Macadamia breeding variety recommendation system and method based on hybrid collaborative filtering algorithm

Abstract

The present invention provides a macadamia nut breeding variety recommendation system and method based on a hybrid collaborative filtering algorithm, and relates to the technical field of macadamia nut breeding variety recommendation. The present invention introduces a hybrid collaborative filtering algorithm, comprehensively considers breeding environment data, user tendency data and historical price data, and can not only effectively standardize and process multi-source data, but also accurately predict future prices through a time series algorithm model. The system uses the calculation of adaptation scores and user tendency scores, combined with the prediction results of historical price data, to generate a comprehensive breeding recommendation index, thereby recommending the best varieties under different breeding environments, which can improve the accuracy of breeding recommendations and user satisfaction, optimize the configuration of breeding resources, make the breeding process more scientific and efficient, and improve the economic benefits of users.

Description

Technical Field

[0001] The present invention relates to the technical field of macadamia nut breeding variety recommendation, and specifically to a macadamia nut breeding variety recommendation system and method based on a hybrid collaborative filtering algorithm. Background Art

[0002] In the traditional macadamia breeding process, breeding experts mainly rely on their own experience and limited breeding data to make decisions. Commonly used data include basic environmental conditions (such as temperature and humidity) and some simple variety performance data. However, environmental factors are multidimensional and dynamically changing, and it is difficult to comprehensively and accurately evaluate the adaptability of different varieties by relying solely on limited data. In addition, expert decisions are highly subjective, and different personal experiences may lead to inconsistent decision results, affecting breeding results. At the same time, traditional breeding methods often focus on improving yield and quality, ignoring market demand and user preferences. Changes in the consumer market have a direct impact on the economic benefits of crops. Failure to consider this factor will result in excellent varieties being obtained in production, but unpopular in the market, ultimately affecting the economic benefits of breeders. User purchasing behavior and rating data can reflect market demand and preferences, and become an important basis for variety selection, but they have not been fully utilized in traditional methods.

[0003] In summary, there is currently a lack of a systematic solution that can comprehensively consider multi-dimensional data in the market. Some existing breeding decision-making systems may process a single type of data, such as environmental data or user data, but lack the ability to integrate and comprehensively analyze multi-source data. Especially when facing complex environmental conditions and changing markets, breeders need a system that can integrate various types of data and provide scientific decision-making support to reduce decision-making uncertainty and economic risks.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The object of the present invention is to provide a macadamia nut breeding variety recommendation system and method based on a hybrid collaborative filtering algorithm to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm includes:

[0008] An environmental data collection module, which is electrically connected to the data processing module and is used to collect breeding data of macadamia nuts and send the collected breeding data to the data processing module;

[0009] A user data collection module, the user data collection module is in communication with the data processing module, and is used to collect user's tendency data on purchasing macadamia nuts, and send the collected tendency data to the data processing module;

[0010] A data processing module, the data processing module is electrically connected to the comprehensive processing module, and is used to perform standardization processing on the breeding data and the tendency data, and to construct an adaptability function, and to calculate the adaptability scores of different macadamia nuts according to the standardized breeding data and the tendency data, and then send the calculated scores to the comprehensive processing module;

[0011] A data analysis module, which is in communication with the comprehensive processing module and is used to collect historical price data of different macadamia nuts, then construct a time series algorithm model to analyze the historical price data, generate predicted price data and send it to the comprehensive processing module;

[0012] A comprehensive processing module is used to generate a breeding recommendation index based on the predicted price data and adaptation scores of different macadamia nuts, select recommended varieties according to the size of the breeding recommendation index, and generate a recommendation table.

[0013] Preferably, the breeding data include breeding temperature, breeding humidity, breeding density, ambient temperature, ambient humidity, and number of insect pests; and the tendency data include number of purchases and user ratings.

