WO2021259197A1

WO2021259197A1 - File processing method and apparatus, storage medium, and terminal

Info

Publication number: WO2021259197A1
Application number: PCT/CN2021/101220
Authority: WO
Inventors: 李文娟; 易明
Original assignee: 中兴通讯股份有限公司
Priority date: 2020-06-22
Filing date: 2021-06-21
Publication date: 2021-12-30
Also published as: CN113900996A

Abstract

The present application provides a file processing method and apparatus, a storage medium, and a terminal. The method comprises: using a second cleanup model to analyze a file stored in a terminal to determine whether the file needs to be cleaned up, wherein the second cleanup model is obtained by training a first cleanup model using training data, the training data comprises a first operation feature by a target object for a first cleanup result, the first cleanup result is obtained by analyzing, by means of the first cleanup model, the file stored in the terminal, the first cleanup result indicates a file to be cleaned up, the first operation feature is used for indicating the feature of an operation performed by the target object for a file of a specified type in the first cleanup result, and files of the same type have the same file feature; and if it is determined that the file needs to be cleaned up, displaying a second cleanup result, wherein the second cleanup result indicates the file to be cleaned up.

Description

File processing method and device, storage medium and terminal

Cross-references to related applications

This application claims the priority of the patent application No. 202010575231.0 submitted to the Chinese Patent Office on June 22, 2020, the entire content of which is incorporated herein by reference.

Technical field

The present disclosure relates to, but is not limited to, the field of communications.

Background technique

In related technologies, the cleaning model in the terminal can only mechanically scan the files in the terminal and display it to the user in the form of a list without distinction. It cannot push the list of files to be cleaned in a targeted manner, resulting in the accuracy of the cleaning model. Very low, users can only manually select the content they need to clean up one by one, and the cleaning efficiency is low.

Summary of the invention

The present disclosure provides a file processing method, including: using a second cleaning model to analyze a file stored in a terminal to determine whether the file needs to be cleaned, wherein the second cleaning model is using training data pair The training data is obtained by training the first cleaning model, and the training data includes: the first operation feature of the target object on the first cleaning result, where the first cleaning result is the first cleaning model to the terminal It is obtained by analyzing the stored files, the first cleaning result indicates the file to be cleaned, and the first operation feature is used to indicate the operation of the target object on the file of the specified type in the first cleaning result Features: files of the same type have the same file feature; in the case where it is determined that the file needs to be cleaned, a second cleaning result is displayed, where the second cleaning result indicates the file to be cleaned.

The present disclosure also provides a file processing device, including: an analysis module configured to analyze a file stored in a terminal using a second cleaning model to determine whether the file needs to be cleaned, wherein the second cleaning The model is obtained by training a first cleaning model using training data. The training data includes: the first operation feature of the target object on the first cleaning result, wherein the first cleaning result is obtained through the first cleaning result. The model is obtained by analyzing the files stored in the terminal, the first cleaning result indicates the file to be cleaned, and the first operating feature is used to instruct the target object to check the first cleaning result. Operating characteristics of files of a specified type, files of the same type have the same file characteristics; the display module is configured to display a second cleaning result when it is determined that the file needs to be cleaned, wherein the second cleaning result indicates The files to be cleaned up.

The present disclosure also provides a computer-readable storage medium in which a computer program is stored, wherein the computer program implements any of the methods described herein when executed by a processor.

The present disclosure also provides a terminal, including a memory and a processor, and a computer program is stored in the memory, and the processor is configured to run the computer program to execute any method described herein.

Description of the drawings

FIG. 1 is a block diagram of the hardware structure of a terminal that implements the file processing method of the present disclosure;

Figure 2 is a flowchart of a file processing method according to the present disclosure;

Fig. 3 is a structural block diagram of a file processing device according to the present disclosure;

Figure 4 is a schematic structural diagram of a terminal cleaning system according to the present disclosure;

FIG. 5 is a schematic diagram of recording user behavior according to the input module of the present disclosure;

Fig. 6 is a schematic diagram of analyzing and processing user behaviors according to the processing module of the present disclosure;

FIG. 7 is a schematic diagram of outputting cleaning results according to the output module of the present disclosure;

FIG. 8 is a schematic flowchart of a file cleaning method according to the present disclosure.

detailed description

Hereinafter, the embodiments of the present disclosure will be described in detail with reference to the drawings and in combination with the embodiments.

