Baby rabbit production performance prediction method based on big data processing

Abstract

The invention belongs to the technical field of breeding, and particularly relates to a newborn rabbit production performance prediction method based on big data processing. According to the method, the innate and postnatal physical states and growth data are obtained as characteristic factors for predicting the production performance of the young rabbits, and the production performance of the young rabbits is accurately predicted by using the sequence data LSTM model which is good at processing and has long-time dependence, so that the nutrient substance level suitable for young rabbit feed is determined; the feeding standards of the meat rabbits in different physiological stages are perfected.

Description

Technical Field

[0001] The present invention belongs to the field of breeding technology, and in particular relates to a method for predicting the production performance of young rabbits based on big data processing. Background Art

[0002] With the continuous advancement of technology, intelligent management has become a development trend in various industries. Rabbit farming is no exception. Rabbit farming and management requires a significant investment of time and effort. Farmers must not only take care of the rabbits' daily needs but also constantly monitor their health. Traditional farming and management methods are often inefficient and prone to oversight.

[0003] Production performance, also known as productivity, refers to a livestock's ability to produce livestock products most economically and efficiently. It's a crucial component of individual livestock assessment and the most meaningful indicator of individual quality. Predicting the production performance of rabbit pups not only helps determine the optimal nutrient level in their diets and improve feeding standards for meat rabbits at different physiological stages, but also further improves their survival rates. Therefore, accurately predicting the production performance of rabbit pups is a pressing issue. Summary of the Invention

[0004] In response to the technical problem of the lack of existing methods for predicting the production performance of young rabbits, the present invention proposes a method for predicting the production performance of young rabbits based on big data processing, which has a reasonable design, a simple structure, is easy to process and can effectively realize the prediction of the production performance of young rabbits.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is as follows: the present invention provides a method for predicting the production performance of young rabbits based on big data processing, comprising the following steps:

[0006] a. First, select a certain number of newborn rabbits and record the birth weight of each selected rabbit;

[0007] b. Then, the daily activity distance of each rabbit is obtained by attaching a motion tracking device to the rabbit's leg;

[0008] c. Obtain data on daily weight gain, feed intake, frequency of feed intake, frequency of water intake, amount of water consumed, and frequency of defecation of the young rabbits on the day before slaughter;

[0009] d. Obtain the growth performance of young rabbits before slaughter;

[0010] e. Preprocess the acquired data to suit model training;

[0011] f. Divide the preprocessed data into a training set and a validation set in a ratio of 7:3;

[0012] g. Initialize the LSTM model parameters, input the training set, and continuously adjust the model parameters until the expected accuracy is achieved, thus building an LSTM-based rabbit production performance prediction model;

[0013] h. Use the validation set to verify the prediction performance of the LSTM-based rabbit production performance prediction model to achieve accurate prediction of rabbit production performance.

[0014] As an advantage, the hidden unit h at the previous moment in the LSTM model t-1 Input after Kalman filtering.

[0015] Preferably, in the step e, the preprocessing method is to remove outliers, fill missing values and perform normalization.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are:

[0017] The present invention provides a method for predicting the production performance of young rabbits based on big data processing. The method obtains congenital and acquired physical conditions and growth data as characteristic factors for predicting the production performance of young rabbits, and uses the LSTM model that is good at processing sequence data with long-term dependencies to achieve accurate prediction of the production performance of young rabbits, thereby determining the appropriate nutrient level in the young rabbit feed and improving the feeding standards of meat rabbits at different physiological stages. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0019] Figure 1 This is a structural diagram of the improved LSTM model. DETAILED DESCRIPTION

[0020] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described below in conjunction with the accompanying drawings and embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways than those described herein. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Example 1: This example aims to scientifically predict the production performance of young rabbits in the later stages, so as to facilitate the determination of feeding standards for meat rabbits at different physiological stages and the rapid selection of breeding rabbits through accurate prediction results. To this end, the method for predicting the production performance of young rabbits based on big data processing provided in this example includes the following steps:

[0023] Considering that prediction requires a large amount of data support, in this embodiment, four batches of 1,000 newborn rabbits were selected for data collection. Furthermore, considering that the future development of breeding rabbits is influenced by both innate and acquired factors, the birth weights of the 4,000 newborn rabbits from the four batches were recorded.

[0024] Like newborn babies, rabbits of normal weight are healthier, while rabbits with lighter weight are often weaker. The lighter weight of these rabbits may be partly due to genetic factors, but more is caused by the influence of the mother during pregnancy. As they grow, their bodies can make up for it and even become stronger. Therefore, when predicting the production performance of rabbits, both their innate and acquired factors need to be considered.

[0025] Considering that exercise is a way to enhance physical health, the daily activity distance of rabbits is beneficial to their future growth. Therefore, the daily activity distance of each rabbit is obtained by binding a motion track device to the rabbit's legs.

[0026] A reasonable diet is also one of the ways to enhance physical health after birth. By determining their daily weight gain, feed intake, feed intake frequency, water drinking frequency, water drinking volume and defecation frequency, it is useful to determine the feeding standards for meat rabbits in different physiological stages. Reasonable addition of feed and water can not only save costs but also obtain meat rabbits with better production performance. To this end, data such as daily weight gain, feed intake, feed intake frequency, water drinking frequency, water drinking volume and defecation frequency of the baby rabbits were obtained the day before slaughter. It should be noted that during the period of obtaining this data, the baby rabbits' feed and water were unlimited to ensure accurate data.

