Port geographic information service platform construction method based on GIS

Abstract

The invention discloses a GIS-based port geographic information service platform construction method, and belongs to the technical field of information service platform construction, and the method comprises the steps: constructing a GIS data model; performing multi-source acquisition on the port geographic information data, extracting features related to berth demand prediction in the multi-source data, and fusing the features; storing the fused data into a GIS model, analyzing the fused data, selecting a first main component column according to the size of the characteristic value, and constructing a parking space demand model based on the first main component column; adjusting the parking space demand model based on an SHAP value analysis method; and the final berth demand model is deployed to a port environment for real-time prediction, and a prediction result is visualized to a GIS platform for decision support. According to the method, multiple aspects of data processing, model construction, safety factors, real-time prediction, personalized recommendation and the like are comprehensively considered, and the overall efficiency and service quality of port management are remarkably improved.

Description

Technical Field

[0001] The invention relates to the technical field of information service platform construction, and in particular to a method for constructing a port geographic information service platform based on GIS. Background Art

[0002] With the rapid development of port business and the increasing requirements for logistics and transportation efficiency, port management is facing increasing challenges. Especially in berth allocation, how to efficiently and reasonably arrange berth resources to meet the needs of ship berthing and improve port operation efficiency has become an urgent problem to be solved. Most of the existing berth allocation methods rely on manual experience or simple rules, which are difficult to adapt to the complex and changing port operation environment. In order to improve port operation efficiency and service quality, many ports have begun to adopt advanced information technology and data analysis technology to optimize berth management and resource allocation. Among them, geographic information system (GIS), as a powerful spatial data management and analysis tool, plays an important role in port geographic information management.

[0003] In the prior art, port geographic information data comes from various sources, including but not limited to ship location information, cargo information, meteorological data, port facility data, etc. These data usually come from different data sources and in different formats. The prior art has challenges in data management. It is difficult to effectively integrate these multi-source data and capture the factors affecting berth demand, resulting in inaccurate prediction results, which in turn affects the effective utilization of berth resources. Summary of the invention

[0004] The purpose of the present invention is to provide a method for constructing a port geographic information service platform based on GIS to solve the problems raised in the above background technology.

[0005] One aspect of the present application provides a method for constructing a port geographic information service platform based on GIS, comprising: S5000: Build GIS data model; S200: receiving multi-source collection of port geographic information data from multiple collectors, respectively extracting m features related to berth demand prediction from the multi-source data, fusing the data extracted from the multiple data sources, and storing the fused data in a GIS data model; S300: Analyze the fused data, select k eigenvectors corresponding to the largest eigenvalues ​​as the first principal component column according to the eigenvalue size, and build a berth demand model based on the first principal component column; S400: Based on the SHAP value analysis method, identify the key factors and their interactions that affect the port berth demand, and determine the impact of the key factors and the interactions on the berth demand model, and provide a second principal component column based on the key factors and their interactions; S500: determining whether the eigenvectors corresponding to the second principal component column and the first principal component column are consistent, if so, still using the original berth demand model; if not, adjusting the berth demand model; S600: deploying the final berth demand model to the port environment for real-time prediction, and visualizing the prediction results on the GIS platform for decision support; S700: Providing optimal berth allocation suggestions based on the prediction results and the user's historical behaviors and preferences.

[0006] As an optional solution of the technical solution of the present application, analyzing the data in the time series database includes: S310: Standardize the data in the time series database; S320: Calculate a covariance matrix, and calculate corresponding eigenvalues ​​and eigenvectors based on the covariance matrix; S330: Arrange the eigenvalues ​​in descending order based on their sizes, and select eigenvectors corresponding to k largest eigenvalues ​​that can explain at least 80% of the variance as principal components to form the first principal component column; S340: Select the first principal component column as a new feature axis, and project the original data into a low-dimensional space corresponding to the k largest eigenvalues.

[0007] As an optional solution of the technical solution of the present application, in the step S200, after the fused data is stored in the GIS data model, an indexing mechanism for the corresponding data based on the GIS data model is established.

[0008] As an optional solution of the technical solution of the present application, the construction of the berth demand model includes: S350: creating input-output pairs by projecting the data onto the low-dimensional space corresponding to the k largest eigenvalues, wherein the input corresponds to the value of the past p time steps and the output is the value of the next time step; and dividing the input-output pairs into a training data set Y1 and a test data set Y2; S360: Build LSTM model; S370: Train the LSTM model based on the training data set.

