A redundant file cleanup system, method, apparatus, and computer readable storage medium

Abstract

The application provides a redundant file cleaning system, method, device and computer readable storage medium, the method obtains the multidimensional characteristics of all associated files of the uninstalled application in response to the cleaning operation triggered by the application uninstallation event; the multidimensional characteristics at least include basic characteristics, semantic characteristics and behavior characteristics; based on the multidimensional characteristics, a preset file evaluation model is called to evaluate the cleaning necessity of each associated file, and an evaluation result is obtained; and a corresponding file cleaning operation is performed based on the evaluation result. The application realizes accurate identification and cleaning of the residual files after the application is uninstalled by collecting multidimensional characteristics in advance and using the multidimensional characteristics of the files to evaluate the necessity of file cleaning.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically, to a redundant file cleanup system, method, apparatus, and computer-readable storage medium. Background Technology

[0002] With the development of the internet, the number of applications (APPs) installed on users' mobile devices is increasing daily. When a user uninstalls an application, they often need to simultaneously clean up the residual data it generated in order to free up storage space and protect privacy and security.

[0003] However, existing cleaning technologies typically rely on single-dimensional features such as file extensions, file path keywords, or last access time for judgment, resulting in inaccurate cleaning results: files that should be cleaned are not deleted, while important files that should not be deleted are mistakenly deleted. Summary of the Invention

[0004] In view of this, the purpose of the present invention is to provide a redundant file cleaning system, method, apparatus and computer-readable storage medium to improve the accuracy of redundant file cleaning.

[0005] Firstly, a redundant file cleanup system is provided for use in terminal devices, including: a data acquisition module, a multi-dimensional feature database, and an intelligent cleanup engine; The data acquisition module is used to collect multi-dimensional features of various files in the terminal device in real time; A multidimensional feature database is used to store the mapping relationship between each file and its corresponding multidimensional features; The intelligent cleanup engine is used to respond to cleanup operations triggered by application uninstallation events. It retrieves all associated files of the uninstalled application and their corresponding multidimensional features from the multidimensional feature database, evaluates the necessity of cleanup for each associated file based on the multidimensional features, and performs cleanup operations based on the evaluation results.

[0006] Optionally, the data acquisition module includes: a basic feature acquisition unit, a semantic feature acquisition unit, and a behavioral feature acquisition unit; The basic feature acquisition unit is used to collect the basic features of a file, including at least one of the following: file path, file size, creation time, modification time, and file extension. The semantic feature acquisition unit is used to analyze the file content through a local semantic recognition model to obtain the semantic category label of the file; The behavior feature collection unit is used to record user operation behavior data on files, including at least one of the following: access frequency, cumulative browsing time, whether the file is saved, and whether it is shared.

[0007] Secondly, a method for cleaning up redundant files is provided, including: In response to the cleanup operation triggered by the application uninstallation event, obtain multi-dimensional features of all associated files of the uninstalled application; the multi-dimensional features include at least basic features, semantic features, and behavioral features; Based on multidimensional features, a preset file evaluation model is invoked to assess the necessity of cleaning up each associated file, and the evaluation results are obtained. Perform the corresponding file cleanup operations based on the evaluation results.

[0008] Optionally, in response to the cleanup operation triggered by the application uninstallation event, the multidimensional characteristics of all associated files of the uninstalled application are obtained, including: Get the application package name of the uninstalled application; Based on the application package name, query the multidimensional feature database for a set of files that are associated with the application package name, where each file in the set is configured with a unique file identifier; Based on the mapping relationship between file identifiers and multidimensional features, the multidimensional features corresponding to each associated file in the file set are obtained.

[0009] Optionally, based on multi-dimensional features, a pre-defined file evaluation model is invoked to assess the necessity of cleaning each associated file, and the evaluation results include: Based on fundamental features, a large model is used to determine the redundancy risk value of each associated file; Based on semantic and behavioral features, a large model is used to determine the file value score of each associated file; The necessity of cleaning up each associated file is assessed based on the redundancy risk value and file value score, and the assessment results are obtained.

[0010] Optionally, the necessity of cleaning up each associated file is assessed based on redundancy risk value and file value score, and the assessment results include: Construct a two-dimensional evaluation coordinate system with redundancy risk value on the horizontal axis and document value on the vertical axis; Based on preset redundancy risk thresholds and document value thresholds, the two-dimensional evaluation coordinate system is divided into four quadrants, each corresponding to a different cleanup level. Based on the redundancy risk value and document value score of each related document, determine its quadrant; The cleanup level is determined based on the quadrant to which each associated file belongs, and the cleanup level is used as the evaluation result.

