CN117892008A - A tool item recommendation method, device, electronic device, storage medium and computer program product - Google Patents

Abstract

The disclosure provides a tool item recommending method, a device, electronic equipment, a storage medium and a computer program product, and relates to the technical field of artificial intelligence, in particular to the technical field of machine learning and intelligent recommending. The method comprises the following steps: responding to the operation of triggering a tool item display interface entering a target application by a user, and acquiring user portraits, file browsing behaviors of the user and a history record of the use of the tool item by the user; predicting the recommendation probability of each tool item provided by the target application locally according to the user portrait, the file browsing behavior of the user and the history record of the user using the tool item; and selecting target tool items meeting preset conditions according to the recommendation probability of each tool item, and displaying the target tool items in the tool item display interface. In the scheme, the recommendation of the tool item is realized locally in the application program, the intervention of a back-end server is not needed, the network bandwidth cost is reduced to a certain extent, and the recommendation calculation is performed locally, so that the network transmission time is not needed, the time delay caused by network blockage is avoided, and the efficiency of recommending the tool item is further ensured.

Description

A tool item recommendation method, device, electronic device, storage medium and computer program product

Technical Field

The present disclosure relates to the field of artificial intelligence technology, in particular to the field of machine learning and intelligent recommendation technology, and specifically to a tool item recommendation method, device, electronic device, storage medium and computer program product.

Background technique

Mobile applications (Mobile Application, App for short) refer to applications designed for mobile devices (such as smartphones, tablets, etc.). These applications can run on corresponding operating systems, such as Android systems.

Currently, mobile applications typically provide a variety of tools and functions to improve user needs and experience, enhance user stickiness, collect and analyze data, gain competitive advantages, build ecosystems, and achieve commercialization goals.

Summary of the invention

The present disclosure provides a tool item recommendation method, device, electronic device, storage medium and computer program product.

According to one aspect of the present disclosure, a tool item recommendation method is provided, comprising:

In response to a user triggering an operation to enter a tool item display interface of a target application, obtaining a user portrait, a user's file browsing behavior, and a user's tool item usage history;

Predicting the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior, and the user's history of using the tool item;

According to the recommendation probability of each tool item, a target tool item that meets the preset conditions is selected and displayed in the tool item display interface.

According to another aspect of the present disclosure, a tool item recommendation device is provided, comprising:

A data acquisition module, for acquiring a user portrait, a user's file browsing behavior, and a user's history of using tool items in response to a user triggering an operation to enter a tool item display interface of a target application;

A prediction module, configured to predict the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior and the user's history of using the tool item;

The recommendation module is used to select target tool items that meet preset conditions according to the recommendation probability of each tool item, and display them in the tool item display interface.

According to another aspect of the present disclosure, there is provided an electronic device, comprising:

at least one processor; and

a memory communicatively connected to the at least one processor; wherein,

The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the tool item recommendation method described in any embodiment of the present disclosure.

According to another aspect of the present disclosure, a non-transitory computer-readable storage medium storing computer instructions is provided, wherein the computer instructions are used to enable a computer to execute the tool item recommendation method described in any embodiment of the present disclosure.

According to another aspect of the present disclosure, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the tool item recommendation method of any embodiment of the present disclosure is implemented.

According to the technology disclosed in the present invention, the recommendation of tool items is implemented locally in the application without the intervention of the backend server, which reduces the network bandwidth cost to a certain extent. Moreover, since the recommendation calculation is performed locally, there is no need to borrow the network, so there is no need for network transmission time, which avoids the delay caused by network congestion, thereby ensuring the efficiency of recommending tool items. In addition, in the process of recommending tool items, in addition to referring to the user portrait, the user's history of using tool items and the user's file browsing behavior can be referenced to ensure that the recommended tool items are more likely to be clicked and used by the user.

It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description.

BRIEF DESCRIPTION OF THE DRAWINGS

The accompanying drawings are used to better understand the present solution and do not constitute a limitation of the present disclosure.