[0009] As an optional solution of the technical solution of the present application, the step S400 includes: S410: Calculate a SHAP value based on the test data set; S420: Determine k key features based on the absolute value order of the SHAP values; and calculate the SHAP interaction values ​​between the k key features. If the SHAP interaction values ​​between any two key features are not greater than a preset value, the k key features determined based on the absolute value order of the SHAP values ​​are used as the second principal component column. S430: If the SHAP interaction value between two key features A and B is greater than a preset threshold, a new key feature AB is established based on the key feature A and the key feature B; after the new key feature is established, the key feature is selected again based on the size of the feature value, so that the number of key features remains k, and the k key features constitute the second principal component column.

[0010] As an optional solution of the technical solution of the present application, the determination in step S500 as to whether the corresponding eigenvectors of the second principal component column and the first principal component column are consistent includes: Whether the eigenvectors in the second principal component column are all the eigenvectors in the first principal component column; Whether the order of the eigenvectors in the second principal component column is consistent with the order of the eigenvectors in the first principal component column; If the eigenvectors in the second principal component column are not all of the eigenvectors in the first principal component column, and / or the order of the eigenvectors in the second principal component column is inconsistent with the eigenvectors in the first principal component column, it is determined that the corresponding eigenvectors of the second principal component column are inconsistent with those of the first principal component column.

[0011] As an optional solution of the technical solution of the present application, the step S700 includes: S710: Collecting the user's berth usage history and preference information; S720: Determine the berth availability status in the future based on the prediction result of the berth demand model; formulate a berth allocation strategy for a specific user in combination with the collected user berth usage history and preference information; S730: Berth allocation optimization based on linear programming:

[0012] subject to , , ; in, is the cost vector, is the decision variable vector, and are the coefficient matrices of equality constraints and inequality constraints, respectively, and are the right-hand side vectors of the equality and inequality constraints, respectively. and are the lower and upper bounds of the variable.

[0013] As an optional solution of the technical solution of this application, it also includes: S800: Generate a recommendation system based on historical user preferences; Based on the prediction result of the berth demand model, the berth availability status in the future period is determined, and the berth allocation strategy for new users is specified in combination with the recommendation system.

[0014] Another aspect of the present application provides a device for constructing a port geographic information service platform, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the aforementioned method when executing the computer program.

[0015] A third aspect of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the aforementioned method are implemented.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The technical solution recorded in this application establishes a time series database and index mechanism through multi-source data collection and fusion, which greatly improves the integrity, accuracy and availability of data, and provides a high-quality data foundation for subsequent analysis; the berth demand model is constructed using the principal component analysis method, which can effectively capture the main factors affecting demand and improve the accuracy of prediction; by introducing the SHAP value analysis method, the key factors affecting the port berth demand and their interactions are identified, so that the model takes safety factors into account more comprehensively and improves the reliability of prediction; by comparing the two principal component columns, the self-inspection and dynamic adjustment of the model are realized, and the adaptability and stability of the model are enhanced; the model is deployed to the actual environment for real-time prediction and visualized on the GIS platform, providing managers with intuitive and timely decision support tools; based on the prediction results and user historical behavior, the optimal berth allocation recommendation is given, which realizes the efficient use of resources and improves user satisfaction.

[0017] This method comprehensively considers multiple aspects such as data processing, model building, security factors, real-time prediction and personalized recommendations, significantly improving the overall efficiency and service quality of port management; through the application of data-driven and intelligent algorithms, it lays the foundation for the intelligent and automated port operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1This is a workflow diagram of the method for constructing a port geographic information service platform based on GIS recorded in the embodiment of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] The present application embodiment records a method for constructing a port geographic information service platform based on GIS, and the construction method mainly includes the following steps: S100: Build GIS data model; In this step, it is necessary to define entities, attributes and the relationships between them in the GIS data model. In a specific implementation of the present application, in order to facilitate the construction of the subsequent berth demand model and give the optimal berth allocation suggestion, the entities in the GIS data model are defined as berths, users, ships and weather conditions, and their attributes are correspondingly defined as the location coordinates of the berth, berth type, berth capacity, user ID, user preference, etc.; the relationships include the relationship between berths and users, the relationship between berths and berth demand, etc.