[0011] Optionally, performing corresponding file cleanup operations based on the evaluation results includes: Generate a tiered cleanup list based on the cleanup level of each associated file; Perform corresponding file cleanup operations based on the tiered cleanup list.

[0012] Optionally, the method further includes: Obtain the storage space status of the terminal device; The redundancy risk threshold and file value threshold are dynamically adjusted based on the storage space status of the terminal device.

[0013] Thirdly, a redundant file cleanup device is provided, comprising: The acquisition unit is used to acquire multi-dimensional features of all associated files of the uninstalled application in response to the cleanup operation triggered by the application uninstallation event; the multi-dimensional features include at least basic features, semantic features and behavioral features; The evaluation unit is used to evaluate the necessity of cleaning each associated file based on multi-dimensional features and by calling a preset file evaluation model to obtain the evaluation results. The cleanup unit is used to perform corresponding file cleanup operations based on the evaluation results.

[0014] Fourthly, a computer-readable storage medium is provided, wherein a computer program is stored therein, and the computer program, when executed by a processor, implements any of the methods described in the second aspect.

[0015] This invention provides a redundant file cleanup system, method, apparatus, and computer-readable storage medium. The method acquires multi-dimensional features of all associated files of the uninstalled application through a cleanup operation triggered by an application uninstallation event. These multi-dimensional features include at least basic features, semantic features, and behavioral features. Based on these features, a preset file evaluation model is invoked to assess the necessity of cleanup for each associated file, yielding an evaluation result. Based on the evaluation result, corresponding file cleanup operations are performed. This invention achieves accurate identification and cleanup of residual files after application uninstallation by pre-collecting multi-dimensional features and utilizing these features to assess the necessity of file cleanup.

[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 The diagram shows a structural diagram of a redundant file cleanup system provided in an embodiment of the present invention; Figure 2 The flowchart of a redundant file cleanup method provided by an embodiment of the present invention is shown; Figure 3 This diagram illustrates the structure of a risk-value assessment coordinate system provided by an embodiment of the present invention. Figure 4 A schematic diagram of a redundant file cleanup device provided in an embodiment of the present invention is shown. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0020] With the development of the internet, the number of applications (APPs) installed on users' mobile devices is increasing daily. When a user uninstalls an application, they often need to simultaneously clean up the residual data it generated in order to free up storage space and protect privacy and security.

[0021] However, existing cleaning technologies typically rely on single-dimensional features such as file extensions, file path keywords, or last access time for judgment, resulting in inaccurate cleaning results: files that should be cleaned are not deleted, while important files that should not be deleted are mistakenly deleted.

[0022] For example, cleaning up files based on file extensions: important documents (such as a game save file in .dat format) may be incorrectly cleaned up because they match a simple file extension; while junk files generated by many applications do not use standard file extensions and thus escape cleaning.

[0023] For example, file path-based cleanup methods only scan files in the application's own installation directory. However, many applications create associated files in other external directories (such as / sdcard / Android / data / and / sdcard / Android / obb / ) during operation. These files are not deleted because they are not in the application's installation directory.

[0024] Based on this, embodiments of the present invention provide a redundant file cleanup system, applied in terminal devices such as mobile phones, tablets, and computers. Its operating system is, for example, the Android system. Figure 1As shown, the system includes a data acquisition module, a multi-dimensional feature database, and an intelligent cleaning engine. The intelligent cleaning engine includes a file retrieval unit, a cleaning strategy decision-making unit, and a cleaning execution unit.

[0025] The data acquisition module includes a basic feature acquisition unit, a semantic feature acquisition unit, and a behavioral feature acquisition unit, which are used to collect the basic features, semantic features, and behavioral features of each file in the mobile terminal, providing a multi-dimensional data foundation for subsequent determination of whether to perform file cleaning.

[0026] In one feasible implementation, the basic feature acquisition unit can obtain the basic features of a file by accessing the file system interface of the terminal device, including: file path, size, modification time, creation time, extension, and directory.

[0027] The semantic feature acquisition unit extracts semantic features from files using a built-in lightweight semantic recognition model. This allows for the differentiation of semantic categories within the file content, thereby determining its value or redundancy to the user. Semantic categories include, for example, family photos, work screenshots, online advertising images, emojis, and ID photos.

