Function recommendation method and device, storage medium and computer program product

Abstract

The invention discloses a function recommendation method and device, a storage medium and a computer program product, and relates to the technical field of data analysis. The method comprises the steps that a login request of a target object is received, and the login request is sent by the target object of different or the same job position to request to log in a target device; in response to the login request, behavior information of a corresponding target object is obtained, the behavior information is used for handling target services for the to-be-departed object, and the target services comprise check-in, endorsement, consignment, ticket refunding and ticket booking; determining a user group type to which the target object belongs based on the behavior information; and based on the user group type, generating recommendation information of the common function control suitable for the target object. The technical problem that an existing departure system is low in business handling efficiency is solved.

Description

Technical Field

[0001] The present invention relates to the field of data analysis technology, and in particular to a function recommendation method, device, storage medium and computer program product. Background Art

[0002] With the economic development of today's society, the rapid development of the civil aviation industry, and the increasing living standards of the people, airplanes have become the preferred mode of transportation for people to work and travel. Airport staff need to handle a large number of passenger service operations every day, and departure operations also have high requirements for the timeliness of staff handling. In the aviation departure system, staff need to complete the departure procedures of passengers quickly and accurately.

[0003] At present, traditional departure systems often only provide fixed function menus, lacking in specificity and flexibility, which causes staff to search among numerous functions when handling different businesses, reducing business processing efficiency.

[0004] To address the above-mentioned problems, no effective solution has been proposed yet. Summary of the invention

[0005] The embodiments of the present invention provide a function recommendation method, device, storage medium and computer program product to at least solve the technical problem of low efficiency of business handling in the current departure system.

[0006] According to one aspect of an embodiment of the present invention, a function recommendation method is provided, comprising: receiving a login request from a target object, wherein the login request is issued by a target object of a different or the same job position to request to log in to a target device; in response to the login request, obtaining behavior information of the corresponding target object, wherein the behavior information is used to handle target services for the object to be departed, wherein the target services include check-in, ticket change, consignment, ticket refund, and ticket booking; based on the behavior information, determining the type of user group to which the target object belongs; based on the user group type, generating recommendation information of commonly used function controls suitable for the target object.

[0007] Furthermore, it includes: generating a display interface based on the recommendation information, the display interface including one or more recommendation function controls for displaying the preferences of the target object.

[0008] Furthermore, based on the behavior information, the user group type is determined, including: obtaining a target cluster analysis model; and based on the target cluster analysis model, analyzing the behavior information to determine the user group type.

[0009] Furthermore, based on the user group type, recommendation information of commonly used function controls suitable for the target object is generated, including: determining commonly used function controls of the user group type based on the user group type and the target cluster analysis model; and generating recommendation information based on the commonly used function controls.

[0010] Furthermore, before obtaining the target cluster analysis model, it includes: obtaining historical user behavior information, wherein the historical user behavior information includes operation record information, stay time information, click count information, user role information, and business attribute information; preprocessing the operation record information, stay time information, click count information, user role information, and business attribute information to obtain preprocessed historical behavior information; extracting user feature vectors based on the preprocessed historical behavior information; constructing an initial cluster analysis model based on the user feature vector; and optimizing the initial cluster analysis model to obtain a target cluster analysis model.

[0011] Furthermore, the initial cluster analysis model is optimized to obtain a target cluster analysis model, including: determining the effectiveness in response to an operation on a feedback control in a display interface, wherein the feedback control is a control for providing feedback on whether the recommended information is effective; based on the effectiveness, the initial cluster analysis model is optimized to obtain a target cluster analysis model.

[0012] Furthermore, the initial clustering analysis model is optimized to obtain a target clustering analysis model, including: determining a click-through rate in response to an operation on a function control in a display interface, where the click-through rate is the frequency with which a target object clicks on a recommended function control; based on the click-through rate, the initial clustering analysis model is optimized to obtain a target clustering analysis model.

[0013] According to another aspect of an embodiment of the present invention, a function recommendation device is also provided, including: a receiving module, the receiving module is used to receive a login request from a target object, wherein the login request is issued by a target object of a different or the same job position to request to log in to a target device; an acquisition module, the acquisition module is used to obtain behavior information of the corresponding target object in response to the login request, wherein the behavior information is used to handle target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking; a determination module, the determination module is used to determine the type of user group to which the target object belongs based on the behavior information; and a generation module, the generation module is used to generate recommendation information suitable for commonly used function controls of the target object based on the user group type.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program is running, the device where the computer-readable storage medium is located is controlled to execute the methods in various embodiments of the present invention.

[0015] According to another aspect of an embodiment of the present invention, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the method in each embodiment of the present invention is implemented.

[0016] In the embodiment of the present invention, by determining the user group type based on user behavior information, the purpose of generating personalized recommendation information is achieved, and user needs can be effectively identified and personalized recommendations can be provided to enhance the operational efficiency and user satisfaction of the departure system, thereby solving the technical problem of low business processing efficiency of the current departure system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0018] Figure 1 is a flow chart of an optional function recommendation method according to an embodiment of the present invention;

[0019] Figure 2 is a flow chart of an optional function recommendation method according to an embodiment of the present invention;

[0020] Figure 3 is a structural block diagram of an optional function recommendation device according to an embodiment of the present invention;

[0021] Figure 4 4 is an architectural diagram of an optional function recommendation device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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 should fall within the scope of protection of the present invention.