[0021] S200: Collect port geographic information data from multiple sources, extract m features related to berth demand prediction from the multi-source data, and fuse the data extracted from multiple data sources; connect the fused data to the GIS data model, and establish a data indexing mechanism.

[0022] Among them, the multi-source collection of port geographic information includes sensor data, satellite images, aerial photos, meteorological data, etc. Due to the diversity of the above data sources, the data needs to be pre-processed before data fusion, including removing erroneous data and duplicate data, aligning the time series of data, etc., and then integrating the multi-source data together to form a unified data set, and converting the above unified data set into a format that can be processed by GIS, such as converting coordinates into a coordinate system supported by GIS, so as to facilitate the connection of the fused data to the GIS data model.

[0023] In addition, in the above steps, the establishment of the data index mechanism is mainly to speed up the query speed of data, so as to facilitate the subsequent collection of port geographic information and attach it to the corresponding position of the GIS data model; the data index mechanism can be a spatial index of R-tree or Quadtree type, or a time index that can quickly locate a specific time point or time period. In a specific embodiment of the present application, in order to facilitate the subsequent creation of input and output pairs based on time sequence, its index mechanism is a time index, for example, it can be a sparse index that only records the index when the data changes.

[0024] S300: Analyze the fused data, select k eigenvectors corresponding to the largest eigenvalues ​​as the first principal component column according to the size of the eigenvalues, and build a berth demand model based on the first principal component column. It can be understood that in this process, k≤m.

[0025] Among them, the construction process of the first principal component column is as follows: S310: Standardize the fused data: ; Where x is the original data point, μ is the standard deviation of the feature; S320: Calculate a covariance matrix, and calculate corresponding eigenvalues ​​and eigenvectors based on the covariance matrix; Among them, the covariance matrix reflects the relationship between each feature. For n samples with m features, the calculation method of the covariance matrix C is: ; Where X is the standardized data matrix, and each column corresponds to a feature.

[0026] Then the eigenvalues ​​λ1,λ2,...,λ of the covariance matrix C are calculated m and the corresponding eigenvectors v1, v2, ..., v m . Among them, the eigenvalue represents the variance along the corresponding feature direction; S330: Arrange in descending order according to the size of the eigenvalues. The eigenvector with a larger eigenvalue represents a more important principal component. Arrange in descending order based on the size of the eigenvalues, and select the eigenvectors corresponding to the k largest eigenvalues ​​that can explain at least 80% of the variance as the first principal component column.

[0027] In a specific embodiment, from the above v1, v2, ..., v m From the m eigenvectors, select the k eigenvectors with the largest eigenvalues ​​to form the first principal component column.

[0028] Then, step S340 is performed, that is, the first principal component column selected above is used as a new characteristic axis to project the original data into a low-dimensional space corresponding to the k largest eigenvalues: For each sample x, the new feature vector y can be expressed as: ; Where V is a matrix consisting of the selected k eigenvectors; The projected data is: ; X is the standardized data matrix.

[0029] After the projected data Y is obtained, a berth demand model needs to be constructed based on the projected data, which specifically includes the following steps: S350: Create input-output pairs by projecting the data into the low-dimensional space corresponding to the k largest eigenvalues, where the input corresponds to the value of the past p time steps and the output is the value of the next time step: In a specific scenario, for a time series Y = [y1, y2, y3, ..., y t ], if you want to predict the value from time t+1 to time t, the input is the value from tp to time t, and the output is the value at time t+1; that is, when you need to predict the value of a future time step, the input and output pairs are created as follows: Input: [y1, y2, y3, …, y t-p ]; Output: [y t-p+1 ].

[0030] Then, the input-output pair needs to be divided into a training data set Y1 and a test data set Y2, wherein the training data set Y1 contains earlier data and the test data set Y2 contains later data. In a specific example of the present application, according to the time corresponding to the data, Y1 selects 80% of the data at the earlier time and Y2 selects 20% of the data at the later time.

[0031] S360: Build an LSTM model. The LSTM model is a special RNN (Recurrent Neural Network). Because it can memorize long-term dependent information, it is more suitable for time series processing and is more applicable to berth demand forecasting.

[0032] In the specific process of building the LSTM model, a deep learning framework such as Keras or TensorFlow can be used. Those skilled in the art can build the LSTM model with the help of languages ​​such as Python, so it will not be described in detail in the specific embodiments of this application.

