Method and device for recommending data application scenarios, electronic equipment and storage medium

Abstract

The application provides a data application scenario recommendation method and device, electronic equipment and storage medium, and belongs to the technical field of computer information processing. The method comprises the following steps: obtaining original text data and a subject label corresponding to the original text data; performing subject extraction processing on the original text data to obtain first subject information; obtaining a first application scenario according to the first subject information; obtaining a second application scenario according to the subject label; if the first application scenario is the same as the second application scenario, taking the first application scenario or the second application scenario as a recommended application scenario; wherein the first application scenario, the second application scenario and the recommended application scenario comprise a high-security scenario, a medium-security scenario and a low-security scenario. By combining the first application scenario and the second application scenario, the accuracy of the application scenario obtained by the subject model is judged, and the accuracy of the obtained recommended application scenario is improved.

Description

Technical Field

[0001] This application relates to the field of computer information processing technology, and in particular to a method, apparatus, electronic device and storage medium for recommending data application scenarios. Background Technology

[0002] With the continuous development of information technology and big data technology, market themes have collected and mastered a large amount of user data information. However, for general market themes, mastering a large amount of raw data itself is not valuable. What is truly valuable is the data products and specific data application scenarios obtained after analyzing and mining the data. Therefore, how to extract themes from massive amounts of data and obtain their suitable application scenarios is an urgent problem to be solved at this stage.

[0003] Most related technologies use topic models to obtain topic information from data. However, topic models are statistical models that cluster the implicit semantic structure of documents in an unsupervised learning manner. They have the problem that the extracted topics do not match the actual topics, so the application scenarios obtained are not accurate enough. Summary of the Invention

[0004] The main objective of this application is to propose a method, apparatus, electronic device, and storage medium for recommending data application scenarios, so as to improve the accuracy of the obtained application scenarios.

[0005] To achieve the above objectives, a first aspect of this application proposes a method for recommending data application scenarios, comprising: acquiring raw text data and corresponding topic tags for the raw text data; performing topic extraction processing on the raw text data to obtain first topic information; obtaining a first application scenario based on the first topic information; obtaining a second application scenario based on the topic tags; if the first application scenario is the same as the second application scenario, then the first application scenario or the second application scenario is used as the recommended application scenario; wherein the first application scenario, the second application scenario, and the recommended application scenario all include: high-security scenario, medium-security scenario, and low-security scenario.

[0006] The data application scenario recommendation method proposed in this application extracts topics from the original text data using a topic model, obtains a corresponding first application scenario based on the extracted topic information, and then compares it with a second application scenario obtained based on topic tags. Only when the comparison results are the same will the corresponding recommended application scenario be obtained. By combining the first and second application scenarios, the accuracy of the application scenarios obtained by the topic model is judged, thereby improving the accuracy of the obtained recommended application scenarios.

[0007] In some embodiments, the step of performing topic extraction processing on the original text data to obtain first topic information includes: performing word segmentation processing on the original text data to obtain segmented text data; performing stop word removal processing on the segmented text data to obtain key text data; and inputting the key text data into a topic model for topic extraction to obtain the first topic information.

[0008] In some embodiments, the step of inputting the key text data into a topic model for topic extraction to obtain the first topic information includes: inputting the key text data into multiple initial models with different numbers of topics to obtain multiple initial topic information; calculating the similarity between any two topics in the initial topic information to obtain model similarity information; obtaining an optimal model based on the multiple model similarity information; and inputting the key text data into the optimal model to obtain the first topic information.

[0009] In some embodiments, calculating the similarity between any two topics in the initial topic information to obtain model similarity information includes: obtaining a preset number of keyword information in the topics to obtain corresponding information to be compared; constructing word frequency vectors of the information to be compared respectively; calculating the similarity between any two word frequency vectors to obtain the model similarity information.

[0010] In some embodiments, obtaining the optimal model based on multiple model similarity information includes: obtaining multiple initial models based on multiple model similarity information, using the multiple initial models as a candidate model group; and selecting the initial model with the smallest number of topics in the candidate model group as the optimal model.

[0011] In some embodiments, obtaining the first application scenario based on the first topic information includes: matching the first topic information with a first application scenario lookup table to obtain the first application scenario; wherein the first application scenario lookup table includes topic information and application scenarios corresponding to the topic information.