[0023] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0024] According to an embodiment of the present invention, an embodiment of a function recommendation method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0025] The method embodiment may be executed in an electronic device or a similar computing device including a memory and a processor. Taking the departure system terminal as an example, the departure system terminal may include one or more processors (processors may include but are not limited to central processing unit (CPU), graphics processing unit (GPU), digital signal processing (DSP) chip, microcontroller unit (MCU), programmable logic device (Field Programmable Gate Array, FPGA), neural network processor (Neural-network Processor Unit, NPU), tensor processor (TensorProcessing Unit, TPU), artificial intelligence (Artificial Intelligence, AI) type processor and other processing devices) and a memory for storing data. Optionally, the departure system terminal may also include a transmission device, input and output device and display device for communication functions. It can be understood by those skilled in the art that the above structural description is only for illustration and does not limit the structure of the departure system terminal. For example, the departure system terminal may also include more or fewer components than the above structural description, or have a configuration different from the above structural description.

[0026] The memory can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the function recommendation method in the embodiment of the present invention. The processor executes various functional applications and data processing by running the computer program stored in the memory, that is, the above-mentioned function recommendation method is realized. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory may further include a memory remotely arranged relative to the processor, and these remote memories may be connected to the mobile terminal via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0027] The transmission device is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device can be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet wirelessly.

[0028] The display device may be, for example, a touch screen type liquid crystal display (LCD) and a touch display (also referred to as a "touch screen" or "touch display screen"). The liquid crystal display may enable a user to interact with a user interface of a mobile terminal. In some embodiments, the mobile terminal has a graphical user interface (GUI), and a user may interact with the GUI by finger contacts and / or gestures on a touch-sensitive surface, wherein the human-computer interaction functions here may optionally include the following interactions: creating web pages, drawing, word processing, making electronic documents, games, video conferencing, instant messaging, sending and receiving emails, call interfaces, playing digital videos, playing digital music and / or web browsing, etc. The executable instructions for executing the above human-computer interaction functions are configured / stored in a computer program product or readable storage medium executable by one or more processors.

[0029] Figure 1 is a function recommendation method according to an embodiment of the present invention, such as Figure 1 As shown, the method comprises the following steps:

[0030] Step S10, receiving a login request from a target object, wherein the login request is sent by a target object of a different or the same job position to request to log in to a target device.

[0031] In step S10, a login request is first received from a target object (i.e., a staff member using the system). These target objects may be from different job positions, such as ordinary check-in clerks, check-in supervisors, boarding clerks, etc., or from the same job position. The login request is the first step for the system to identify and respond to user needs.

[0032] Step S12, in response to the login request, obtain the behavior information of the corresponding target object, wherein the behavior information is used to handle the target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking.

[0033] In step S12, after the target object successfully logs into the system, the behavior information of the object begins to be collected and recorded. The behavior information may include: operation frequency (i.e., the number of times a specific operation is performed), operation type (such as check-in, ticket change, consignment, ticket refund, ticket booking, etc.), operation time (for example, busy or non-busy hours of the day), and other information that may affect the operation, such as the role of the target object, business attributes, and behavioral characteristics of the customer group. This information is used for subsequent user group type analysis and personalized recommendation generation.

[0034] Step S14, determining the type of user group to which the target object belongs based on the behavior information;

[0035] In step S14, the behavior pattern of the target object is analyzed using the collected behavior information in combination with a preset clustering model to determine the type of user group to which it belongs. User group types may be grouped based on similar behavior characteristics, operation preferences, business needs, etc. For example, the system may identify a user group type such as "ordinary check-in clerks at large airports who often handle checked baggage business."

[0036] Step S16: Generate recommendation information of commonly used function controls suitable for the target object based on the user group type.

[0037] In step S16, based on the determined user group type, recommended information of commonly used functional controls suitable for the target object is generated, and the recommended information will be customized according to the specific needs and preferences of the target object. For example, if the target object belongs to the user group type of "ordinary check-in staff at large-scale airports who often handle checked baggage business", the system may give priority to recommending functional controls related to checked baggage, such as "adding luggage" and "viewing passenger information". This helps the target object find and use the functions they need most quickly, thereby improving work efficiency and user experience.

[0038] Based on steps S10 to S16, the user group type is determined based on user behavior information, and the purpose of generating personalized recommendation information is achieved. It can effectively identify user needs and provide personalized recommendations to enhance the operational efficiency and user satisfaction of the departure system.

[0039] Furthermore, the function recommendation method includes:

[0040] Step S18: generating a display interface based on the recommendation information, wherein the display interface includes one or more recommendation function controls for displaying the target object's preferences.

[0041] In step S18, the following key steps are generally included:

[0042] Analyze the recommended information: Before generating the display interface, it is necessary to conduct an in-depth analysis of the recommended information obtained in step S16) to understand the specific needs and preferences of the target object. These recommended information are generated based on the characteristics of the user group type to which the target object belongs and past behavior data, and include the most commonly used or most relevant functional modules.