[0027] After obtaining the above data, it is necessary to obtain the final growth performance results of the rabbits. To this end, the growth performance of the rabbits before slaughter is obtained. Of course, the rabbits with good growth performance can be kept as breeding rabbits, and the rabbits before slaughter can also be determined as slaughter-period rabbits.

[0028] The acquired data is preprocessed to adapt to model training. The preprocessing of the collected data in this embodiment includes removing outliers, filling missing values and normalizing the data to make it suitable for model training.

[0029] During sensor operation, there is a possibility of data anomalies due to noise interference. The criteria for identifying outliers are as follows: an outlier is identified when the absolute value of the difference between a measured value and its mean is greater than three times the standard deviation of the measured value, or when an illegal character value occurs that cannot be detected in the data. For outlier points where the absolute value of the difference between the measured value and its mean is greater than three times the standard deviation, specific processing measures are implemented, replacing the data with the average value of the data on both sides of the outlier. Illegal character values are treated as missing values. In this study, missing values are primarily caused by sensor communication failures, and the missing value filling method uses the KNN algorithm.

[0030] To mitigate convergence difficulties caused by varying dimensions of environmental parameters, we use the maximum-minimum normalization method to scale the data to a specific range, typically [0, 1]. This method is simple to understand, preserves data distribution information, and is applicable to most data distributions.

[0031] To this end, the normalization formula is:

[0032]

[0033] Where X is the original data, Xmin is the minimum value in the original data set, and Xmax is the maximum value in the original data set.

[0034] Next, the data set can be constructed. To facilitate the verification of the results, in this embodiment, the preprocessed data is divided into a training set and a validation set in a ratio of 7:3.

[0035] Predicting production performance is a time-series problem, and a complex, nonlinear one. Factors influencing production performance include both innate and acquired factors, and these factors exhibit multicollinearity. For example, rabbits born at a normal birth weight may travel farther daily and be healthier, forming a virtuous cycle. This collinearity complicates performance prediction.

[0036] Traditional neural networks, such as BP neural networks and extreme learning machines, lack memory and are unable to record the impact of historical environmental values on future production performance. Consequently, these models cannot accurately simulate the impact of environmental factors on production performance, resulting in low prediction accuracy and a suboptimal solution for time series problems. The LSTM model is suitable for time series problems. As a variant of the recurrent neural network (RNN), LSTM inherits the RNN's memory function while effectively addressing the vanishing gradient problem that RNNs often experience with long-term memory.

[0037] Therefore, in this embodiment, the LSTM model is used for prediction. At the same time, considering that the mortality rate of young rabbits is high, and these mortality rates are generally caused by accidental factors, such as being accidentally stepped on by the mother rabbit due to limited mobile environment, resulting in damage to body functions and accidental death, etc. Therefore, in this embodiment, the Kalman filter algorithm is introduced to the information h transmitted at the last moment. t-1 The results are corrected to improve the dynamic prediction ability of the model. In this way, by initializing the LSTM model parameters, inputting the training set, and continuously adjusting the model parameters until the expected accuracy is obtained, a LSTM-based rabbit production performance prediction model is constructed.

[0038] Finally, a validation set was used to verify the predictive performance of the LSTM-based model for predicting the growth performance of young rabbits, achieving accurate predictions of their growth performance. The LSTM-based model for predicting the growth performance of young rabbits provided in this embodiment was verified to have an accuracy rate of over 95%, meeting the design requirements.

[0039] The above description is merely a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any person skilled in the art may utilize the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes for application in other fields. However, any simple modification, equivalent change, and modification of the above embodiments made in accordance with the technical essence of the present invention without departing from the technical solution of the present invention shall still fall within the scope of protection of the technical solution of the present invention.

Claims

1. A method for predicting the production performance of young rabbits based on big data processing, characterized in that: The following steps are involved: a. First, select a certain number of newborn rabbits and record the birth weight of each selected rabbit; b. Then, the daily activity distance of each rabbit is obtained by attaching a motion tracking device to the rabbit's leg; c. Obtain data on daily weight gain, feed intake, frequency of feed intake, frequency of water intake, amount of water consumed, and frequency of defecation of the young rabbits on the day before slaughter; d. Obtain the growth performance of baby rabbits before slaughter; e. Preprocess the acquired data to suit model training; f. Divide the preprocessed data into a training set and a validation set in a ratio of 7:3; g. Initialize the LSTM model parameters, input the training set, and continuously adjust the model parameters until the expected accuracy is achieved, thus building an LSTM-based rabbit production performance prediction model; h. Use the validation set to verify the prediction performance of the LSTM-based rabbit production performance prediction model to achieve accurate prediction of rabbit production performance.

2. The method for predicting the production performance of young rabbits based on big data processing according to claim 1, wherein The hidden unit h of the previous moment in the LSTM model t-1 Input after Kalman filtering.

3. The method for predicting the production performance of young rabbits based on big data processing according to claim 2, wherein: In the step e, the preprocessing method is to remove outliers, fill missing values and perform normalization.

Images