[0028] In a specific instance, the semantic recognition model is a lightweight convolutional neural network model (such as MobileNetV3-small or EfficientNet-Lite) deployed locally on the mobile terminal.

[0029] The input to this semantic recognition model is the raw data of the file (such as images, documents, or videos); the output is a probability distribution vector containing semantic labels and their confidence scores.

[0030] During the training phase, the model uses a large-scale labeled dataset for supervised learning. This labeled data covers common mobile device file types, such as: images: family photos, work screenshots, online advertising images, emoticons, ID photos (ID cards, invoices), wallpapers, application cache thumbnails, etc.; documents: PDF contracts, PPT presentations, Excel spreadsheets, etc.

[0031] In another implementation, the behavior feature collection unit can collect user operations on files through system log monitoring, application tracking, or interfaces such as Android UsageStatsService. This includes access frequency, cumulative browsing time, last access time, whether the file is saved, shared, or renamed. This behavioral data reflects the intensity of user interaction with the file and is an important basis for assessing the file's potential value.

[0032] All collected basic features, semantic features, and behavioral features are combined into a feature set for the corresponding file. Then, the file ID, file path, feature set, and source are used to construct a mapping relationship and write it into a local multidimensional feature database (such as SQLite or Room database).

[0033] It should be noted that the data acquisition module collects the basic features, semantic features, and behavioral features of new files generated on the terminal device in real time, and stores the extracted multidimensional features into the local multidimensional feature database to save time during subsequent file cleaning and improve cleaning efficiency.

[0034] The intelligent cleanup engine is used to respond to cleanup trigger events after an application is uninstalled and to perform file cleanup operations.

[0035] Specifically, in response to the cleanup trigger event, the file retrieval unit obtains the package name of the uninstalled application, retrieves the set of files associated with the package name in the multidimensional feature database based on the package name, extracts the corresponding multidimensional feature data, and inputs it into the cleanup strategy decision unit.

[0036] This invention establishes a logical mapping relationship between files and source application package names in a multidimensional feature database, ensuring that even if a file is located in a non-private directory, it can be accurately associated and included in the cleanup scope, thus significantly improving the detection rate of residual files.

[0037] The cleanup strategy decision unit is equipped with a file evaluation model. Based on the file evaluation model and the multi-dimensional characteristics of the files, the cleanup strategy decision unit evaluates whether the files need to be cleaned and generates a cleanup list. The cleanup list is then sent to the cleanup execution unit, which calls the file system interface of the terminal device to perform file cleanup.

[0038] This invention achieves accurate identification and cleanup of residual files after application uninstallation by collecting multi-dimensional features in advance and using these features to assess the necessity of file cleanup.

[0039] Based on the system constructed above, this invention provides a method for cleaning up redundant files, applied to an intelligent cleaning engine, such as... Figure 2 As shown, it includes the following steps: Step S201: In response to the cleanup operation triggered by the application uninstallation event, obtain the multidimensional characteristics of all associated files of the uninstalled application.

[0040] In this step, the cleanup operations triggered by the application uninstallation event fall into two categories: one is the cleanup operation manually clicked by the user on the UI interface of the terminal device; the other is the application uninstallation event listened to by the system.

[0041] In a specific example, the steps to obtain the multidimensional characteristics of the associated files of the uninstalled application include: Step S201A: Obtain the application package name of the uninstalled application.

[0042] The package name of the uninstalled application is a unique identifier for each application in the Android system.

[0043] Step S201B: Based on the application package name, query the multidimensional feature database for a set of files that are associated with the application package name.

[0044] Each file in the file set is configured with a unique file identifier.

[0045] Based on the above embodiments, it can be understood that a mapping relationship between file ID (file identifier), file path, feature set and source application is constructed in the multidimensional feature database. The source application is first located based on the package name of the uninstalled application, and then all files associated with the uninstalled application can be determined based on the correspondence between the source application and the file ID.

[0046] It's important to note that a single file may correspond to multiple source applications. For example, an image might be taken by the camera on an Android device, enhanced by a photo editing app, and then uploaded to a social media app. In this case, the image originates from three different applications.

[0047] Step S201C: Based on the mapping relationship between file identifiers and multidimensional features, obtain the multidimensional features corresponding to each associated file in the file set.