[0043] Design the display interface: Design a user-friendly display interface based on the recommended information. The display interface should be clear and intuitive, and effectively highlight the recommended functional modules so that they can be seen at a glance among many functions. The interface design may include the layout of functional modules, color coding, icons, shortcut key prompts, etc. to improve ease of use and attractiveness.

[0044] Function module display: On the display interface, the recommended function modules are presented to the target audience in the form of lists, icons, heat maps or other visual elements. The function modules should be displayed in a way that ensures that the target audience can quickly identify and access the functions they use most frequently, thereby improving operation speed and efficiency.

[0045] Personalized adjustment: The display interface may need to be adjusted according to the real-time behavior of the target object. For example, if the system detects that the target object spends more time on a specific function or clicks more times, it may indicate that they have additional needs for this function, and the display interface should highlight or prioritize these functions.

[0046] User feedback integration: The display interface should include a user feedback mechanism, allowing the target audience to provide feedback on the effectiveness of the recommendation function. This can be a simple "valid / invalid" button, or a more detailed evaluation system. User feedback will be used for subsequent recommendation strategy optimization.

[0047] Testing and optimization: After the display interface is generated, user testing should be conducted to collect feedback and analyze the usage and click-through rate of the interface to determine which recommendations actually improve efficiency in actual operations. Based on the test results, the system may need to adjust the recommendation algorithm or interface design to better meet the needs of the target audience.

[0048] Through step S18, a personalized and intuitive function recommendation interface can be provided for each target object, thereby optimizing its operating experience and efficiency in the departure system. The personalized and real-time feedback mechanism of this function recommendation method is an important means to improve airport operation efficiency and passenger service quality.

[0049] Optionally, based on the behavior information, the user group type is determined, including:

[0050] Step S201, obtaining a target cluster analysis model.

[0051] In step S201, a target cluster analysis model suitable for the current situation is obtained from a pre-trained model library. This model is obtained by learning and training a large amount of historical user behavior data, and can identify and distinguish the behavioral characteristics and preferences of different user groups. When obtaining the model, it is necessary to consider the latest data, the accuracy of the model, and the special needs of the current business scenario to ensure that the selected model can accurately reflect the user group type of the target object.

[0052] Step S202: Analyze the behavior information based on the target cluster analysis model to determine the user group type.

[0053] In step S202, after determining the target cluster analysis model, the system will analyze and process the collected behavior information. The specific process includes:

[0054] Data preprocessing: Clean and standardize behavioral information to ensure that data quality and structure meet model requirements.

[0055] Feature extraction: Extract features related to user group types from behavioral information, such as operation frequency, dwell time, number of clicks, user roles, business attributes, etc.

[0056] Cluster assignment: The extracted feature vector is input into the target cluster analysis model, and the model will assign the target object to the corresponding user group type according to the similarity of the features.

[0057] Result output: The system outputs the type of user group to which the target object belongs, which will be used for subsequent function recommendations.

[0058] Based on steps S201 to S202, the system can accurately determine the user group type based on the target cluster analysis model by analyzing the behavior information of the target object. This determination method is the basis for personalized function recommendation, ensuring that the recommended content can truly meet the needs and preferences of the target object, thereby improving the efficiency of airport operators and user satisfaction.

[0059] Optionally, based on the user group type, recommendation information suitable for commonly used functional controls of the target object is generated, including:

[0060] Step S211, based on the user group type and the target cluster analysis model, determine the commonly used function controls of the user group type.

[0061] In step S211, the previously determined user group type and target cluster analysis model are used to identify and determine the function controls most commonly used by members of the user group. Specifically, the following sub-steps are included:

[0062] Function frequency statistics: Analyze all historical behavior data belonging to the same user group type and count the usage frequency of each function control.

[0063] High-frequency function identification: Based on statistical results, the most frequently used function controls in a user group type are identified. These functions are usually closely related to the job responsibilities, business needs, and operating preferences of the user group.

[0064] Feature analysis: Analyze the usage scenarios and user behavior characteristics of high-frequency function controls to further understand what factors make these function controls particularly popular among this user group.

[0065] Step S212: Generate recommendation information based on the commonly used function controls.

[0066] In step S212, after determining the commonly used function controls of the user group type, personalized recommendation information will continue to be generated so that these functions are preferentially displayed when the target object uses the system. The generation of recommendation information may include the following steps:

[0067] Recommendation list construction: The system arranges high-frequency function controls in order of priority and constructs a recommendation list.

[0068] Recommendation information formatting: Convert the recommendation list into the information format displayed in the user interface, which may include icons, names, shortcut key prompts, etc.

[0069] Recommended information display: When the target object logs in to the system and uses it, the recommended information will be displayed in a prominent position on the user interface, such as a toolbar, drop-down menu, or quick access area.

[0070] Real-time update: Based on the real-time operations and feedback of the target object, the system may need to update the recommendation information in real time to ensure that the recommended content is consistent with the user's current needs.

[0071] Based on steps S211 to S212, recommendation information that meets the specific needs of the target object can be generated, significantly improving the work efficiency and user experience of airport operators. This process reflects the application of personalized recommendation technology in the civil aviation departure system. Through data analysis and machine learning technology, it can intelligently adapt to the work modes of different user groups and provide more accurate and efficient services.

