmentation, and the results of the abnormal physiological fluctuation analysis, the reproductive status information, behavioral characteristics, and physiological condition data of the female animals are matched by individuals and time series. The data within the time point are organized, and the changing trends of body temperature, heart rate, and activity frequency during the reproductive cycle are analyzed to obtain reproductive-related information of the female animals.

[0041] Based on the aforementioned female animal reproductive correlation information, the following formula is used:

[0042]

[0043] Calculate the reproductive success probability GS of the female livestock population to obtain recommended mating information;

[0044] Among them, R f A is the reproductive status score of the f-th female animal. f G is the health status score of the f-th female animal, N′ represents the total number of female animals participating in the analysis, and G f It is the behavioral characteristic fit score of the f-th female animal.

[0045] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0046] This invention analyzes data from various states of the female animal's reproductive cycle, clarifying the path of state changes at different stages. This enables accurate prediction and quantitative assessment of the state within the reproductive cycle, leading to more precise breeding plans. By evaluating the dwell time and transition trends at each stage, the optimal mating time can be predicted, improving overall reproductive efficiency. Based on similarity analysis of behavioral data such as activity frequency, body temperature, and number of matings, individual female animals are divided into different behavioral groups, allowing for more targeted reproductive management strategies for each group. By continuously recording fluctuations in physiological parameters and monitoring them using historical thresholds, accurate identification and early warning of abnormal physiological fluctuations are achieved. This enables early detection of health problems, reduces reproductive risks, and ensures the stability of the reproductive process. Furthermore, by integrating information on reproductive status, behavioral characteristics, and physiological conditions, analysis of mating suitability and reproductive success rates provides managers with decision-making support and improves the overall reproductive efficiency of the herd. Attached Figure Description

[0047] Figure 1 This is a system flowchart of the present invention;

[0048] Figure 2 A flowchart illustrating the construction of the transition probability matrix for this invention;

[0049] Figure 3 This is a flowchart illustrating the prediction of future state transition paths for female livestock according to the present invention;

[0050] Figure 4 This is a flowchart illustrating the analysis of the distance between feature vectors in this invention;

[0051] Figure 5 This is a flowchart illustrating the process of obtaining behavioral characteristic group segmentation results according to the present invention;

[0052] Figure 6 This is a flowchart illustrating the basic fluctuations in physiological information analyzed in this invention.

[0053] Figure 7 This is a flowchart illustrating the process of identifying abnormal fluctuations in physiological information according to the present invention.

[0054] Figure 8 A flowchart for selecting the optimal mating time for this invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0056] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0057] Please see Figure 1 A livestock and veterinary mating data analysis system includes:

[0058] The state transition analysis module collects state data of female animals during their reproductive cycle, including estrus, pregnancy, lactation, and recovery. It arranges the state data of each female animal in chronological order throughout the entire reproductive cycle into a state sequence. By analyzing the sequence data of state changes, it calculates the transition probability between states, integrates the data of all female animals to construct a transition probability matrix, evaluates the dwell time and transition probability of each state based on the transition probability matrix, predicts the future state transition path of the female animals, and generates the state transition assessment results of the female animals.

[0059] The behavioral clustering and grouping module is based on the assessment results of the female animal state transition and behavioral characteristic data such as the female animal activity frequency, body temperature, weight, and number of reproductions. It unifies the behavioral characteristic data into feature vectors, calculates the similarity between female animals by analyzing the distance between feature vectors, and divides the female animals into groups based on the similarity. It performs feature difference analysis on each group, establishes typological grouping of female animals, and obtains behavioral characteristic grouping results.

[0060] The abnormal fluctuation detection module is based on the physiological information of female animals, such as body temperature, heart rate, and activity frequency, in the behavioral characteristic group segmentation results. It continuously records the physiological data of female animals in specific behavioral groups to analyze the basic fluctuation of physiological information. By calculating the change value of the parameters, comparing the change value with the normal range, referring to the threshold range of past data statistics, it identifies the physiological information of abnormal fluctuations, records the time and frequency of abnormal fluctuations, and generates abnormal physiological fluctuation analysis results.

[0061] The mating data analysis module integrates and analyzes the reproductive data of female animals by combining the results of female animal status transformation assessment, behavioral characteristic group segmentation results, and abnormal physiological fluctuation analysis results. It extracts key information from reproductive status, behavioral characteristics, and physiological conditions, calculates the suitability and reproductive success probability of the female animal group based on past mating success rate data, selects the best mating time, and generates recommended mating information.

[0062] The results of the female animal state transition assessment include the average time spent in each state, the probability of transition between states, and the predicted path of future state transitions. The results of the behavioral characteristics group segmentation include the mean activity frequency, body temperature variation range, and weight fluctuation range of each group. The results of the abnormal physiological fluctuation analysis include the physiological parameters of abnormal fluctuations, the time of occurrence, and the frequency of fluctuations. The recommended mating information includes the suitability score, the probability of reproductive success, and the optimal mating time.