[0014] Preferably, the method for standardizing the breeding data and the tendency data is:

[0015] At breeding time T y Different varieties of macadamia nuts were sampled N times, and the breeding temperature, breeding humidity, breeding density, environmental temperature, environmental humidity, and number of pests in the breeding data of each variety of macadamia nuts were calibrated as Subscript i represents the sampling number, i = 1, 2, 3, ..., N, and superscript p represents the variety number of macadamia nuts, p = 1, 2, 3, ... P;

[0016] At observation time T g Sample m groups of users, and mark the purchase times and user ratings in the tendency data of each group of users as The superscript q represents the user number, q = 1, 2, 3, ..., M;

[0017] A standardized model is constructed, and the breeding temperature, breeding humidity, ambient temperature, ambient humidity in the breeding data, and the number of purchases and user scores in the tendency data are respectively used as the input data of the standardized model for calculation, and the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, number of purchases, and user scores are generated as the corresponding output data. The calculation method of the standardized model is:

[0018]

[0019] Where X, Represent the input data and output data of the standardized model, X max , X min Represent the maximum and minimum values ​​of the standardized model input data, respectively.

[0020] Preferably, the method for calculating the adaptation scores of different macadamia nuts is:

[0021] The average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, purchase times, and user ratings are calculated as follows:

[0022]

[0023]

[0024] In the formula They represent the average values ​​of breeding temperature, breeding humidity, environmental temperature, environmental humidity, purchase times, and environmental score after standardized processing. They represent the breeding temperature, breeding humidity, environmental temperature, environmental humidity, purchase times, and environmental score after standardized treatment respectively;

[0025] The breeding cost score is calculated based on the average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, breeding density and pest frequency. The calculation method is:

[0026]

[0027] Then, the user propensity score θ is calculated based on the average of the number of purchases and user ratings after standardization. p , the calculation method is:

[0028]

[0029] According to the breeding cost score and user propensity score θ p Calculate the adaptation score ω together p , the calculation method is:

[0030]

[0031] Where w 1 、w 2 Represent the first weight and the second weight respectively, 0≤w 1 ≤w 2 ≤1,μ 1 , μ 2 Represent the first adjustment coefficient and the second adjustment coefficient respectively, 0≤μ 1 ≤μ 2 ≤1.

[0032] Preferably, the historical price data of different macadamia nuts are calibrated as Subscript j represents the time series number, j = 1, 2, 3, …, J. After the historical price data of different macadamia nuts are converged, they are divided into training set and verification set according to the time period. The training set is used to train the time series algorithm model, and the training set is used to verify the model error. When the model error is less than the preset error threshold, the model training is considered to be completed.

[0033] Preferably, the method for performing convergence processing on historical price data is:

[0034] Calculate the mean and standard deviation of historical price data using the following calculation methods:

[0035]

[0036]

[0037] In the formula Represents historical price data after convergence processing.

[0038] Preferably, the method for generating a breeding recommendation index based on the predicted price data and adaptation scores of different macadamia nuts is:

[0039] According to user needs, set the forecast time period T yc , using the time series algorithm model to generate the predicted price data at each time point in the prediction period The time and predicted price data are used as the horizontal and vertical axes to form a curve, and then the time period T yc Divide into K subintervals and label them as T k , T k =[t k-1 ,t k ], where the subscript k represents the number of the subinterval, t k-1 ,t k Indicates the starting time point and the ending time point of the sub-interval;

[0040] Calculate the breeding recommendation index corresponding to each sub-interval The calculation method is:

[0041]

[0042] The macadamia nut variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table.

[0043] The macadamia nut breeding variety recommendation method based on the hybrid collaborative filtering algorithm is applicable to the above recommendation system. The specific steps include:

[0044] S1: Collect breeding data of different varieties of macadamia nuts during the breeding process, user purchasing tendency data of macadamia nuts, and historical price data of different varieties of macadamia nuts;

[0045] S2: Generate adaptation scores for different varieties of macadamia nuts based on breeding data and tendency data, build a time series algorithm model, use historical price data to train and optimize the time series algorithm model, and use the trained time series algorithm model to generate predicted price data;

[0046] S3: setting a prediction time period according to user needs, dividing the prediction time period into several sub-intervals, and generating a breeding recommendation index based on the adaptation scores and predicted price data of different varieties of macadamia nuts in each sub-interval;

