Resource allocation methods, apparatus, equipment and computer storage media

Abstract

This application discloses a resource allocation method, apparatus, device, and computer storage medium, relating to the field of big data technology. The resource allocation method includes: obtaining target keywords from question-and-answer text; querying a keyword index library for core keywords matching the target keywords; if a first core keyword in a first core field matches the target keyword, configuring resources for the target object based on the resource content in the first core field; if no core keyword matching the target keyword can be found, querying a similar keyword in the keyword index library that matches the target keyword; and if any similar keyword in a second core field matches the target keyword, configuring resources for the target object based on the resource content in the second core field. According to embodiments of this application, personalized configuration of target object resources can be achieved according to different marketing scenarios, effectively meeting the diverse needs of different scenarios.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of big data, and particularly relates to a resource configuration method and device, equipment and a computer storage medium. BACKGROUND

[0002] In the online application programs (APPs) of major banks today, many virtual digital spokespersons have appeared. Virtual digital people are virtual service assistants with real images, and can interact with customers in real time through language, body and expression movements. They can often meet the business consultation needs of customers, and are exclusive one-on-one virtual customer service assistants, which can solve the problems of insufficient staff and insufficient capacity of physical customer managers, so as to enable the banking industry to better serve customers through technological means. Therefore, at present, virtual digital people have been gradually applied to various marketing scenarios.

[0003] However, although the business handling type digital people are relatively popular at present, the decoration such as background and costume of the digital people is relatively simple, and cannot meet the diversity needs of enterprise institutions in different scenarios, and cannot form sufficient visual attraction for users, which is easy to cause visual fatigue. SUMMARY

[0004] The embodiments of the present application provide a resource configuration method, device, equipment and computer storage medium, which can realize personalized configuration of target object resources according to different marketing scenarios, and effectively meet the diversity needs of different scenarios.

[0005] In a first aspect, the embodiments of the present application provide a resource configuration method, which comprises:

[0006] obtaining a target keyword in a question and answer text;

[0007] querying a core keyword matching the target keyword in a keyword index library; wherein the keyword index library stores a plurality of core fields; any core field of the plurality of core fields includes a core keyword, at least one approximate keyword and resource content;

[0008] in a case where a first core keyword in a first core field matches the target keyword, configuring resources for the target object according to the resource content in the first core field; the plurality of core fields include the first core field;

[0009] in a case where the core keyword matching the target keyword cannot be queried, querying an approximate keyword matching the target keyword in the keyword index library;

[0010] in a case where any approximate keyword in a second core field matches the target keyword, configuring resources for the target object according to the resource content in the second core field.

[0011] In some possible implementation manners, the first word sense identifier is further included in any of the plurality of core fields; and the resource configuration method further includes:

[0012] In a case where an approximate keyword matching the target keyword cannot be queried, the question and answer text is parsed into the first word sense identifier based on a natural language processing technology;

[0013] The keyword index library is queried for a word sense identifier matching the first word sense identifier;

[0014] In a case where the second word sense identifier included in the third core field matches the first word sense identifier, the target object is configured with the resource content in the third core field.

[0015] In some possible implementation manners, in a case where an approximate keyword matching the target keyword cannot be queried, the resource configuration method further includes:

[0016] The question and answer text is parsed based on a natural language processing technology to obtain N target approximate keywords corresponding to the first word sense identifier, N being a positive integer;

[0017] In a case where the second word sense identifier included in the third core field matches the first word sense identifier, the M target approximate keywords are added to the third core field;

[0018] The M target approximate keywords are approximate keywords in the N target approximate keywords that do not coincide with at least one approximate keyword in the third core field, and M is a positive integer.

[0019] In some possible implementation manners, the resource configuration method further includes:

[0020] In a case where a word sense identifier matching the first word sense identifier cannot be queried in the keyword index library, the question and answer text or / and the first word sense identifier is stored;

[0021] If the storage times of the question and answer text or / and the first word sense identifier are greater than a preset threshold value within a preset time period, a fourth core field is generated; the fourth core field includes a core keyword, at least one approximate keyword, and resource content determined based on the question and answer text or / and the first word sense identifier;

[0022] The fourth core field is stored in the keyword index library.