[0012] In some embodiments, obtaining the second application scenario based on the topic tag includes: matching the topic tag with a second application scenario lookup table to obtain the second application scenario; wherein the second application scenario lookup table includes topic tags and application scenarios corresponding to the topic tags.

[0013] To achieve the above objectives, a second aspect of this application proposes a data application scenario recommendation device, comprising: a data acquisition module for acquiring raw text data and corresponding topic tags; a topic extraction module for performing topic extraction processing on the raw text data to obtain first topic information; an application scenario acquisition module for obtaining a first application scenario based on the first topic information and a second application scenario based on the topic tags; and a comparison module for recommending either the first application scenario or the second application scenario as the recommended application scenario when the first application scenario and the second application scenario are the same; wherein the first application scenario, the second application scenario, and the recommended application scenario include: high-security scenarios, medium-security scenarios, and low-security scenarios.

[0014] To achieve the above objectives, a third aspect of this application provides an electronic device comprising: at least one memory; at least one processor; at least one program; said program being stored in the memory, and the processor executing said at least one program to implement the method described in the first aspect of this application.

[0015] To achieve the above objectives, a fourth aspect of this application provides a storage medium that is a computer-readable storage medium storing computer-executable instructions for causing a computer to perform the method described in the first aspect above. Attached Figure Description

[0016] Figure 1 This is a flowchart of a recommended method for data application scenarios provided in the embodiments of this application;

[0017] Figure 2 This is a classification diagram of application scenarios in the embodiments of this application;

[0018] Figure 3 yes Figure 1 Flowchart of S200;

[0019] Figure 4 yes Figure 3 Flowchart of S230 in China;

[0020] Figure 5 yes Figure 4 Flowchart of S232 in China;

[0021] Figure 6 yes Figure 4 Flowchart of S233 in China;

[0022] Figure 7 This is a graph showing the relationship between model similarity and the number of topics in the embodiments of this application;

[0023] Figure 8 This is a schematic diagram of the hardware structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0025] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0026] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0027] First, let's analyze some of the terms used in this application:

[0028] Natural Language Processing (NLP): NLP uses computers to process, understand, and utilize human language (such as Chinese and English). NLP is a branch of artificial intelligence and an interdisciplinary field of computer science and linguistics, often referred to as computational linguistics. NLP includes syntactic analysis, semantic analysis, and discourse understanding. It is commonly used in machine translation, handwritten and printed character recognition, speech recognition and text-to-speech conversion, information retrieval, information extraction and filtering, text classification and clustering, sentiment analysis, and opinion mining. It involves data mining, machine learning, knowledge acquisition, knowledge engineering, artificial intelligence research, and linguistic research related to language computation.

[0029] Latent Dirichlet Allocation (LDA): LDA is a topic model, a type of unsupervised learning algorithm that does not require manually labeled training sets during training. It only requires a document set and a specified number of topics. It can output the topics of each document in the document set according to a probability distribution, and for each topic, it can find some words to describe it. LDA has wide applications in natural language processing, including text mining and its subfields such as text topic recognition, text classification, and text similarity calculation.

[0030] Application Scenarios: The application scenarios for the data in this application are based on the classification of data confidentiality to determine the applicable application environments. Application scenarios can include high-confidentiality, medium-confidentiality, and low-confidentiality scenarios. When data falls under a high-confidentiality scenario, data processing and application must be carried out in a high-confidentiality environment; when data falls under a low-confidentiality environment, data processing and application do not require a high level of confidentiality.

[0031] Related technologies only use topic models to obtain topic information from data. However, topic models are statistical models that cluster the implicit semantic structure of a document set in an unsupervised learning manner. They have the problem that the extracted topics do not match the actual topics, so the application scenarios obtained are not accurate enough.

[0032] Based on this, embodiments of this application provide a method, apparatus, electronic device, and storage medium for recommending data application scenarios, which can improve the accuracy of the obtained recommended application scenarios.

[0033] This application provides a method, apparatus, electronic device, and storage medium for recommending data application scenarios. The specific details are illustrated in the following embodiments. First, the method for recommending data application scenarios in this application is described.

