File storage path recommendation method and device, terminal equipment and storage medium

Abstract

The application provides a file storage path recommendation method and device, terminal equipment and a storage medium, and belongs to the technical field of communication. The terminal equipment determines a semantic feature vector and a subject feature vector of a to-be-downloaded file respectively, determines a storage path of the to-be-downloaded file based on a fusion feature vector obtained by splicing the semantic feature vector and the subject feature vector, and performs storage path recommendation, so that files of different categories are saved under different storage paths, facilitating subsequent management and searching of the downloaded files by a user. Moreover, the storage path of the to-be-downloaded file is determined based on the fusion feature vector obtained by splicing the semantic feature vector and the subject feature vector, which can improve the accuracy of the recommended storage path.

Description

Technical Field

[0001] This application relates to the field of communication technology, and in particular to a method, apparatus, terminal device, and storage medium for recommending file storage paths. Background Technology

[0002] With the rapid development of the Internet, various data resources such as files, programs, music, and videos can be stored online, and users can download the data files they need through software such as browsers.

[0003] Currently, when downloading files through a browser, the browser generally has a default download location. However, all the files that need to be downloaded are stored in the default location after downloading, resulting in a messy and disordered file storage that makes it difficult for users to find the downloaded files. Summary of the Invention

[0004] This application provides a file storage path recommendation method, apparatus, terminal device, and storage medium to recommend file storage paths when a user downloads a file, facilitating the user's subsequent management and retrieval of the downloaded file.

[0005] To achieve the above objectives, the technical solution of this application embodiment is implemented as follows:

[0006] In a first aspect, embodiments of this application provide a file storage path recommendation method, the method comprising:

[0007] Generate a semantic feature vector of the file name based on the predicate headword contained in the file name of the file to be downloaded and the content governed by the predicate headword;

[0008] Based on the topic-based word segments contained in the filename of the file to be downloaded, and the topic to which each topic segment belongs, generate a topic feature vector for the filename;

[0009] The semantic feature vector and the topic feature vector are concatenated to obtain a fused feature vector;

[0010] Based on the fused feature vector, the storage path of the file to be downloaded is determined and the storage path is output.

[0011] In one optional implementation, generating a semantic feature vector for the filename based on the predicate headword contained in the filename of the file to be downloaded and the content governed by the predicate headword includes:

[0012] Determine the predicate headword contained in the file name and the content governed by the predicate headword;

[0013] The controlled content is segmented into multiple content segments;

[0014] Based on the word vector of each content segment, the distance between each content segment and the predicate headword, and the word vector of the predicate headword, a semantic feature vector of the file name is generated.

[0015] In one optional implementation, determining the predicate head word contained in the filename and the content governed by the predicate head word includes:

[0016] By performing syntactic structure analysis on the filename using a skeleton dependency tree, the predicate headword and the content governed by the predicate headword are obtained.

[0017] In one optional implementation, generating a topic feature vector for the filename based on topic segmentation words contained in the filename of the file to be downloaded, and the topic to which each topic segmentation word belongs, includes:

[0018] Determine the topic segments contained in the filename and the topic to which each topic segment belongs;

[0019] Based on the cosine distance between the high-frequency words of each topic and the file name, and the topic word segmentation, a topic feature vector of the file name is generated.

[0020] In one optional implementation, determining the topic segments contained in the filename and the topic to which each topic segment belongs includes:

[0021] The topic segments contained in the file name and the topic to which each topic segment belongs are determined using the LDA model.

[0022] In one optional implementation, determining the storage path of the file to be downloaded based on the fused feature vector includes:

[0023] Based on the fused feature vector, the target category to which the file to be downloaded belongs is determined;

[0024] The storage path of the target category is used as the storage path of the file to be downloaded, and the storage path of the file to be downloaded is output.

[0025] In one optional implementation, determining the target category of the file to be downloaded based on the fused feature vector includes:

[0026] The target category to which the file to be downloaded belongs is determined by clustering the fused feature vector with the feature vectors of each pre-saved category.

[0027] In one alternative implementation, after outputting the storage path, the method further includes:

[0028] If a user confirms the path input, the file to be downloaded is saved according to the storage path.