[0023] In some possible implementation manners, before the core keyword matching the target keyword in the keyword index library is queried, the resource configuration method further includes:

[0024] determine at least one approximate keyword corresponding to each core keyword in the plurality of core keywords and resource content based on the plurality of core keywords;

[0025] generate a plurality of core fields based on the plurality of core keywords, the at least one approximate keyword corresponding to each core keyword in the plurality of core keywords, and the resource content;

[0026] store the plurality of core fields into a keyword index library.

[0027] In some possible implementation manners, the target object is a digital person; the resource content is classified into background type resource and clothing type resource; and the resource configuration method further includes:

[0028] in a case where the resource content is the background type resource, configuring background content of the digital person based on the resource content and re-rendering a picture; or,

[0029] in a case where the resource content is the clothing type resource, configuring clothing content of the digital person based on the resource content and re-rendering a picture.

[0030] In a second aspect, an embodiment of the present application provides a resource configuration device, which includes:

[0031] a first obtaining module configured to obtain a target keyword in a question and answer text;

[0032] a first querying module configured to query a core keyword matched with the target keyword in a keyword index library; wherein the keyword index library stores a plurality of core fields; any core field in the plurality of core fields includes a core keyword, at least one approximate keyword, and resource content;

[0033] a first configuring module configured to, in a case where a first core keyword in a first core field matches the target keyword, configure a target object based on resource content in the first core field; the plurality of core fields include the first core field;

[0034] a second querying module configured to, in a case where the core keyword matched with the target keyword cannot be queried, query an approximate keyword matched with the target keyword in the keyword index library;

[0035] a second configuring module configured to, in a case where any approximate keyword in a second core field matches the target keyword, configure the target object based on resource content in the second core field.

[0036] In a third aspect, an embodiment of the present application provides a resource configuration device, which includes:

[0037] a processor and a memory storing computer program instructions;

[0038] The processor implements the resource configuration method provided in any one of the embodiments of the application when executing the computer program instructions.

[0039] In a fourth aspect, the embodiments of the application provide a computer readable storage medium, which stores computer program instructions. The computer program instructions are executed by a processor to implement the resource configuration method provided in any one of the embodiments of the application.

[0040] In a fifth aspect, the embodiments of the application provide a computer program product. Instructions in the computer program product are executed by a processor of an electronic device to enable the electronic device to perform the resource configuration method provided in any one of the embodiments of the application.

[0041] The resource configuration method, device, equipment and computer storage medium provided in the embodiments of the application can match the core keyword in the keyword index library by extracting the target keyword in the question and answer text, and can further match the approximate keyword to expand the matching range in the case that the core keyword cannot be matched. Therefore, the target object can be configured with resources according to the resource content corresponding to the matched core keyword or approximate keyword, and the personalized configuration of the target object resources in different marketing scenarios can be realized, and the diversity demand of different scenarios can be effectively met. BRIEF DESCRIPTION OF DRAWINGS

[0042] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings needed in the embodiments of the application will be briefly introduced. Those skilled in the art can obtain other drawings according to these drawings without creating any creative labor.

[0043] Figure 1 is a flowchart of the resource configuration method provided in an embodiment of the application;

[0044] Figure 2is a scenario embodiment flow schematic diagram of a resource configuration method provided by an embodiment of the present application;

[0045] Figure 3 is a structural schematic diagram of a resource configuration apparatus provided by an embodiment of the present application;

[0046] Figure 4 is a structural schematic diagram of a resource configuration device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0047] The features and exemplary embodiments of various aspects of the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are intended to explain the present application, but not to limit the present application. The present application can be implemented without some of the specific details for those skilled in the art. The following description of the embodiments is merely intended to provide a better understanding of the present application by showing examples of the present application.

[0048] It should be noted that, in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "contain" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or device. Without more limitations, the elements defined by the statement "include" do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0049] It should be noted that the acquisition, storage, use, processing and other data in the technical solutions of the present application comply with the relevant provisions of national laws and regulations.

[0050] Digital human streaming: digital human streaming service (including three-dimensional digital human streaming service and two-dimensional digital human streaming service) outputs the video picture rendered by the digital human in the cloud to the client through streaming media, and pushes the video stream rendered by the digital human to the server.