[0048] Multidimensional features include the basic features, semantic features, and behavioral features of a document.

[0049] Step S202: Based on multi-dimensional features, call the preset file evaluation model to evaluate the necessity of cleaning each associated file and obtain the evaluation results.

[0050] As mentioned in the previous example, a file may have multiple source applications. Therefore, when uninstalling one of these applications, such as a social media app, the accuracy of the cleanup can be improved by assessing whether the file needs to be cleaned up, rather than simply deleting it.

[0051] Step S203: Perform the corresponding file cleanup operation based on the evaluation results.

[0052] In this step, the cleanup execution unit can obtain the file path of each file in the multidimensional feature database based on the file ID in the cleanup list, and then find the file location based on the file path for deletion.

[0053] This invention achieves accurate identification and cleanup of residual files after application uninstallation by collecting multi-dimensional features in advance and using these features to assess the necessity of file cleanup.

[0054] Based on the above embodiments, and using multi-dimensional features, a preset file evaluation model is invoked to assess the necessity of cleaning up each associated file, and the evaluation results are obtained in the following steps: Step S202A: Based on the basic features, use a large model to determine the redundancy risk value of each associated file.

[0055] In this embodiment of the invention, the document evaluation model includes a redundancy risk evaluation model and a document value evaluation model.

[0056] In this step, basic characteristics include file path, size, modification time, creation time, file extension, and directory.

[0057] Based on these characteristics, a pre-defined redundancy risk assessment model is used to calculate its redundancy risk value (range: 1 to 10, the higher the value, the more likely it is to be a redundant file).

[0058] In one feasible implementation, the redundancy risk assessment model is a locally deployed lightweight large language model (LLM) or structured inference engine, with the input being the basic features of the file and preset prompt words. These prompt words contain evaluation rules, such as: Please assess the redundancy risk of the document according to the following rules (1-10 points): For every 10MB increase in file size beyond 1MB, the risk increases by 1 point. Paths containing / cache / , / temp / , or / Download / will incur a risk score of +3. If the last revision time exceeds 180 days, the risk score is increased by 2 points. Files with extensions like .tmp, .log, or .apk (not in an installed state) are considered a risky item, earning +2 points. Based on a comprehensive analysis of all characteristics, a final risk score is generated.

[0059] After weighting and fusing the various features, the model outputs a comprehensive redundancy risk value.

[0060] In a specific example: A file is 500MB in size, located at / Android / data / com.xingin.xhs / cache / , and was modified 200 days ago. The model outputs a redundancy risk value of 8.5.

[0061] Step S202B: Based on semantic and behavioral features, a large model is used to determine the file value score of each associated file.

[0062] Similarly, the system extracts the semantic features (such as family photos, work screenshots, online advertisement images, emoticons, ID photos, etc.) and behavioral features (such as access frequency, browsing time, whether to collect, etc.) of the files, inputs them into the file value assessment model, and outputs its user value score (range: 1 to 10, the higher the value, the more likely the user is to keep it).

[0063] The document value assessment model is also a locally deployed lightweight large language model or structured reasoning engine. It can use the same model as the redundancy risk assessment model, only the input prompts and input features are different.

[0064] The following are the key words used in the valuation of this document: Please evaluate the document's potential value to users according to the following rules (1-10 points): The content type is family photos, documents, and contracts, value +4 points; Content type is online advertising image / emoticon, value +2 points; ≥ 5 visits, +3 points value; Total browsing time > 1 minute, value +2 points; Already added to favorites, value +3 points; Based on a comprehensive assessment, a final value score is output.

[0065] For example, if an image is identified as a "downloaded emoji" (low semantic score), and the user only opens it once and does not save it, the model outputs a user value score of 2.0.

[0066] Step S202C: Evaluate the necessity of cleaning up each associated file based on the redundancy risk value and file value score to obtain the evaluation result.

[0067] In this step, two dynamic thresholds can be set: a redundancy risk threshold and a file value threshold.

[0068] The redundancy risk value of each file determined in step S202A is compared with the redundancy risk threshold; and the value score determined in step S202B is compared with the file value threshold. Based on the comparison results, it is determined whether the file needs to be cleaned up. For example, if the redundancy risk value is greater than the redundancy risk value but less than the file value threshold, then the file needs to be cleaned up.