It should be noted that the terms "first" and "second" in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence.

The method provided in the present disclosure can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, FIG. 1 is a hardware structure block diagram of a terminal that implements the file processing method of the present disclosure. As shown in FIG. 1, the mobile terminal may include one or more (only one is shown in FIG. 1) processor 102 (the processor 102 may include, but is not limited to, a processing device such as a microprocessor MCU or a programmable logic device FPGA) And the memory 104 for storing data, wherein the above-mentioned mobile terminal may also include a transmission device 106 and an input/output device 108 for communication functions. Those of ordinary skill in the art can understand that the structure shown in FIG. 1 is only for illustration, and does not limit the structure of the above-mentioned mobile terminal. For example, the mobile terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration from that shown in FIG.

The memory 104 may be used to store computer programs, for example, software programs and modules of application software, such as the computer programs corresponding to the file processing method in the present disclosure. The processor 102 executes various computer programs by running the computer programs stored in the memory 104. Functional application and data processing, that is, to achieve the above-mentioned methods. The memory 104 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 examples, the memory 104 may further include a memory remotely provided with respect to the processor 102, and these remote memories may be connected to the mobile terminal through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of a mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, NIC for short), which can be connected to other network devices through a base station to communicate with the Internet. In an example, the transmission device 106 may be a radio frequency (Radio Frequency, referred to as RF) module, which is used to communicate with the Internet in a wireless manner.

The present disclosure provides a method for processing files running on the above-mentioned terminal. FIG. 2 is a flowchart of the method for processing files according to the present disclosure. As shown in FIG. 2, the method may include the following steps S202 and S204.

In step S202, a second cleaning model is used to analyze a file stored in the terminal to determine whether the file needs to be cleaned. The second cleaning model is obtained by training the first cleaning model using training data. The training data includes: the first operation feature of the target object on the first cleaning result, where the first cleaning result is obtained by analyzing the files stored in the terminal through the first cleaning model, and the first cleaning result The file to be cleaned is indicated, and the first operation characteristic is used to indicate the operation characteristic of the target object for the file of the specified type in the first clean-up result, and files of the same type have the same file characteristic.

In step S204, in the case where it is determined that the file needs to be cleaned, a second cleanup result is displayed, where the second cleanup result indicates the file to be cleaned.

Through the above steps, since the cleaning model is trained according to the operating characteristics of the cleaning result of the target object, the cleaning model is continuously optimized with the operating characteristics of the target object, so that the final cleaning result determined by the cleaning model is more in line with the behavior habits of the target object. Therefore, It can solve the problem of low accuracy of pushing the files to be cleaned in related technologies, and achieve the technical effect of improving the accuracy of pushing the files to be cleaned.

In an exemplary embodiment, the cleanup model may be a cleanup application or cleanup code.

In an exemplary embodiment, files of one type with the same file characteristics may be referred to as the same type of file or the same type of file. Among them, the same file feature can be the same file name, belonging to the same application, belonging to the same type of application, belonging to the same webpage, belonging to the same type of webpage, and the content of the file containing the same object (for example, it contains characters). , For another example, including a specific person), belonging to the same contact object, etc.

In an exemplary embodiment, the cleaning model may be trained through machine learning using training data. The operating characteristics included in the training data may include: operating behavior and the file type to which the operating behavior points.

In the present disclosure, a terminal (also referred to as a terminal device) may be a terminal with a communication function, and the terminal may communicate with other devices through a network or a connection line or a connection interface. The terminal in the present disclosure may include but is not limited to at least one of the following: mobile phones (such as Android phones, iOS phones, etc.), notebook computers, tablet computers, handheld computers, MID (Mobile Internet Devices), PAD, desktop computers , Smart TV, smart home equipment, etc. The aforementioned networks may include, but are not limited to: wired networks, wireless networks, where the wired networks include: local area networks, metropolitan area networks, and wide area networks, and the wireless networks include: Bluetooth, WIFI, and other networks that implement wireless communication. The above is only an example, and this disclosure does not make any limitation on this.

In an exemplary embodiment, before using the second cleaning model to analyze the files stored in the terminal, the method further includes: using training data to train the first cleaning model to obtain the second cleaning model, Wherein, the training of the first cleaning model using the training data includes: training the first cleaning model when the first operation feature indicates that the probability of the first type of file being retained is higher than the probability of the second type of file being retained. The model preferentially determines the second type of file as the file to be cleaned; and/or, in the case that the first operating characteristic indicates that the probability of the third type of file being cleaned is higher than the probability of the fourth type of file being cleaned, train the The first cleaning model preferentially determines the file of the third type as the file to be cleaned.