FIG1 is a schematic diagram of a flow chart of a tool item recommendation method according to an embodiment of the present disclosure;

FIG2 is a flow chart of another tool item recommendation method according to an embodiment of the present disclosure;

FIG. 3 is a flow chart of another tool item recommendation method according to an embodiment of the present disclosure.

FIG4 is a schematic diagram of the structure of a tool item recommendation device according to an embodiment of the present disclosure;

FIG. 5 is a block diagram of an electronic device for implementing the tool item recommendation method according to an embodiment of the present disclosure.

Detailed ways

The following is a description of exemplary embodiments of the present disclosure in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

When displaying the tool items provided by the mobile application, the display order of the tool items is recommended by the mobile terminal according to a simple strategy provided by the product manager, or recommended by the backend according to a preset strategy, and then the determined recommendation results are sent to the mobile terminal through the network for display. These strategies can be simple or complex. Simple strategies can be directly written as a fixed order or recommended based on the name or frequency of use of big data; complex recommendation strategies can be recommended through collaborative filtering methods, where collaborative filtering methods refer to pushing tool item usage records of users with the same characteristics to the other party as a recommendation tool. This complex recommendation requires the backend server to calculate the usage records of other users and needs to communicate through the network. However, this method may have the problem of untimely recommendations due to network delays. Based on this, the present disclosure proposes a method for recommending tool items locally in an application. The process of this method can be seen in the following embodiment.

Figure 1 is a schematic diagram of a tool item recommendation method according to an embodiment of the present disclosure, which can be applied to a scenario where tool items are recommended locally in an application. The method can be performed by a tool item recommendation device, which is implemented in software and/or hardware and configured in an electronic device.

As shown in FIG1 , the method specifically includes the following steps:

S101. In response to a user triggering an operation of entering a tool item display interface of a target application, a user portrait, a user's file browsing behavior, and a user's tool item usage history are obtained.

In this embodiment, the target application refers to an application installed on a mobile device such as a smart phone or a tablet computer. Exemplarily, the target application can be a network disk application or a browser application, which is not specifically limited here. The tool items provided by the target application refer to the functions within the target application, and can be functional components within the target application. Taking the target application as an example of a network disk application, the tool items provided by the target application can exemplarily include PDF watermark removal, conversion to PDF, image watermark removal, image compression, etc. When the target application is a network disk application, the disclosed solution can accurately and timely recommend the tool items needed by network disk users.

It should be noted that the tool items provided by the target application are different from those of the mini-program. Taking the target application as an example, the tool items of the network disk application are part of the network disk application and belong to the internal functions of the network disk application. They are mainly used to help users manage and maintain their files in the network disk. In addition, they can also be used to manipulate local files on the mobile terminal. Mini-programs are lightweight applications that run on specific platforms such as WeChat and Alipay. Therefore, the tool items provided by the target application are not the same as mini-programs, which makes the method of recommending mini-programs locally different from the method of recommending tool items locally.

For example, for easy management, a tool item display interface is set up in the target application to display tool items recommended to users. When the user enters the tool item display interface of the target application through touch operation, the recommendation of tool items is triggered. In order to implement tool item recommendations locally in the target application, it is necessary to obtain user portraits, user file browsing behaviors, and user tool item usage history records; and user portraits, user file browsing behaviors, and user tool item usage history records can usually be saved as user data locally in the target application, so these three types of data can be obtained locally in the target application. Wherein, the user portrait includes at least one of the user identity (e.g., student, office worker), user preference (e.g., study, entertainment) and user rights (e.g., membership rights purchased by the user); the user's file browsing behavior includes the current time (e.g., the time when the user triggers the tool item recommendation), the file type browsed by the user within the preset time before the current time, and the target file type determined according to the browsing frequency corresponding to the file type; wherein the target file type refers to the file type most frequently browsed by the user determined according to the browsing frequency; the user's history of using tool items includes the tool items used by the user within the preset time before the current time, the type of the tool item, the commonly used tools determined according to the use frequency of the tool item, and the commonly used tool item type determined according to the use frequency corresponding to each type; that is, the user's history of using tool items mainly includes the tool items used by the user recently, the type of the recently used tool item, the tool item used most frequently by the user, and the tool item type used most frequently by the user. It can be understood that a tool item type can correspond to a tool item. For example, the PDF type of the tool item type corresponds to a tool item such as PDF reading, PDF watermark removal, and conversion to PDF.

