Fishing field fish returning management method and system based on big data analysis

Abstract

The invention provides a fishing field fish returning management method and system based on big data analysis, and the method comprises the following steps: a user selects a fishing field and a fishing target fingerling, obtains a fishing brand, and inputs the initial user fishing information; acquiring historical fishing information of a fishing field user, and performing data preprocessing to obtain a fishing feature data set of the corresponding fishing field; constructing a recommendation model based on a machine learning network, respectively inputting the fishing feature data sets of the corresponding fishing courses into the recommendation model for iterative training, and outputting recommended fish species and recommended fishing position information of the current time of the corresponding fishing courses by the recommendation model; the user selects a fishing position in the recommended fishing position information for fishing, the fish is weighed according to the fingerling, the unit price of the fingerling is obtained, the total price of the returned fish is calculated, and a payment interface is called to perform remote settlement on a user account; according to the method, by collecting the fishing information of the user, analyzing the historical data, constructing the recommendation model and realizing the automatic settlement process, the fishing field management efficiency and the fishing experience of the user are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and in particular, to a fishing ground fish return management method and system based on big data analysis. Background Art

[0002] Fishing ground fish return management refers to the scientific regulation and sustainable utilization of fish resources in fishing places, aiming to balance fishery ecology, economic benefits and user experience. With the development of recreational fishery, fishing grounds often face problems such as decreasing fish numbers and species imbalance, resulting in a decline in the fishing experience and an increase in ecological risks. Fish return management ensures the ecological stability of the fishing ground by regularly releasing fry, controlling the fishing volume, and monitoring water quality and fish growth conditions.

[0003] A method and system for weighing live fish with the publication number CN114812767A. The method includes obtaining the beating frequency of the live fish through a weighing device, determining the weighing time according to the beating frequency, weighing the live fish using the weighing time to obtain the standard weight of the fish; taking the weight of the live fish falling on the weighing device as the initial weight of the fish, storing the initial weight, collecting multiple weight data of the live fish on the weighing device as the temporary weight of the fish, comparing the temporary weight with the initial weight, eliminating the temporary weight less than the initial weight, and performing an average operation on the remaining temporary weights to obtain the effective weight of the fish; and determining whether the effective weight is equal to the standard weight.

[0004] In traditional fishing ground management, due to the lack of effective means for collecting and analyzing user fishing information, it is difficult for fishing grounds to provide accurate fish species and fishing position recommendations for users based on real-time environmental parameters and fish conditions, resulting in low fishing success rates and poor experiences for users; at the same time, the traditional settlement process is cumbersome and requires manual operation, which is not only inefficient but also prone to errors and disputes, thus reducing the fishing ground management efficiency and user fishing experience. Summary of the Invention

[0005] In view of this, the present invention proposes a fishing ground fish return management method and system based on big data analysis. By collecting user fishing information, analyzing historical data, constructing a recommendation model, and implementing an automated settlement process, it can not only recommend the optimal fish species and fishing positions for users based on real-time environmental parameters and fish conditions prediction, but also automatically perform remote settlement, simplify the transaction process, and significantly improve the fishing ground management efficiency and user fishing experience.

[0006] In the first aspect, the present invention provides a fishing ground fish return management method based on big data analysis, including the following steps:

[0007] S1, the user selects a fishing ground and the target fish species for fishing, obtains a fishing license and enters the initial user fishing information;

[0008] S2. Obtain the historical fishing information of users in the fishing ground, and perform data preprocessing to obtain the fishing feature dataset corresponding to the fishing ground;

[0009] S3. Build a recommendation model based on the machine learning network, and input the fishing feature datasets of each corresponding fishing ground into the recommendation model for iterative training. The recommendation model outputs the recommended fish species and recommended fishing position information at the current time of the corresponding fishing ground;

[0010] S4. The user selects a fishing position in the recommended fishing position information for fishing, weighs the fish caught according to the fish species, obtains the unit price of the fish species to calculate the total price of the fish returned, and calls the payment interface to remotely settle the user's account.

[0011] On the basis of the above technical solutions, preferably, in step S1, the user selects a fishing ground and the target fishing fish species, obtains a fishing card and enters the initial user fishing information, including the user obtaining the fishing card of the fishing ground and entering the initial user fishing information through the mobile terminal. The user fishing information includes personal information, fishing ground information, target fish species, fishing time, environmental information and fishing position information. The personal information includes name, age, contact information, fishing age, account and preferred fish species; the fishing ground information includes fishing ground name, geographical location, water quality type and development time; the fishing time includes season, time period and holiday marking status, and the environmental information includes water temperature, air pressure, temperature and humidity, weather condition, pH value and dissolved oxygen; the fishing position information includes fishing price, fishing position number, the caught weight of each fish species and fishing position status.

[0012] On the basis of the above technical solutions, preferably, in step S2, obtain the historical fishing information of users in the fishing ground, and perform data preprocessing to obtain the corresponding standard training dataset of fishing ground users, including the following sub-steps:

[0013] S21. Obtain the historical fishing information of all users of the corresponding fishing ground from the database according to the fishing ground name;

[0014] S22. Perform data cleaning, conversion and merging processing on the historical fishing information of all users of the corresponding fishing ground to obtain standard user fishing data;

[0015] S23. Perform feature extraction and calibration on the fishing age, preferred fish species, season, time period, holiday marking status, water temperature, air pressure, temperature and humidity, weather condition, pH value, dissolved oxygen, fishing price, fishing position number, fish species type and the caught weight of each fish species information in the standard user fishing data to obtain standard feature data;

[0016] S24. Add the standard feature data to the dataset to obtain the fishing feature dataset corresponding to the fishing ground.