[0072] Optionally, before obtaining the target cluster analysis model, the following steps are included:

[0073] Step S221, obtaining historical user behavior information, wherein the historical user behavior information includes operation record information, stay time information, click count information, user role information, and business attribute information.

[0074] In step S221, various user behavior indicators are covered, including operation record information (i.e., each operation performed by the user and its results), dwell time information (the time the user spends on a specific page or function), click count information (the frequency with which a specific function or operation is clicked), user role information (such as check-in clerk, boarding clerk, etc.), and business attribute information (such as flight status, passenger type, etc.). This information provides a comprehensive perspective to gain a deep understanding of the user's operating habits and business needs.

[0075] Step S222, pre-processing the operation record information, stay time information, click count information, user role information, and business attribute information to obtain pre-processed historical behavior information.

[0076] In step S222, the preprocessing stage aims to clean and format the data to ensure the accuracy of subsequent analysis. The preprocessing steps may include: data cleaning, removing outliers, missing values ​​and duplicate data; data conversion, converting non-numeric data into numeric data for machine learning model processing; data standardization, unifying data dimensions to avoid certain features from having a dominant influence on the model due to excessive numerical ranges; feature engineering, selecting and constructing features that are helpful for clustering analysis.

[0077] Step S223: extracting user feature vectors based on the pre-processed historical behavior information.

[0078] In step S223, after data preprocessing, a series of features related to user behavior and needs are extracted based on the behavior information, and these features will be organized into a feature vector for subsequent cluster analysis. Features may include: operation frequency, stay time, number of clicks, user role code, business attribute code, etc. Each historical user will be represented as a feature vector.

[0079] Step S224: construct an initial clustering analysis model based on the user feature vector.

[0080] In step S224, the user feature vectors are analyzed using a machine learning algorithm (such as K-means clustering) to divide the users into different groups. The construction of the initial model may require setting the number of clusters, selecting an initialization method, and defining a distance metric. The system will try to find natural groupings in the data based on the feature vectors and preliminarily determine the outline of the user group.

[0081] Step S225, optimizing the initial cluster analysis model to obtain a target cluster analysis model.

[0082] In step S225, the optimization strategy includes: adjusting the number of clusters, using indicators such as the elbow method or silhouette coefficient to evaluate and select the optimal number of clusters; retraining the model, retraining the model according to the optimized parameters (such as cluster centers); feature selection, selecting more effective features for clustering based on the contribution or correlation of the features; model evaluation, evaluating the generalization ability and stability of the model through cross-validation or holdout method to ensure the performance of the model on new data.

[0083] Based on steps S221 to S225, the system can build and optimize a target cluster analysis model to accurately identify and distinguish the behavior patterns of different user groups. This model will be used for real-time user behavior analysis and function recommendation to improve the operational efficiency and user experience of the departure front-end system. The whole process embodies the concept of data-driven decision-making and personalized service.

[0084] Optionally, the initial cluster analysis model is optimized to obtain a target cluster analysis model, including:

[0085] Step S231, determining the validity in response to an operation on a feedback control in the display interface, wherein the feedback control is a control for providing feedback on whether the recommendation information is valid.

[0086] In step S231, the display interface will include a feedback control for users to directly provide feedback on the effectiveness of the recommended information. When the user completes the operation, the system will display a series of recommended functions, and the user can express his evaluation of the recommended functions through the feedback control, such as "effective" or "ineffective". The system will record these feedbacks and calculate the effectiveness of the recommended information based on the ratio of the user's positive feedback (such as "effective") to the total number of feedbacks. For example, if a recommended function is marked as "effective" by users 70 times out of 100 displays, its effectiveness is 70%. The statistics of effectiveness enable the system to quantify the practicality of the recommended information and user satisfaction, providing data support for model optimization.

[0087] Step S232: Based on the effectiveness, the initial cluster analysis model is optimized to obtain a target cluster analysis model.

[0088] In step S232, the collected effectiveness data will be used to guide the optimization of the clustering analysis model. The specific operations are as follows: Analyze the recommended functions with low effectiveness and identify problems with their matching with the user group and the recommendation strategy. For example, if the effectiveness of a certain recommended function is low in multiple user groups, it may be necessary to re-evaluate the general demand for the function, or check whether there are problems in the data preprocessing and feature extraction. Adjust model parameters, such as the number of clusters, the initialization method of the clustering algorithm, or the distance measurement method, to improve the effectiveness of the recommendation. In some cases, it may be necessary to reselect or train the clustering model to more accurately capture user behavior patterns. Feature selection and weight adjustment, based on the effectiveness analysis, the system may need to re-examine which features contribute most to the accuracy of the model and the recommendation effect, and adjust the weights of these features.

[0089] Based on steps S231 and S232, user feedback is continuously collected, and the model is regularly evaluated and adjusted to continuously improve the effectiveness of the recommended information. Through such a mechanism, the system can more intelligently adapt to changes in user needs and provide more accurate and personalized function recommendations, thereby improving the work efficiency of airport operators, reducing operation time, and improving user experience and airport operation efficiency.