In an exemplary embodiment, the first operation feature is obtained according to the target object's operation behavior on the first cleaning result and the file pointed to by the operation behavior.

In an exemplary embodiment, the training data further includes: a second operating feature of the target object on the file stored in the terminal, wherein the second operating feature is used to indicate that the target object has the target object on the file stored in the terminal. The operating characteristics of the stored files of the specified type.

In an exemplary embodiment, before using the second cleaning model to analyze the files stored in the terminal, the method further includes: using training data to train the first cleaning model to obtain the second cleaning model, Wherein, the training of the first cleaning model by using the training data includes: training the first cleaning model in the case that the second operating feature indicates that the fifth type of file is accessed more frequently than the sixth type of file is accessed. The model prioritizes this sixth type of file as the file to be cleaned up.

In an exemplary embodiment, the second operating feature is obtained according to the target object's operating behavior on the file stored in the terminal and the file pointed to by the operating behavior.

In an exemplary embodiment, after the second cleaning result is displayed, the method further includes: using a third cleaning model to analyze a file stored in the terminal to determine whether the file needs to be cleaned, wherein the first cleaning model The third cleaning model is obtained by training the second cleaning model using training data. The training data includes: a third operation feature of the target object on the second cleaning result, where the third operation feature is used to indicate the The target object's operating characteristics of the file of the specified type in the second cleaning result; in the case where it is determined that the file needs to be cleaned, the third cleaning result is displayed, where the third cleaning result indicates the file to be cleaned.

It should be noted that in the present disclosure, the training of the cleaning model can be iterative. For example, each time the cleaning model outputs the cleaning result and receives the operation of the target object on the cleaning result, it can be based on this time (or as of the current time). The operation feature of the target object retrains the cleaning model, where the operation feature may be the operation feature of the cleaning result and/or the operation feature of the file stored in the terminal.

In an exemplary embodiment, the method further includes: using the second cleaning model to analyze a file stored in the terminal to determine whether the file in the second cleaning result needs to be sent to other than the terminal. External storage device; in the case where it is determined that the file needs to be sent to a storage device other than the terminal, the analysis result is displayed, where the analysis result indicates the file to be sent.

It should be noted that, in some cases, some files may need to be stored in other storage devices other than the terminal. After sending such files to other storage devices, such files will be deleted on this terminal. Therefore, in the present disclosure, when it is determined that a certain file is a file to be deleted, it can also be distinguished whether the file needs to be sent to other storage devices.

In an exemplary embodiment, before using the second cleaning model to analyze the files stored in the terminal, the method further includes: using training data to train the first cleaning model to obtain the second cleaning model, Wherein, the training of the first cleaning model using the training data includes: before the first operating feature and the second operating feature indicate that the seventh type of file is to be cleaned, the situation is sent to a storage device other than the terminal Next, train the first cleaning model to preferentially determine the seventh type of file as the file to be sent.

In an exemplary embodiment, before using the second cleaning model to analyze the files stored in the terminal, the method further includes: receiving a start signal through the terminal, wherein the start signal is used to instruct to start the second cleaning model. 2. Clean up the model.

In an exemplary embodiment, in the second cleaning result, the order of the file to be cleaned is determined according to the probability of the file to be cleaned being cleaned.

In an exemplary embodiment, after the displaying of the second cleaning result, the method further includes: receiving a target operation of the target object on the second cleaning result; and executing a corresponding operation on the second cleaning result according to the target operation operate.

Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, of course, it can also be implemented by hardware, but in many cases the former is Better implementation. Based on this understanding, the technical solution of the present disclosure essentially or the part that contributes to the existing technology can be embodied in the form of a software product, and the computer software product is stored in a storage medium (such as ROM/RAM, magnetic disk, The optical disc) includes several instructions to make a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) execute the methods described in the various embodiments of the present disclosure.

The present disclosure also provides a file processing device, which is used to implement any of the above-mentioned methods, and those that have been explained will not be repeated. As used below, the term "module" can implement a combination of software and/or hardware with predetermined functions. Although the devices described in the following embodiments are preferably implemented by software, implementation by hardware or a combination of software and hardware is also possible and conceived.