[0034] The data application scenario recommendation method provided in this application embodiment can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, or smartwatch, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms; the software can be an application that implements the data application scenario recommendation method, etc., but is not limited to the above forms.

[0035] Figure 1 This is an optional flowchart of a recommended method for data application scenarios provided in the embodiments of this application. Figure 1 The method may include, but is not limited to, steps S100 to S500.

[0036] S100, retrieve the original text data and the corresponding topic tags for the original text data;

[0037] S200, perform topic extraction processing on the original text data to obtain the first topic information;

[0038] S300, based on the first theme information, obtains the first application scenario;

[0039] S400, based on the topic tags, obtains the second application scenario;

[0040] S500: If the first application scenario is the same as the second application scenario, then the first application scenario or the second application scenario will be used as the recommended application scenario.

[0041] Specifically, in step S100, the user-inputted raw text data is first obtained. This raw text data is a document set containing at least one document. For example, if the raw text data consists of all reviews for a product, each review is treated as a document, and all reviews form a document set. Application scenario recommendations are then made based on this document set.

[0042] After obtaining the raw text data, it is necessary to tag the raw text data with topic tags to obtain tagged text data. For example, each document in the document set can be manually tagged with a topic to obtain topic tags. It is understood that the topic in this application refers to keywords describing the content of a document; therefore, topic tagging involves tagging the keywords in the document. During the tagging process, a single document can have multiple topics, and different documents can have the same topic. After tagging is completed, the topic tags for the corresponding raw text data can be obtained.

[0043] Then, in step S200, the original text data is processed using a topic model to extract topics, thereby obtaining the first topic information. Specifically, the LDA topic model can be used for topic extraction. When using the LDA topic model, it is only necessary to set the number of topics to be extracted from the document set. The obtained first topic information includes: the probability that each document belongs to a specific topic, and the distribution of topic words in the specific topic. From this, the topic word distribution of each document can be obtained, thus obtaining the topic of each document.

[0044] After obtaining the first topic information, step S300 is used to obtain the first application scenario based on the first topic information. It can be understood that when there is only one document in the original text data, the obtained first application scenario is the application scenario corresponding to that document; when there are multiple documents in the original text data, the obtained first application scenario includes the application scenarios corresponding to each document individually.

[0045] Simultaneously, the topic tags in the tagged text data are obtained. In step S400, the second application scenario of the corresponding document is obtained based on the topic tags. The second application scenario can be obtained based on the correspondence between topic tags and application scenarios, or it can be obtained by manually tagging the application scenarios of the corresponding documents while simultaneously tagging the original text data with topic tags, thus obtaining the second application scenario for the original text data.

[0046] After obtaining the first and second application scenarios, step S500 compares the first and second application scenarios. If the comparison result is the same, it means that the topic extracted by the topic model is the same as or similar to the topic obtained through manual annotation, and therefore the application scenarios are the same. This mutual verification improves the accuracy of the recommended application scenarios. If the comparison result is different, the first application scenario calculated by the system can be used as the recommended application scenario, or the matching degree of the relevant raw data in the corresponding application scenario can be output to the user for reference.

[0047] The first application scenario, second application scenario, and recommended application scenario in this application all include: high-security scenario, medium-security scenario, and low-security scenario. (See reference...) Figure 2 This is a diagram illustrating the specific application scenarios. A high-security scenario occurs when the original text data needs to be stored in a private cloud and the data does not leave the database; a medium-security scenario occurs when the data needs to be stored in a private cloud and the data leaves the database; a low-security scenario occurs when the data needs to be stored in a public cloud and the data leaves the database; and a medium-security scenario occurs when the data needs to be stored in a public cloud and the data does not leave the database. For example, if the user-inputted original text data is a collection of product review documents, and the document topics obtained through topic extraction and topic tagging are keywords such as "satisfied," "easy to use," and "shipped," then based on the topics, it can be determined that the corresponding document collection contains review data. Product review data is generally stored in a public cloud and the data leaves the database for user viewing, which falls under the low-security scenario. In some other embodiments, the application scenario can also be determined through information such as the meaning of the topics, and can be set according to specific needs.