[0090] Optionally, the initial cluster analysis model is optimized to obtain a target cluster analysis model, including:

[0091] Step S241 , in response to the operation action on the function control in the display interface, determine the click rate, where the click rate is the frequency with which the target object clicks on the recommended function control.

[0092] In step S241, the target object's interaction with the function control on the display interface is monitored in real time, especially the clicks on the recommended function control. Each click event is recorded and the click rate of a specific function control is calculated, that is, the frequency of the target object clicking on a certain function within a certain period of time. The statistics of the click rate help the system evaluate which functions are most popular with users and which functions may not be fully used.

[0093] Step S242: Based on the click rate, the initial cluster analysis model is optimized to obtain a target cluster analysis model.

[0094] In step S242, the optimization process may include: feature weight adjustment: If the click-through rate of certain functions is significantly higher than that of other functions, this may indicate that these functions are more important to specific user groups. The system can adjust the feature weights accordingly so that functions with high click-through rates play a more important role in cluster analysis. Cluster quantity evaluation: By analyzing the click-through rate distribution of different functions, the system may need to re-evaluate the number of clusters. If certain functions show high click-through rates in multiple user groups, this may mean that the existing number of clusters or clustering methods need to be adjusted in order to more accurately identify differences in user needs. Model retraining: Based on the updated feature weights and number of clusters, the system retrains the cluster analysis model to reflect the latest user behavior data and click-through rate information. After optimization, the target cluster analysis model will more accurately reflect user behavior patterns, thereby generating a function recommendation list that better meets user needs.

[0095] Based on steps S241 to S242, the cluster analysis model can be continuously optimized to provide more personalized and efficient function control recommendations. This dynamic optimization mechanism based on actual user feedback is the key to improving recommendation quality, meeting user needs, and improving airport operation efficiency. In actual application scenarios, this mechanism can not only adapt to changes in user needs, but also improve system performance through real-time data feedback, thereby achieving better user experience and business processing efficiency.

[0096] Figure 2 is a flowchart of another code checking method according to one embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0097] Step S201, obtaining a target cluster analysis model.

[0098] Step S202: Analyze the behavior information based on the target cluster analysis model to determine the user group type.

[0099] Step S211, based on the user group type and the target cluster analysis model, determine the commonly used function controls of the user group type.

[0100] Step S212: Generate recommendation information based on the commonly used function controls.

[0101] Step S221, obtaining historical user behavior information, wherein the historical user behavior information includes operation record information, stay time information, click count information, user role information, and business attribute information.

[0102] Step S222, pre-processing the operation record information, stay time information, click count information, user role information, and business attribute information to obtain pre-processed historical behavior information.

[0103] Step S223: extracting user feature vectors based on the pre-processed historical behavior information.

[0104] Step S224: construct an initial clustering analysis model based on the user feature vector.

[0105] Step S225, optimizing the initial cluster analysis model to obtain a target cluster analysis model.

[0106] Step S231, determining the validity in response to an operation on a feedback control in the display interface, wherein the feedback control is a control for providing feedback on whether the recommendation information is valid.

[0107] Step S232: Based on the effectiveness, the initial cluster analysis model is optimized to obtain a target cluster analysis model.

[0108] Step S241 , in response to the operation action on the function control in the display interface, determine the click rate, where the click rate is the frequency with which the target object clicks on the recommended function control.

[0109] Step S242: Based on the click rate, the initial cluster analysis model is optimized to obtain a target cluster analysis model.

[0110] Based on the above steps S201 to S203, in an embodiment of the present invention, a personalized and intuitive function recommendation interface can be provided for each target object, thereby optimizing its operating experience and efficiency in the departure system, thereby solving the technical problem of low efficiency of business processing in the current departure system.

[0111] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0112] In an embodiment of the present invention, a function recommendation device is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As used below, the term "module" can implement a combination of software and / or hardware of a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0113] Figure 3 is a structural block diagram of a function recommendation device according to one embodiment of the present invention. Figure 3 As shown, the device comprises:

[0114] The receiving module 301 is used to receive a login request from a target object, wherein the login request is sent by a target object of a different or the same job position to request to log in to a target device;

[0115] The acquisition module 302 is used to obtain the behavior information of the corresponding target object in response to the login request, wherein the behavior information is used to handle the target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking;

[0116] Determination module 303, the determination module 303 is used to determine the type of user group to which the target object belongs based on the behavior information;

[0117] The generating module 304 is used to generate recommendation information suitable for commonly used function controls of the target object based on the user group type.

[0118] Optionally, the generating module 304 is further used to generate a display interface based on the recommendation information, where the display interface includes one or more recommendation function controls for displaying the preferences of the target object.

[0119] Optionally, the acquisition module 302 is further used to acquire a target cluster analysis model; the determination module 303 is further used to analyze the behavior information based on the target cluster analysis model to determine the user group type.

[0120] Optionally, the generating module 304 is further configured to determine common function controls of the user group type based on the user group type and the target clustering analysis model; and generate recommendation information based on the common function controls.

[0121] Optionally, the function recommendation device also includes a training module 305, which is used to obtain historical user behavior information, wherein the historical user behavior information includes operation record information, stay time information, number of clicks information, user role information, and business attribute information; pre-processing the operation record information, stay time information, number of clicks information, user role information, and business attribute information to obtain pre-processed historical behavior information; extracting user feature vectors based on the pre-processed historical behavior information; constructing an initial clustering analysis model based on the user feature vector; and optimizing the initial clustering analysis model to obtain a target clustering analysis model.