[0051] As described in the background section, at present, in the bank business scene, the virtual digital human as a virtual service assistant with real image can interact with customers in real time through language, body and expression actions, and can enable the banking industry to better serve customers through technological means.

[0052] However, although the current business type digital human is relatively popular, the resource content such as the background and the costume of the digital human push stream is single, and cannot meet the diversity demand of the enterprise institution in different scenes, and cannot form enough visual attraction to the user, and is easy to cause visual fatigue.

[0053] In view of the above, in order to solve the problems in the prior art, the embodiments of the present application provide a resource configuration method, device, equipment, storage medium and computer program product. It should be noted that the embodiments provided by the present application are not used to limit the scope of the present application.

[0054] Firstly, the resource configuration method provided by the embodiments of the present application is introduced.

[0055] Figure 1 The flowchart of the resource configuration method provided by an embodiment of the present application is shown. The resource configuration method is applied to an electronic device, which can include a server or a user terminal, etc. As shown in the figure, the resource configuration method includes the following steps: Figure 1

[0056] S110, obtaining a target keyword in a question and answer text;

[0057] S120, querying a core keyword matching the target keyword in a keyword index library; wherein the keyword index library stores a plurality of core fields; any core field in the plurality of core fields includes a core keyword, at least one approximate keyword and resource content;

[0058] S130, in the case that the first core keyword in the first core field matches the target keyword, configuring resources for the target object according to the resource content in the first core field; the plurality of core fields include the first core field;

[0059] S140, in the case that the core keyword matching the target keyword cannot be queried, querying the approximate keyword matching the target keyword in the keyword index library;

[0060] S150, in the case that any approximate keyword in the second core field matches the target keyword, configuring resources for the target object according to the resource content in the second core field.

[0061] ​The resource configuration method of this application embodiment involves: obtaining target keywords from question-and-answer text; querying core keywords in a keyword index library that match the target keywords; configuring resources for the target object based on the resource content in the first core field if the first core keyword in the first core field of the keyword index library matches the target keywords; querying approximate keywords in the keyword index library if no core keywords matching the target keywords can be found; and configuring resources for the target object based on the resource content in the second core field if any approximate keyword in the second core field matches the target keywords.

[0062] This application provides a resource configuration method that can extract target keywords from question and answer text to match core keywords in a keyword index library. Furthermore, it can further match similar keywords to expand the matching range when core keywords cannot be matched. This allows for resource configuration of the target object based on the resource content corresponding to the queried core keywords or similar keywords, thereby enabling personalized configuration of target object resources in different marketing scenarios and effectively meeting the diverse needs of different scenarios.

[0063] The specific implementation methods of steps 110 to 150 above will be described in detail below.

[0064] In S110, the target keywords are obtained from the question and answer text.

[0065] The following example illustrates this using a bank's multi-marketing business context. For instance, in a certain business scenario, the system can obtain the text "I want to transfer money" by receiving the user's manual or voice input on the corresponding display interface.

[0066] After obtaining the above question and answer text, the target keyword "transfer" in the question and answer text "I want to transfer money" can be determined and extracted by relevant keyword extraction algorithms or pre-trained keyword extraction models.

[0067] It should be noted that in some other embodiments, if multiple keywords are extracted from a question-and-answer text, the weights of these keywords can be evaluated by referring to the marketing practices of banks, and the keyword with the highest weight can be selected as the target keyword. Furthermore, if there are still multiple keywords with the highest weight, they can be randomly selected from among them. This application does not impose strict limitations on this; the specific method for extracting the target keyword can be determined based on actual business needs.

[0068] In S120, during the specific implementation, the core keywords that match the target keyword are queried in the keyword index library; the keyword index library stores multiple core fields; any one of the multiple core fields may include the core keyword, at least one approximate keyword, and resource content.

[0069] For example, the keyword index library includes a core field, in which the core keyword is "transfer", and similar keywords are "money transfer", "remittance", etc. The resource content can be set according to actual marketing needs. For example, when the target audience in the current text is a digital human, the above resource content can be a pre-set background resource presented when the digital human interacts in the "transfer" scenario. This application does not impose any restrictions on this.