Fig. 3 is a structural block diagram of a file processing device according to the present disclosure. As shown in Fig. 3, the device includes: a first analysis module 31 configured to use a second cleaning model to analyze files stored in the terminal to determine the Whether the file needs to be cleaned, wherein the second cleanup model is obtained by training the first cleanup model using training data, and the training data includes: the first operation feature of the target object on the first cleanup result, where the first cleanup model A cleaning result is obtained by analyzing the file stored in the terminal through the first cleaning model, the first cleaning result indicates the file to be cleaned, and the first operating feature is used to indicate that the target object has The operating characteristics of files of a specified type in the cleaning result, and files of the same type have the same file characteristics; the first display module 33 is configured to display the second cleaning result when it is determined that the file needs to be cleaned. The second cleaning result indicates the file to be cleaned.

In an exemplary embodiment, the device further includes: a training module configured to use the training data to train the first cleaning model before using the second cleaning model to analyze the files stored in the terminal to obtain the The second cleaning model, wherein the training the first cleaning model using the training data includes: in the case that the first operating feature indicates that the probability of the first type of file being retained is higher than the probability of the second type of file being retained, Training the first cleaning model to prioritize determining the second type of file as a file to be cleaned; and/or, where the first operating feature indicates that the probability of the third type of file being cleaned is higher than the probability of the fourth type of file being cleaned In this case, the first cleaning model is trained to first determine the third type of files as files to be cleaned.

In an exemplary embodiment, the training module is further configured to use training data to train the first cleaning model to obtain the second cleaning model before analyzing the files stored in the terminal using the second cleaning model. Model, wherein the training the first cleaning model using the training data includes: training the first cleaning model when the second operating feature indicates that the fifth type of file is accessed more frequently than the sixth type of file is accessed. A cleaning model prioritizes the sixth type of file as a file to be cleaned.

In an exemplary embodiment, after the display of the second cleaning result, the device further includes: a second analysis module configured to use the third cleaning model to analyze the file stored in the terminal to determine whether the file is required Is cleaned up, where the third clean-up model is obtained by training the second clean-up model using training data, and the training data includes: the third operation feature of the target object on the second clean-up result, wherein the first The third operating feature is used to indicate the operating feature of the target object on the file of the specified type in the second cleaning result; the second display module is configured to display the third cleaning result when it is determined that the file needs to be cleaned, where: The third cleaning result indicates the file to be cleaned.

In an exemplary embodiment, the device further includes: a third analysis module configured to use the second cleaning model to analyze the file stored in the terminal to determine whether the file in the second cleaning result is required Is sent to a storage device other than the terminal; the third display module is configured to display the analysis result when it is determined that the file needs to be sent to a storage device other than the terminal, where the analysis result indicates The file to be sent.

In an exemplary embodiment, the training module is further configured to use training data to train the first cleaning model to obtain the second cleaning model before analyzing the files stored in the terminal using the second cleaning model. A model, wherein the training of the first cleaning model using the training data includes: sending to a storage device other than the terminal before the first operating feature and the second operating feature indicate that the seventh type of file is to be cleaned In the case of training, the first cleaning model is trained to preferentially determine the seventh type of file as the file to be sent.

In an exemplary embodiment, the device further includes: a receiving module configured to receive a start signal before analyzing the file stored in the terminal using the second cleaning model, wherein the start signal is used to indicate the start The second cleaning model.

In an exemplary embodiment, after the second cleaning result is displayed, the device further includes: an operation module configured to receive a target operation of the target object on the second cleaning result; and the second cleaning result according to the target operation Perform the corresponding operation on the cleanup result.

It should be noted that each of the above-mentioned modules can be implemented by software or hardware. For the latter, it can be implemented in the following way, but not limited to this: the above-mentioned modules are all located in the same processor; or, the above-mentioned modules are in any combination The forms are located in different processors.

Example implementation

The following further explains the embodiments of the present disclosure in combination with specific scenarios.

The method provided in this embodiment can be executed in a smart mobile terminal or a personal computer. It should be pointed out that the execution of the method in this embodiment requires the read and write permissions of the terminal, that is, the system permissions within the security range, so that the user's operation information can be obtained, similar to the permissions of a mobile phone housekeeper.