[0017] Based on the above technical solutions, preferably, in S23, the fish species type, fish catch weight, and fishing position utility of the standard user's fishing data are calibrated. Among them, the fishing position utility calibration includes the following sub-steps:

[0018] Preset the time window length, obtain the caught weight of each fish species at each fishing position every day within the past time window, and form the catch sequence of each fish species;

[0019] Sort the total fish catch within the time window in ascending order to obtain the ascending catch sequence W s,f ={w1, w2,..., w T};

[0020] Preset the percentile, and calculate the percentile position index of the catch sequence according to the catch sequence of each fish species and the preset percentile. The expression is:

[0021] S p =(p / 100)×(T + 1)

[0022] In the formula, S p is the percentile position index, p is the preset percentile, and T is the number of days in the time window;

[0023] Judge whether S p is an integer. If S p is an integer, then take the S p -th value in the ascending catch sequence of the corresponding fish species to obtain the fishing position catch calibration value;

[0024] P p =W s,f [S p -1]

[0025] In the formula, P p is the fishing position catch calibration value, and W s,f [] is the ascending catch sequence of fish species f at fishing position s;

[0026] If S p is a non-integer, then use the linear interpolation method to calculate the fishing position catch calibration value. The expression is:

[0027]

[0028] In the formula, is the floor function, is the ceiling function;

[0029] According to the preset percentile, the fishing position catch calibration value, and the ascending catch sequence of the corresponding fish species, calculate the utility value corresponding to the fishing position. The expression is:

[0030] U s,f =(W s,f -P(100-p) (W s,f )) / (P p (W s,f )-P (100-p) (W s,f ))

[0031] where U s,f is the utility value of fish species f at fishing position s, P (100-p) is the calibrated value of fish catch at the fishing position corresponding to the 100 - p percentile. If P p =P (100-p) , then U s,t is 0;

[0032] According to the preset recommended classification level based on the utility value, if the utility value is in the interval [0, 0.4), it is marked as a low - efficiency fishing position; if the utility value is in the interval [0.4, 0.7), it is marked as an average fishing position; if the utility value is in the interval [0.7, 1], it is marked as an excellent fishing position, and the fishing position number and fish species type are associated.

[0033] On the basis of the above technical solutions, preferably, in step S3, the recommendation model is constructed based on a machine - learning network. The fishing - characteristic data sets of each corresponding fishing ground are respectively input into the recommendation model for iterative training. The recommendation model outputs the recommended fish species and recommended fishing - position information of the current time of the corresponding fishing ground, including the following sub - steps:

[0034] S31, construct a recommendation model based on a machine - learning network. The recommendation model includes an input layer, a shared layer, a fish - species branch module, and a fishing - position branch module. The output end of the input layer is connected to the input end of the shared layer. The fishing - characteristic data sets of each corresponding fishing ground are respectively input into the recommendation model for iterative training. The output end of the shared layer is respectively connected to the input ends of the fish - species branch module and the fishing - position branch module, which are used to extract features from the fishing - characteristic data sets of each corresponding fishing ground. The fish - species branch module outputs the fish - catching rates of each fish species, and the fishing - position branch module outputs the utility values of each fish species at the fishing positions;

[0035] S32, obtain the top three fish species with the highest fish - catching rates as the recommended fish species. According to the recommended fish species, obtain the utility values of the fish species corresponding to the fishing positions, and select the fishing positions where the utility values of the fish species are in the excellent - fishing - position interval as the recommended fishing positions;

[0036] S33, obtain the historical total fish catch, the number of fishing times, and the fishing price of the recommended fishing positions for comprehensive calculation to obtain the evaluation value Score s , and the expression is:

[0037] Score s =α×T s / max(T)+β×(1 - C s / max(C))+γ×((1-(Ps -min(P)) /

[0038] (max(P)-min(P)))

[0039] where T s is the total historical catch of fishing spot s, C s is the number of fishing times at fishing spot s, P s is the current fishing price of fishing spot s, T is the total historical catch of the fishing ground, C is the maximum number of times of the fishing spot, min(P) is the lowest fishing price of the fishing ground, max(P) is the highest fishing price of the fishing ground, α is the weight coefficient of the catch, β is the weight coefficient of the number of fishing times, and γ is the weight coefficient of the fishing price;

[0040] S34. Sort the evaluation values of each recommended fishing spot in descending order and associate them with the fishing spot number and the recommended fish species to obtain the recommended fishing list information for the user to view.

[0041] Based on the above technical solution, preferably, after the step S34, it further includes: judging whether the recommended fish species at the current time of the fishing ground is consistent with the fishing target fish species selected by the user. If they are consistent, obtain the idle fishing spots in the recommended fishing list information according to the fishing spot status information, and select the fishing spot with the highest evaluation value among the idle fishing spots as the recommended fishing spot to be assigned to the user; if they are inconsistent, judge whether the user changes the fishing target fish species to the recommended fish species. If it is changed, obtain the fishing spot with the highest evaluation value among the idle fishing spots as the recommended fishing spot to be assigned to the user; otherwise, obtain the utility values corresponding to the fishing target fish species of each fishing spot from the time window to the current time, and select the fishing spots whose fish species utility values are in the excellent fishing spot interval as the candidate fishing spots; obtain the idle fishing spots in the candidate fishing spots according to the fishing spot status information and calculate the evaluation values of the idle fishing spots in each candidate fishing spot, and select the fishing spot with the highest evaluation value as the recommended fishing spot to be assigned to the user.

[0042] Based on the above technical solution, preferably, the sharing layer includes a first feature extraction unit, a second feature extraction unit, and a third feature extraction unit. The first feature extraction unit, the second feature extraction unit, and the third feature extraction unit are connected in sequence, and the third feature extraction unit outputs shared feature data, where

[0043] The expression of the first feature extraction unit is:

[0044] h1 = σ(W1X + b1)

[0045] where X is the standard feature data corresponding to each fishing ground, σ is the Sigmoid function, W1 is the weight matrix, with a dimension of R 256×64 , and b1 is the bias vector;

[0046] The expression of the second feature extraction unit is:

[0047] h2 = LayerNorm(W2h1 + b2)

[0048] Where h1 is the output feature of the first feature extraction unit, W2 is the weight matrix with dimension R 128×256 , b2 is the bias vector, and LayerNorm() is the layer normalization function;

[0049] The expression of the third feature extraction unit is:

[0050] h3 = Dropout(σ(W3h2 + b3), c = 0.3)

[0051] Where h2 is the output feature of the second feature extraction unit, W3 is the weight matrix with dimension R 64×128 , b3 is the bias vector, Dropout() is the regularization function, and c is the neuron dropout probability.

[0052] Based on the above technical solutions, preferably, the fish species branch module includes a first fully connected layer, a second fully connected layer, and a first output layer. Among them, the input of the first fully connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional feature to 32 dimensions. The expression is:

[0053] z1 = W4h3 + b4

[0054] Where W4 is the weight matrix with dimension R 32×64 , b4 is the bias vector with dimension R 32 ;

[0055] The output of the first fully connected layer is connected to the input of the second fully connected layer, mapping the 32-dimensional feature to the fish species number m dimensions. The expression is:

[0056] z2 = RELU(z1)W5 + b5

[0057] Where RELU() is the ReLU function, W5 is the weight matrix with dimension R m×32 , b5 is the bias vector with dimension R m ;

[0058] The output of the second fully connected layer is connected to the first output layer, and the first output layer is used to output the probabilities of each fish species. The expression is:

[0059]

[0060] Where p i is the predicted probability of the i-th fish species, and the sum of the predicted probabilities of all fish species is 1.