[0029] If a path modification operation is received from the user, the file to be downloaded will be saved according to the modified storage path.

[0030] Secondly, embodiments of this application also provide a file storage path recommendation device, the device comprising:

[0031] The filename processing unit is configured to generate a semantic feature vector of the filename based on the predicate headword contained in the filename of the file to be downloaded and the content governed by the predicate headword; generate a topic feature vector of the filename based on the topic segmentation words contained in the filename and the topic to which each topic segmentation word belongs; and concatenate the semantic feature vector and the topic feature vector to obtain a fused feature vector.

[0032] The storage path determination unit is used to determine the storage path of the file to be downloaded based on the fused feature vector, and output the storage path.

[0033] In one optional implementation, the filename processing unit is specifically used for:

[0034] Determine the predicate headword contained in the file name and the content governed by the predicate headword;

[0035] The controlled content is segmented into multiple content segments;

[0036] Based on the word vector of each content segment, the distance between each content segment and the predicate headword, and the word vector of the predicate headword, a semantic feature vector of the file name is generated.

[0037] In one optional implementation, the filename processing unit is specifically used for:

[0038] By performing syntactic structure analysis on the filename using a skeleton dependency tree, the predicate headword and the content governed by the predicate headword are obtained.

[0039] In one optional implementation, the filename processing unit is specifically used for:

[0040] Determine the topic segments contained in the filename and the topic to which each topic segment belongs;

[0041] Based on the cosine distance between the high-frequency words of each topic and the file name, and the topic word segmentation, a topic feature vector of the file name is generated.

[0042] In one optional implementation, the filename processing unit is specifically used for:

[0043] The topic segments contained in the file name and the topic to which each topic segment belongs are determined using the LDA model.

[0044] In one optional implementation, the filename processing unit is specifically used for:

[0045] Based on the fused feature vector, the target category to which the file to be downloaded belongs is determined;

[0046] The storage path of the target category is used as the storage path of the file to be downloaded, and the storage path of the file to be downloaded is output.

[0047] In one optional implementation, the filename processing unit is specifically used for:

[0048] The target category to which the file to be downloaded belongs is determined by clustering the fused feature vector with the feature vectors of each pre-saved category.

[0049] In one optional implementation, the storage path determination unit may also be used for:

[0050] If a user confirms the path input, the file to be downloaded is saved according to the storage path.

[0051] If a path modification operation is received from the user, the file to be downloaded will be saved according to the modified storage path.

[0052] Thirdly, embodiments of this application provide a terminal device, including a memory and a processor, wherein a computer program is stored in the memory; the processor is configured to read the computer program stored in the memory and execute it, so that any of the methods provided in the first aspect are executed.

[0053] Fourthly, this application provides a computer-readable storage medium storing a computer program or instructions, which, when executed by a terminal device, causes the terminal device to perform any of the methods provided in the first aspect above.

[0054] Fifthly, this application provides a computer program product comprising a computer program or instructions that, when executed by a terminal device, implement any of the methods provided in the first aspect above.

[0055] The file storage path recommendation method provided in this application determines the semantic feature vector and topic feature vector of the file to be downloaded, respectively. Based on the fused feature vector obtained by concatenating the semantic feature vector and the topic feature vector, the storage path of the file to be downloaded is determined, and storage path recommendations are made so that different types of files are saved to different storage paths, making it easier for users to manage and find downloaded files later. Furthermore, determining the storage path of the file to be downloaded based on the fused feature vector obtained by concatenating the semantic feature vector and the topic feature vector can improve the accuracy of the recommended storage path. Attached Figure Description

[0056] Figure 1 This application provides an illustration of a file storage path recommendation method.