[0048] By using the recommended method for the data application scenario of the present application, users can obtain the recommended application scenarios that conform to the specific content of the data by inputting the original text data, which is convenient for users to select different data processing methods according to the recommended application scenarios. For example, when the recommended application scenario is a high-security scenario, users should pay attention to protecting the confidentiality and security of the data usage environment when using the data to prevent data leakage; when the recommended application scenario is a low-security scenario, users can use the data publicly to reduce the costs incurred due to data confidentiality.

[0049] In some embodiments, referring to Figure 3 , the step of performing topic extraction processing on the original text data to obtain the first topic information specifically includes:

[0050] S210, performing word segmentation processing on the original text data to obtain word-segmented text data;

[0051] S220, performing stop-word removal processing on the word-segmented text data to obtain key text data;

[0052] S230, inputting the key text data into a topic model for topic extraction to obtain the first topic information.

[0053] The data input into the topic model needs to go through natural language processing first. Specifically, in step S210, first perform word segmentation processing on the obtained original text data. When the original text data has multiple documents, perform word segmentation processing on each document separately. The word segmentation method can adopt dictionary-based word segmentation methods (such as the maximum forward matching method, reverse maximum matching method, etc.), statistics-based word segmentation methods, and knowledge-understanding-based word segmentation methods (training word segmentation models through machine learning), etc.

[0054] Through word segmentation processing, word-segmented text data is obtained. Since there will be meaningless words such as "de" and "děi" in the word-segmented text data after word segmentation, therefore, through step S220, use a stop-word list to perform stop-word removal processing on the word-segmented text data to remove the meaningless words in the word-segmented text data and obtain key text data. Finally, through step S230, input the processed key text data into the topic model for topic extraction, thereby obtaining the first topic information.

[0055] In some embodiments, referring to Figure 4 , the step of inputting the key text data into a topic model for topic extraction to obtain the first topic information includes:

[0056] S231, inputting the key text data into multiple initial models with different numbers of topics respectively to obtain multiple initial topic information;

[0057] S232, calculate the similarity between any two topics in the initial topic information to obtain model similarity information;

[0058] S233, the optimal model is obtained based on the similarity information of multiple models;

[0059] S234. Input the key text data into the optimal model to obtain the first topic information.

[0060] Since topic models with different numbers of topics extract different topics, it is necessary to select a topic model with the optimal number of topics for topic extraction. Step S231 generates multiple topic models with different numbers of topics as initial models. For example, based on the size of the input document set, ten topic models with 1 to 10 topics are selected as initial models. Then, key text data is input into each initial model to obtain the corresponding ten initial topic information. For example, an initial model with two topics yields initial topic information including two types of topics and their corresponding keyword information.

[0061] Then, in step S232, the similarity between any two topics in the initial topic information is calculated to obtain the model similarity information. The similarity in this application can be calculated using cosine similarity. It can be understood that when the initial model has 1 topic, it obtains only one topic, and therefore the model similarity information is 1; when the initial model has 3 topics, it obtains 3 topics, so it is necessary to calculate the similarity between any two topics and then take their average as the model similarity information.

[0062] Then, in step S233, the optimal model is obtained based on the obtained similarity information from multiple models. When the number of topics in the topic model is small, the differences between the multiple topics are insufficient, resulting in relatively similar topics. Therefore, it is necessary to select an initial model with an appropriate number of topics as the optimal model. When using cosine similarity to calculate the similarity between topics, the smaller the result, the less similar the topics are, indicating significant differences between them. Therefore, the model with the least similar topics among the initial models is selected as the optimal model for topic extraction from the key text data.

[0063] In some embodiments, refer to Figure 5 The steps for calculating the similarity between any two topics in the initial topic information to obtain model similarity information include:

[0064] S235, Obtain a preset number of keyword information from the topic to obtain the corresponding comparison information;

[0065] S236, construct the word frequency vectors of the information to be compared respectively;

[0066] S237, calculate the similarity between any two word frequency vectors to obtain model similarity information.

[0067] Specifically, since the initial topic information calculated in step S231 contains varying numbers of topics, and each topic has multiple corresponding keywords, step S235 retrieves a preset number of keyword information from each topic, using these keywords as comparison information. For example, the first 100 keywords from each topic are retrieved and merged into a single set, which is then used as comparison information. It is understood that the number of comparison information items is the same as the number of topics in the model. The acquisition of keywords can be adjusted based on parameters such as the size of the document set.