[0033] S370: training the LSTM model; the LSTM model will learn how to predict the value of the next time step based on a given input sequence, for example, when the input sequence is [y1, y2, y3], the goal of the LSTM model is to predict the value of y4. In an embodiment of the present application, the training of the LSTM model is based on a training data set.

[0034] In the traditional model building process, a training data set is often used to train the model, and then a test data set is used to test the trained model to evaluate the performance of the model. In this scheme, since the training data set and the test data set come from the same source, when they train and test the same model respectively, the interaction between different features is often difficult to reflect. Therefore, in the embodiment of the present application, the interaction between features and the adjustment of the LSTM model are based on the SHAP value analysis in steps S400 to S500.

[0035] S400: Based on the SHAP value analysis method, identify the key factors and their interactions that affect the port berth demand, and determine the impact of the key factors and the interactions on the berth demand model, and give a second principal component column based on the key factors and their interactions. Specifically, the following steps are included: S410: For a single prediction, the SHAP value is expressed as: ; Where f(y) is the predicted value of the LSTM model; is the benchmark prediction value, and in this embodiment, the average value of the LSTM model prediction value is selected; is the SHAP value of feature i; y i is the value of feature i; k is the total number of eigenvalues.

[0036] According to the above calculation method, the SHAP values ​​of the factors represented by the k eigenvectors selected based on the maximum eigenvalue are calculated, and the k key features are sorted according to the absolute value of each SHAP value.

[0037] Wherein, step S420 mainly includes the following sub-steps: S421: Calculate the SHAP interaction value between k features: For two features i and j, the SHAP interaction value between them is expressed as: ; in, represents the contribution of feature i given the existence of feature j; represents the contribution of feature j given the existence of feature i; and Represent the independent SHAP values ​​of features i and j respectively.

[0038] In the above calculation method, the contribution of feature i given the existence of feature j is The calculation method is: ; Among them, Y is the set of all features, Represents the set of all features except features i and j; is the predicted value when the feature subset S is added with features i and j; It is the predicted value when feature subset S is added with feature j; The feature subset S represents all subsets after excluding features i and j, and .

[0039] The same calculation principle is used to calculate the contribution of j given that feature i already exists. .

[0040] S422: Determine whether the SHAP interaction value between any two key features is greater than a preset value, if so, proceed to step S430, if not, proceed to step S500; S430: If the SHAP interaction value between two key features A and B is greater than a preset threshold, a new key feature AB is established based on the key feature A and the key feature B; after establishing the new key feature, the original key feature A and the original key feature B are discarded and replaced with the new key feature AB. At this time, the number of key features is reduced, so it is necessary to supplement the key features. In the embodiment of the present application, it is necessary to select the k+1th key feature from the features arranged in descending order based on the eigenvalue size in step S330, so that the number of key features remains k, and the k key features constitute the second principal component column; if the SHAP interaction value between the other two key features is also greater than the preset threshold, a new feature is created according to the above steps, and then the feature vectors sorted according to the eigenvalue size in step S330 are selected in turn; if the number of feature vectors in step S330 is difficult to meet the reduction in the number of key features caused by the SHAP interaction value being greater than the preset threshold, it is necessary to consider adjusting the preset SHAP interaction value threshold.

[0041] The creation method of the above-mentioned new key feature AB is based on the specific situation. For example, in some embodiments, the interaction between feature A and feature B can be captured in a multiplication manner. In this case, the new feature AB=A×B; in some embodiments, new features can also be created through function combination. In this case, the new feature AB=g(A)+h(B).

[0042] After the new key features are created, the adjusted features are used as the second principal component column. At this time, due to the generation of new key features, the feature vectors corresponding to the original first principal component column are different. Then, the second principal component column is used as the new feature axis, and the original data is projected to construct the berth demand model.

[0043] S500: The k key features determined based on the absolute size sorting of the SHAP values ​​are used as the second principal component column. At this time, the k key features in the second principal component column are the same as the k key features in the first principal component column. Therefore, it is only necessary to determine whether the order of the k key features in the second principal component column is the same as the order of the k key features in the first principal component column. If the ordering of the two is the same, the original berth demand model is used. If the ordering of the two is different, the second principal component column is used as the new feature axis, and the original data is projected to construct the berth demand model.