[0061] Based on the above technical solutions, preferably, the fishing position branch module includes a third fully connected layer, a fourth fully connected layer, and a second output layer, where,

[0062] The input of the third fully connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional feature to 16 dimensions. The expression is:

[0063] u1 = W6h3 + b6

[0064] In the formula, W6 is the weight matrix with a dimension of R 16×64 , and b6 is the bias vector with a dimension of R 16 ;

[0065] The output of the third fully connected layer is connected to the input of the fourth fully connected layer, mapping the 16-dimensional feature to a 1-dimensional scalar. The expression is:

[0066] y = σ(u1)W7 + b7

[0067] In the formula, σ is the Sigmoid function, W7 is the weight matrix with a dimension of R 1×16 , and b7 is the bias vector with a dimension of R;

[0068] The output of the fourth fully connected layer is connected to the input of the second output layer, and the second output layer outputs the fishing position utility value, where y ∈ [0, 1].

[0069] In a second aspect, the present invention also provides a fishing ground fish return management system as described above, which is implemented by using the fishing ground fish return management method based on big data analysis. The system includes:

[0070] An information entry module for the user to select a fishing ground and the target fish species for fishing, obtain a fishing card, and enter the initial user fishing information;

[0071] A data processing module for obtaining the historical fishing information of the fishing ground users and performing data preprocessing to obtain the fishing feature data set corresponding to the fishing ground;

[0072] A recommendation module for constructing a recommendation model based on a machine learning network, inputting the fishing feature data sets corresponding to each fishing ground into the recommendation model for iterative training, and the recommendation model outputs the recommended fish species and recommended fishing position information for the current time of the corresponding fishing ground;

[0073] A settlement module for the user to select a fishing position in the recommended fishing position information for fishing, weigh the fish catch according to the fish species, obtain the unit price of the fish species to calculate the total price of the fish return, and call the payment interface to remotely settle the user's account.

[0074] The fishing ground fish return management method and system of the present invention have the following beneficial effects compared with the prior art:

[0075] (1) By collecting users' fishing information, analyzing historical data, constructing a recommendation model, and implementing an automated settlement process, not only can the optimal fish species and fishing positions be recommended to users based on real-time environmental parameters and fish conditions prediction, increasing the fishing success rate; but also the automated weighing, pricing, and remote settlement simplify the transaction process, ensure the convenience of transactions, enhance user stickiness, thus significantly improving the fishing ground management efficiency and users' fishing experience;

[0076] (2) By calculating the utility value independently for each fish species, the fishing effect of each fishing position for different fish species can be accurately evaluated, avoiding the ambiguity of overall evaluation, providing more accurate fishing position recommendations for users, and calculating the utility value based on percentiles to effectively filter out data interference from extreme weather or accidental bumper catches, improving the recommendation accuracy; (3) By constructing a recommendation model through a machine learning network, complex factors such as the fishing ground environment and fish species distribution can be accurately captured, improving the accuracy of recommended fish species and fishing positions; and by comprehensively evaluating fishing positions, the fishing positions with the highest cost performance and most suitable for the current fishing conditions are selected to provide more accurate and personalized fishing recommendations for users, improving the fishing success rate and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0077] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0078] Figure 1 is a flowchart of the fishing ground fish return management method based on big data analysis of the present invention;

[0079] Figure 2 is a structural diagram of the recommendation model of the fishing ground fish return management method based on big data analysis of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0080] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention belong to the scope of protection of the present invention.

[0081] In the first aspect, as Figure 1 shown, the present invention provides a fishing ground fish return management method based on big data analysis, including the following steps:

[0082] S1. The user selects a fishing ground and the target fish species to be fished, obtains a fishing permit and enters the initial user fishing information.

[0083] In this embodiment, step S1 includes the user obtaining a fishing permit for the fishing ground and entering the initial user fishing information through the mobile terminal. The user fishing information includes personal information, fishing ground information, target fish species, fishing time, environmental information and fishing position information. The personal information includes name, age, contact information, fishing age, account and preferred fish species; the fishing ground information includes fishing ground name, geographical location, water quality type and development time; the fishing time includes season, time period and holiday mark status; the environmental information includes water temperature, air pressure, temperature and humidity, weather condition, pH value and dissolved oxygen content; the fishing position information includes fishing price, fishing position number, the caught weight of each fish species and fishing position status.

[0084] It should be noted that the user selects the target fishing ground and the fish species to be fished through the fishing ground management platform or the mobile application. After the user completes the selection of the fishing ground, the user needs to go to the fishing ground management office or obtain a fishing permit through the online reservation system. The fishing permit, as the access credential for the fishing activity, not only records the basic information and fishing permissions of the user, but also is associated with the resource allocation and management of the fishing ground; after the user obtains the fishing permit, the user needs to enter the initial user fishing information through the mobile application, and can enter it through the fishing ground APP, mini-program or third-party fishing service platform. These information comprehensively cover all aspects of the fishing activity and provide strong support for fishing ground management, user service and data analysis.

[0085] Among them, the personal information includes name, age, contact information, fishing age, account information and preferred fish species, which helps the fishing ground understand the user background and provide personalized services; the fishing ground information includes fishing ground name, geographical location, water quality type (the water quality type includes fresh water, salt water, mixed water, etc.) and development time, etc.; the accurate entry of the fishing ground information helps the fishing ground to carry out resource allocation, environmental maintenance and user guidance to ensure the smooth progress of the fishing activity; the user needs to clarify the target fish species for this fishing, which is convenient for the fishing ground to arrange fishing positions, prepare bait and provide fishing suggestions. The fishing time includes season (such as spring, summer, autumn and winter), time period (such as morning, afternoon and evening) and holiday mark status (whether it is a weekend or a holiday); the record of the fishing time helps the fishing ground analyze the user behavior pattern, optimize the fishing position allocation and activity arrangement. The environmental information details the environmental parameters such as water temperature, air pressure, temperature and humidity, weather condition (sunny, rainy and cloudy), pH value and dissolved oxygen content; these environmental information are convenient for the system to predict the fish situation and remind the user to adjust the fishing strategy. The fishing position information includes fishing price (i.e., fishing position fee), fishing position number, the caught weight of each fish species (historical data) and fishing position status (idle, reserved and under maintenance, etc.). The real-time update of the fishing position information helps the user quickly understand the situation of the fishing ground and make a reasonable choice.