[0068] Then, in step S236, word frequency vectors are constructed based on the frequency of keywords in the information to be compared. Finally, the similarity between the multiple word frequency vectors is calculated in step S237 to obtain the similarity information of the model. For example, cosine similarity can be used to calculate the similarity between word frequency vectors.

[0069] In some embodiments, refer to Figure 6 The steps to obtain the optimal model based on the similarity information of multiple models include:

[0070] S238. Based on the similarity information of multiple models, multiple initial models are obtained, and these multiple initial models are used as a candidate model group.

[0071] S239. Select the initial model with the smallest number of topics in the candidate model group as the optimal model.

[0072] When the range of topics selected for the initial model is large, multiple initial models may be obtained based on model similarity information. For example, two initial models with 3 and 4 topics have the same model similarity and both have the minimum calculated result (obtained by cosine similarity). In this case, the two initial models with 3 and 4 topics are placed in the candidate model group. When the number of topics for a model is large, the computational cost of topic extraction is also large. To reduce the computational cost of topic extraction, the initial model with the smallest number of topics is selected as the optimal model for subsequent topic extraction.

[0073] Reference Figure 7This is an embodiment of the present application, showing the relationship between model similarity and the number of topics. The model similarity is calculated using cosine similarity. When the number of topics exceeds 3, the model similarity is the average of multiple cosine similarities. The figure shows that when the initial model has 2 topics, its model similarity is the lowest. That is, the difference between topics obtained by the topic model with 2 topics is the greatest. Compared with other topic models with a larger number of topics, this topic model has a smaller number of topics, resulting in better topic discrimination with less computation. The cosine similarity can be calculated using equation (1). Assuming there are two n-dimensional vectors, represented by Ai and Bi respectively, the cosine similarity between Ai and Bi is:

[0074]

[0075] Where cosθ represents cosine similarity, n represents the dimension of the vector, and i represents the i-th data in the n-dimensional vector.

[0076] In some embodiments, the step of obtaining a first application scenario based on first topic information includes: matching the first topic information with a first application scenario lookup table to obtain the first application scenario; wherein the first application scenario lookup table includes topic information and application scenarios corresponding to the topic information. The first application scenario lookup table of this application is pre-set. By matching the first topic information obtained after topic extraction, the application scenario applicable to the corresponding document in the original text data can be identified, thus improving the efficiency of identifying the application scenario.

[0077] In some embodiments, the step of obtaining a second application scenario based on topic tags includes: matching the topic tags using a second application scenario lookup table to obtain the second application scenario; wherein the second application scenario lookup table includes topic tags and application scenarios corresponding to the topic tags. The second application scenario lookup table in this application is pre-set. By matching topic tags, the applicable application scenario for the corresponding document in the original text data can be quickly obtained, improving the efficiency of identifying application scenarios. It is understood that if the topic tags are the same as the keyword library of the first topic information, the first application scenario lookup table and the second application scenario lookup table can be merged.

[0078] This application embodiment also provides a recommendation device for data application scenarios, including:

[0079] The data acquisition module is used to acquire raw text data and the corresponding topic tags for the raw text data;

[0080] The topic extraction module is used to extract topics from the original text data to obtain the first topic information.

[0081] The application scenario acquisition module is used to obtain the first application scenario based on the first topic information and the second application scenario based on the topic tags.

[0082] The comparison module is used to recommend the first application scenario or the second application scenario when the first application scenario and the second application scenario are the same.

[0083] The first application scenario, the second application scenario, and the recommended application scenario include: high-security scenario, medium-security scenario, and low-security scenario.

[0084] In some embodiments, the process by which the topic extraction module performs topic extraction processing on the original text data to obtain first topic information includes: performing word segmentation processing on the original text data to obtain segmented text data; performing stop word removal processing on the segmented text data to obtain key text data; and inputting the key text data into the topic model for topic extraction to obtain first topic information.

[0085] In some embodiments, the process of the topic extraction module inputting key text data into a topic model to extract topics and obtain first topic information includes: inputting key text data into multiple initial models with different numbers of topics to obtain multiple initial topic information; calculating the similarity between any two topics in the initial topic information to obtain model similarity information; obtaining an optimal model based on the multiple model similarity information; and inputting key text data into the optimal model to obtain the first topic information.