S102: Predicting the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior, and the user's history of using the tool items.

Optionally, locally in the target application, based on the user portrait, the user's file browsing behavior and the user's historical records of using tool items, a machine learning method or a data statistical method is used to predict the recommendation probability of each tool item provided by the target application; wherein the recommendation probability is used to measure the possibility of a tool item being recommended and displayed in the tool item display interface.

It is understandable that when using machine learning to predict the recommendation probability of a tool item, a tool recommendation model can be trained through machine learning methods and deployed locally in the target application, and the user portrait, the user's file browsing behavior and the user's history of using tool items can be used as model inputs. The recommendation probability of the tool item can then be determined based on the model output.

When using data statistics to predict the recommendation probability of a tool item, you can also directly perform data analysis on the user portrait, the user's file browsing behavior, and the user's history of using the tool item to determine the tool items that the user may use, and determine the possibility of being used by setting a probability value.

It should be noted that the reason why the disclosed solution chooses to use user portraits, user file browsing behaviors and user tool item usage history records to predict the recommendation probability of each tool item provided by the target application, rather than using a single user portrait to predict the recommendation probability of each tool item, is that the user portrait will basically not change in a short period of time. If the tool item recommendation probability is predicted based only on the user portrait, there will be problems such as untimely and inaccurate recommended tool items. This problem can be avoided by predicting based on the three types of data disclosed in the present disclosure.

S103: According to the recommendation probability of each tool item, select target tool items that meet preset conditions, and display them in the tool item display interface.

In an optional implementation, the tool items can be reordered according to the recommendation probability of each tool item; for example, they can be reordered in descending order of recommendation probability; and then based on the reordering results, target tool items that meet preset conditions can be selected; wherein the preset conditions can be that the recommendation probability is ranked in the top N, or that the recommendation probability is greater than a preset probability threshold; thus, when selecting target tool items based on the reordering results, the tool items with a recommendation probability greater than the preset probability threshold can be selected as target tool items, or the tool items with a recommendation probability ranked in the top N can be selected as target tool items. In this way, the tool items that users are most likely to use can be selected and displayed to users, thereby ensuring that users can easily and quickly reach the tool items they need.

In this embodiment, after the locally recommended tool items are displayed in the tool item display interface of the target application, the tool items actually used by the user later, the types of tool items, and the files browsed can also be recorded. At the same time, it can also be determined whether the user has used tool items or tool item types that have not been used before, or has browsed files or file types that have not been seen before, that is, whether the user's file browsing behavior and the user's history of using tool items have changed. It should be noted that the recording of the tool items used by the user and the files browsed is based on the user's authorization. On this basis, if the user exits the tool item display interface and triggers to enter the tool item display interface of the target application again in a short period of time, it is first determined whether the user's file browsing behavior and the user's history of using tool items have changed. If no change occurs, the target tool item determined last time is recommended to the user. If a change occurs, a new user portrait, the user's file browsing behavior, and the user's history of using tool items are obtained, and the recommendation is re-performed according to the steps S102-S103.

In this embodiment, the recommendation of tool items is implemented locally in the application, without the intervention of the backend server, which reduces the network bandwidth cost to a certain extent. In addition, since the recommendation calculation is performed locally, there is no need to borrow the network, so there is no need for network transmission time, avoiding delays caused by network congestion, thereby ensuring the efficiency of recommending tool items. In addition, in the process of recommending tool items, in addition to referring to the user portrait, the user's history of using tool items and the user's file browsing behavior can be referenced to ensure that the recommended tool items are more likely to be clicked and used by the user.