[0057] Figure 2 A flowchart illustrating a file storage path recommendation method provided in this application embodiment;

[0058] Figure 3 A schematic diagram of a display interface provided in an embodiment of this application;

[0059] Figure 4 A schematic diagram of a file storage path recommendation device provided in an embodiment of this application;

[0060] Figure 5 This is a schematic diagram of the structure of a terminal device provided in an embodiment of this application. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the embodiments of this application will be described in detail below with reference to the accompanying drawings. The terminology used in the implementation section of this application is only used to explain specific embodiments of this application and is not intended to limit this application. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] In this application embodiment, "multiple" refers to two or more. Therefore, in this application embodiment, "multiple" can also be understood as "at least two". "At least one" can be understood as one or more, such as one, two, or more. For example, "including at least one" means including one, two, or more, and it does not limit which ones are included. For example, including at least one of A, B, and C, then it could include A, B, C, A and B, A and C, B and C, or A and B and C. "And / or" describes the association relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / ", unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0063] Unless otherwise stated, the ordinal numbers such as "first" and "second" mentioned in the embodiments of this application are used to distinguish multiple objects, and are not used to limit the order, sequence, priority or importance of multiple objects.

[0064] Currently, when downloading files through a browser, the browser generally has a default download location. However, all downloaded files are stored in the same default location, resulting in disorganized and disorganized file storage, which is inconvenient for users to find downloaded files. Therefore, this application provides a file storage path recommendation method.

[0065] Figure 1 The following examples illustrate application scenarios of the file storage path recommendation method provided in the embodiments of this application. For instance... Figure 1 As shown, in this application scenario, terminal device 100 can connect to server 200 via a network. Server 200 can be a cloud storage server, which can store music files, video files, and other files of various types. In practical applications, terminal device 100 can connect to different servers via a network, and different servers can be used to store files from different fields.

[0066] Terminal device 100 can be a mobile phone, vehicle-mounted terminal, portable terminal, desktop computer, or personal computer, etc. Terminal device 100 has a browser or other software that provides download functionality installed. Terminal device 100 can download files from server 200 using the software and save the downloaded files locally.

[0067] To avoid downloaded files being stored haphazardly in a default location, this application provides a file storage path recommendation method that saves different types of files to different locations on the terminal device.

[0068] The following section uses downloading files via a browser as an example to provide a detailed explanation of the file storage path recommendation method provided in this application embodiment. Figure 2 An exemplary flowchart of a file storage path recommendation method provided in an embodiment of this application is shown. This file storage path recommendation method can be... Figure 1 The terminal device 100 executes, such as Figure 2 As shown, the method may include the following steps:

[0069] S201, Generate a semantic feature vector of the file name based on the predicate headword and the content governed by the predicate headword contained in the file name of the file to be downloaded.

[0070] When a terminal device receives a user's instruction to download a file via a browser, it obtains the filename of the file to be downloaded and determines the predicate headword and the content governed by the predicate headword in the filename. For example, a filename can be considered a sentence, and the predicate headword can refer to the verb in the sentence. In a sentence, the predicate verb is the core of the sentence and governs other words in the sentence, while the verb itself is not governed by other words. Therefore, the content governed by the predicate headword can refer to the subject before the verb, the object after the verb, etc. For example, suppose the filename is titled "Methods for Extracting Chemical Raw Materials from Petroleum." In this filename, the predicate headword is "extract," and the content governed by the predicate headword includes the following participles: "petroleum," "chemical raw materials," and "method."

[0071] In some embodiments, filenames can be syntactically analyzed using a skeleton dependency tree to obtain a two-layer structure: the predicate head and the content governed by the predicate head. The skeleton dependency tree can analyze the relationships between words in a sentence according to predefined grammatical rules, automatically transforming a linear sequence of sentences into a word-level tree structure. All modifiers in a sentence are directly or indirectly subordinate to the predicate head in some dependency relationship; that is, except for the predicate head, all other words in the sentence directly depend on another word, and no single word in the sentence can simultaneously depend on two or more other words. In other words, except for the predicate head, each word in the sentence has its own parent node, and the words in the sentence can form a tree structure that reflects the sentence's grammatical structure.

[0072] After determining the content governed by the predicate head, the governing content can be segmented into multiple content segments. Based on the word vector of each content segment, the distance between each content segment and the predicate head, and the word vector of the predicate head, a semantic feature vector for the filename is generated. The distance between any content segment and the predicate head can be understood as the number of content segments located between that content segment and the predicate head, i.e., the position of that content segment. Word vectors, also known as word vector features, describe the semantic features of words in natural language text. They typically refer to dense vectors or matrices that can be understood by machines, obtained by vectorizing words represented in natural language. Word vectors are the numerical representation of words in natural language text within a machine. The word vector of any word can be obtained using the Word2Vec model.