[0086] In addition, obtain an electronic fishing tag. The ID corresponding to the electronic fishing tag has unique identification. When a user needs to fish, the obtained electronic fishing tag is associated with the user's information. After the fishing settlement, the system automatically unbinds the electronic fishing tag from the user's information.

[0087] It can be understood that the fishing ground can, through the collected user fishing information, carry out resource allocation, environmental maintenance and activity arrangement, improve the operation efficiency of the fishing ground, and provide strong data support for the data analysis of the fishing ground.

[0088] S2. Obtain the historical fishing information of the fishing ground users and perform data preprocessing to obtain a fishing feature data set corresponding to the fishing ground.

[0089] Among them, step S2 includes the following sub-steps:

[0090] S21. Obtain the historical fishing information of all users of the corresponding fishing ground from the database according to the fishing ground name;

[0091] S22. Perform data cleaning, conversion and merging processing on the historical fishing information of all users of the corresponding fishing ground to obtain standard user fishing data;

[0092] S23. Perform feature extraction and calibration on the fishing age, preferred fish species, season, time period, holiday marking status, water temperature, air pressure, temperature and humidity, weather conditions, pH value, dissolved oxygen, fishing price, fishing position number, fish species type and the caught weight information of each fish species in the standard user fishing data to obtain standard feature data;

[0093] S24. Add the standard feature data to the data set to obtain a fishing feature data set corresponding to the fishing ground.

[0094] It should be noted that using the fishing ground name as the query keyword, accessing the database storing user fishing information, retrieving and extracting the historical fishing records of all users of the fishing ground, providing comprehensive basic data for subsequent analysis, performing data cleaning on the historical fishing information of all users of the fishing ground, removing duplicate records, ensuring the uniqueness of each piece of data, filling in missing values using the median, and correcting incorrect data; performing standardization processing on the cleaned data, and merging relevant data from different data sources or tables, integrating data from multiple fishing activities to form a panoramic view of the user's fishing history, improving data quality, ensuring data consistency and availability, and laying a solid foundation for subsequent feature extraction.

[0095] In step S23 of this embodiment, the fish species type, fish catch weight and fishing position utility calibration of the standard user fishing data are carried out. Among them, the fishing position utility calibration includes the following sub-steps:

[0096] Preset the time window length, obtain the fishing weights of each fish species at each fishing position every day within the past time window, and form the catch sequences of each fish species;

[0097] Sort the total catch within the time window in ascending order to obtain the ascending catch sequence W s,f ={w1, w2,..., w T};

[0098] Preset the percentile, and calculate the percentile position index of the catch sequence according to the catch sequence of each fish species and the preset percentile. The expression is:

[0099] S p =(p / 100)×(T + 1)

[0100] In the formula, S p is the percentile position index, p is the preset percentile, and T is the number of days in the time window;

[0101] Judge whether S p is an integer. If S p is an integer, then take the S p th value in the ascending catch sequence of the corresponding fish species to obtain the catch calibration value of the fishing position;

[0102] P p =W s,f [S p - 1]

[0103] In the formula, P p is the catch calibration value of the fishing position, and W s,f [] is the ascending catch sequence of fish species f at fishing position s;

[0104] If S p is a non - integer, then use the linear interpolation method to calculate the catch calibration value of the fishing position. The expression is:

[0105]

[0106] In the formula, is the floor function, is the ceiling function;

[0107] According to the preset percentile, the catch calibration value of the fishing position, and the ascending catch sequence of the corresponding fish species, calculate the utility value corresponding to the fishing position. The expression is:

[0108] U s,f =(W s,f - P (100-p) (W s,f )) / (P p (W s,f ) - P (100-p) (W s,f))

[0109] In the formula, U s,f is the utility value of fish species f at fishing position s, and P (100-p) is the calibrated value of the fish catch at the fishing position corresponding to the 100 - p percentile. If P p = P (100-p) , then U s,t is 0;

[0110] Preset the recommended classification level according to the utility value. If the utility value is in the interval [0, 0.4), it is marked as a low - efficiency fishing position. If the utility value is in the interval [0.4, 0.7), it is marked as an average fishing position. If the utility value is in the interval [0.7, 1], it is marked as an excellent fishing position, and the fishing position number is associated with the fish species type.

[0111] It should be noted that the time window length is the past 7 days. Obtain the fishing weights of each fish species at each fishing position every day within this time window to form the fish catch sequence of each fish species, reflecting the fishing effect of the fishing position at different time periods. Sort the total fish catch within the time window in ascending order to obtain the ascending fish catch sequence of each fish species, which is prepared for subsequent calculation of percentiles and fishing position utility values. The preset percentile is 75. Calculate the percentile position index of the fish catch sequence based on the fish catch sequence of each fish species and the preset percentile, determine the position corresponding to the percentile in the ascending fish catch sequence, so as to obtain the calibrated fish catch value, reflecting the fishing effect of the fishing position at this percentile. Quantify the fishing effect of the fishing position for different fish species through the calculated utility value of the fishing position, form comparable utility values, and convert the fishing position utility value into an intuitive recommended level, and associate it with the fishing position number and fish species type for the convenience of users' understanding and selection.

[0112] It should be noted that by calculating the utility value independently for each fish species, the fishing effect of each fishing position for different fish species can be accurately evaluated, avoiding the ambiguity of overall evaluation, and providing more accurate fishing position recommendations for users. Calculating the utility value based on percentiles effectively filters out the data interference of extreme weather or accidental bumper catches, improving the recommendation accuracy.

[0113] In addition, the sliding time window design can quickly capture the migration law of fish schools. When the sudden rise in temperature causes the fish school to move to the deep water area, the utility value of the deep - water fishing position is automatically updated within 24 hours. Furthermore, the length of the time window can be changed in real - time in response to environmental changes, which can improve the accuracy of model recommendations.

[0114] As Figure 2 shown in S3, construct a recommendation model based on a machine - learning network. Input the fishing feature datasets of each corresponding fishing ground into the recommendation model for iterative training. The recommendation model outputs the recommended fish species and recommended fishing position information for the current time of the corresponding fishing ground.