In this embodiment, an original file cleaning model is first placed in the terminal. This model can be a cleaning mechanism model. The default judgment parameters of the model include, but are not limited to: identifying images cached by the application, and images edited and modified by the user. Pictures with high browsing frequency, pictures in chat groups, pictures downloaded and saved by users, etc., can also identify various other file types, such as video, audio, and files.

This model will repeatedly collect and process the user's usage habits, so as to achieve long-term training, extract the characteristic points of the user's usage habits, and generate a new cleaning model after a stage. When the cleanup is triggered, the terminal outputs a cleanup list for the user to choose according to the new cleanup model, where the cleanup list includes a list of content to be cleaned determined by the cleanup model. The data generated by the user in the process of using the cleaning model can be used as input information in the embodiment of the present disclosure.

It should be noted that the embodiments of the present disclosure cannot accurately determine which data the user wants to delete and which data to retain at one time. Instead, it is more and more close to the user through repeated training. The real intention is the purpose, so as to achieve the purpose of self-learning and intelligence.

The user can perform direct operations on the output results (that is, the cleanup list output by the terminal) (for example, check, uncheck, retain, one-click cleanup, delete one or more pieces of data, etc.), and the terminal records the user's operations on the output results and The content pointed to by the operation, for example, record the content information that is directly cleaned up (that is, the information of the data stored in the terminal), and also record the content information that the user checks or keeps, as part of the input data for the next retraining , And the method basis of retraining. Exemplarily, the terminal may record all the user's operations on the output result and the content pointed to by all operations; it may also record part of the user's operation on the output result and the content pointed to by the part of the operation. Part of the operations that need to be recorded by the terminal may be preset, for example, may be operations related to the retention or deletion of content information, such as deletion operations and retention operations.

In an exemplary embodiment, the input data for the next retraining consists of two parts: one part is the generation of new user data between the last and the next output, for example, after the last output is generated until the current The new user data generated in the time period before the output result is generated; the other part is the data that has not been deleted by the user in the last output result. It should be noted that this part of the content (that is, the aforementioned two kinds of data) can be determined to be that the user is not satisfied with the training result, or the training result is not the result that the user wants, so the next training is required to make The training model is more mature to achieve the goal of approaching the user's expected result next time.

In an exemplary embodiment, the cleanup model provided in the present disclosure can provide the user to package the content that needs to be retained after outputting the result, so that the user can copy the content that needs to be retained in other storage devices but needs to be deleted locally to other storage devices. After the locale, you can safely delete these contents locally.

Fig. 4 is a schematic structural diagram of the terminal cleaning system according to the present disclosure. As shown in Fig. 4, the framework for implementing the method of the present disclosure may include: an input module, a processing module, an output module, and a user operation recording module.

Among them, in an exemplary embodiment, the input module may record user behavior, and the user's operation and use behavior of the terminal is used as the initial input data of the cleaning model. The input module has a dedicated big data storage space, which is configured to save or record the user's usage of each terminal application and the usage of each contact within a period of time. Exemplary, including but not limited to these content: the time period of using a certain application, the way of use, the generated cache file, the storage path of the file, the deletion of the file, and so on.

In an exemplary embodiment, the processing module is configured to extract content that is frequently accessed by the user after training and learning the cleaning model for a period of time, such as contacts that frequently chat, web pages that are frequently browsed, and Watch videos, frequently listened to music apps, frequently used camera apps, etc., as well as different types of files generated after these corresponding apps and content are used, the path where the files exist, the size of the files, and so on. After the user uses it for one month, two months, three months or more, according to the user’s usage habits, the commonality of these file operations is extracted, for example: the downloaded video can be deleted after watching it, and the music file that is often listened to is not deleted. The cache of frequently browsed web pages can be deleted, the cached videos of Moments can be deleted, the chat videos or pictures with whom are not deleted, the photos and videos of children, family, and friends can not be deleted, and the cached results of entertainment applications can be deleted, etc.

In an exemplary implementation, it needs to be specifically noted that if the algorithm learns that the user has a high probability of saving and copying a specific file or sending it to other places, the processing module needs to perform processing on these files. Refine it and finally show it to users. For example, the user regularly packs and saves the children’s photos, videos, or travel photos, selfies, etc., before cleaning up the content on the terminal. Then, when the cleaning of this type of content is triggered, the content or the content needs to be cleaned up. The list of is displayed to the user, and allows the user to have an actionable plan.