[0073] For example, weights can be assigned to the word vectors of each content segment based on its position, generating a semantic feature vector containing the predicate head word, the word vectors of each content segment, and the weights. The semantic feature vector G can be represented as: G = {w, Aσi}, where w is the word vector of the predicate head word in the filename, i.e., the vectorized representation of the predicate head word; A is the weight vector, A = {ω}. i |i∈{1,2,…,n}};ω i The distance between the i-th content segment and the predicate headword i∈n,;σ i Let be the word vector of the i-th content segment, i.e., the vectorized representation of the i-th content segment. The greater the distance between the content segment and the predicate head word, i.e., the farther the content segment is from the predicate head word, the smaller its weight, with a maximum weight of 1. For example, in the phrase "methods for extracting chemical raw materials from petroleum," the predicate head word is "extract," and the content governed by the predicate head word is "petroleum," "chemical raw materials," and "methods." Their weights are 6 / 11, 3 / 11, and 2 / 11, respectively.

[0074] By using the distance between the content segment and the predicate headword as the weight of the content segment, the importance of different content segments can be reflected, which can express the semantic features of the file name more meticulously and accurately.

[0075] S202: Generate a topic feature vector for the file name based on the topic segmentation words contained in the file name of the file to be downloaded, and the topic to which each topic segmentation word belongs.

[0076] Terminal devices can use the Latent Dirichlet Allocation (LDA) model to determine the topic words contained in filenames and the topic to which each topic word belongs. The LDA model is a topic model used to identify topic words in a sentence or paragraph. For example, suppose a filename D contains N... i Each topic is segmented into words, w ij This represents the j-th topic segment, where each topic segment corresponds to a potential topic, z. ij Let the j-th topic be the topic to which it belongs. Input the filename D into the LDA model, and the LDA model will output a topic distribution matrix, which can be represented as w. i ={w ij |j∈{1,2,…,N} i}},z i ={z ij |j∈{1,2,…,N} i The vector dimension of the topic distribution matrix is ​​the same as the dimension of the semantic feature vector G mentioned above.

[0077] The terminal device stores pre-defined datasets on multiple topics, or it can download datasets on multiple topics via a network. Each topic dataset includes high-frequency words for that topic. For example, in a dataset on the topic of "cat," high-frequency words might include words like "cat" and "fish." After determining the topic to which each topic segmentation word in a filename belongs, the terminal device can obtain the high-frequency words for each topic and determine the cosine distance between the high-frequency words of each topic and the filename. For example, using the j-th topic segmentation word w... ij The corresponding theme w ij For example, let's take topic w ij The high-frequency words are represented as n-dimensional feature vectors, and the word segments contained in the file name D are also represented as n-dimensional feature vectors. Then the topic z ij The cosine distance between the high-frequency words and the filename D can be expressed as: Among them, z ijk Indicates topic w ij The k-th dimension of the feature vector, D k Let represent the k-th dimension of the feature vector of filename D. The terminal device generates a topic feature vector P for each filename based on topic segmentation, the cosine distance between the high-frequency words of each topic and the filename, and the topic segmentation itself. The topic feature vector P can be represented as...

[0078] The cosine distance between the high-frequency words of the topic to which the topic segment belongs and the file name is used as the weight of the topic segment. Different topic segments have different weights, which can make the obtained topic feature vector more accurately reflect the topic features of the file name.

[0079] In some embodiments, the execution order of steps S201 and S202 can be interchanged, or steps S201 and S202 can be executed simultaneously.

[0080] S203, concatenate the semantic feature vector and the topic feature vector to obtain the fused feature vector.

[0081] Terminal devices can use vector concatenation to superimpose semantic feature vectors and topic feature vectors to form a fused input vector G′={P;G} that includes word meaning features and overall semantic features. G′ represents the text vectorization representation of the fused semantic feature vector and topic feature vector, and “;” is the vector concatenation symbol.