[0122] Optionally, the training module 305 is further used to determine the validity in response to an operation on a feedback control in the display interface, wherein the feedback control is a control for providing feedback on whether the recommendation information is valid;

[0123] Based on the effectiveness, the initial cluster analysis model is optimized to obtain the target cluster analysis model.

[0124] Optionally, the training module 305 is also used to determine the click-through rate in response to the operation action on the function control in the display interface, where the click-through rate is the frequency of the target object clicking the recommended function control; based on the click-through rate, the initial clustering analysis model is optimized to obtain the target clustering analysis model.

[0125] It should be noted that the above modules can be implemented by software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0126] Figure 4 is a structural block diagram of a function recommendation device according to one embodiment of the present invention. Figure 4 As shown, the device comprises:

[0127] 1) Departure user data collection module (receiving module 301 and acquisition module 302): Use the departure system client to transmit back the user's operation times, stay time, click times, user role, business attributes and other data, and parse and store them in the following dimensions.

[0128] a. User role: the user's identity or role (such as general check-in clerk, check-in supervisor, boarding clerk, etc.).

[0129] b. Business attributes: corresponding to flight and passenger attributes in the departure system, such as flight status such as open check-in, check-in completed, boarding started, etc., and characteristic information such as passengers in wheelchairs, VIP, checked-in but no luggage added, etc. For example: flight status - open check-in, passenger check-in status - checked-in, passenger type - ordinary / elderly / infant / high-end, luggage status - with / without free baggage allowance, ticket status - unused / checked-in / refunded, itinerary type - domestic one-way / domestic connecting / international one-way / mixed connecting, seat status - reserved / not reserved, etc.

[0130] c. Function operation frequency: the number of times users use each function.

[0131] d. Function usage / residence time: the average time users spend on each function.

[0132] e. Functional operation path: the path that a user follows from start to finish when using a specific function to handle a business.

[0133] In addition, the following information may be included:

[0134] User company size: airports with tens of millions of employees / airports with millions of employees;

[0135] Counter attributes: check-in / baggage check-in / high-end / staff;

[0136] Business attributes: check-in / boarding;

[0137] Airport area: South China / Southwest China, etc.;

[0138] User account creation time: 2013.3.1; User login time: 2013.5.1 08:30; User system exit time: 2013.5.1 20:28;

[0139] Operation behavior: operation business ID, operation content, operation method, operation time, operation result, error content, etc.

[0140] 2) Departure user grouping module (determination module 303): based on the collected behavior data and in combination with the user's business role and business attributes, the user is grouped using a clustering algorithm to obtain different user groups.

[0141] a. Data preprocessing: Clean the collected behavioral data, such as cleaning up some missing, duplicate and dirty data, to facilitate subsequent analysis.

[0142] b. Feature extraction: Extract the user's feature vector from the preprocessed data. For example:

[0143] Basic information characteristics: Employee attributes: new employee / old employee; Employee activity: active employee / general employee / inactive employee; Employee airport scale: large scale / small scale; Employee business scope: many businesses / few businesses; Employee business attributes: check-in / boarding, etc.

[0144] Behavioral characteristics: operation preferences: mouse / shortcut keys; operation ID, operation content, operation method, etc. of each business attribute.

[0145] c. Clustering algorithm processing: Since all types of data are discrete data, the K-Means clustering algorithm is selected. After a large number of historical features are extracted and trained as mentioned above, the clustering results are output in the following format:

[0146] 1) Customer group name: Large-scale airport general check-in new employees handling checked baggage business customers;

[0147] 2) Customer group characteristics: Employee attributes - new employees; Employee airport scale - large scale; Employee role - ordinary employee; Employee business - check-in; Service object check-in status - checked-in; Itinerary status - domestic one-way; Seat selection status - seat selected; Baggage status - free baggage allowance;

[0148] 3) Behavioral characteristics: High-frequency operation business IDs - Add luggage 11110 / View passenger information 23423; High-frequency operation paths - Extract passenger information 34353@Add luggage 11110 / Ticket counter page 879898.

[0149] It should be noted that the K-means algorithm is the most commonly used clustering algorithm. The main idea is: Given the value of K and K initial cluster centers, each point (i.e., data record) is assigned to the cluster represented by the cluster center closest to it. After all points are assigned, the center of the cluster is recalculated (taking the average value) based on all points within a cluster, and then the steps of assigning points and updating the cluster center are iterated until the change in the cluster center is very small or the specified number of iterations is reached.

[0150] The basic principle of the K-Means clustering algorithm is as follows:

[0151] Assume that the given data sample X contains n objects X = {X1, X2, X3,..., Xn}, where each object has attributes of m dimensions. The goal of the Kmeans algorithm is to cluster the n objects into the specified k clusters based on the similarity between objects, and each object belongs to and only belongs to one cluster with the smallest distance to the cluster center. For Kmeans, first, k clustering centers {C1, C2, C3,..., Ck} need to be initialized, 1 < k ≤ n, and then the Euclidean distance from each object to each clustering center is calculated as shown in the following formula:

[0152]

[0153] In the above formula, Xi represents the i-th object 1 ≤ i ≤ n, Cj represents the j-th clustering center 1 ≤ j ≤ k, Xit represents the t-th attribute of the i-th object 1 ≤ t ≤ m, and Cjt represents the t-th attribute of the j-th clustering center.