[0063] Please see Figure 2 The specific steps for constructing the transition probability matrix are as follows:

[0064] Based on the estrus, pregnancy, lactation and recovery period status data of female animals in the reproductive cycle, the time and status identifier of the status changes of each female animal are recorded, and the data are organized in chronological order to obtain the status sequence data of female animals.

[0065] Based on data from different states of female animals during their reproductive cycle, including estrus, pregnancy, lactation, and recovery, data acquisition devices (such as RFID tags or biosensors) are used to monitor the state changes of each female animal in real time, recording the specific time and state identifier of each state change. All state data is collected and imported into a Python Pandas library or an SQL database, and then sorted chronologically to generate a complete state sequence for each female animal. First, the imported data is cleaned to remove duplicate or missing values ​​to ensure accuracy and consistency. Then, the data is sorted according to timestamps to form a time series of the female animal throughout its reproductive cycle. Records with abnormal data (such as significantly abnormal state time spans) are identified and cleaned using an anomaly detection method based on time intervals to ensure the rationality of the state sequence.

[0066] Based on the female animal state sequence data, the following formula is used:

[0067]

[0068] Calculate the probability P of transitioning from state i to state j. ij This allows us to obtain the probability information of state transitions.

[0069] Among them, P ijThis reflects the tendency of female animals to switch between different reproductive states (such as estrus, pregnancy, lactation, or recovery period), N ij N represents the total number of times the female animal transitions from state i to state j, indicating the frequency of this transition. It is used to quantify the actual frequency of each state transition. The Pandas library in Python is used to statistically analyze the female animal state data sequence. ik T is the number of transitions the female animal makes from state i to state k, calculated from the state sequence data. k is the average dwell time in state k, where k is an index variable representing all possible state numbers, and m is the total number of possible states;

[0070] T i and T j T represents the average dwell time in state i and state j, respectively, reflecting the average dwell time of female animals in various reproductive states (such as estrus, pregnancy, etc.), helping to analyze the stability and transition time distribution of female animals in different states. i The calculation method is as follows:

[0071]

[0072] In the formula, t in The duration of the female animal's stay in state i during the nth time is captured by sensors, N. i The total number of times the female animal appears in state i is obtained from the sequence data. n represents the observation number of each stay and is used to calculate the total stay time. The average stay time can be obtained using data analysis tools such as SPSS and Excel.

[0073] T j The calculation method is as follows:

[0074]

[0075] In the formula, t jn The duration of the female animal's stay in state j for the nth time, N j It is the total number of times the female animal appears in state j;

[0076] σ T This is the standard deviation of the time the female animal spends in state i, representing the fluctuation in the time spent in different states. It is used to measure the consistency of the time spent by the female animal in different states and reflects the degree of change in the time distribution of the female animal during state transitions. The calculation method is as follows:

[0077]

[0078] For example, in a dataset of female animal reproductive cycles: if the number of transitions N from estrus (state i) to gestation (state j) is...ij =8, the average duration T of estrus (state i) i =5 days, the average stay T during pregnancy (state j) j =10 days, standard deviation σ of the duration of estrus state T =3, the total number of times the estrus cycle transitions to all states multiplied by the duration of each state.

[0079] Calculate the numerator:

[0080] N ij ×(T i +T j )=8×(5+10)=8×15=120

[0081] Calculate the denominator:

[0082]

[0083] Calculate the final probability P ij :

[0084]

[0085] Calculation result P ij The value of ≈0.9756 indicates that there is approximately a 97.56% probability that a female animal will transition into gestation after entering estrus.

[0086] Based on the transition probability information between states, the number of times the female animal transitions from each state to other states is counted, and the state transition frequencies are integrated. Based on the state transition data of the female animal during the reproductive cycle, a transition probability matrix is ​​constructed.

[0087] After integrating the state sequence data of all female livestock, a transition probability matrix is ​​constructed to accurately analyze the transition relationships between different states in the female livestock's reproductive cycle. The specific steps are as follows: First, the state sequence of each female livestock is analyzed one by one, recording each instance of the female livestock transitioning from one state to another. For example, when a female livestock transitions from estrus to pregnancy, it is recorded as one "estrus to pregnancy" transition. The number of times each female livestock transitions from estrus to pregnancy, lactation, or recovery during the entire reproductive cycle is counted sequentially, and the same information is recorded for other state transitions until the complete state transition data distribution of the female livestock throughout the entire reproductive cycle is obtained. This process is repeated for the state sequence of each female livestock, accumulating the number of transitions between each state. Next, the number of transitions for all female livestock is summarized to obtain a total state transition count matrix. The rows and columns of this matrix represent the four states in the reproductive cycle (e.g., estrus, pregnancy, lactation, and recovery), and each element value in the matrix represents the total number of transitions from the corresponding row state to the corresponding column state. For example, the first row and first column of the matrix represent the cumulative number of transitions from estrus to pregnancy for all female animals; the second row and second column represent the cumulative number of transitions from pregnancy to lactation; and so on, filling all elements of the matrix until the entire matrix of transition counts is constructed. Finally, the transition probabilities P calculated in paragraph 2 are substituted into the matrix. ij This converts the transition frequency of each state into the corresponding probability matrix elements. For example, if paragraph 2 has already calculated the transition probability P from estrus to pregnancy... ij =0.9756, so this value is filled into the corresponding position in the transition probability matrix to represent the probability of transitioning from estrus to gestation. All calculated transition probabilities are then filled in sequentially, so that each element of the matrix represents the specific probability of transitioning from one state to another. The final generated transition probability matrix provides a detailed structure of the state transitions in the female animal's reproductive cycle, where each element quantifies the probability of transitioning from one state to another.