[0069] This invention uses a risk-value approach to assess the necessity of file cleanup, enabling a more refined classification of the necessity. This effectively avoids the problems of accidentally deleting important files or missing large amounts of junk files caused by the traditional "one-size-fits-all" approach, and improves the accuracy of redundant file cleanup.

[0070] Based on the above embodiments, the necessity of cleaning up each associated file is assessed based on redundancy risk value and file value score, and the assessment results include: Step S202C1: Construct a two-dimensional evaluation coordinate system with redundancy risk value on the horizontal axis and document value on the vertical axis.

[0071] Step S202C2: Based on the preset redundancy risk threshold and document value threshold, the two-dimensional evaluation coordinate system is divided into four quadrants.

[0072] Each quadrant corresponds to a different level of cleanup.

[0073] In one feasible implementation, such as Figure 3 As shown, a two-dimensional risk-value assessment coordinate system can be constructed: X-axis: Redundancy risk value (1~10), the larger the value, the more likely the file is to be junk; Y-axis: User value score (1~10), the higher the value, the more important the file is to the user.

[0074] The system is divided into four quadrants based on two dynamic thresholds (redundancy risk threshold (X0): default value 6.0; file value threshold (Y0): default value 5.0), as shown in the table below:

[0075] Step S202C3: Determine the quadrant to which each associated file belongs based on its redundancy risk value and file value score.

[0076] Based on the redundancy risk value and value score of each file determined in steps S202A and S202B, the (X, Y) coordinates of each file are determined, and then the quadrant in which it is located is determined.

[0077] Step S202C4: Determine the cleanup level based on the quadrant to which each associated file belongs, and use the cleanup level as the evaluation result.

[0078] Based on the above embodiments, performing corresponding file cleanup operations based on the evaluation results includes: Step S203A: Generate a tiered cleanup list based on the cleanup level of each associated file.

[0079] Step S203B: Perform the corresponding file cleanup operation based on the tiered cleanup list.

[0080] The list of items to be cleaned up will be sent to the user via a pop-up window for confirmation. After the user confirms, the cleanup execution unit will carry out the cleanup.

[0081] This invention further categorizes the cleanup list into four levels: priority cleanup, no cleanup, optional cleanup, and pending confirmation. This enhances users' awareness of the redundancy risks and value of each file, preventing accidental or missed deletion of valuable user content (such as screenshots and saved images).

[0082] In another embodiment, the aforementioned redundancy risk threshold and file value threshold can be dynamically adjusted based on the storage space of the infected female device.

[0083] In one example, when the available storage space on an Android device is less than 10%, X0 is lowered to 5.0 and Y0 is increased to 6.0 to expand the scope of cleanup.

[0084] When available space is greater than 20%, increase X0 to 7.0 and decrease Y0 to 4.0 to narrow the cleanup scope and ensure data security.

[0085] The embodiments of the present invention can achieve a balance between ensuring safety and meeting emergency needs through dynamic threshold adjustment.

[0086] In this embodiment of the invention, after generating the cleanup list, the intelligent cleanup engine records the user's final operation behavior on each file. If a file is determined by the system to be "cleanable" (fourth quadrant), but the user chooses to "keep" or "restore," this behavior is sent back to the file evaluation model as a negative feedback signal. The file evaluation model then adjusts the semantic weight or behavioral weight of that type of file accordingly.

[0087] For example, if a user saves "emoji" files multiple times, the value score of the "meme" category will automatically increase by +2.

[0088] The embodiments of the present invention enable the document value assessment model to have personalized adaptive capabilities, and the assessment accuracy continues to improve after long-term use.

[0089] Based on the same inventive concept, embodiments of the present invention provide a redundant file cleanup device, such as... Figure 4 As shown, it includes: The acquisition unit 401 is used to acquire multi-dimensional features of all associated files of the uninstalled application in response to the cleanup operation triggered by the application uninstallation event; the multi-dimensional features include at least basic features, semantic features and behavioral features; The evaluation unit 402 is used to evaluate the necessity of cleaning each associated file based on multi-dimensional features and by calling a preset file evaluation model to obtain the evaluation results. Cleaning unit 403 is used to perform corresponding file cleanup operations based on the evaluation results.

[0090] The computer program product for cleaning up redundant files provided in this embodiment of the invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0091] The redundant file cleanup device provided in this embodiment of the invention can be specific hardware on a device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this embodiment of the invention are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiments can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the above method embodiments, and will not be repeated here.