[0086] In some embodiments, the process by which the topic extraction module calculates the similarity between any two topics in the initial topic information to obtain model similarity information includes: obtaining a preset number of keyword information in the topics to obtain the corresponding information to be compared; constructing word frequency vectors for the information to be compared respectively; and calculating the similarity between any two word frequency vectors to obtain model similarity information.

[0087] In some embodiments, the process by which the topic extraction module obtains the optimal model based on multiple model similarity information includes: obtaining multiple initial models based on multiple model similarity information, and using the multiple initial models as a candidate model group; selecting the initial model with the smallest number of topics in the candidate model group as the optimal model.

[0088] In some embodiments, the process by which the application scenario acquisition module obtains the first application scenario based on the first topic information includes: matching the first topic information with a first application scenario lookup table to obtain the first application scenario; wherein the first application scenario lookup table includes topic information and application scenarios corresponding to the topic information.

[0089] In some embodiments, the process by which the application scenario acquisition module obtains a second application scenario based on topic tags includes: matching topic tags with a second application scenario lookup table to obtain the second application scenario; wherein the second application scenario lookup table includes topic tags and application scenarios corresponding to topic tags.

[0090] The data application scenario recommendation device proposed in this application is used to implement the data application scenario recommendation method in the above embodiments. Its specific implementation process is the same as the data application scenario recommendation method, and will not be described in detail here.

[0091] This application also provides an electronic device, including:

[0092] At least one memory;

[0093] At least one processor;

[0094] At least one program;

[0095] The program is stored in a memory, and the processor executes the at least one program to implement the recommended method for implementing the data application scenarios described above. The electronic device can be any smart terminal, including mobile phones, tablets, personal digital assistants (PDAs), and in-vehicle computers.

[0096] Please see Figure 8 , Figure 8 The hardware structure of an electronic device according to another embodiment is illustrated. The electronic device includes:

[0097] The processor can be implemented using a general-purpose CPU (Central Processing Unit), microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to achieve the technical solutions provided in the embodiments of this application.

[0098] The memory can be implemented in the form of ROM (Read-Only Memory), static storage device, dynamic storage device, or RAM (Random Access Memory). The memory can store the operating system and other applications. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory and the processor calls and executes the recommended methods for data application scenarios in the embodiments of this application.

[0099] Input / output interfaces are used to implement information input and output;

[0100] The input / communication interface is used to enable communication and interaction between this device and other devices. Communication can be achieved through wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0101] A bus is used to transfer information between various components of a device, such as processors, memory, input / output interfaces, and input / communication interfaces.

[0102] The processor, memory, input / output interface, and input / communication interface are interconnected within the device via a bus.

[0103] This application also provides a storage medium, which is a computer-readable storage medium storing computer-executable instructions for causing a computer to execute the recommended method for the above-described data application scenario.

[0104] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs. Furthermore, memory may include high-speed random access memory, and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, memory may optionally include memory remotely located relative to the processor, and these remote memories can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0105] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

[0106] Those skilled in the art will understand that the technical solutions shown in the figures do not constitute a limitation on the embodiments of this application, and may include more or fewer steps than shown, or combine certain steps, or different steps.

[0107] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] Those skilled in the art will understand that all or some of the steps in the methods disclosed above, as well as the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, or suitable combinations thereof.

[0109] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0110] It should be understood that in this application, "at least one (item)" means one or more, and "more than" means two or more. "And / or" is used to describe the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: only A exists, only B exists, and both A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple.

[0111] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0112] 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.

[0113] Furthermore, the functional units in the various embodiments of this application 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. The integrated unit can be implemented in hardware or as a software functional unit.

[0114] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes multiple 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 application. The aforementioned storage medium includes various media capable of storing programs, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0115] The embodiments of this application have been described in detail above with reference to the accompanying drawings. However, this application is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of this application. Furthermore, unless otherwise specified, the embodiments and features described in the embodiments of this application can be combined with each other.