[0088] Please see Figure 3 The specific steps for obtaining the predicted future state transition path of female livestock are as follows:

[0089] Based on the transition probability matrix, the dwell time and probability of state transition of female animals in each state during the reproductive cycle are analyzed. The probability value of each state in the matrix is ​​extracted in turn and the data are correlated to obtain the state transition trend assessment results.

[0090] By constructing a transition probability matrix, we further analyze the average dwell time and transition probability of female livestock in each state during the reproductive cycle. First, we extract the self-transition probability of each state from the matrix; that is, the probability that the female livestock will remain in its current state without transitioning to another state within a cycle. The average dwell time for each state can be estimated based on the self-transition probability. Specifically, a higher self-transition probability indicates a longer dwell time in that state, while a lower self-transition probability indicates a shorter dwell time. Based on this relationship between probability and time, we analyze the self-transition probability of each state sequentially to calculate its average dwell time. After obtaining the average dwell time for each state, we combine it with the transition probabilities between states to further refine the transition probability of the female livestock. Specifically, we obtain the probability value of each state transitioning to another state in the transition probability matrix and combine it with the average dwell time of the current state; that is, the dwell time of each state corresponds to its transition probability. For example, if the estrus period lasts for 5 days and the transition probability is 0.2, then there is a 20% probability of transitioning to the next state within 5 days; and so on, the duration and transition probability of other states are evaluated one by one, and finally the distribution data of the duration and transition probability of each state within the cycle are obtained.

[0091] Based on the state transition trend assessment results, the node of the current state is taken as the starting point of the path, the probability of the next state transition is analyzed, and the nodes are gradually added to the transition path to generate the state transition assessment results of the female animal.

[0092] Based on the transition probabilities derived from the transition probability matrix and the dwell time in each state, the state transition path of the female animal in the future reproductive cycle is predicted. First, the current state is selected as the starting point, and the row for that state is read from the transition probability matrix to obtain the transition probability distribution to other states. For each possible transition state, its corresponding probability value is extracted and recorded, and a path node is established. Then, in the next step, the current node state is used as the new starting state, and the transition probability distribution for that state is searched again in the matrix. Through a chain-like probability accumulation method, the probability of transitioning to the next state at each step is calculated step by step, updating the transition path for each state node. To predict multi-step transition paths, a set prediction step size is introduced for recursion. Each time, the state result of the previous step is updated and substituted into the transition matrix to generate new transition probabilities. Throughout the path prediction process, the path with the highest transition probability is selected based on the cumulative probability, and the duration of the female animal in each state is determined by combining the dwell time in each state. By recursively calling the rows and columns of the transition matrix, the next step for each state is continuously predicted, ultimately generating a path containing multiple transition nodes. This path is a prediction sequence of female livestock transitions in future reproductive cycles. By using the cumulative probability of multiple state nodes, it provides probabilistic assessment data on the future state changes of female livestock during the reproductive cycle, providing a reliable predictive basis for the reproductive management and state monitoring of female livestock.

[0093] Please see Figure 4 The specific steps for obtaining the distance between feature vectors are as follows:

[0094] Based on the assessment results of the female animal's state transition and behavioral characteristic data, data on the female animal's activity frequency, body temperature, weight, and number of reproductions were extracted. The data were then standardized and integrated into feature vectors to obtain the feature vector set of the female animal.

[0095] Feature vectors are constructed by combining behavioral characteristic data of female animals (including activity frequency, body temperature, weight, and reproductive frequency). Each feature data point is converted into a unified numerical representation. First, behavioral characteristic data is extracted from the data acquisition system. Activity frequency data is recorded by the number of daily activities, ensuring data clarity and no duplication, and the average activity value is extracted. Body temperature data is acquired through a body temperature sensor, with the average of the daily maximum and minimum body temperatures representing the daily temperature. The body temperature records over the period are further averaged. Weight data uses a dataset of each weighing, with the monthly average weight of each female animal representing the feature value. Reproductive frequency data is directly extracted from historical records as cumulative values. To unify these feature values, the data is input into a standardization system, employing a min-max normalization method to map each feature to a range between 0 and 1. For example, Python's MinMaxScaler is used for normalization. In this way, the resulting feature vector includes standardized activity frequency, body temperature, weight, and reproductive frequency, forming a unified feature vector format.