[0082] S204. Based on the fused feature vector, determine the storage path of the file to be downloaded and output the storage path.

[0083] The terminal device pre-stores feature vectors for multiple file categories, with different categories stored in different locations within the device (i.e., different categories correspond to different storage paths). The K-means clustering algorithm is used to cluster the fused feature vectors with the pre-stored feature vectors for each category, determining the target category of the file to be downloaded. Where a j Referring to the j-th category, iterative calculations are performed on the fused feature vector and the pre-saved feature vectors of each category until a... j Convergence, i.e., a j Until the minimum error is reached, the category of the feature vector that belongs to the same class as the fused feature vector at this point is taken as the target category of the file to be downloaded. The target category is the topic category with the closest similarity to the filename of the file to be downloaded. In other embodiments, the terminal device can also determine the target category with the closest similarity to the filename of the file to be downloaded by other means, for example, by comparing the fused feature vector with the feature vectors of each pre-saved category one by one.

[0084] The terminal device has multiple categories of storage paths pre-stored. The terminal device obtains the storage path of the target category, uses the storage path of the target category as the storage path of the file to be downloaded, and outputs the storage path of the file to be downloaded.

[0085] For example, the terminal device can display the storage path of the file to be downloaded. Figure 3As shown, the display interface also includes two buttons: "One-Click Download" and "Save As". If the user agrees with the storage path recommended by the terminal device, they can click the "One-Click Download" button. The terminal device receives the user's confirmation of the path entered by clicking the "One-Click Download" button and saves the file to be downloaded according to the confirmed storage path. If the user does not agree with the storage path recommended by the terminal device, they can click the "Save As" button. The terminal device receives the user's modification of the path entered by clicking the "Save As" button, obtains the modified storage path, and saves the file to be downloaded according to the modified storage path.

[0086] This application provides a method for intelligently recommending download file location paths by fully considering word meaning and overall semantic features. When downloading a file, the method obtains the filename, analyzes the overall syntactic structure of the filename using a skeleton dependency tree, assigns positional weights to content segmentation words, and marks the dependency relationship between the predicate head word of the filename and the content segmentation words it governs, as well as the relative position of the content segmentation words and the predicate head word, to determine a semantic feature vector representing the sentence structure of the filename. Based on the fused feature vector obtained by concatenating the semantic feature vector and the topic feature vector, the storage path of the file to be downloaded is determined, enabling intelligent recommendation of download location paths. This facilitates "one-click download" for users, accurately stores downloaded files, and facilitates user operation and subsequent management and retrieval of downloaded files. Furthermore, by using skeleton dependency tree semantic and LDA topic fusion extraction technology, the storage path of the most similar topic category is obtained, which can improve the accuracy of the recommended storage path, reduce the tedious operation of manually adding file storage paths, and improve the intelligence and flexibility of the file download process.

[0087] Based on the same inventive concept as the above embodiments, this application also provides a file storage path recommendation device, which can be applied to... Figure 1 In the terminal device 100 shown, the file storage path recommendation device can be used to implement the functions of the above method embodiments, and thus can achieve the beneficial effects of the above method embodiments.

[0088] like Figure 4 As shown, the file storage path recommendation device may include a filename processing unit 401 and a storage path determination unit 402. This file storage path recommendation device is used to implement the above... Figure 2 The function shown in the method embodiment. When the file storage path recommendation device is used to implement Figure 2The function of the method embodiment shown is as follows: the filename processing unit 401 can be used to execute S201 to S203, and the storage path determination unit 402 can be used to execute S204. For example, the filename processing unit 401 is used to generate a semantic feature vector of the filename based on the predicate headword and the content governed by the predicate headword contained in the filename of the file to be downloaded; generate a topic feature vector of the filename based on the topic segmentation words contained in the filename and the topic to which each topic segmentation word belongs; and concatenate the semantic feature vector and the topic feature vector to obtain a fused feature vector. The storage path determination unit 402 is used to determine the storage path of the file to be downloaded based on the fused feature vector and output the storage path.