Claims

1. A recommendation method for a data application scenario, characterized in that, include: Obtain the original text data and the corresponding topic tags for the original text data; The original text data is subjected to topic extraction processing to obtain the first topic information; The first application scenario is obtained based on the first topic information; The second application scenario is obtained based on the aforementioned topic tags; If the first application scenario is the same as the second application scenario, then the first application scenario or the second application scenario shall be used as the recommended application scenario. The first application scenario, the second application scenario, and the recommended application scenario all include: high security scenario, medium security scenario, and low security scenario. The high security scenario is characterized by the original text data being stored in a private cloud and the data not leaving the database. The medium security scenario is characterized by the original text data being stored in a private cloud and the data leaving the database. The low security scenario is characterized by the original text data being stored in a public cloud and the data not leaving the database. The step of performing topic extraction processing on the original text data to obtain first topic information includes: The original text data is segmented into words to obtain segmented text data; The segmented text data is processed to remove stop words, resulting in key text data; The key text data is input into the topic model for topic extraction to obtain the first topic information; The step of inputting the key text data into a topic model for topic extraction to obtain the first topic information includes: The key text data is input into multiple initial models with different numbers of topics to obtain multiple initial topic information; wherein, the initial topic information includes two types of topics and their corresponding keyword information; Calculate the similarity between any two topics in the initial topic information to obtain model similarity information; The optimal model is obtained based on multiple model similarity information; wherein, the optimal model is the least similar model among the topics; The key text data is input into the optimal model to obtain the first topic information.

2. The method according to claim 1, characterized in that, The calculation of the similarity between any two topics in the initial topic information to obtain model similarity information includes: Obtain a preset number of keyword information from the topic to obtain the corresponding information to be compared; Construct the word frequency vectors of the information to be compared respectively; Calculate the similarity between any two of the word frequency vectors to obtain the model similarity information.

3. The method according to claim 2, characterized in that, The step of obtaining the optimal model based on multiple model similarity information includes: Based on the similarity information of multiple models, multiple initial models are obtained, and the multiple initial models are used as a candidate model group; The initial model with the smallest number of topics in the candidate model group is selected as the optimal model.

4. The method according to any one of claims 1 to 3, characterized in that, The step of obtaining the first application scenario based on the first topic information includes: The first application scenario is obtained by matching the first topic information with a first application scenario lookup table; wherein, the first application scenario lookup table includes topic information and application scenarios corresponding to the topic information.

5. The method according to any one of claims 1 to 3, characterized in that, The process of obtaining the second application scenario based on the topic tags includes: The second application scenario is obtained by matching the topic tags using a second application scenario lookup table; wherein the second application scenario lookup table includes topic tags and application scenarios corresponding to the topic tags.

6. A recommendation device for a data application scenario, characterized in that, include: The data acquisition module is used to acquire raw text data and corresponding topic tags for the raw text data; A topic extraction module is used to perform topic extraction processing on the original text data to obtain first topic information; An application scenario acquisition module is used to obtain a first application scenario based on the first topic information and a second application scenario based on the topic tags. A comparison module is used to recommend either the first application scenario or the second application scenario when the first application scenario and the second application scenario are the same. The first application scenario, the second application scenario, and the recommended application scenario include: high security scenario, medium security scenario, and low security scenario. The high security scenario is characterized by the original text data being stored in a private cloud and the data not leaving the database. The medium security scenario is characterized by the original text data being stored in a private cloud and the data leaving the database. The low security scenario is characterized by the original text data being stored in a public cloud and the data not leaving the database. The step of performing topic extraction processing on the original text data to obtain first topic information includes: The original text data is segmented into words to obtain segmented text data; The segmented text data is processed to remove stop words, resulting in key text data; The key text data is input into the topic model for topic extraction to obtain the first topic information; The step of inputting the key text data into a topic model for topic extraction to obtain the first topic information includes: The key text data is input into multiple initial models with different numbers of topics to obtain multiple initial topic information; wherein, the initial topic information includes two types of topics and their corresponding keyword information; Calculate the similarity between any two topics in the initial topic information to obtain model similarity information; The optimal model is obtained based on multiple model similarity information; wherein, the optimal model is the least similar model among the topics.

7. An electronic device, characterized in that, include: At least one memory; At least one processor; At least one program; The program is stored in the memory, and the processor executes the at least one program to achieve the following: The method as described in any one of claims 1 to 5.

8. A storage medium, said storage medium being a computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions for causing a computer to perform: The method as described in any one of claims 1 to 5.

Images