[0096] Based on the feature vector set of the female animal, the distance between the feature vectors is analyzed using the following formula:

[0097]

[0098] Calculate the Euclidean distance d between female animals q and r across all behavioral characteristics (activity frequency, body temperature, weight, number of reproductions). qr Establish groupings based on the behavioral characteristics of female livestock;

[0099] Where, d qr This is used to quantify the similarity between two female animals in these behavioral traits. The smaller the distance value, the closer the two are in each behavioral trait. qt′ and z rt′ These represent the values ​​of female livestock q and r on the behavioral characteristic at the t′ point, including activity frequency, body temperature, weight, and number of reproductions. Activity frequency is recorded by activity monitoring equipment, using the average of daily activity frequency as the feature value. Body temperature is the average of the daily maximum and minimum values ​​extracted from the body temperature monitoring equipment, and the average value over the period is taken. Weight is recorded periodically by weighing equipment, using the average weight over the period as the feature. Number of reproductions is the cumulative value extracted from mating records. s is the dimension of the feature vector, representing the number of behavioral characteristics. Currently, there are four features: activity frequency, body temperature, weight, and number of reproductions.

[0100] For example, if the standardized feature vectors of female animal q and female animal r are respectively and

[0101] Calculate the square of the difference for each feature:

[0102] Activity frequency: (0.6-0.5) 2 =0.01; body temperature: (0.7-0.6) 2 =0.01; weight: (0.5-0.4) 2 =0.01; Number of reproductions: (0.8-0.7) 2 =0.01.

[0103]

[0104] The results indicate that the similarity distance between female animals q and r in behavioral characteristics such as activity frequency, body temperature, weight, and reproductive frequency is 0.2. This is based on the distance d between each pair of female animals. qr Hierarchical clustering or K-means clustering methods are used to group female animals with high similarity into the same group. First, by setting a distance threshold, all female animals are compared pairwise, and the distance for each d is recorded. qr The similarity of female animals is calculated, and then based on a distance threshold, female animals with smaller distances are grouped into the same initial group. In hierarchical clustering, starting from the initial group, groups with high similarity are gradually merged until a suitable number of groups are formed. In K-means clustering, based on the feature vectors of each female animal, centroids are selected and iterative optimization is performed to assign female animals to the group represented by the nearest centroid. The final partitioning results in several groups, each composed of female animals with high similarity in behavioral characteristics, providing a clear grouping structure for subsequent feature difference analysis.

[0105] Please see Figure 5 The specific steps for obtaining the behavioral characteristic group segmentation results are as follows:

[0106] Based on the behavioral characteristics of female livestock, we group them into groups and extract behavioral characteristic data from the female livestock groups, including activity frequency, body temperature, weight, and number of reproductions. We then perform statistical analysis on the data and determine the characteristic differences between groups to obtain the characteristic difference analysis results.

[0107] First, behavioral characteristic data for each female animal in the divided groups, including activity frequency, body temperature, weight, and reproductive frequency, are extracted. This data is imported into statistical analysis software, such as SPSS or the pandas library in Python, to generate descriptive statistics for each characteristic, including mean, standard deviation, and distribution, to clarify the overall characteristic profile of each group. Then, analysis of variance (ANOVA) is used to test the significance of differences in behavioral characteristic data between different groups. Specifically, the mean difference between groups is analyzed for each characteristic item separately, and the significance index of each characteristic between groups is recorded to verify whether there are significant differences in each behavioral characteristic among groups. For correlation analysis between characteristics, the Pearson correlation coefficient is used to test the correlation between each combination of characteristics to understand the degree of association between, for example, activity frequency and body temperature and weight, and to determine the dependence between characteristics. Finally, the results of ANOVA and correlation analysis are summarized to form a characteristic difference analysis report for each group, clarifying the unique characteristic structure of each group.

[0108] Based on the results of the characteristic difference analysis, the behavioral characteristic range of each group is determined, and the grouping is further optimized by merging groups with similar characteristic ranges into type groups, adjusting the boundary characteristics of each type group, and optimizing the consistency of female animal characteristics within the same type group to obtain the behavioral characteristic group division results.