[0070] In some possible implementations, to construct the keyword index library more reasonably and facilitate keyword matching queries in the keyword index library in subsequent steps, the resource configuration method may further include the following before querying the core keywords in the keyword index library that match the target keyword:

[0071] Based on multiple core keywords, at least one approximate keyword and resource content corresponding to each of the multiple core keywords are determined;

[0072] Based on multiple core keywords, at least one similar keyword corresponding to each of the multiple core keywords, and resource content, multiple core fields are generated.

[0073] Store multiple core fields in a keyword index library.

[0074] In practice, relevant business personnel can pre-determine multiple core keywords based on different business marketing scenarios to support coverage of diverse business needs. After determining multiple core keywords, approximate keywords corresponding to each keyword are determined using appropriate natural language processing methods or by considering user conversation habits. For example, after determining a core keyword as "transfer," relevant business personnel can configure "money transfer" or "remittance" as approximate keywords based on users' actual spoken language habits. The above resource content can be set according to the actual marketing needs corresponding to different core keywords, and this application does not impose specific restrictions on this.

[0075] Thus, after determining multiple core keywords, at least one approximate keyword corresponding to each of the multiple core keywords, and resource content, multiple corresponding core fields are generated respectively.

[0076] For example, when generating core fields, the content of the core keyword field, the content of its corresponding similar keyword field, and the address link corresponding to the resource content can be simply merged, etc. This application does not impose specific restrictions on this.

[0077] Thus, after generating multiple core fields, these core fields are stored in a keyword index library to construct the keyword index library of this application. This keyword index library can be implemented using a relevant database, such as Access, MySQL, or BD2; this application does not impose any specific restrictions on this.

[0078] In S130, if the first core keyword in the first core field matches the target keyword, the target object is configured with resources based on the resource content in the first core field; multiple core fields may include the first core field.

[0079] In practice, the core keywords that match the target keywords are queried in the keyword index. If the keyword index contains multiple core fields, including a first core field, and that first core field includes a first core keyword that matches the target keywords (e.g., both are "transfer"), then the target object is configured with resources based on the resource content in that first core field.

[0080] The specific target objects mentioned above can be determined according to actual business needs, such as virtual customer service, digital human marketing assistants, etc. This application does not impose specific restrictions on this.

[0081] In S140, if the core keyword matching the target keyword cannot be found, the approximate keyword matching the target keyword is queried from the keyword index.

[0082] In practice, in the keyword index library, a core keyword often corresponds to one or more similar keywords. Therefore, the number of similar keywords is more than that of the core keyword, and correspondingly, the matching complexity is also higher than that of the core keyword.

[0083] Therefore, in this application, the core keywords that match the target keyword are first queried in the keyword index library. If no core keywords that match the target keyword are found, the keyword index library is then queried to determine whether the keyword index library includes similar keywords that match the target keyword.

[0084] In S150, if any approximate keyword in the second core field matches the target keyword, the target object is configured with resources based on the resource content in the second core field.

[0085] In practice, the core keywords that match the target keywords are queried in the keyword index. If the keyword index contains multiple core fields including a second core field, and any approximate keyword in the second core field matches the target keyword, then the target object can be configured with resources based on the resource content in the second core field.

[0086] As an example, the target keyword mentioned above could be "remittance". The second core field includes the core keyword "transfer" and several similar keywords such as "transfer money" and "remittance". The target keyword obviously matches the similar keyword "remittance" in the second core field.

[0087] In some possible implementations, considering the possibility that the keyword index library does not contain similar keywords matching the target keywords, furthermore, to fully ensure the personalized configuration of target object resources in different marketing scenarios and effectively meet the diverse needs of different scenarios, any one of the above core fields may also include a semantic identifier; the resource configuration method may also include:

[0088] When no similar keywords matching the target keyword can be found, the question-and-answer text is parsed into first semantic identifiers based on natural language processing technology.

[0089] Query the keyword index for word meaning identifiers that match the first word meaning identifier;

[0090] If the second semantic identifier, which may be included in the third core field, matches the first semantic identifier, the target object is configured with resources based on the resource content in the third core field.

[0091] Natural language processing (NPL) is an important branch of artificial intelligence. Its purpose is to use computers to intelligently process natural language. It can typically be used for operations such as stemming, lemmatization, sentiment analysis, text semantic similarity, and text summarization.