FIG2 is a flow chart of another tool item recommendation method according to an embodiment of the present disclosure. As shown in FIG2 , the tool item recommendation method specifically includes the following steps:

S201. In response to a user triggering an operation of entering a tool item display interface of a target application, a user portrait, a user's file browsing behavior, and a user's history of using tool items are obtained.

In this embodiment, the process of predicting the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior and the user's history of using the tool items can be seen in steps S202-S203.

S202: Perform feature encoding on the user portrait, the user's file browsing behavior, and the user's tool item usage history in the target application to obtain corresponding feature values.

In this embodiment, a preset encoding algorithm (such as a one-hot encoding algorithm) may be used to perform feature encoding on the user portrait, the user's file browsing behavior, and the user's history of using tool items to obtain corresponding feature values.

S203 : Predict the recommendation probability of each tool item provided by the target application based on the obtained feature value and in combination with a tool recommendation model pre-deployed locally on the target application.

In this embodiment, the tool recommendation model pre-trained and deployed locally in the target application may be a decision tree model. Since the decision tree model is a model that allows multiple feature inputs, the feature values of different data obtained in step S202 may be input into the tool recommendation model in a preset format, and then the recommendation probability of the tool item may be determined based on the output of the tool recommendation model.

It should be noted that the decision tree model is chosen as the tool recommendation model because it is more accurate than other logistic regression models and there is no need to worry about overfitting. In addition, the decision tree model is smaller in size and faster in inference than the support vector machine model or the neural network model, which can ensure the efficiency of tool item recommendation.

S204: According to the recommendation probability of each tool item, select target tool items that meet preset conditions, and display them in the tool item display interface.

In this embodiment, the tool item recommendation is performed locally by machine learning, which can ensure the efficiency of tool item recommendation. In addition, the decision tree model is selected as the tool recommendation model to avoid model overfitting, and the model size is small, and the space occupied by local deployment is small.

FIG3 is a flow chart of another tool item recommendation method according to an embodiment of the present disclosure. As shown in FIG3 , the tool item recommendation method specifically includes the following steps:

S301: Train the tool recommendation model using pre-acquired training samples.

Each training sample includes the tool items used by the user, as well as the user's history of using other tool items before using the tool item (including the user's frequently used tool items, frequently used tool item types, recently used tool items, and recently used tool item types), user portraits (including user identity, preferences, etc.), and user file browsing behaviors (including the file types frequently browsed by the user, the file types recently used, etc.). In the training project, frequently used tool items, frequently used tool item types, recently used tool items, and recently used tool item types are each a list, and the list can contain N tool items, which need to be encoded. The others are all single values and only need to be mapped to a data value. In this way, the encoded and mapped data is used as the input of the model for training, and the recommendation probability of each tool item output by the model is compared with the tool items used by the user included in the sample, and the model parameters are adjusted according to the comparison results. In this way, through continuous training, a tool recommendation model that predicts the recommendation probability of tool items can be obtained.

In this embodiment, the trained tool recommendation model can be trimmed and optimized to reduce the model size and avoid occupying a large amount of space when the tool recommendation model is deployed locally in the target application.

S302: Convert the trained tool recommendation model into a model file in a target format according to the system type of the terminal where the target application is located.

The system type running on the terminal where the target application is located may be a common IOS system, an Android system, or other systems that can run on mobile terminals, which is not specifically limited here.

In this embodiment, for the sake of convenience, the system type of the terminal running is an IOS system. For the IOS system, the tool recommendation model can be converted into a model file in a target format using the format conversion tool coremltools, where the target format can be mlmodel format.

S303: Compile the model file in the target format to obtain a compiled model file.

Exemplarily, after obtaining the model file in the mlmodel format, in order to ensure that the model can be deployed in the target application, the model file in the mlmodel format can be compiled to obtain a compiled model file (mlmodelc file).

S304: compress the compiled model file to obtain a model file compression package corresponding to the tool recommendation model, and save the model file compression package.

Exemplarily, a compression algorithm may be used to compress the mlmodelc file to obtain a model file compression package, which may be a zip file package, and then the model file compression package is saved. The saved model file compression package is essentially an offline package.