[0092] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and method can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0093] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0094] In addition, the functional units in the embodiments provided by the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0095] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0096] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0097] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. All should be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A redundant file cleanup system, characterized in that, Applied to terminal devices, including: data acquisition module, multi-dimensional feature database and intelligent cleaning engine; The data acquisition module is used to collect multidimensional features of each file in the terminal device in real time; The multidimensional feature database is used to store the mapping relationship between each file and its corresponding multidimensional features; The intelligent cleaning engine is used to respond to the cleaning operation triggered by the application uninstallation event, retrieve all associated files of the uninstalled application and their corresponding multidimensional features in the multidimensional feature database, evaluate the necessity of cleaning each associated file based on the multidimensional features, and perform the cleaning operation based on the evaluation result.

2. The system according to claim 1, characterized in that, The data acquisition module includes: a basic feature acquisition unit, a semantic feature acquisition unit, and a behavioral feature acquisition unit; The basic feature acquisition unit is used to acquire basic features of a file, including at least one of the following: file path, file size, creation time, modification time, and file extension. The semantic feature acquisition unit is used to analyze the file content through a local semantic recognition model to obtain the semantic category label of the file; The behavior feature collection unit is used to record user operation behavior data on files, including at least one of the following: access frequency, cumulative browsing time, whether the file is saved, and whether it is shared.

3. A method for cleaning up redundant files based on the redundant file cleaning system of claim 1 or 2, characterized in that, include: In response to the cleanup operation triggered by the application uninstallation event, obtain multidimensional features of all associated files of the uninstalled application; The multidimensional features include at least basic features, semantic features, and behavioral features; Based on the multidimensional features, a preset file evaluation model is invoked to evaluate the necessity of cleaning each associated file, and the evaluation results are obtained. Based on the evaluation results, perform the corresponding file cleanup operations.

4. The method according to claim 3, characterized in that, The cleanup operation triggered in response to the application uninstallation event, which obtains multidimensional features of all associated files of the uninstalled application, includes: Obtain the application package name of the uninstalled application; Based on the application package name, a set of files that are associated with the application package name is queried in the multidimensional feature database, wherein each file in the set of files is configured with a unique file identifier; Based on the mapping relationship between file identifiers and multidimensional features, the multidimensional features corresponding to each associated file in the file set are obtained.

5. The method according to claim 3, characterized in that, Based on the multidimensional features, a preset file evaluation model is invoked to assess the necessity of cleaning each associated file, and the evaluation results include: Based on the aforementioned fundamental characteristics, a large model is used to determine the redundancy risk value of each associated file; Based on the aforementioned semantic and behavioral features, a large model is used to determine the file value score of each associated file; The necessity of cleaning up each associated file is assessed based on the redundancy risk value and file value score, and the assessment result is obtained.

6. The method according to claim 5, characterized in that, The assessment of the necessity of cleaning up each associated file based on the redundancy risk value and file value score yields the following results: Construct a two-dimensional evaluation coordinate system with redundancy risk value on the horizontal axis and document value on the vertical axis; Based on preset redundancy risk thresholds and file value thresholds, the two-dimensional evaluation coordinate system is divided into four quadrants, each corresponding to a different cleanup level. Based on the redundancy risk value and document value score of each related document, determine its quadrant; The cleanup level is determined based on the quadrant to which each of the associated files belongs, and the cleanup level is used as the evaluation result.

7. The method according to claim 6, characterized in that, The process of performing corresponding file cleanup operations based on the evaluation results includes: Generate a tiered cleanup list based on the cleanup level of each associated file; Perform corresponding file cleanup operations based on the tiered cleanup list.

8. The method according to claim 6, characterized in that, The method further includes: Obtain the storage space status of the terminal device; The redundancy risk threshold and file value threshold are dynamically adjusted based on the storage space status of the terminal device.

9. A redundant file cleanup device, characterized in that, include: The acquisition unit is used to acquire multidimensional features of all associated files of the uninstalled application in response to the cleanup operation triggered by the application uninstallation event. The multidimensional features include at least basic features, semantic features, and behavioral features; The evaluation unit is used to evaluate the necessity of cleaning each associated file based on the multidimensional features by calling a preset file evaluation model and obtaining the evaluation result. The cleaning unit is used to perform corresponding file cleaning operations based on the evaluation results.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the method described in any one of claims 3-8.

Images