[0089] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0090] Determine the predicate headword and the content governed by the predicate headword in the file name;

[0091] The dominant content is segmented into multiple content segments;

[0092] Based on the word vector of each content segment, the distance between each content segment and the predicate headword, and the word vector of the predicate headword, a semantic feature vector of the file name is generated.

[0093] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0094] By performing syntactic structure analysis on filenames using a skeleton dependency tree, we can obtain the predicate headword and the content governed by the predicate headword.

[0095] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0096] Determine the topic segments contained in the filename and the topic to which each topic segment belongs;

[0097] Based on the cosine distance between the high-frequency words of each topic and the file name, as well as topic word segmentation, a topic feature vector for the file name is generated.

[0098] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0099] The LDA model is used to determine the topic segments contained in the file name and the topic to which each topic segment belongs.

[0100] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0101] Based on the fused feature vector, determine the target category to which the file to be downloaded belongs;

[0102] Use the storage path of the target category as the storage path of the file to be downloaded, and output the storage path of the file to be downloaded.

[0103] In one alternative implementation, the filename processing unit 401 can be specifically used for:

[0104] The target category of the file to be downloaded is determined by clustering the fused feature vector with the feature vectors of each pre-saved category.

[0105] In one alternative implementation, the storage path determination unit 402 can also be used for:

[0106] If the user confirms the path input, the file to be downloaded will be saved according to the storage path.

[0107] If a path modification operation is received from the user, the file to be downloaded will be saved according to the modified storage path.

[0108] Based on the same inventive concept as the above embodiments, this application also provides a terminal device, which can be... Figure 1 The terminal device 100 shown is used to implement the functions of the above method embodiments, and thus can achieve the beneficial effects of the above method embodiments.

[0109] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; as shown below. Figure 5 As shown, in this embodiment of the application, the electronic device 100 includes: a processor 101, a display 102, a memory 103, an input device 106, a bus 105, and a communication module 104; the processor 101, memory 103, input device 106, display 102, and communication module 104 are all connected through the bus 105, which is used for data transmission between the processor 101, memory 103, display 102, communication module 104, and input device 106.

[0110] The memory 103 can be used to store software programs and modules, such as the program instructions / modules corresponding to the text processing method in this embodiment. The processor 101 executes various functional applications and data processing of the electronic device 100 by running the software programs and modules stored in the memory 103, such as the file storage path recommendation method provided in this embodiment. The memory 103 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs of at least one application, etc.; the data storage area may store data created according to the use of the electronic device 100 (storage paths corresponding to different types of files), etc. In addition, the memory 103 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0111] The processor 101 is the control center of the electronic device 100. It connects various parts of the electronic device 100 via the bus 105 and various interfaces and lines. It performs various functions of the electronic device 100 and processes data by running or executing software programs and / or modules stored in the memory 103 and calling data stored in the memory 103. Optionally, the processor 101 may include one or more processing units, such as a CPU, GPU, digital processing unit, etc.

[0112] In this embodiment of the application, the processor 101 is configured to read program instructions from the memory 102 and execute the following steps: generate a semantic feature vector of the file name based on the predicate headword contained in the file name of the file to be downloaded and the content governed by the predicate headword; generate a topic feature vector of the file name based on the topic segmentation words contained in the file name and the topic to which each topic segmentation word belongs; concatenate the semantic feature vector and the topic feature vector to obtain a fused feature vector; determine the storage path of the file to be downloaded based on the fused feature vector, and output the storage path.

[0113] The processor 101 can display the recommended storage path to the user via the display 102.

[0114] The processor 101 can also connect to a network via the communication module 104 to download files from a server.

[0115] Input device 106 is mainly used to obtain operation commands input by the user. The input device 106 may also be different depending on the electronic device. For example, when the electronic device is a computer, the input device 106 can be an input device such as a mouse or keyboard; when the electronic device is a smartphone, tablet computer or other portable device, the input device 106 can be a touch screen.

[0116] In some possible implementations, various aspects of the file storage path recommendation method provided in this application can also be implemented as a program product comprising program code. When the program product is run on a computer device, the program code causes the computer device to perform the steps of the file storage path recommendation method according to the various exemplary embodiments of this application described above. For example, the computer device can perform actions such as... Figure 2 The flowchart of the file storage path recommendation method shown in steps S201 to S204 is as follows.