It should be noted that steps S301 to S304 can be performed in a local machine or in a server, and are not specifically limited here.

S305. When the user logs in to the target application for the first time, the model file compression package is obtained; the model file compression package is decompressed so as to locally deploy the tool recommendation model in the target application.

In order to avoid the target application installation package being too large, the disclosed solution does not directly integrate the model file compression package into the target application installation package. Instead, when the user installs the target application and logs in to the target application for the first time, the model file compression package is obtained from the server or the local machine, and the model file compression package is decompressed to obtain the mlmodelc file. In this way, the target application can use the mlmodelc file to perform recommendation predictions for tool items locally according to steps S306-S308.

S306: In response to the user triggering an operation to enter the tool item display interface of the target application, obtain a user portrait, a user's file browsing behavior, and a history record of the user using the tool items.

S307: Predict the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior, and the user's history of using the tool items.

S308: Select target tool items that meet preset conditions according to the recommendation probability of each tool item, and display them in the tool item display interface.

In this embodiment, the trained model file is converted into an offline package, and when the target application is installed and logged in for the first time, the offline package is obtained and the local deployment of the tool recommendation model is completed, avoiding the problem of directly integrating the model file into the installation package, which leads to the problem of the installation package being too large.

FIG4 is a schematic diagram of the structure of a tool item recommendation device according to an embodiment of the present disclosure. The present embodiment is applicable to a scenario where tool item recommendation is performed locally in an application. The device can implement the tool item recommendation method described in any embodiment of the present disclosure. As shown in FIG4 , the device 400 specifically includes:

The data acquisition module 401 is used to acquire the user portrait, the user's file browsing behavior and the user's history of using the tool items in response to the user triggering the operation of entering the tool item display interface of the target application;

Prediction module 402, used to predict the recommendation probability of each tool item provided by the target application based on the user portrait, the user's file browsing behavior and the user's history of using the tool item locally in the target application;

The recommendation module 403 is used to select target tool items that meet preset conditions according to the recommendation probability of each tool item, and display them in the tool item display interface.

In an optional implementation, the prediction module is further used to:

Performing feature encoding on the target application locally for the user portrait, the user's file browsing behavior, and the user's history of using tool items to obtain corresponding feature values;

According to the obtained feature values, combined with a tool recommendation model pre-deployed locally on the target application, the recommendation probability of each tool item provided by the target application is predicted.

In an optional implementation, the method further includes:

A training module, used to train the tool recommendation model using pre-acquired training samples; wherein each training sample includes a tool item used by a user, as well as a history of the user's use of the tool item before the user used the tool item, a user portrait, and the user's file browsing behavior;

A conversion module, used to convert the trained tool recommendation model into a model file of a target format according to the system type of the terminal where the target application is located;

A compiling module, used for compiling the model file in the target format to obtain a compiled model file;

The compression and saving module is used to compress the compiled model file, obtain the model file compression package corresponding to the tool recommendation model, and save the model file compression package.

In an optional implementation, the method further includes:

When the user logs in to the target application for the first time, obtaining the model file compressed package;

The model file compression package is decompressed to deploy the tool recommendation model locally in the target application.

In an optional implementation, the tool recommendation model is a decision tree model.

In an optional implementation, the recommendation module is further used to:

Re-rank the tool items according to their recommendation probability;

Based on the reordering results, target tool items that meet preset conditions are selected.

In an optional implementation, the user portrait includes at least one of user identity, user preference and user rights;

The file browsing behavior of the user includes the current time, the file types browsed by the user within a preset time period before the current time, and the target file type determined according to the browsing frequency corresponding to the file type;

The history record of the user's use of tool items includes the tool items used by the user within a preset time before the current time, the types of the tool items, common tools determined based on the use frequency of the tool items, and common tool types determined based on the use frequency corresponding to each type.

In an optional implementation, the target application is a network disk application.

The above-mentioned product can execute the method provided by any embodiment of the present disclosure, and has the corresponding functional modules and beneficial effects of the execution method.