[0154] Compare the distance from each object to each clustering center in turn, and assign the object to the cluster of the clustering center with the closest distance, obtaining k clusters {S1, S2, S3,..., Sk}.

[0155] The K-means algorithm defines the prototype of the cluster with the center. The cluster center is the mean value of all objects within the cluster in each dimension, and its calculation formula is as follows:

[0156]

[0157] Where Cl represents the center of the l-th cluster, 1≤l≤k1≤l≤k, |Sl| represents the number of objects in the l-th cluster, Xi represents the i-th object in the l-th cluster, 1≤i≤|Sl|.

[0158] d. Clustering result evaluation: The elbow method is used to evaluate the quality of clustering results to ensure the quality of clustering. For example, by observing the curve, find the inflection point where the SSE value drops sharply as the number of clusters increases and then tends to stabilize. The number of clusters corresponding to this inflection point is considered to be the optimal number of clusters.

[0159] The core indicator of the elbow method is SSE (sum of the squared errors):

[0160]

[0161] Among them, Ci is the i-th cluster, p is the sample point in Ci, mi is the centroid of Ci (the mean of all samples in Ci), and SSE is the clustering error of all samples, which represents the quality of clustering effect. The core idea of ​​the elbow method is: as the number of clusters k increases, the sample division will be more refined, and the degree of aggregation of each cluster will gradually increase, so the error square sum SSE will naturally gradually decrease. In addition, when k is less than the actual number of clusters, since the increase in k will greatly increase the degree of aggregation of each cluster, the decrease in SSE will be large. When k reaches the actual number of clusters, the return of the degree of aggregation obtained by increasing k will rapidly decrease, so the decrease in SSE will drop sharply, and then tend to be flat as the k value continues to increase. In other words, the relationship between SSE and k is in the shape of an elbow, and the k value corresponding to this elbow is the actual number of clusters of the data.

[0162] 3) User group function recommendation module (generation module 304): for each user group, analyze their common function requirements and generate a corresponding function recommendation list. The specific implementation is as follows:

[0163] a. Automatic identification of customer groups: When departing users use the system, the trained clustering model is used to automatically identify user groups. For example, if employee Zhang San logs into the system, extracts service object information, and calls the user clustering model to identify that the user group currently using the system belongs to the customer group of new employees of large-scale airport general check-in who handle checked baggage business (the passenger is a one-way domestic itinerary, has selected a seat in advance, has checked in, has a free baggage allowance, and the flight is about to take off); if it is still early before the flight takes off, the customer group is new employees of large-scale airport general check-in who handle seat change business, etc.

[0164] b. Recommendation list generation: Based on the user group classification, a corresponding common function recommendation list is generated for each user group based on the function operation frequency, operation time and operation frequency of each user group. For example, the recommended function list for the customer group of the large-scale airport general check-in new employee handling checked baggage business is to add luggage and view passenger information; while the recommended function list for the user group of the large-scale airport general check-in new employee handling seat change business is to modify seats and view flight information (number of booked persons).

[0165] 4) Departure front-end display module: Display the recommended function list to users at the front end of the departure system, so that users can quickly find the required functions.

[0166] 5) Feedback learning module: Collect user feedback on the recommendation function and continuously optimize the clustering model and function recommendation strategy. For example, the efficiency of the recommendation function is counted by counting the click rate of users of the recommendation function and displaying the "valid / invalid" feedback entry in the front-end function recommendation interface of the departure system. The recommended content with low effectiveness is combined with the customer group and the latest historical behavior data to train the user clustering model again, output and display new recommended content on the front end, and repeatedly evaluate and train.

[0167] According to one embodiment of the present invention, there is also provided an electronic device, comprising: a memory storing an executable program; and a processor for running the program, wherein the above-mentioned function recommendation method is executed when the program is running.

[0168] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0169] Step S10, receiving a login request from a target object, wherein the login request is sent by a target object of a different or the same job position to request to log in to a target device.

[0170] Step S12, in response to the login request, obtain the behavior information of the corresponding target object, wherein the behavior information is used to handle the target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking.

[0171] Step S14: determining the type of user group to which the target object belongs based on the behavior information.

[0172] Step S16: Generate recommendation information of commonly used function controls suitable for the target object based on the user group type.

[0173] According to one embodiment of the present invention, a computer-readable storage medium is further provided. The computer-readable storage medium includes a stored executable program, wherein when the executable program runs, the device where the storage medium is located is controlled to execute the above-mentioned function recommendation method.

[0174] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0175] Step S10, receiving a login request from a target object, wherein the login request is sent by a target object of a different or the same job position to request to log in to a target device.

[0176] Step S12, in response to the login request, obtain the behavior information of the corresponding target object, wherein the behavior information is used to handle the target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking.

[0177] Step S14: determining the type of user group to which the target object belongs based on the behavior information.