[0047] S4: The macadamia nut variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] The present invention introduces a hybrid collaborative filtering algorithm, comprehensively considers breeding environment data, user propensity data and historical price data, and can not only effectively standardize and process multi-source data, but also accurately predict future prices through a time series algorithm model. The system uses the calculation of adaptation scores and user propensity scores, combined with the prediction results of historical price data, to generate a comprehensive breeding recommendation index, thereby recommending the best varieties under different breeding environments, which can improve the accuracy of breeding recommendations and user satisfaction, optimize the configuration of breeding resources, make the breeding process more scientific and efficient, and improve the economic benefits of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the module structure of the present invention;

[0051] Figure 2 It is a schematic diagram of the overall method flow of the present invention. Specific implementation methods

[0052] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0053] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0054] Example:

[0055] See also Figure 1 , the present invention provides a technical solution:

[0056] The macadamia nut breeding variety recommendation system based on the hybrid collaborative filtering algorithm includes: an environmental data collection module, a user data collection module, a data processing module, a data analysis module, and a comprehensive processing module.

[0057] The environmental data acquisition module can use the temperature and humidity sensor of the Arduino MKR ENV Shield series, and match the data transmitter of the Arduino MKR1000 WiFi series, and can receive and send data through WIFI. The environmental data acquisition module is electrically connected to the data processing module, and is used to collect the breeding data of macadamia nuts, and send the collected breeding data to the data processing module, the breeding data includes breeding temperature, breeding humidity, breeding density, ambient temperature, ambient humidity, and the number of pests, wherein the breeding temperature and breeding humidity refer to the temperature and humidity of the culture environment in the process of macadamia nut cultivation, the ambient temperature and ambient humidity represent the overall temperature and humidity of the region, the ambient temperature and ambient humidity can be monitored by sensors, and can also be queried through an external network, the number of pests can be recorded manually by the user, and can also be monitored by using an image recognition device of the Raspberry Pi Camera Module v2 series.

[0058] The user data collection module is a web-based user interface developed using the AngularJS framework. It is used to collect user purchase times and ratings, and is connected to the data processing module for communication. It is used to collect user tendency data on purchasing macadamia nuts, and send the collected tendency data to the data processing module. The tendency data includes the number of purchases and user ratings. At the same time, the MongoDB database is used to store user purchase records and rating data.

[0059] The data processing module is electrically connected to the comprehensive processing module, and uses an Amazon Web Services (AWS) EC2t2.large series computing platform to run Python and Pandas libraries to perform data standardization processing and construct an adaptability function, which is used to standardize the breeding data and the tendency data, and construct an adaptability function, and calculate the adaptability scores of different macadamia nuts based on the standardized breeding data and the tendency data, and then send them to the comprehensive processing module;

[0060] The data analysis module is connected to the comprehensive processing module by communication, and uses the Statsmodels library in the Python environment to construct a time series algorithm model to collect historical price data of different macadamia nuts, and then construct a time series algorithm model to analyze the historical price data, generate predicted price data and send it to the comprehensive processing module;

[0061] The comprehensive processing module uses a server-side application developed in Python based on the Flask framework and is hosted on a Microsoft Azure VM (virtual machine) Standard B2s model. It is used to generate a breeding recommendation index based on the predicted price data and adaptation scores of different macadamia nuts, select recommended varieties based on the size of the breeding recommendation index, and generate a recommendation table.

[0062] The method for standardizing breeding data and tendency data is:

[0063] At breeding time T y Different varieties of macadamia nuts were sampled N times, and the breeding temperature, breeding humidity, breeding density, environmental temperature, environmental humidity, and number of pests in the breeding data of each variety of macadamia nuts were calibrated as Subscript i represents the sampling number, i = 1, 2, 3, ..., N, and superscript p represents the variety number of macadamia nuts, p = 1, 2, 3, ... P;

[0064] At observation time T g M groups of users are sampled, and the purchase times and user ratings in the tendency data of each group of users are marked as The superscript q represents the user number, q = 1, 2, 3, ..., M;

[0065] A standardized model is constructed, and the breeding temperature, breeding humidity, ambient temperature, ambient humidity in the breeding data, and the number of purchases and user scores in the tendency data are respectively used as the input data of the standardized model for calculation, and the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, number of purchases, and user scores are generated as the corresponding output data. The calculation method of the standardized model is:

[0066]

[0067] Where X, Represent the input data and output data of the standardized model, X max , X min Represent the maximum and minimum values ​​of the standardized model input data, respectively.