[0109] First, based on the characteristic differences of each group obtained from the aforementioned analysis, a numerical range for each characteristic is set. Groups with similar characteristic ranges are merged into a single type group to ensure that female animals within the same type group maintain a high degree of consistency in behavioral characteristics. Then, hierarchical clustering or K-means clustering is used to further optimize the groups. By selecting the centroids of the type groups and performing multiple iterations of optimization, clustering parameters are set in the sklearn library of Python. The initial centroids are used to reassign the existing groups, gradually grouping female animals with more similar characteristic values ​​into the corresponding type groups. In each iteration, the centroids of each type group are updated, and the range of boundary features is adjusted to ensure that each female animal matches the core characteristics of its type group to the greatest extent. Finally, the typified grouping structure of female animals is formed, making the structure of each group clear and the characteristics consistent.

[0110] Please see Figure 6 The specific steps for obtaining the baseline fluctuations in physiological information are as follows:

[0111] Based on the behavioral characteristics of the group, physiological information such as body temperature, heart rate and activity frequency of the female animals were extracted from the group. The data were organized in chronological order, and the basic fluctuation of physiological information was observed to obtain the physiological fluctuation record of the group.

[0112] First, body temperature, heart rate, and activity frequency of each female animal are continuously collected and automatically uploaded using electronic monitoring devices. This ensures that each collected data is linked to a unique identifier for long-term tracking. When exporting data through the system interface, it is organized chronologically to generate a continuous time-series data table. Subsequently, the platform processes the data, including data cleaning to remove outliers and missing data, and grouping and classifying physiological fluctuations at different times. By comparing physiological characteristics with historical baseline values, the system filters out individuals or groups with significant fluctuations and marks female animals with potential health risks. Finally, based on daily and periodic fluctuation analysis, the monitoring results are visualized.

[0113] Based on the physiological fluctuation records of the population, the following formula is used:

[0114]

[0115] Calculate the change value ΔP′ of the y-th physiological information item. y The results of the analysis of changes in physiological parameters were obtained;

[0116] Where, ΔP′ y P′ is used to capture the changes in specific physiological parameters at different time points. y,h P′ represents the y-th physiological parameter value at the current time h, collected by monitoring equipment. y,h-1 Let W represent the y-th physiological parameter value at the previous time h-1, and D represent the difference in days between the two measurements, calculated directly by comparing the date stamps of the two sampling times. y This is the weighting coefficient for the y-th physiological parameter, used to balance the importance of different parameters. It is derived from historical data analysis. The specific process includes performing multiple regression analysis on each physiological parameter or using correlation analysis tools (such as SPSS) to calculate its importance under different conditions, and assigning appropriate weights based on the parameter's impact on mating success. For parameters with high correlation, such as the importance of heart rate in estrus cycle prediction, a higher weight W can be assigned. y To improve the accuracy of the results, S y This is the seasonally adjusted value of the y-th physiological parameter, derived from historical data analysis. Time series analysis or comparison with historical means (such as using Python's time series analysis tools) is used to determine the baseline levels of each parameter in different seasons, and these baseline differences are quantified as the adjustment value S. y This is to ensure that the calculation results can reflect normal fluctuations in different seasons and reflect the impact of seasonality on physiological data.

[0117] For example, if the body temperature, heart rate, and activity frequency of the female animal at the current moment are recorded as P′ temp,h =39.0℃, P′ hr,h =80 bpm, P′act,h = 50 times / day, the previous record is P′ temp,h-1 =37.8℃, P′ hr,h-1 =75 bpm, P′ act,h-1 = 48 times / day, measurement interval D = 2 days, weighting parameter W temp =1.1, W hr =1.05, W act =1.2, seasonally adjusted value is S temp =0.2, S hr =0.1, S act =0.15.

[0118] Body temperature change calculation:

[0119]

[0120] Heart rate variability calculation:

[0121]

[0122] Calculation of activity frequency variation:

[0123]

[0124] The results showed that the body temperature changed by 0.86℃, the heart rate changed by 2.725 bpm, and the activity frequency changed by 1.35 times / day.

[0125] Please see Figure 7 The specific steps for obtaining physiological information to identify abnormal fluctuations are as follows:

[0126] Based on the analysis results of changes in physiological parameters, the daily recorded changes in body temperature, heart rate, and activity frequency are compared with the conventional range of previous statistics to determine whether each data exceeds the normal fluctuation range and generate the screening results of abnormal data.