Based on the above results, a new cleaning model is generated after training, and the cleaning model calculates and presents the terminal user's operating data through a certain cleaning algorithm.

In an exemplary embodiment, the output module is configured to provide an interface on the output terminal to display the output result of the training algorithm, that is, to provide a user cleanup list, and the cleanup list can be displayed according to a certain rule. For example, it can be displayed according to the user's possible cleaning priority: files that the user is likely to delete are the first priority, which can be ranked higher; the files that may be deleted are the second priority, and the ranking is slightly lower. And so on. It can also be displayed in accordance with the convergence accuracy level of the algorithm: the file that the algorithm estimates the most accurately can be safely deleted as the first priority, which can be ranked higher; the second is accurate, and the file that may require a little judgment from the user is the second Priority, sorted slightly later, and so on. The display mode is not limited here, and all similar solutions fall within the protection scope of the present disclosure.

In an exemplary embodiment, in addition to displaying the clearable content, the output module may also display to the user the files extracted by the algorithm, which can be saved or packaged by the user, and provided for the user to operate.

In an exemplary embodiment, the user can perform various operations on the cleanup list displayed on the output terminal, and the user operation recording module is configured to record user operations. If the user can clean up the displayed results with confidence, one-click cleaning can be performed; if the output result of the cleaning list displayed on the output terminal may need to be judged, after confirming, one or more items of content can be cleaned up or canceled.

The content that the user cancels to clean up loops to the input module, and enters the next round of training and learning as part of the input data. At the same time, the marking of this part of the content is particularly important, because it can be used as a parameter of the factor of the next revision of the algorithm, making the algorithm closer and closer to the user's intention. For example, the result of deleting the chat picture obtained by the algorithm is not the user's intention, so in the next algorithm, this part of the content may not appear in the content deleted by the user. The list provided by the cleanup model is the content that the cleanup model thinks can be cleaned up. If the user cancels the cleanup, the algorithm determines that this part of the content should be deleted, but the user thinks it should be kept, so this part of the content will not appear in the user deleted content afterwards .

In an exemplary implementation, the content directly deleted by the user may be deleted once or kept for a period of time, so as to prevent the user from deleting it by mistake and wanting to retrieve it.

At the same time, in order to facilitate users to clean up at ease, when the algorithm shows that it is likely to be the file that the user wants to save, it is necessary to provide a channel for the user to package and save these files to other storage places. The specific implementation method is not specifically limited here. For example, it can be checked, unchecked, combined and packaged, one-click packaged, saved to, sent to, etc.

In an exemplary embodiment, the present disclosure provides a training-based terminal space cleaning system, including an input module, a processing module, and an output module.

Fig. 5 is a schematic diagram of the input module recording user behavior according to the present disclosure. As shown in Fig. 5, the input module is configured to record user behavior. The user operates in the foreground and different databases in the background store various types of files generated during use. The fields that need to be used include but are not limited to the fields in Figure 5, which can be specified according to actual needs. At the same time, in the input stage, the tables in the database are used as initial values. With the user's use, the content in each table is constantly updated, and there are records of different users' operations for subsequent training algorithms.

Fig. 6 is a schematic diagram of analyzing and processing user behaviors by a processing module according to the present disclosure. As shown in Fig. 6, the processing module is configured to analyze and process user behaviors. The processing module uses the database to store all the user's processing behaviors of the background files, including the time of being deleted, saved, forwarded, and operation, etc., through these operations, according to the built-in cleaning algorithm, the user's usage habits and some The tendentious behavior is to generate a new database. In the new database, the files in the old database are marked and classified, and finally a database file that can be basically used for subsequent output is formed.

FIG. 7 is a schematic diagram of the output module outputting the cleaning result according to the present disclosure. As shown in FIG. 7, the output module is configured to output the cleaning result. When the cleanup action is triggered, the cleanup result obtained by the training algorithm is displayed on the terminal user interface. According to the previous explanation, the cleanup list can be displayed according to certain rules. For example: it can be displayed according to the cleanup priority that the user may perform; it can also be displayed according to the convergence accuracy level of the algorithm, etc., without specific restrictions. The user can delete part of the content, copy part of the content, and deselect part of the content when operating on the displayed interface. The reserved content back-end database is specially marked as part of the input data for the next training.