[0068] In this step, the purpose of standardizing breeding data and tendency data is to eliminate dimensional differences. Through standardization, all data can be normalized to the same scale, so that reasonable comparison and analysis can be performed between data. The standardized data has a similar scale and range, which can simplify the complexity of subsequent data processing and analysis, facilitate the application of algorithms and models, and accelerate convergence. At the same time, standardization allows breeding data and user tendency data to be integrated and analyzed under the same framework, which is convenient for building a comprehensive recommendation model.

[0069] The method for calculating the adaptation scores for different macadamia nuts is:

[0070] The average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, purchase times, and user ratings are calculated as follows:

[0071]

[0072] In the formula They represent the average values ​​of breeding temperature, breeding humidity, environmental temperature, environmental humidity, purchase times, and environmental score after standardized processing. They respectively represent the breeding temperature, breeding humidity, ambient temperature, ambient humidity, number of purchases and environmental score after standardized processing. Calculating the average value by using the standardized data can effectively reduce the impact of data noise and outliers and ensure the reliability and stability of the calculation results.

[0073] The breeding cost score is calculated based on the average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, breeding density and pest frequency. The calculation method is:

[0074]

[0075] It can be seen from the formula that the greater the difference between the breeding temperature and breeding humidity in the macadamia breeding process and the local ambient temperature and ambient humidity, the more means are needed, such as covering, greenhouses, etc. to adjust the process, which means that the higher the cost required in the breeding process, the greater the corresponding cost score. Conversely, the greater the planting density, the lower the cost required and the corresponding cost score.

[0076] Then, the user propensity score θ is calculated based on the average of the number of purchases and user ratings after standardization. p , the calculation method is:

[0077]

[0078] According to the breeding cost score and user propensity score θ p Calculate the adaptation score ω together p , the calculation method is:

[0079]

[0080] Where w 1 、w 2 Represent the first weight and the second weight respectively, 0≤w 1 ≤w 2 ≤1,μ 1 , μ 2 Represent the first adjustment coefficient and the second adjustment coefficient respectively, 0≤μ 1 ≤μ 2 ≤1, the second weight and the second adjustment coefficient are larger here, because when breeding macadamia nuts, the selection of varieties needs to consider the user market first, and then the cultivation cost. The two are causally related. The fluctuation of the user market has a greater impact on the selection of macadamia nut varieties. The user propensity score is used to reflect the user market, so a larger weight and adjustment coefficient are required.

[0081] In this step, breeding data and user preference data are taken into consideration at the same time, and multi-dimensional data such as environmental factors, breeding costs, user purchasing behavior and ratings are integrated together to evaluate the adaptability of macadamia varieties more comprehensively and comprehensively. The number of user purchases and ratings are also taken into consideration, and the recommendation results are more likely to meet user preferences, thereby improving user satisfaction and loyalty. By comprehensively considering breeding costs and user needs, it helps breeders better control and manage breeding costs, select the optimal breeding environment and variety combination, and maximize economic benefits.

[0082] The historical price data of different macadamia nuts are calibrated as Subscript j represents the time series number, j = 1, 2, 3, …, J. After the historical price data of different macadamia nuts are converged, they are divided into training set and verification set according to the time period. The training set is used to train the time series algorithm model, and the training set is used to verify the model error. When the model error is less than the preset error threshold, the model training is considered to be completed.

[0083] The method for convergence processing of historical price data is:

[0084] Calculate the mean and standard deviation of historical price data using the following calculation methods:

[0085]

[0086] In the formula Represents the historical price data after convergence processing. By calculating the mean and standard deviation of the historical price data and performing convergence processing, it can effectively reduce the impact of data noise and outliers, improve the quality and reliability of the data, and also improve the operation speed.

[0087] The method to generate breeding recommendation index based on predicted price data and adaptation scores for different macadamia nuts is:

[0088] According to user needs, set the forecast time period T yc , using the time series algorithm model to generate the predicted price data at each time point in the prediction period The time and predicted price data are used as the horizontal and vertical axes to form a curve, and then the time period T yc Divide into K subintervals and label them as T k , T k =[t k-1 ,t k ], where the subscript k represents the number of the subinterval, t k-1 ,t k Indicates the start and end time points of the subinterval.