[0115] Step S3 includes the following sub-steps:

[0116] S31. Build a recommendation model based on a machine learning network. The recommendation model includes an input layer, a shared layer, a fish species branch module, and a fishing position branch module. The output end of the input layer is connected to the input end of the shared layer. The fishing feature data sets of each corresponding fishing ground are respectively input into the recommendation model for iterative training. The output end of the shared layer is respectively connected to the input ends of the fish species branch module and the fishing position branch module, and is used to extract features from the fishing feature data sets of each corresponding fishing ground. The fish species branch module outputs the fish catch rates of each fish species, and the fishing position branch module outputs the utility values of each fish species at the fishing positions;

[0117] S32. Obtain the top three fish species with the highest fish catch rates as the recommended fish species. Obtain the utility values of the fish species corresponding to the fishing positions according to the recommended fish species, and select the fishing positions where the utility values of the fish species are in the excellent fishing position interval as the recommended fishing positions;

[0118] S33. Obtain the historical total fish catch, the number of fishing times, and the fishing price of the recommended fishing positions for comprehensive calculation to obtain the evaluation value Score of each fishing position s , and the expression is:

[0119] Score s = α×T s / max(T)+β×(1 - C s / max(C))+γ×((1-(P s -min(P)) /

[0120] (max(P)-min(P)))

[0121] In the formula, T s is the historical total fish catch of fishing position s, C s is the number of fishing times of fishing position s, P s is the current fishing price of fishing position s, T is the historical total fish catch of the fishing ground, C is the maximum number of times of the fishing position, min(P) is the lowest fishing price of the fishing ground, max(P) is the highest fishing price of the fishing ground, α is the weight coefficient of the fish catch, β is the weight coefficient of the number of fishing times, and γ is the weight coefficient of the fishing price;

[0122] S34. Associate the evaluation values of each recommended fishing position with the fishing position numbers and the recommended fish species in descending order to obtain the recommended fishing list information for the user to view.

[0123] It should be noted that the input layer receives the fishing feature datasets of each corresponding fishing ground, including information such as fishing age, preferred fish species, season, time period, holiday marking status, water temperature, air pressure, temperature and humidity, weather conditions, pH value, dissolved oxygen content, fishing price, fishing position number, fish species type, and the caught weight of each fish species; the sharing layer conducts preliminary feature extraction and shared representation on the input feature datasets to provide basic features for subsequent branch modules. The fish species branch module further extracts fish species-related features based on the output of the sharing layer and outputs the fish catching rates of each fish species. The fishing position branch module also extracts fishing position-related features based on the output of the sharing layer and outputs the utility values of each fish species at the fishing positions; the fishing feature datasets of each corresponding fishing ground are respectively input into the recommendation model for iterative training, and the model parameters are optimized through the backpropagation algorithm to improve the prediction accuracy of the model. Considering the historical total fish catch, fishing times, and fishing price of the recommended fishing positions comprehensively, the evaluation values of each fishing position are calculated by means of weighted summation. The weight coefficient α of the fish catch represents the relative importance of the historical total fish catch in the evaluation value. When α is larger, the evaluation value focuses more on reflecting the historical fish catch situation of the fishing position, which is suitable for users who value the fish catch more. The weight coefficient β of the fishing times represents the relative importance of the fishing times in the evaluation value. When β is larger, the evaluation value focuses more on reflecting the popularity or fishing activity of the fishing position, which is suitable for users who like lively places and pursue the fishing experience; the weight coefficient γ of the fishing price represents the relative importance of the fishing price in comprehensively evaluating the quality of the fishing position. The larger the γ value, the greater the impact of the fishing price on the evaluation value. α + β + γ = 1. By comprehensively evaluating the fishing positions, the fishing position with the highest cost performance and most suitable for the current fishing conditions is selected.

[0124] It can be understood that by constructing a recommendation model through a machine learning network, complex factors such as the fishing ground environment and fish species distribution can be accurately captured, improving the accuracy of recommended fish species and fishing positions; and by comprehensively evaluating the fishing positions, the fishing position with the highest cost performance and most suitable for the current fishing conditions is selected to provide more accurate and personalized fishing recommendations for users, improving the fishing success rate and satisfaction.

[0125] After step S34, it also includes: judging whether the recommended fish species at the current time of the fishing ground is consistent with the fishing target fish species selected by the user. If they are consistent, the idle fishing positions in the recommended fishing list information are obtained according to the fishing position status information, and the fishing position with the highest evaluation value among the idle fishing positions is selected as the recommended fishing position to be assigned to the user; if they are not consistent, it is judged whether the user changes the fishing target fish species to the recommended fish species. If it is changed, the fishing position with the highest evaluation value among the idle fishing positions is obtained and assigned to the user as the recommended fishing position; otherwise, the utility values corresponding to the fishing target fish species of each fishing position from the time window to the current time are obtained, and the fishing positions with fish species utility values in the excellent fishing position interval are selected as candidate fishing positions; the idle fishing positions among the candidate fishing positions are obtained according to the fishing position status information, and the evaluation values of the idle fishing positions among the candidate fishing positions are calculated, and the fishing position with the highest evaluation value is selected as the recommended fishing position to be assigned to the user.

[0126] Among them, the system first determines whether the recommended fish species at the current time of the fishing ground is the same as the target fish species selected by the user, ensures that the recommended fishing positions match the user's fishing preferences, and improves user satisfaction; if the recommended fish species is the same as the user's target fish species, the system preferentially selects the idle fishing position with the highest evaluation value to provide the best fishing experience. If the recommended fish species is not the same as the user's target fish species, the system determines whether the user is willing to change the target fish species to the recommended fish species. In the case where the user is willing to adjust the fishing target, the system still provides the optimal fishing position recommendation; in the case where the user adheres to the target fish species, by calculating the utility value and evaluation value of the target fish species, the most suitable fishing position is found for the user, improving satisfaction.

[0127] In this embodiment, the sharing layer includes a first feature extraction unit, a second feature extraction unit, and a third feature extraction unit. The first feature extraction unit, the second feature extraction unit, and the third feature extraction unit are connected in sequence. The third feature extraction unit outputs shared feature data, where

[0128] The expression of the first feature extraction unit is:

[0129] h1 = σ(W1X + b1)

[0130] In the formula, X is the standard feature data corresponding to each fishing ground, σ is the Sigmoid function, W1 is the weight matrix, and the dimension is R 256×64 , b1 is the bias vector;

[0131] The expression of the second feature extraction unit is:

[0132] h2 = LayerNorm(W2h1 + b2)

[0133] In the formula, h1 is the output feature of the first feature extraction unit, W2 is the weight matrix, and the dimension is R 128×256 , b2 is the bias vector, and LayerNorm() is the layer normalization function;

[0134] The expression of the third feature extraction unit is:

[0135] h3 = Dropout(σ(W3h2 + b3), c = 0.3)

[0136] In the formula, h2 is the output feature of the second feature extraction unit, W3 is the weight matrix, and the dimension is R 64×128 , b3 is the bias vector, Dropout() is the regularization function, and c is the neuron dropout probability.