[0117] This application also provides a computer storage medium storing computer-executable instructions for implementing the file storage path recommendation method described in any embodiment of this application.

[0118] The storage medium can be any available medium that a computer can access. For example, but not limited to, a computer-readable medium can include RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer.

[0119] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0120] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0121] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0122] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0123] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of protection of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for recommending file storage paths, characterized in that, include: Generate a semantic feature vector of the file name based on the predicate headword contained in the file name of the file to be downloaded and the content governed by the predicate headword; Based on the topic-based word segments contained in the filename of the file to be downloaded, and the topic to which each topic segment belongs, generate a topic feature vector for the filename; The semantic feature vector and the topic feature vector are concatenated to obtain a fused feature vector; Based on the fused feature vector, the target category to which the file to be downloaded belongs is determined from a plurality of preset different categories, wherein the different categories correspond to different storage paths; Use the storage path of the target category as the storage path of the file to be downloaded, and output the storage path of the file to be downloaded; The step of generating a semantic feature vector for the filename based on the predicate headword contained in the filename of the file to be downloaded and the content governed by the predicate headword includes: By performing syntactic structure analysis on the filename using the skeleton dependency tree, the predicate headword and the content governed by the predicate headword are obtained; The controlled content is segmented into multiple content segments; Based on the word vector of each content segment, the word order interval between each content segment and the predicate head word, and the word vector of the predicate head word, a semantic feature vector of the file name is generated, wherein the word order interval is the number of content segments located between the content segment and the predicate head word.

2. The method according to claim 1, characterized in that, The step of generating a topic feature vector for the filename based on topic segmentation contained in the filename of the file to be downloaded and the topic to which each topic segmentation belongs includes: Determine the topic segments contained in the filename and the topic to which each topic segment belongs; Based on the cosine distance between the feature vectors of high-frequency words in each topic and the feature vector of the file name, and the topic segmentation, a topic feature vector of the file name is generated.

3. The method according to claim 2, characterized in that, The step of determining the topic segments contained in the filename and the topic to which each topic segment belongs includes: The topic segments contained in the file name and the topic to which each topic segment belongs are determined using the LDA model.

4. The method according to claim 1, characterized in that, The step of determining the target category to which the file to be downloaded belongs from a set of preset different categories based on the fused feature vector includes: The target category to which the file to be downloaded belongs is determined by clustering the fused feature vector with the feature vectors of each pre-saved category.

5. The method according to claim 1, characterized in that, After outputting the storage path of the file to be downloaded, the method further includes: If a user confirms the path input, the file to be downloaded is saved according to the storage path. If a path modification operation is received from the user, the file to be downloaded will be saved according to the modified storage path.

6. A file storage path recommendation device, characterized in that, include: The filename processing unit is configured to generate a semantic feature vector of the filename based on the predicate headword contained in the filename of the file to be downloaded and the content governed by the predicate headword; and to generate a topic feature vector of the filename based on the topic segmentation words contained in the filename and the topic to which each topic segmentation word belongs. The semantic feature vector and the topic feature vector are concatenated to obtain a fused feature vector; The target category determination unit is used to determine the target category to which the file to be downloaded belongs from a plurality of preset different categories based on the fused feature vector, wherein the different categories correspond to different storage paths; A storage path determination unit is used to determine the storage path of the target category as the storage path of the file to be downloaded, and output the storage path of the file to be downloaded; The filename processing unit is specifically used for: By performing syntactic structure analysis on the filename using the skeleton dependency tree, the predicate headword and the content governed by the predicate headword are obtained; The controlled content is segmented into multiple content segments; Based on the word vector of each content segment, the word order interval between each content segment and the predicate head word, and the word vector of the predicate head word, a semantic feature vector of the file name is generated, wherein the word order interval is the number of content segments located between the content segment and the predicate head word.

7. A terminal device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program that can run on the processor, and when the computer program is executed by the processor, it implements the method of any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the method of any one of claims 1 to 5.

9. A computer program product, characterized in that, It includes computer-executable instructions for causing a computer to perform the method as described in any one of claims 1 to 5.

Images