[0089] Calculate the breeding recommendation index corresponding to each sub-interval The calculation method is:

[0090]

[0091] The macadamia nut variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table.

[0092] Assuming that a prediction is needed for the next two years, the sub-intervals can be divided according to the maturity cycle of macadamia nuts (generally 5 to 7 months, and 6 months in this embodiment), and 4 corresponding sub-intervals can be obtained, thereby obtaining the macadamia nut variety with the highest breeding recommendation index in each maturity cycle, which is convenient for users to select varieties and obtain higher economic benefits.

[0093] In this step, the future price is predicted through the time series algorithm model, which can capture the dynamic changes of market prices and provide a scientific basis for breeding recommendations; the prediction time period is divided into multiple sub-intervals, which can more carefully analyze the market performance of macadamia varieties in different time periods and improve the pertinence and accuracy of recommendations; by calculating the breeding recommendation index, the predicted price and adaptation score are comprehensively considered to make the recommendation results more comprehensive and reasonable; according to market changes and user needs, the division of sub-intervals and recommendation strategies can be flexibly adjusted to improve the adaptability and flexibility of recommendations, which can effectively reduce resource waste, optimize the breeding process, and improve resource utilization efficiency.

[0094] See also Figure 2 , the macadamia nut breeding variety recommendation method based on hybrid collaborative filtering algorithm is applicable to the above recommendation system, and the specific steps include:

[0095] S1: Collect breeding data of different varieties of macadamia nuts during the breeding process, user purchasing tendency data of macadamia nuts, and historical price data of different varieties of macadamia nuts;

[0096] S2: Generate adaptation scores for different varieties of macadamia nuts based on breeding data and tendency data, build a time series algorithm model, use historical price data to train and optimize the time series algorithm model, and use the trained time series algorithm model to generate predicted price data;

[0097] S3: setting a prediction time period according to user needs, dividing the prediction time period into several sub-intervals, and generating a breeding recommendation index based on the adaptation scores and predicted price data of different varieties of macadamia nuts in each sub-interval;

[0098] S4: The macadamia variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table

[0099] In summary, the present invention introduces a hybrid collaborative filtering algorithm, comprehensively considers breeding environment data, user propensity data and historical price data, and can not only effectively standardize and process multi-source data, but also accurately predict future prices through a time series algorithm model. The system uses the calculation of adaptation scores and user propensity scores, combined with the prediction results of historical price data, to generate a comprehensive breeding recommendation index, thereby recommending the best varieties under different breeding environments, which can improve the accuracy of breeding recommendations and user satisfaction, optimize the configuration of breeding resources, make the breeding process more scientific and efficient, and improve the economic benefits of users.

[0100] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0101] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0102] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0103] The above is only a specific implementation method of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be covered by the protection scope of the present application.

Claims

1. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm is characterized by: include: An environmental data collection module, which is electrically connected to the data processing module and is used to collect breeding data of macadamia nuts and send the collected breeding data to the data processing module; A user data collection module, the user data collection module is in communication with the data processing module, and is used to collect user's tendency data on purchasing macadamia nuts, and send the collected tendency data to the data processing module; A data processing module, the data processing module is electrically connected to the comprehensive processing module, and is used to perform standardization processing on the breeding data and the tendency data, and to construct an adaptability function, and to calculate the adaptability scores of different macadamia nuts according to the standardized breeding data and the tendency data, and then send the calculated scores to the comprehensive processing module; A data analysis module, which is in communication with the comprehensive processing module and is used to collect historical price data of different macadamia nuts, then construct a time series algorithm model to analyze the historical price data, generate predicted price data and send it to the comprehensive processing module; A comprehensive processing module is used to generate a breeding recommendation index based on the predicted price data and adaptation scores of different macadamia nuts, select recommended varieties according to the size of the breeding recommendation index, and generate a recommendation table.

2. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 1, characterized in that: The breeding data include breeding temperature, breeding humidity, breeding density, ambient temperature, ambient humidity, and number of insect infestations; the tendency data include number of purchases and user ratings.

3. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 2, characterized in that: The method for standardizing breeding data and tendency data is: At breeding time T y Different varieties of macadamia nuts were sampled N times, and the breeding temperature, breeding humidity, breeding density, environmental temperature, environmental humidity, and number of pests in the breeding data of each variety of macadamia nuts were calibrated as ρ y , c y , subscript i represents the sampling number, i=1,2,3,…,N, superscript p represents the variety number of macadamia nuts, p=1,2,3,…P; At observation time T g M groups of users are sampled, and the purchase times and user ratings in the tendency data of each group of users are marked as The superscript q represents the user number, q = 1, 2, 3, ..., M; A standardized model is constructed, and the breeding temperature, breeding humidity, ambient temperature, ambient humidity in the breeding data, and the number of purchases and user scores in the tendency data are respectively used as the input data of the standardized model for calculation, and the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, number of purchases, and user scores are generated as the corresponding output data. The calculation method of the standardized model is: Where X, Represent the input data and output data of the standardized model, X max , X min Represent the maximum and minimum values ​​of the standardized model input data, respectively.

4. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 3, characterized in that: The method for calculating the adaptation scores for different macadamia nuts is: The average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, purchase times, and user ratings are calculated as follows: In the formula They represent the average values ​​of breeding temperature, breeding humidity, environmental temperature, environmental humidity, purchase times, and environmental score after standardized processing. They represent the breeding temperature, breeding humidity, environmental temperature, environmental humidity, purchase times, and environmental score after standardized treatment respectively; The breeding cost score is calculated based on the average values ​​of the standardized breeding temperature, breeding humidity, ambient temperature, ambient humidity, breeding density and pest frequency. The calculation method is: Then, the user propensity score θ is calculated based on the average of the number of purchases and user ratings after standardization. p , the calculation method is: According to the breeding cost score and user propensity score θ p Calculate the adaptation score ω together p , the calculation method is: Wherein w1 and w2 represent the first weight and the second weight respectively, 0≤w1≤w2≤1, μ1 and μ2 represent the first adjustment coefficient and the second adjustment coefficient respectively, 0≤μ1≤μ2≤1.

5. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 4, characterized in that: The historical price data of different macadamia nuts are calibrated as Subscript j represents the time series number, j = 1, 2, 3, …, J. After the historical price data of different macadamia nuts are converged, they are divided into training set and verification set according to the time period. The training set is used to train the time series algorithm model, and the training set is used to verify the model error. When the model error is less than the preset error threshold, the model training is considered to be completed.

6. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 5, characterized in that: The method for convergence processing of historical price data is: Calculate the mean and standard deviation of historical price data using the following calculation methods: In the formula Represents historical price data after convergence processing.

7. The macadamia nut breeding variety recommendation system based on hybrid collaborative filtering algorithm according to claim 6, characterized in that: The method to generate breeding recommendation index based on predicted price data and adaptation scores for different macadamia nuts is: According to user needs, set the forecast time period T yc , using the time series algorithm model to generate the predicted price data at each time point in the prediction period The time and predicted price data are used as the horizontal and vertical axes to form a curve, and then the time period T yc Divide into K subintervals and label them as T k , T k =[t k-1 ,t k ], where the subscript k represents the number of the subinterval, t k-1 ,t k Indicates the starting time point and the ending time point of the sub-interval; Calculate the breeding recommendation index corresponding to each sub-interval The calculation method is: The macadamia variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table.

8. A method for recommending macadamia nut breeding varieties based on a hybrid collaborative filtering algorithm, characterized in that: The recommendation method is applicable to the recommendation system according to any one of claims 1 to 7, and the specific steps include: S1: Collect breeding data of different varieties of macadamia nuts during the breeding process, user purchasing tendency data of macadamia nuts, and historical price data of different varieties of macadamia nuts; S2: Generate adaptation scores for different varieties of macadamia nuts based on breeding data and tendency data, build a time series algorithm model, use historical price data to train and optimize the time series algorithm model, and use the trained time series algorithm model to generate predicted price data; S3: setting a prediction time period according to user needs, dividing the prediction time period into several sub-intervals, and generating a breeding recommendation index based on the adaptation scores and predicted price data of different varieties of macadamia nuts in each sub-interval; S4: The macadamia nut variety with the largest breeding recommendation index in each sub-interval is taken as the recommended variety, and the recommended varieties are sorted according to the sub-interval number to obtain the final recommendation table.

Images