[0044] The construction of the berth demand model in step S500 and step S430 is carried out using the ideas in step S350 to step S370, but the method is slightly different. The re-construction of the berth demand model is mainly carried out in the following way: 1) Create input-output pairs with data projected to the low-dimensional space corresponding to the k eigenvalues ​​corresponding to the second principal component column, where the input still corresponds to the value of the past p time steps, and the output still corresponds to the value of the next time step; the time step is selected according to actual needs, such as in some ports with more ship traffic, the time step is shorter, which may be 0.5h; in some ports with less ship traffic, the time step may be 4h. The created input-output pairs are no longer divided into training data set and test data set, and the data corresponding to these input-output pairs are all training data.

[0045] 2) Build a new LSTM model.

[0046] 3) Train a new LSTM model based on the input-output pairs created in 1).

[0047] The model obtained in 3) above is the final berth demand model, which can then be deployed to the port environment for real-time prediction, and the prediction results can be visualized on the GIS platform.

[0048] When the final berth demand model is deployed to the port for real-time prediction, the optimal berth allocation recommendation can be given based on the prediction results and the user's historical behavior and preferences, which is step S700.

[0049] This step is specifically performed as follows: S710: Collect the user's berth usage history and preference information; the usage history includes the frequency and time period of the user's past berth usage; the preference generally refers to the user's preferred berth location, berth type, etc.

[0050] S720: Determine the berth availability status in the future based on the prediction result of the berth demand model; formulate a berth allocation strategy for a specific user in combination with the collected user berth usage history and preference information; S730: Berth allocation optimization based on linear programming:

[0051] subject to , , ; in, is the cost vector, is the decision variable vector, and and are the coefficient matrices of the equality and inequality constraints respectively, and are the right-hand side vectors of the equality and inequality constraints, respectively. and are the lower and upper bounds of the variable.

[0052] In the above linear program: is the objective function, which expresses the quantity we wish to minimize; Equality constraints means that the berth demand is equal to the forecasted demand, where is a matrix, each row corresponds to a berth position, and each column corresponds to a user; Each row of is a unit vector, indicating the allocation of berth positions. is a vector representing the demand forecast value for each berth location.

[0053] Inequality constraints Indicates that the berth allocation does not exceed the available berths, where is a matrix, each row corresponds to a berth position, and each column corresponds to a user; Each row of is a unit vector, indicating the allocation of berth positions; is a vector representing the maximum number of available berths at each berth position; The Nether is a vector with each element being 0, indicating that the number of allocated berths cannot be less than 0; Upper Bound is a vector, each element of which represents the maximum number of available berths for each berth position.

[0054] The above step S700 can give the optimal berth allocation suggestion based on the historical behavior and preferences of historical users. However, for a certain port, there will inevitably be some new users. The technical solution of this application takes into account the problem that new users have no historical behavior and preferences, and provides an allocation strategy for new users.

[0055] The berth allocation strategy for new users is recorded in step S800. Specifically, it can be implemented as follows: S810: Data collection and preprocessing Collect all historical user behavior data, including berth reservation records, berth usage frequency, berth duration, etc.; Based on the collected data, a user preference model is constructed, and a clustering algorithm can be used to divide users into different groups, each with similar berth demand preferences.

[0056] S820: New user feature collection For new users, get a preliminary understanding of their needs by asking them about their vessel type, cargo type, expected docking time, etc.

[0057] S830: User similarity calculation Use the information provided by the new user to match the preference model of historical users to find the most similar group of users; If the new user already has some initial behavior data (such as past berth reservations), collaborative filtering can be used to find other users with similar behaviors to the new user.

[0058] S840: Berth recommendation generation Use the previously established berth demand model to predict berth demand; Combining the characteristics of new users with the behavioral data of historical users, a list of recommended berths is generated for new users, which can be based on factors such as berth availability, location, and facility configuration.

[0059] S850: Recommendation result optimization Further adjust the recommendation list based on additional information provided by new users (such as special needs, time windows, etc.), and encourage new users to provide feedback to further optimize the recommendation algorithm in the future.