In an exemplary embodiment, the present disclosure provides a method for a training-based space clearing function of a mobile terminal that automatically triggers a scan in the background. FIG. 8 is a schematic flowchart of a file cleaning method according to the present disclosure. As shown in FIG. 8, the method may include steps S1 to S9.

In step S1, the original cleaning algorithm is prefabricated.

In step S2, the user operates the terminal.

In step S3, the user's operation information on the file is obtained and stored.

In step S4, commonality is extracted according to the user's operation over a period of time.

In step S5, the pre-made original algorithm is used for algorithm training according to the extracted commonality to form a new algorithm, and the algorithm will be continuously revised according to the user's continuous operation in order to achieve the real goal of the user;

In step S6, the trained algorithm has a result that can be output at any time. When the user initiates a cleanup or triggers the cleanup in the background, the result of the algorithm is displayed to the user. It's just that this result may change over time, but it's all for better satisfying user needs.

In step S7, the user deletes the given cleaning result, which means that the user approves the deletion of this part of the content, that is, the algorithm converges better for the part of the data.

In step S8, the user cancels the deletion of the given cleaning result, which means that the user does not approve the deletion of this part of the content, that is to say, the part of the data algorithm needs to be continuously revised. Therefore, it enters the next cycle and continues to perform algorithm training.

In step S9, the user agrees to the copy data provided, the user directly copies, and the packaging operation is performed in the background, and then the user deletes it.

In an exemplary embodiment, the precondition is that the terminal has system permissions and a suitable cleaning algorithm is built-in, which can be the current mainstream algorithm. The present disclosure is based on the original algorithm and is based on user habits. The algorithm is continuously trained to reach an intelligent algorithm that is infinitely close to the real intention of the user.

In an exemplary implementation, in the user-operable interface, check, uncheck, delete, copy, etc. operations are performed according to the displayed cleanup list, and the retained files continue to be trained for the next time and are used as algorithm corrections. factor. The content user interface of the algorithm is not visible.

In this embodiment, through the continuous training and learning of the algorithm, the user’s worries that users have not dared to clean up many pictures, videos, and files have been gradually solved to a certain extent; at the same time, it is convenient for the user to automatically integrate the required files for the user, and Clean up after copying or sharing to other places, users will not worry about things being lost, and a lot of space can be freed up.

The present disclosure also provides a computer-readable storage medium in which a computer program is stored, wherein the computer program is configured to execute any method described herein when running.

In an exemplary embodiment, the foregoing computer-readable storage medium may include, but is not limited to: U disk, Read-Only Memory (Read-Only Memory, ROM for short), Random Access Memory (Random Access Memory, RAM for short) , Mobile hard drives, magnetic disks or optical discs and other media that can store computer programs.

In an exemplary embodiment, the aforementioned electronic device may further include a transmission device and an input-output device, wherein the transmission device is connected to the aforementioned processor, and the input-output device is connected to the aforementioned processor.

For specific examples in this embodiment, reference may be made to the examples described in the foregoing embodiment and exemplary embodiments, and this embodiment will not be repeated here.

Obviously, those skilled in the art should understand that the above-mentioned modules or steps of the present disclosure can be implemented by a general computing device, and they can be concentrated on a single computing device or distributed in a network composed of multiple computing devices. Above, they can be implemented with program codes executable by a computing device, so that they can be stored in a storage device for execution by the computing device, and in some cases, they can be executed in a different order than shown here. Or the described steps, or fabricate them into individual integrated circuit modules respectively, or fabricate multiple modules or steps of them into a single integrated circuit module to achieve. In this way, the present disclosure is not limited to any specific combination of hardware and software.

The foregoing descriptions are only preferred embodiments of the present disclosure, and are not intended to limit the present disclosure. For those skilled in the art, the present disclosure may have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the principles of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

A file processing method, including:

Use the second cleaning model to analyze the files stored in the terminal to determine whether the files need to be cleaned, where the second cleaning model is obtained by training the first cleaning model using training data, and the training The data includes: the first operation characteristic of the target object on the first cleaning result, wherein the first cleaning result is obtained by analyzing the files stored in the terminal through the first cleaning model, and the first cleaning result A cleaning result indicates a file to be cleaned, and the first operating characteristic is used to indicate the operating characteristic of the target object on a file of a specified type in the first cleaning result, and files of the same type have the same file characteristics;