[0137] It should be noted that by connecting the first feature extraction unit, the second feature extraction unit, and the third feature extraction unit in sequence, the system can gradually extract more abstract and high-level feature representations from the original standard feature data; this hierarchical feature extraction method is conducive to capturing complex patterns and relationships in the data; and the shared feature data output by the third feature extraction unit is jointly used by the fish species branch module and the fishing position branch module, realizing the sharing and reuse of features, which is conducive to reducing repeated calculations, improving the model efficiency, and enabling the two branch modules to make predictions based on the same feature basis.

[0138] In addition, the layer normalization function is used in the second feature extraction unit, which helps to accelerate the convergence speed of the model and improve the training efficiency of the model; at the same time, layer normalization can also enhance the robustness of the model, making the model insensitive to small changes in the input data. And the regularization function is used in the third feature extraction unit, which helps to prevent the model from overfitting, improve the generalization ability of the model, and enable the model to maintain good prediction performance when facing new and unseen data.

[0139] In this embodiment, the fish species branch module includes a first fully connected layer, a second fully connected layer, and a first output layer. Among them, the input of the first fully connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional feature to 32 dimensions. The expression is:

[0140] z1 = W4h3 + b4

[0141] In the formula, W4 is the weight matrix, with a dimension of R 32×64 , b4 is the bias vector, with a dimension of R 32 ;

[0142] The output of the first fully connected layer is connected to the input of the second fully connected layer, mapping the 32-dimensional feature to the m-dimensional fish species quantity. The expression is:

[0143] z2 = RELU(z1)W5 + b5

[0144] In the formula, RELU() is the ReLU function, W5 is the weight matrix, with a dimension of R m×32 , b5 is the bias vector, with a dimension of R m ;

[0145] The output of the second fully connected layer is connected to the first output layer, and the first output layer is used to output the probabilities of each fish species. The expression is:

[0146]

[0147] In the formula, p i is the predicted probability of the i-th fish species, and the sum of the predicted probabilities of all fish species is 1.

[0148] It should be noted that the fish species branch module transforms the shared feature data into the predicted probabilities of each fish species through the layer-by-layer mapping of the first fully connected layer, the second fully connected layer, and the first output layer, enabling the model to learn the complex relationships between different fish species and features, thereby achieving accurate fish species prediction; the first output layer outputs the predicted probabilities of each fish species, and the sum of the predicted probabilities of all fish species is 1, which is conducive to users understanding the occurrence possibilities of different fish species and providing more comprehensive information for fishing decision-making.

[0149] In this embodiment, the fishing position branch module includes a third fully connected layer, a fourth fully connected layer, and a second output layer, where

[0150] the input of the third fully connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional feature to 16 dimensions, and the expression is:

[0151] u1 = W6h3 + b6

[0152] where W6 is the weight matrix with a dimension of R 16×64 and b6 is the bias vector with a dimension of R 16 ;

[0153] the output of the third fully connected layer is connected to the input of the fourth fully connected layer, mapping the 16-dimensional feature to a 1-dimensional scalar, and the expression is:

[0154] y = σ(u1)W7 + b7

[0155] where σ is the Sigmoid function, W7 is the weight matrix with a dimension of R 1×16 and b7 is the bias vector with a dimension of R;

[0156] the output of the fourth fully connected layer is connected to the input of the second output layer, and the second output layer outputs the fishing position utility value, y ∈ [0, 1].

[0157] It should be noted that the fishing position branch module transforms the shared feature data into the fishing position utility value through the layer-by-layer mapping of the third fully connected layer, the fourth fully connected layer, and the second output layer, reflecting the attractiveness and potential value of the fishing position for fishing, which helps users select the optimal fishing position. The fishing position utility value output by the second output layer is normalized to the interval [0, 1], making the utility values between different fishing positions comparable and facilitating users to more intuitively understand the advantages and disadvantages of each fishing position.

[0158] S4. The user selects a fishing position in the recommended fishing position information, weighs the fish catch according to the fish species, obtains the unit price of the fish species to calculate the total price of the returned fish, and calls the payment interface to remotely settle the user's account.

[0159] It should be noted that the user selects a fishing spot for fishing according to the fishing spot information recommended by the system. The system records the fishing spot information selected by the user. After the user finishes fishing, the user takes the catch to the designated weighing area for weighing, inserts the electronic fishing tag onto the electronic scale, and selects the fish species type on the electronic scale. The weighing device records the weight of the catch and transmits the data to the system. The system queries the corresponding unit price of the fish species according to the fish species information transmitted by the weighing device. The unit price of the fish species can be dynamically adjusted according to factors such as market conditions and seasonal changes. The system calculates the total price of the returned fish according to the weight of the catch and the unit price of the fish species, calls the payment interface, and transmits the total price of the returned fish to the payment platform. The payment platform sends the total price of the returned fish to the fishing ground management staff for confirmation according to the user's account information. After confirmation, the payment platform deducts the amount according to the user's account information and feeds back the settlement result to the system. The system records the settlement result for subsequent query and statistics, and stores the fishing information of the current user in the database for preservation.

[0160] In this embodiment, by comprehensively collecting user fishing information, accurately analyzing historical data, constructing an intelligent recommendation model, and realizing an automated settlement process, the fishing ground management efficiency and user fishing experience are significantly improved; the system can not only provide a scientific basis for resource allocation for the fishing ground, but also recommend the optimal fish species and fishing spots for users according to real-time environmental parameters and fish situation prediction, increasing the fishing success rate; at the same time, the automated weighing, pricing, and remote settlement functions simplify the transaction process, ensure the fairness and convenience of the transaction, and enhance user stickiness; in addition, this method continuously optimizes the recommendation algorithm through data accumulation and analysis, promotes the continuous improvement of the fishing ground service quality, and realizes a win-win situation between the fishing ground and users.

[0161] In the second aspect, the present invention also provides a fishing ground returned fish management system based on big data analysis, which is implemented by using the fishing ground returned fish management method based on big data analysis. The system includes:

[0162] An information entry module, which is used for the user to select a fishing ground and the target fish species for fishing, obtain a fishing tag and enter the initial user fishing information;

[0163] A data processing module, which is used to obtain the historical fishing information of the fishing ground users and perform data preprocessing to obtain a fishing feature data set corresponding to the fishing ground;

[0164] A recommendation module, which is used to construct a recommendation model based on a machine learning network, input the fishing feature data sets of each corresponding fishing ground into the recommendation model for iterative training, and the recommendation model outputs the recommended fish species and recommended fishing spot information of the corresponding fishing ground at the current time;

[0165] A settlement module, which is used for the user to select a fishing spot in the recommended fishing spot information for fishing, weigh the catch according to the fish species, obtain the unit price of the fish species to calculate the total price of the returned fish, and call the payment interface to remotely settle the user's account.