In the technical solution of the present disclosure, the collection, storage, use, processing, transmission, provision and disclosure of user personal information involved are in compliance with the provisions of relevant laws and regulations and do not violate public order and good morals.

According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium and a computer program product.

FIG5 shows a schematic block diagram of an example electronic device 500 that can be used to implement an embodiment of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workbenches, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present disclosure described and/or required herein.

As shown in FIG5 , the device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. In the RAM 503, various programs and data required for the operation of the device 500 can also be stored. The computing unit 501, the ROM 502, and the RAM 503 are connected to each other via a bus 504. An input/output (I/O) interface 505 is also connected to the bus 504.

A number of components in the device 500 are connected to the I/O interface 505, including: an input unit 506, such as a keyboard, a mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a disk, an optical disk, etc.; and a communication unit 509, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 509 allows the device 500 to exchange information/data with other devices through a computer network such as the Internet and/or various telecommunication networks.

The computing unit 501 may be a variety of general and/or special processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special artificial intelligence (AI) computing chips, various computing units that execute machine learning model algorithms, digital signal processors (DSPs), and any appropriate processors, controllers, microcontrollers, etc. The computing unit 501 performs the various methods and processes described above, such as the tool item recommendation method. For example, in some embodiments, the tool item recommendation method may be implemented as a computer software program that is tangibly contained in a machine-readable medium, such as a storage unit 508. In some embodiments, part or all of the computer program may be loaded and/or installed on the device 500 via the ROM 502 and/or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the tool item recommendation method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the tool item recommendation method in any other appropriate manner (e.g., by means of firmware).

Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and/or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

The program code for implementing the method of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, so that the program code, when executed by the processor or controller, implements the functions/operations specified in the flow chart and/or block diagram. The program code may be executed entirely on the machine, partially on the machine, partially on the machine and partially on a remote machine as a stand-alone software package, or entirely on a remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, device, or equipment. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

The systems and techniques described herein may be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

A computer system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The relationship between the client and the server is generated by computer programs that are executed on the corresponding computers and have a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services. The server may also be a server of a distributed system, or a server combined with a blockchain.

Artificial intelligence is a discipline that studies how to use computers to simulate certain human thought processes and intelligent behaviors (such as learning, reasoning, thinking, planning, etc.). It includes both hardware-level and software-level technologies. Artificial intelligence hardware technologies generally include technologies such as sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, and big data processing; artificial intelligence software technologies mainly include computer vision technology, speech recognition technology, natural language processing technology, machine learning/deep learning technology, big data processing technology, knowledge graph technology, and other major directions.

Cloud computing refers to a technology system that uses network access to elastically scalable shared physical or virtual resource pools. Resources can include servers, operating systems, networks, software, applications, and storage devices, and can be deployed and managed on demand and in a self-service manner. Cloud computing technology can provide efficient and powerful data processing capabilities for technical applications such as artificial intelligence and blockchain, as well as model training.

It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions provided by this disclosure can be achieved, and this document does not limit this.

The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims (19)