In the case where it is determined that the file needs to be cleaned up, a second cleanup result is displayed, where the second cleanup result indicates the file to be cleaned up.
The method according to claim 1, wherein, before said using the second cleaning model to analyze the files stored in the terminal, the method further comprises:

Training the first cleaning model using training data to obtain the second cleaning model, wherein the training the first cleaning model using the training data includes:

In the case where the first operating characteristic indicates that the probability of the first type of file being retained is higher than the probability of the second type of file being retained, the first cleaning model is trained to prioritize the second type of file to be cleaned Documents; and/or,

In the case where the first operating feature indicates that the probability of the third type of file being cleaned is higher than the probability of the fourth type of file being cleaned, the first cleaning model is trained to prioritize the third type of file to be cleaned document.
The method according to claim 1 or 2, wherein the first operation characteristic is obtained according to the operation behavior of the target object on the first cleaning result and the file pointed to by the operation behavior.
The method according to claim 1, wherein the training data further comprises: a second operation characteristic of the target object on the file stored in the terminal, wherein the second operation characteristic is used to indicate the The operating characteristics of the target object on the specified type of file stored in the terminal.
The method according to claim 4, wherein, before said using the second cleaning model to analyze the files stored in the terminal, the method further comprises:

Training the first cleaning model using training data to obtain the second cleaning model, wherein the training the first cleaning model using the training data includes:

In the case where the second operating feature indicates that the frequency of the fifth type of file being accessed is higher than the frequency of accessing the sixth type of file, the first cleaning model is trained to prioritize the sixth type of file to be cleaned document.
The method according to claim 4 or 5, wherein the second operation characteristic is obtained according to the operation behavior of the target object on the file stored in the terminal and the file pointed to by the operation behavior.
The method according to claim 1, wherein, after the displaying the second cleaning result, the method further comprises:

Use the third cleaning model to analyze the files stored in the terminal to determine whether the files need to be cleaned, where the third cleaning model is obtained by training the second cleaning model using training data, so The training data includes: a third operating characteristic of the target object on the second cleaning result, wherein the third operating characteristic is used to indicate that the target object has a specific type of file in the second cleaning result. Operating characteristics;

In the case where it is determined that the file needs to be cleaned up, a third cleanup result is displayed, where the third cleanup result indicates the file to be cleaned up.
The method according to claim 4, further comprising:

Use the second cleaning model to analyze the files stored in the terminal to determine whether the files in the second cleaning result need to be sent to a storage device other than the terminal;

In the case where it is determined that the file needs to be sent to a storage device other than the terminal, the analysis result is displayed, where the analysis result indicates the file to be sent.
8. The method according to claim 8, wherein, before said using the second cleaning model to analyze the files stored in the terminal, the method further comprises:

Training the first cleaning model using training data to obtain the second cleaning model, wherein the training the first cleaning model using the training data includes:

In the case where the first operating feature and the second operating feature indicate that the seventh type of file is sent to a storage device other than the terminal before being cleared, the first cleaning model is trained to transfer the The seventh type of file is prioritized as the file to be sent.
The method according to claim 1, wherein, before using the second cleaning model to analyze the files stored in the terminal, the method further comprises:

A start signal is received through the terminal, where the start signal is used to instruct to start the second cleaning model.
The method according to claim 1, wherein in the second cleaning result, the order of the files to be cleaned is determined according to the probability of the files to be cleaned being cleaned.
The method according to claim 1, wherein after the displaying the second cleaning result, the method further comprises:

Receiving the target operation of the target object on the second cleaning result;

Perform a corresponding operation on the second cleaning result according to the target operation.
A file processing device, including:

The analysis module is configured to use a second cleaning model to analyze files stored in the terminal to determine whether the files need to be cleaned, wherein the second cleaning model is obtained by training the first cleaning model using training data Yes, the training data includes: a first operation feature of the target object on the first cleaning result, wherein the first cleaning result is obtained by analyzing the files stored in the terminal through the first cleaning model , The first cleaning result indicates the file to be cleaned, and the first operating feature is used to indicate the operating feature of the target object on the file of the specified type in the first cleaning result, and files of the same type have the same File characteristics;

The display module is configured to display a second cleaning result when it is determined that the file needs to be cleaned, where the second cleaning result indicates the file to be cleaned.
A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program implements the method described in any one of claims 1 to 12 when executed by a processor.
A terminal includes a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the method described in any one of claims 1 to 12.