[0127] The change value ΔP′ based on physiological information y The values ​​were compared with normal ranges, and anomaly detection was performed using threshold ranges derived from historical data. Specifically, the normal range for body temperature was set at 37.5℃ to 39.0℃, for heart rate at 60 bpm to 80 bpm, and for daily activity frequency at 45 to 55 times per day. Baseline values ​​were obtained from data collected over the past 30 days, such as a baseline average of 38.2℃ for body temperature, 70 bpm for heart rate, and 50 times per day for activity frequency. The threshold range for each physiological parameter was calculated by adding or subtracting three standard deviations from the mean. During the analysis, ΔP′ was used... temp =0.86, ΔP′ hr =2.725 and ΔP′act The change of 1.35 was compared with the corresponding physiological baseline value. Body temperature analysis: The current body temperature change is 0.86℃. When the baseline body temperature is 38.2℃, this change raises the body temperature to 39.06℃, slightly higher than the upper limit of the normal range (39.0℃), and is judged as a slight abnormal fluctuation. Heart rate analysis: The heart rate change is 2.725 bpm, increasing the baseline heart rate from 70 bpm to 72.725 bpm, still within the normal range and not judged as abnormal. Activity frequency analysis: The activity frequency change is 1.35 times / day, increasing the baseline activity frequency from 50 times / day to 51.35 times / day, which is within the normal fluctuation range. During the comparison process, when the change value causes a result exceeding the normal range, the system marks the data as abnormal data and records the reason for the abnormality. Based on the above results, the change in body temperature exceeds the threshold range and should be further observed; other physiological parameters are all within the normal range.

[0128] Based on the screening results of abnormal data, all data marked as abnormal are recorded in chronological order. The frequency and time distribution of abnormal fluctuations are statistically analyzed. The presence of periodic characteristics is checked and analyzed, and abnormal physiological fluctuation analysis results are generated.

[0129] After identifying abnormal physiological fluctuations, the time and frequency of these fluctuations are recorded to generate the final analysis results. First, each temperature data point marked as abnormal is recorded chronologically and marked with a specific timestamp. For example, a temperature of 39.06℃ recorded at 10:00 AM on October 27th is designated as the abnormal fluctuation point. Then, the frequency of abnormal fluctuations over several consecutive days is statistically analyzed to determine if they exhibit periodicity. For instance, if temperature fluctuations occur at the same time each day, they may be related to specific behaviors or environmental factors. Furthermore, the number of times daily temperature exceeds the normal range is cumulatively recorded to assess the severity and frequency trend of the abnormality. If abnormal temperatures are recorded for three consecutive days, on-site intervention or examination may be necessary. The final abnormality analysis results will include the time series, frequency distribution, and potential periodic patterns of the abnormal temperature, providing veterinarians with decision-making support to ensure the health of female animals remains within an appropriate range and to avoid impacting mating success rates.

[0130] Please see Figure 8 The specific steps for determining the optimal mating time are as follows:

[0131] Based on the results of the assessment of the state transition of female animals, the results of the group segmentation of behavioral characteristics, and the results of the analysis of abnormal physiological fluctuations, the reproductive status information, behavioral characteristics, and physiological condition data of female animals are matched by individuals and time series. The data within the time point are sorted out, and the changing trends of body temperature, heart rate, and activity frequency during the reproductive cycle are analyzed to obtain reproductive-related information of female animals.

[0132] First, behavioral data of female animals (such as activity frequency, feed intake, and movement patterns) are arranged in chronological order according to daily records, and aligned with physiological data such as body temperature and heart rate. Then, time series analysis methods, such as the sliding window technique, are used to calculate the trends in behavioral and physiological data over 7-day or 30-day intervals, observing their synchronicity or lag. For example, is an increase in heart rate fluctuation or a decrease in activity frequency accompanied by an increase in body temperature? Next, dynamic time warping is used to compare the fluctuation patterns of behavioral characteristics and physiological data, identifying similar patterns among different individuals to determine the degree to which health status affects reproductive behavior. Similar patterns among different individuals include: synchronized fluctuation patterns of body temperature and heart rate: during the estrous cycle, the period of rising body temperature is accompanied by the same trend of rising heart rate. These individuals typically reach breeding suitability within 23 days of the peak body temperature, and individuals matching this pattern are more suitable for concentrated mating. Inverse correlation patterns between activity frequency and the reproductive cycle: some individuals show decreased activity frequency around ovulation and increased activity frequency during non-breeding periods. Identifying individuals with similar behavioral patterns helps monitor reproductive progress based on activity levels. Abnormal physiological fluctuation patterns: Some individuals exhibit abnormal fluctuations in body temperature or heart rate outside of estrus, such as prolonged elevated body temperature or abnormally rapid heart rate. These patterns indicate that health conditions may affect reproductive behavior, requiring further health assessment. Patterns of reproductive behavior consistent with environmental responses: Different female animals exhibit consistent physiological and behavioral changes under the same environmental changes (such as temperature changes or water fluctuations), for example, high temperatures causing multiple females to experience synchronized increases in body temperature and heart rate, and reduced activity. Discovering these patterns helps assess the impact of environmental factors on reproductive behavior. For periods of abnormal fluctuations, further analysis is needed to determine if the behavioral patterns are inconsistent with reproductive status, such as reduced activity frequency during periods of elevated body temperature, to reveal potential health risks. Ultimately, the analysis results of these interaction patterns provide accurate reference data for predicting reproductive success rates and selecting mating timing.