1. A tool item recommendation method, comprising: In response to a user triggering an operation to enter a tool item display interface of a target application, obtaining a user portrait, a user's file browsing behavior, and a user's tool item usage history; Predicting the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior, and the user's history of using the tool item; According to the recommendation probability of each tool item, a target tool item that meets the preset conditions is selected and displayed in the tool item display interface. 2. The method according to claim 1, wherein the predicting the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior and the user's history of using the tool item comprises: Performing feature encoding on the target application locally for the user portrait, the user's file browsing behavior, and the user's history of using tool items to obtain corresponding feature values; According to the obtained feature values, combined with a tool recommendation model pre-deployed locally on the target application, the recommendation probability of each tool item provided by the target application is predicted. 3. The method according to claim 2, further comprising: The tool recommendation model is trained using pre-acquired training samples; wherein each training sample includes a tool item used by a user, as well as a history of the user's use of the tool item before the user used the tool item, a user portrait, and the user's file browsing behavior; According to the system type running on the terminal where the target application is located, converting the trained tool recommendation model into a model file in a target format; Compiling the model file in the target format to obtain a compiled model file; The compiled model file is compressed to obtain a model file compression package corresponding to the tool recommendation model, and the model file compression package is saved. 4. The method according to claim 3, further comprising: When the user logs in to the target application for the first time, obtaining the model file compressed package; The model file compression package is decompressed to deploy the tool recommendation model locally in the target application. 5. The method according to any one of claims 2 to 4, wherein the tool recommendation model is a decision tree model. 6. The method according to claim 1, wherein the step of selecting a target tool item that meets a preset condition according to the recommendation probability of each tool item comprises: Re-rank the tool items according to their recommendation probability; Based on the reordering results, target tool items that meet preset conditions are selected. 7. The method according to claim 1 or 2, wherein the user portrait includes at least one of user identity, user preference and user rights; The file browsing behavior of the user includes the current time, the file types browsed by the user within a preset time period before the current time, and the target file type determined according to the browsing frequency corresponding to the file type; The history record of the user's use of tool items includes the tool items used by the user within a preset time before the current time, the types of the tool items, common tools determined based on the use frequency of the tool items, and common tool types determined based on the use frequency corresponding to each type. 8. The method according to any one of claims 1-4, wherein the target application is a network disk application. 9. A tool item recommendation device, comprising: A data acquisition module, for acquiring a user portrait, a user's file browsing behavior, and a user's history of using tool items in response to a user triggering an operation to enter a tool item display interface of a target application; A prediction module, configured to predict the recommendation probability of each tool item provided by the target application locally based on the user portrait, the user's file browsing behavior and the user's history of using the tool item; The recommendation module is used to select target tool items that meet preset conditions according to the recommendation probability of each tool item, and display them in the tool item display interface. 10. The apparatus according to claim 9, wherein the prediction module is further configured to: Performing feature encoding on the target application locally for the user portrait, the user's file browsing behavior, and the user's history of using tool items to obtain corresponding feature values; According to the obtained feature values, combined with a tool recommendation model pre-deployed locally on the target application, the recommendation probability of each tool item provided by the target application is predicted. 11. The apparatus according to claim 10, further comprising: A training module, used to train the tool recommendation model using pre-acquired training samples; wherein each training sample includes a tool item used by a user, as well as a history of the user's use of the tool item before the user used the tool item, a user portrait, and the user's file browsing behavior; A conversion module, used to convert the trained tool recommendation model into a model file of a target format according to the system type of the terminal where the target application is located; A compiling module, used for compiling the model file in the target format to obtain a compiled model file; The compression and saving module is used to compress the compiled model file, obtain the model file compression package corresponding to the tool recommendation model, and save the model file compression package. 12. The apparatus according to claim 11, further comprising: When the user logs in to the target application for the first time, obtaining the model file compressed package; The model file compression package is decompressed to deploy the tool recommendation model locally in the target application. 13. The device according to any one of claims 10-12, wherein the tool recommendation model is a decision tree model. 14. The device according to claim 9, wherein the recommendation module is further configured to: Re-rank the tool items according to their recommendation probability; Based on the reordering results, target tool items that meet preset conditions are selected. 15. The device according to claim 9 or 10, wherein the user portrait includes at least one of user identity, user preference and user rights; The file browsing behavior of the user includes the current time, the file types browsed by the user within a preset time period before the current time, and the target file type determined according to the browsing frequency corresponding to the file type; The history record of the user's use of tool items includes the tool items used by the user within a preset time before the current time, the types of the tool items, common tools determined based on the use frequency of the tool items, and common tool types determined based on the use frequency corresponding to each type. 16. The device according to any one of claims 9 to 12, wherein the target application is a network disk application. 17. An electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the tool item recommendation method according to any one of claims 1 to 8. 18. A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to execute the tool item recommendation method according to any one of claims 1-8. 19. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the tool item recommendation method according to any one of claims 1 to 8.