[0166] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0167] Those skilled in the art can clearly understand that for the sake of convenience and brevity of description, the specific working processes of the systems and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0168] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some interfaces, and the indirect couplings or communication connections of devices or units can be in electrical, mechanical or other forms.

[0169] 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, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0170] In addition, the functional units in the various embodiments of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0171] If a function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, ROM, RAM, magnetic disks, or optical discs that can store program codes.

[0172] In addition, it should be noted that in the systems and methods of the present invention, obviously, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. And, the steps of performing the above series of processes can naturally be executed in chronological order according to the described order, but it is not necessary to execute them in chronological order. Some steps can be executed in parallel or independently of each other. For those of ordinary skill in the art, it is understandable that all or any steps or components of the methods and devices of the present invention can be implemented in any computing device (including processors, storage media, etc.) or a network of computing devices in the form of hardware, firmware, software, or a combination thereof, which can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0173] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing system. The computing system can be a well-known general system. Therefore, the object of the present invention can also be achieved only by providing a program product containing program codes for implementing the method or device. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be noted that in the devices and methods of the present invention, obviously, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. And, the steps of performing the above series of processes can naturally be executed in chronological order according to the described order, but it is not necessary to execute them in chronological order. Some steps can be executed in parallel or independently of each other.

[0174] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A fishing ground fish return management method based on big data analysis, characterized in that, It includes the following steps: S1. The user selects a fishing ground and the target fish species for fishing, obtains a fishing card and enters the initial user fishing information; S2. Obtain the historical fishing information of users in the fishing ground, and perform data preprocessing to obtain the fishing feature data set corresponding to the fishing ground; S3. Build a recommendation model based on the machine learning network, input the fishing feature data sets of each corresponding fishing ground into the recommendation model for iterative training, and the recommendation model outputs the recommended fish species and recommended fishing position information at the current time of the corresponding fishing ground; S4. The user selects a fishing position in the recommended fishing position information for fishing, weighs the fish catch according to the fish species, obtains the unit price of the fish species and calculates the total price of the returned fish, and calls the payment interface to remotely settle the user's account.

2. The method for managing the return of fish in a fishing ground based on big data analysis according to claim 1, wherein: In step S1, when the user selects a fishing ground and the target fish species for fishing, obtains a fishing card and enters the initial user fishing information, it includes the user obtaining the fishing card of the fishing ground and entering the initial user fishing information through the mobile terminal. The user fishing information includes personal information, fishing ground information, target fish species, fishing time, environmental information and fishing position information. The personal information includes name, age, contact information, fishing age, account and preferred fish species; the fishing ground information includes fishing ground name, geographical location, water quality type and development time; the fishing time includes season, time period and holiday marking status, and the environmental information includes water temperature, air pressure, temperature and humidity, weather conditions, pH value and dissolved oxygen; the fishing position information includes fishing price, fishing position number, the fishing weight of each fish species and the fishing position status.

3. The fishing ground fish return management method based on big data analysis according to claim 2, characterized in that: In step S2, obtain the historical fishing information of users in the fishing ground and perform data preprocessing to obtain the corresponding standard training data set of users in the fishing ground, including the following sub-steps: S21. Obtain the historical fishing information of all users of the corresponding fishing ground from the database according to the fishing ground name; S22. Perform data cleaning, conversion and merging processing on the historical fishing information of all users of the corresponding fishing ground to obtain the standard user fishing data; S23. Perform feature extraction and calibration on the fishing age, preferred fish species, season, time period, holiday marking status, water temperature, air pressure, temperature and humidity, weather conditions, pH value, dissolved oxygen, fishing price, fishing position number, fish species type and the fishing weight of each fish species information in the standard user fishing data to obtain the standard feature data; S24. Add the standard feature data to the data set to obtain the fishing feature data set corresponding to the fishing ground.

4. The fishing ground fish-return management method based on big data analysis according to claim 3, wherein: In S23, calibrate the fish species type, fish catch weight and fishing position utility of the standard user fishing data. Among them, the fishing position utility calibration includes the following sub-steps: Preset the time window length, obtain the fishing weight of each fish species at each fishing position every day within the past time window, and form the fish catch sequence of each fish species; Sort the total amount of fish caught within the time window in ascending order to obtain the ascending fish catch sequence W of each fish species s,f ={w1, w2,..., w T}; Preset the percentile, and calculate the percentile position index of the fish catch sequence according to the fish catch sequence of each fish species and the preset percentile. The expression is: S p = (p / 100) × (T + 1) Where S p is the percentile position index, p is the preset percentile, and T is the number of days in the time window; Judge S p whether it is an integer. If S p is an integer, then take the S p -th value in the ascending sequence of the catch of the corresponding fish species to obtain the calibrated value of the catch at the fishing position; P p = W s,f [S p - 1] where P p is the calibrated value of the fish catch at the fishing position, and W s,f [] is the ascending sequence of the fish catches of fish species f at fishing position s; If S p is a non-integer, the calibrated value of the catch at the fishing spot is calculated using linear interpolation, and the expression is: In the formula, is the floor function, is the ceiling function; Calculate the utility value corresponding to the fishing position according to the preset percentile, the fishing position fish catch calibration value and the ascending fish catch sequence of the corresponding fish species. The expression is: U s,f = (W s,f - P (100-p) (W s,f )) / (P p (W s,f ) - P (100-p) (W s,f )) Where U s,f is the utility value of fish species f at fishing position s, and P (100-p) is the calibrated fish catch value of the fishing position corresponding to the 100 - p percentile. If P p = P (100-p) , then U s,t is 0; Preset the recommended classification levels according to the utility values. If the utility value is in the interval [0, 0.4), it is marked as a low - efficiency fishing position. If the utility value is in the interval [0.4, 0.7), it is marked as an average fishing position. If the utility value is in the interval [0.7, 1], it is marked as an excellent fishing position, and the fishing position numbers are associated with the fish species types.