[0060] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for constructing a port geographic information service platform based on GIS, applied to a device for constructing a port geographic information service platform based on GIS, characterized in that: include: S100: Build GIS data model; S200: receiving multi-source collection of port geographic information data from multiple collectors, respectively extracting m features related to berth demand prediction from the multi-source data, fusing the data extracted from the multiple data sources, and storing the fused data in a GIS data model; S300: Analyze the fused data, select k eigenvectors corresponding to the largest eigenvalues ​​as the first principal component column according to the eigenvalue size, and build a berth demand model based on the first principal component column; S400: Based on the SHAP value analysis method, identify the key factors and their interactions that affect the port berth demand, and determine the impact of the key factors and the interactions on the berth demand model, and provide a second principal component column based on the key factors and their interactions; S5 00: Determine whether the eigenvectors corresponding to the second principal component column and the first principal component column are consistent; if so, still use the original berth demand model; if not, adjust the berth demand model; S600: deploying the final berth demand model to the port environment for real-time prediction, and visualizing the prediction results on the GIS platform for decision support; S700: Providing optimal berth allocation suggestions based on the prediction results and the user's historical behaviors and preferences.

2. The method according to claim 1, characterized in that In the step S200, after the fused data is stored in the GIS data model, an index mechanism for the corresponding data based on the GIS data model is established.

3. The method according to claim 1, characterized in that The step S500 of determining whether the eigenvectors corresponding to the second principal component column and the first principal component column are consistent includes: Whether the eigenvectors in the second principal component column are all the eigenvectors in the first principal component column; Whether the order of the eigenvectors in the second principal component column is consistent with the order of the eigenvectors in the first principal component column; If the eigenvectors in the second principal component column are not all of the eigenvectors in the first principal component column, and / or the order of the eigenvectors in the second principal component column is inconsistent with the eigenvectors in the first principal component column, it is determined that the corresponding eigenvectors of the second principal component column are inconsistent with those of the first principal component column.

4. The method for constructing a port geographic information service platform based on GIS according to claim 1, characterized in that: The analysis of the fused data includes: S310: performing standardization processing on the fused data; S320: Calculate a covariance matrix, and calculate corresponding eigenvalues ​​and eigenvectors based on the covariance matrix; S330: Arrange the eigenvalues ​​in descending order based on their sizes, and select eigenvectors corresponding to k largest eigenvalues ​​that can explain at least 80% of the variance as principal components to form the first principal component column; S340: Select the first principal component column as a new feature axis, and project the original data into a low-dimensional space corresponding to the k largest eigenvalues.

5. The method according to claim 4, characterized in that The constructing of the berth demand model comprises: S350: creating input-output pairs by projecting the data into the low-dimensional space corresponding to the k largest eigenvalues, wherein the input corresponds to the value of the past p time steps and the output is the value of the next time step; and dividing the input-output pairs into a training data set Y1 and a test data set Y2; S360: Build LSTM model; S370: Train the LSTM model based on the training data set.

6. The method according to claim 5, characterized in that The step S400 includes: S410: Calculate a SHAP value based on the test data set; and determine k key features based on the absolute value order of the SHAP value; S420: Calculate the SHAP interaction values ​​between k key features. If the SHAP interaction values ​​between any two key features are not greater than a preset value, the k key features determined by sorting based on the absolute values ​​of the SHAP values ​​are used as the second principal component column. S430: If the SHAP interaction value between two key features A and B is greater than a preset threshold, a new key feature AB is established based on the key feature A and the key feature B; after the new key feature is established, the key feature is selected again based on the size of the feature value, so that the number of key features remains k, and the k key features constitute the second principal component column.

7. The method according to claim 1, characterized in that The step S700 includes: S710: Collecting the user's berth usage history and preference information; S720: Determine the berth availability status in the future based on the prediction result of the berth demand model; formulate a berth allocation strategy for a specific user in combination with the collected user berth usage history and preference information; S730: Berth allocation optimization based on linear programming: subject to , ， ； in, is the cost vector, is the decision variable vector, and are the coefficient matrices of equality constraints and inequality constraints, respectively, and are the right-hand side vectors of the equality and inequality constraints, respectively. and are the lower and upper bounds of the variable.

8. The method according to claim 1, characterized in that Also includes: S800: Generate a recommendation system based on historical user preferences; Based on the prediction result of the berth demand model, the berth availability status in the future period is determined, and the berth allocation strategy for new users is specified in combination with the recommendation system.

9. A device for constructing a port geographic information service platform based on GIS, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images