[0133] Based on the reproductive correlation information of female animals, the following formula is used:

[0134]

[0135] Calculate the reproductive success probability GS of the female livestock population to obtain recommended mating information;

[0136] Among them, GS is used to assess whether the population is at an appropriate breeding time, and R... f This is the reproductive status score of the f-th female animal, determined by monitoring data on body temperature, heart rate, and ovulation time. For example, when body temperature reaches its peak and heart rate increases, it indicates that ovulation has occurred. This score is derived through logistic regression analysis using historical data and monitoring tools such as Excel. fThis is the health status score of the f-th female animal, determined based on the analysis of abnormal physiological fluctuations. A higher score is given if body temperature and heart rate fluctuations are within the healthy range. N′ represents the total number of female animals participating in the analysis, and G... f The behavioral characteristics fit score of the f-th female animal reflects whether the behavioral characteristics match the behavior during estrus. These behaviors are calculated by associating them with reproductive success rate through multiple regression analysis. The main data include daily activity frequency, changes in food intake, movement trajectory and mating performance. The behavioral data are collected daily by monitoring equipment.

[0137] To calculate the behavioral trait fit score G for each female animal f A regression model needs to be built to correlate behavioral data with reproductive success rate. The specific steps are as follows:

[0138] Collect daily behavioral data of all female livestock, such as daily activity frequency, feed intake, and movement patterns, and pair this data with historical reproductive success rates.

[0139] Use the statsmodels or SPSS tools in Python to build a multiple regression model, setting the target variable as reproductive success rate and the independent variables as various behavioral data.

[0140] Regression model specification:

[0141] Y = β0 + β1X1 + β2X2 + β3X3 + ε

[0142] Where Y represents the reproductive success rate, X1, X2, and X3 represent activity frequency, food intake, and movement trajectory, respectively, β0 is the intercept, β1, β2, and β3 are the regression coefficients of their respective behavioral characteristics, and ε is the error term.

[0143] Model Fitting and Coefficient Calculation: The collected behavioral data is substituted into the regression model for fitting, and the regression coefficients of each behavioral characteristic are calculated using the least squares method. The magnitude of the regression coefficients reflects the degree of influence of the behavioral characteristics on reproductive success rate.

[0144] Behavioral fit score calculation: Based on the fitting results, calculate the behavioral fit score G for each female animal. f The formula is as follows:

[0145] G f =β1X 1f +β2X 2f +β3X 3f

[0146] Among them, X 1f X 2f X 3f These represent the activity frequency, feed intake, and movement trajectory of the f-th female animal, respectively.

[0147] Results Validation and Application: Based on the results of the regression model, the behavioral characteristic score G for each female animal was calculated. f Then, it was used for overall reproductive fitness calculations.

[0148] If there are 3 female animals in the group, the scores are as follows: 1st female animal: R1 = 0.8, G1 = 0.9, A1 = 0.95; 2nd female animal: R2 = 0.7, G2 = 0.85, A2 = 0.9; 3rd female animal: R3 = 0.9, G3 = 0.8, A3 = 0.85; Total number of female animals: N′ = 3.

[0149] Substitute into the formula:

[0150]

[0151] This result indicates that the overall reproductive success rate of the current female livestock population is 61%. By comparing this result with historical reproductive success rates (e.g., 55%), it can be determined whether the current population is in optimal reproductive condition. If it is higher than the baseline, it suggests that the current population is suitable for mating; if it is lower than the baseline, there may be health or behavioral abnormalities requiring further intervention. This result will support reproductive management decisions, ensuring mating at the optimal time, improving reproductive success rates, and optimizing reproductive management strategies.