5. The fishing ground fish return management method based on big data analysis according to claim 4, characterized in that: In step S3, the recommendation model is constructed based on the machine - learning network. The fishing feature data sets of each corresponding fishing ground are respectively input into the recommendation model for iterative training. The recommendation model outputs the recommended fish species and recommended fishing position information for the current time of the corresponding fishing ground, including the following sub - steps: S31, construct a recommendation model based on the machine - learning network. The recommendation model includes an input layer, a shared layer, a fish - species branch module, and a fishing - position branch module. The output end of the input layer is connected to the input end of the shared layer. The fishing feature data sets of each corresponding fishing ground are respectively input into the recommendation model for iterative training. The output end of the shared layer is respectively connected to the input ends of the fish - species branch module and the fishing - position branch module, which are used to extract features from the fishing feature data sets of each corresponding fishing ground. The fish - species branch module outputs the fish - catching rates of each fish species, and the fishing - position branch module outputs the utility values of each fish species at the fishing positions. S32, obtain the top three fish species with the highest fish - catching rates as the recommended fish species. According to the recommended fish species, obtain the utility values of the corresponding fish species at the fishing positions, and select the fishing positions with the utility values in the excellent fishing - position interval as the recommended fishing positions. S33. Obtain the historical total catch amount, fishing times, and fishing prices of the recommended fishing positions for comprehensive calculation to obtain the evaluation value Score of each fishing position s , and the expression is: Score s = α × T s / max(T) + β × (1 - C s / max(C)) + γ × ((1 - (P s - min(P)) / (max(P)-min(P))) Where, T s is the total historical catch of fishing position s, C s is the number of fishing times of fishing position s, P s is the current fishing price of fishing position s, T is the total historical catch of the fishing ground, C is the maximum number of times of the fishing position, min(P) is the lowest fishing price of the fishing ground, max(P) is the highest fishing price of the fishing ground, α is the weight coefficient of the catch, β is the weight coefficient of the number of fishing times, and γ is the weight coefficient of the fishing price; S34, associate the evaluation values of each recommended fishing position with the fishing position numbers and the recommended fish species in descending order to obtain the recommended fishing list information for the user to view.

6. The fishing ground fish-return management method based on big data analysis according to claim 5, wherein: After step S34, it also includes: judging whether the recommended fish species at the current time of the fishing ground is the same as the fishing target fish species selected by the user. If they are the same, obtain the idle fishing positions in the recommended fishing list information according to the fishing position status information, and select the fishing position with the highest evaluation value among the idle fishing positions as the recommended fishing position to be assigned to the user. If they are not the same, judge whether the user changes the fishing target fish species to the recommended fish species. If changed, obtain the fishing position with the highest evaluation value among the idle fishing positions as the recommended fishing position to be assigned to the user. Otherwise, obtain the utility values of the corresponding fish species of the fishing target at each fishing position from the time window to the current time, and select the fishing positions with the utility values in the excellent fishing - position interval as the candidate fishing positions. Obtain the idle fishing positions among the candidate fishing positions according to the fishing position status information and calculate the evaluation values of the idle fishing positions among the candidate fishing positions, and select the fishing position with the highest evaluation value as the recommended fishing position to be assigned to the user.

7. The fishing ground fish return management method based on big data analysis according to claim 5, characterized in that: The shared layer includes a first feature extraction unit, a second feature extraction unit, and a third feature extraction unit. The first feature extraction unit, the second feature extraction unit, and the third feature extraction unit are connected in sequence. The third feature extraction unit outputs the shared feature data. Among them, The expression of the first feature extraction unit is: h1 = σ(W1X + b1) Where X is the standard feature data corresponding to each fishing ground, σ is the Sigmoid function, W1 is the weight matrix with a dimension of R 256×64 , and b1 is the bias vector; The expression of the second feature extraction unit is: h2 = LayerNorm(W2h1 + b2) where h1 is the output feature of the first feature extraction unit, W2 is the weight matrix with a dimension of R 128×256 , b2 is the bias vector, and LayerNorm() is the layer normalization function; The expression of the third feature extraction unit is: h3 = Dropout(σ(W3h2 + b3), c = 0.3) where h2 is the output feature of the second feature extraction unit, W3 is the weight matrix with dimension R 64×128 , b3 is the bias vector, Dropout() is the regularization function, and c is the neuron dropout probability.

8. The method for managing fish return in a fishing ground based on big data analysis according to claim 7, wherein: The fish species branching module includes a first fully-connected layer, a second fully-connected layer, and a first output layer. Among them, the input of the first fully-connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional features to 32 dimensions. The expression is: z1 = W4h3 + b4 Wherein, W4 is a weight matrix with a dimension of R 32×64 , and b4 is a bias vector with a dimension of R 32 ; The output of the first fully-connected layer is connected to the input of the second fully-connected layer, mapping the 32-dimensional features to m dimensions of the number of fish species. The expression is: z2 = RELU(z1)W5 + b5 where RELU() is the ReLU function, W5 is the weight matrix with dimension R m×32 , and b5 is the bias vector with dimension R m ; The output of the second fully-connected layer is connected to the first output layer. The first output layer is used to output the probabilities of each fish species. The expression is: where p i is the predicted probability of the i-th fish species, and the sum of the predicted probabilities of all fish species is 1.

9. The method for managing the fish return in a fishing ground based on big data analysis according to claim 7, wherein: The fishing position branching module includes a third fully-connected layer, a fourth fully-connected layer, and a second output layer. Among them, The input of the third fully-connected layer is connected to the output of the third feature extraction unit, reducing the 64-dimensional features to 16 dimensions. The expression is: u1 = W6h3 + b6 Where, W6 is the weight matrix with a dimension of R 16×64 , and b6 is the bias vector with a dimension of R 16 ; The output of the third fully-connected layer is connected to the input of the fourth fully-connected layer, mapping the 16-dimensional features to a 1-dimensional scalar. The expression is: y = σ(u1)W7 + b7 where σ is the Sigmoid function, W7 is the weight matrix with dimension R 1×16 , and b7 is the bias vector with dimension R; The output of the fourth fully-connected layer is connected to the input of the second output layer. The second output layer outputs the fishing position utility value, y ∈ [0, 1].

10. A fish return management system for fishing grounds based on big data analysis, implemented by using the fish return management method for fishing grounds based on big data analysis according to any one of claims 1-9, characterized in that: The system includes: An information entry module, which is used for the user to select a fishing ground and the target fish species for fishing, obtain a fishing card and enter the initial user fishing information; A data processing module, which is used to obtain the historical fishing information of the fishing ground users and perform data preprocessing to obtain the fishing feature data set corresponding to the fishing ground; A recommendation module, which is used to construct a recommendation model based on a machine learning network, input the fishing feature data sets of each corresponding fishing ground into the recommendation model for iterative training, and the recommendation model outputs the recommended fish species and recommended fishing position information of the corresponding fishing ground at the current time; A settlement module, which is used for the user to select a fishing position in the recommended fishing position information for fishing, weigh the fish catch according to the fish species, obtain the unit price of the fish species to calculate the total price of the returned fish, and call the payment interface to remotely settle the user's account.

Images