[0178] Step S16: Generate recommendation information of commonly used function controls suitable for the target object based on the user group type.

[0179] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0180] According to one embodiment of the present invention, a computer program product is also provided, including a computer program, and the computer program implements the above-mentioned function recommendation method when executed by a processor.

[0181] Optionally, in this embodiment, the computer program product may be configured as a computer program for executing the following steps:

[0182] Step S10, receiving a login request from a target object, wherein the login request is sent by a target object of a different or the same job position to request to log in to a target device.

[0183] Step S12, in response to the login request, obtain the behavior information of the corresponding target object, wherein the behavior information is used to handle the target business for the object to be departed, wherein the target business includes check-in, ticket change, consignment, ticket refund, and ticket booking.

[0184] Step S14: determining the type of user group to which the target object belongs based on the behavior information.

[0185] Step S16: Generate recommendation information of commonly used function controls suitable for the target object based on the user group type.

[0186] This application achieves the following technical effects:

[0187] 1) Personalized recommendation: The device can group users by clustering algorithm based on their actual operation records, stay time, click times and other behavioral data, thereby obtaining different user groups. This clustering method enables the recommendation system to more accurately identify the needs of each user and achieve personalized recommendations.

[0188] 2) Self-learning capability: The device contains a feedback learning module to collect user feedback on recommended features. This feedback information is used to continuously optimize the clustering model and feature recommendation strategy to adapt to changing user needs. This self-learning mechanism enables the recommendation system to become more accurate and efficient over time.

[0189] 3) Improve user experience and airport operation efficiency: Through personalized recommendations and self-learning capabilities, the device can significantly improve user experience, allowing users to find the required functions faster and reduce operation time. At the same time, this also improves the airport's operational efficiency, because users can complete the departure process more efficiently and reduce the intervention and assistance of airport staff.

[0190] 4) Real-time: The device can collect user behavior data in real time, analyze and process it immediately, and generate a corresponding function recommendation list. This real-time feature enables the recommendation system to respond to changes in user needs in a timely manner and provide the latest recommendations.

[0191] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0192] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. 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 mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0193] 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 distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0194] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0195] If the integrated unit 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 all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0196] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A function recommendation method, characterized in that: include: Receiving a login request from a target object, wherein the login request is issued by the target object of a different or the same job position to request to log in to a target device; In response to the login request, obtaining behavior information corresponding to the target object, wherein the behavior information is used to handle target services for the object to be departed, wherein the target services include check-in, ticket change, consignment, ticket refund, and ticket booking; Based on the behavior information, determining the type of user group to which the target object belongs; Based on the user group type, recommendation information of commonly used function controls suitable for the target object is generated.

2. The method according to claim 1, characterized in that include: Based on the recommendation information, a display interface is generated, wherein the display interface includes one or more recommendation function controls for displaying the target object's preferences.

3. The method according to claim 2, characterized in that Determining the user group type based on the behavior information includes: Obtain target cluster analysis model; Based on the target cluster analysis model, the behavior information is analyzed to determine the user group type.

4. The method according to claim 3, characterized in that Generating the recommendation information of the commonly used function controls suitable for the target object based on the user group type includes: Determining the commonly used function controls of the user group type based on the user group type and the target cluster analysis model; The recommendation information is generated based on the commonly used function controls.

5. The method according to claim 3 or 4, characterized in that: Before obtaining the target cluster analysis model, the method includes: Obtaining historical user behavior information, wherein the historical user behavior information includes operation record information, stay time information, click count information, user role information, and business attribute information; Preprocessing the operation record information, stay time information, click count information, user role information, and business attribute information to obtain preprocessed historical behavior information; Extracting a user feature vector based on the preprocessed historical behavior information; Based on the user feature vector, construct an initial clustering analysis model; The initial cluster analysis model is optimized to obtain the target cluster analysis model.

6. The method according to claim 5, characterized in that Optimizing the initial cluster analysis model to obtain the target cluster analysis model includes: In response to an operation on a feedback control in the display interface, obtaining a validity degree, wherein the feedback control is a control for providing feedback on whether the recommendation information is valid; Based on the effectiveness, the initial cluster analysis model is optimized to obtain the target cluster analysis model.

7. The method according to claim 5, characterized in that Optimizing the initial cluster analysis model to obtain the target cluster analysis model includes: In response to an operation on the function control in the display interface, determining a click rate, wherein the click rate is a frequency at which the target object clicks on the recommended function control; Based on the click rate, the initial cluster analysis model is optimized to obtain the target cluster analysis model.

8. A function recommendation device, characterized in that: include: A receiving module, the receiving module is used to receive a login request from a target object, wherein the login request is sent by the target object of a different or the same job position to request to log in to a target device; an acquisition module, the acquisition module being used to acquire behavior information corresponding to the target object in response to the login request, wherein the behavior information is used to handle target services for the object to be departed, wherein the target services include check-in, ticket change, consignment, ticket refund, and ticket booking; A determination module, the determination module is used to determine the type of user group to which the target object belongs based on the behavior information; A generation module is used to generate recommendation information of commonly used function controls suitable for the target object based on the user group type.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored executable program, wherein when the executable program is run, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that The invention comprises a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 7.

Images