[0152] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A livestock veterinarian breeding data analysis system, characterized by, The system includes: The state transition analysis module collects data on the reproductive cycle of female animals, integrates data from all female animals to construct a transition probability matrix, predicts the future state transition path of female animals based on the transition probability matrix, and generates state transition assessment results for female animals. The behavior clustering and grouping module, based on the evaluation results of the female animal state transition and the behavioral characteristic data of the female animal, unifies the behavioral characteristic data into feature vectors, analyzes the distance between feature vectors, divides the groups according to similarity, and obtains the behavioral characteristic group division results. The fluctuation anomaly detection module continuously records the physiological data of female animals in a specific behavioral group based on the physiological information of the female animals in the behavioral feature group segmentation results, analyzes the basic fluctuation of physiological information, identifies abnormal fluctuation physiological information, and generates abnormal physiological fluctuation analysis results. The mating data analysis module integrates and analyzes the reproductive data of female animals by combining the assessment results of female animal state transition, the results of behavioral characteristic group segmentation, and the results of abnormal physiological fluctuation analysis. Based on the previous mating success rate data, it selects the best mating time and generates recommended mating information. The specific steps for obtaining the transition probability matrix are as follows: Based on the estrus, pregnancy, lactation and recovery period status data of female animals in the reproductive cycle, the time and status identifier of the status changes of each female animal are recorded, and the data are organized in chronological order to obtain the status sequence data of female animals. Based on the female animal state sequence data, the following formula is used: ； Calculate from state Transition to state probability This allows us to obtain the probability information of state transitions. wherein is the total number of transitions of a female from state to state , is the number of transitions of a female from state to state , is the average residence time of state , is an index variable, is the total number of states, is the standard deviation of the residence time of a female in state ; Based on the transition probability information between states, the number of times the female animal transitions from each state to other states is counted, and the state transition frequencies are integrated. Based on the state transition data of the female animal during the reproductive cycle, a transition probability matrix is ​​constructed. The specific steps for obtaining the baseline fluctuations of the analyzed physiological information are as follows: Based on the behavioral characteristics of the group segmentation results, physiological information such as body temperature, heart rate and activity frequency of the female animals are extracted from the group. The data is organized in chronological order, the basic fluctuation of physiological information is monitored, and the physiological fluctuation record of the group is obtained. Based on the physiological fluctuation records of the group, the following formula is used: ； calculating a change value of the physiological information a change value of the physiological information obtaining a change analysis result of the physiological parameter wherein, denotes the current time the first physiological parameter value, denotes the first physiological parameter value of the previous time the first physiological parameter value of the previous time is the difference in days between the two measurements, is the weighting coefficient of the first physiological parameter, is the seasonal adjustment value of the first physiological parameter. The specific steps for selecting the optimal mating time are as follows: Based on the assessment results of the female animal state transition, the results of the behavioral characteristic group segmentation, and the results of the abnormal physiological fluctuation analysis, the reproductive status information, behavioral characteristics, and physiological condition data of the female animals are matched by individuals and time series. The data within the time point are organized, and the changing trends of body temperature, heart rate, and activity frequency during the reproductive cycle are analyzed to obtain reproductive-related information of the female animals. Based on the aforementioned female animal reproductive correlation information, the following formula is used: ； Calculating the probability of reproductive success for a dam population to obtain recommended mating information; wherein, is the number of the first is the reproductive status score of the first is the number of the second is the health status score of the second denotes the total number of cows participating in the analysis, is the number of the third is the behavioral characteristic suitability score of the third 2. The livestock breeder breeding data analysis system of claim 1, wherein, The specific steps for obtaining the predicted future state transition path of the female animal are as follows: Based on the transition probability matrix, the dwell time and probability of state transition of female animals in each state during the reproductive cycle are analyzed. The probability value of each state in the matrix is ​​extracted in turn and the data is correlated to obtain the state transition trend evaluation result. Based on the state transition trend assessment results, the node of the current state is taken as the starting point of the path, the probability of the next state transition is analyzed, and the nodes are gradually added to the transition path to generate the state transition assessment results of the female animal.

3. The livestock breeder breeding data analysis system of claim 2, wherein, The specific steps for obtaining the distance between the feature vectors are as follows: Based on the assessment results of the female animal's state transition and behavioral characteristic data, data on the female animal's activity frequency, body temperature, weight, and number of reproductions were extracted. The data were then standardized and integrated into feature vectors to obtain the feature vector set of the female animal. Based on the feature vector set of the female animal, the distance between the feature vectors is analyzed using the following formula: ； Computing dams With dams Euclidean distances between all behavioral features , establishing a cluster grouping of dams behavioral features; wherein, and are respectively represent the sow and the sow the value of the sow on the behavior characteristics, is the dimension number of the feature vector, representing the number of behavior characteristics, which is currently the activity frequency, body temperature, body weight and number of reproduction.

4. The livestock breeder breeding data analysis system of claim 3, wherein, The specific steps for obtaining the behavioral feature group segmentation results are as follows: Based on the grouping of the female livestock's behavioral characteristics, behavioral characteristic data of the female livestock group are extracted, including activity frequency, body temperature, weight, and number of reproductions. The data are statistically analyzed and processed, and the characteristic differences between groups are determined to obtain the characteristic difference analysis results. Based on the results of the feature difference analysis, the behavioral feature range of each group is determined, and the grouping is further optimized by merging groups with similar feature ranges into type groups, adjusting the boundary features of each type group, and optimizing the consistency of female animal features within the same type group to obtain the behavioral feature group division results.

5. The livestock breeder breeding data analysis system of claim 4, wherein, The specific steps for obtaining physiological information to identify abnormal fluctuations are as follows: Based on the analysis results of the changes in the physiological parameters, the daily recorded changes in body temperature, heart rate and activity frequency are compared with the conventional range of previous statistics to determine whether each data exceeds the normal fluctuation range and generate the screening results of abnormal data. Based on the screening results of the abnormal data, all data marked as abnormal are recorded in chronological order, the frequency and time distribution of abnormal fluctuations are statistically analyzed, the presence of periodic characteristics is checked and analyzed, and abnormal physiological fluctuation analysis results are generated.

Images