[0092] In this embodiment, any of the aforementioned core fields in the keyword index library may also include a semantic identifier. When a similar keyword matching the target keyword cannot be found, the question-and-answer text can be parsed into a first semantic identifier based on natural language processing technology.

[0093] Subsequently, the first semantic identifier obtained from the parsing is compared with the semantic identifiers included in multiple core fields to determine whether there is a semantic identifier in the keyword index that matches the first semantic identifier.

[0094] If the keyword index is found to contain a third core field, which in turn contains a second semantic identifier that matches the first semantic identifier, then the target object can be configured based on the resource content in the third core field.

[0095] It should be noted that the aforementioned first semantic identifier can specifically be represented as a semantic ID (Identity document), a string, or an encoded value, etc., and this application does not impose specific limitations on it. The types of the aforementioned second semantic identifier and the first semantic identifier are not elaborated upon in this application.

[0096] In some possible implementations, when no approximate keywords matching the target keyword can be found, in order to more fully guarantee the personalized configuration of target resources in different marketing scenarios and effectively meet the diverse needs of different scenarios, the resource configuration method may further include:

[0097] The question-and-answer text is parsed using natural language processing technology to obtain N target approximate keywords corresponding to the first semantic identifier, where N is a positive integer;

[0098] If the second semantic identifier, which may be included in the third core field, matches the first semantic identifier, then M target approximate keywords are added to the third core field;

[0099] Among them, the M target approximation keywords are the approximation keywords that do not overlap with at least one approximation keyword in the third core field among the N target approximation keywords, and M is a positive integer.

[0100] In this embodiment, when no approximate keywords matching the target keyword can be found, the question and answer text is parsed based on natural language processing technology. In addition to obtaining the aforementioned first semantic identifier, N approximate keywords corresponding to the first semantic identifier can also be determined.

[0101] In this way, after retrieving the second meaning identifier in the third core field based on the first meaning identifier, the above N approximate keywords are compared with the approximate keywords contained in the third core field, and then the M target approximate keywords that do not overlap with the approximate keywords contained in the third core field are determined.

[0102] In this way, adding the M target similar keywords to the third core field makes the similar keywords included in the third core field more comprehensive, which is conducive to improving the success rate of subsequent keyword matching. This ensures more comprehensive personalized configuration of target resources in different marketing scenarios, so as to effectively meet the diverse needs of different scenarios.

[0103] In some possible implementations, considering that there may still be cases where the keyword index library does not contain the aforementioned second semantic identifier, further, in order to expand and optimize the keyword index library to fully ensure the personalized configuration of target object resources in different marketing scenarios, thereby effectively meeting the diverse needs of different scenarios, the resource configuration method may also include:

[0104] If a word sense identifier that matches the first word sense identifier cannot be found in the keyword index, store the question and answer text and / or the first word sense identifier;

[0105] If the number of times the question-and-answer text and / or the first semantic identifier is stored exceeds a preset threshold within a preset time period, a fourth core field is generated; the fourth core field may include core keywords determined based on the question-and-answer text and / or the first semantic identifier, at least one approximate keyword, and resource content;

[0106] Store the fourth core field in the keyword index library.

[0107] In practice, when a word sense identifier that matches the first word sense identifier cannot be found in the keyword index library, the question and answer text and / or the first word sense identifier obtained from parsing the question and answer text are stored.

[0108] Thus, when the number of storage times accumulates to a value greater than a preset threshold within a preset time period, it indicates that the question-and-answer text and its corresponding first semantic identifier appear frequently in the business scenario. It is advisable to expand and optimize the keyword index based on the question-and-answer text and / or the first semantic identifier to more fully and comprehensively cover multiple business scenarios.

[0109] Based on this, the newly added core keyword, at least one approximate keyword, and resource content are determined based on the question-and-answer text and / or the first semantic identifier. The fourth core field is generated based on the newly added core keyword, at least one approximate keyword, and resource content, and the fourth core field is stored in the keyword index library to expand the core fields in the keyword index library. This can more fully guarantee the personalized configuration of target object resources in different marketing scenarios and effectively meet the diverse needs of different scenarios.

[0110] In some possible implementations, considering more specific banking business scenarios and actual marketing needs, the aforementioned target object can specifically be a digital human; the aforementioned resource content can be divided into background type resources and clothing type resources; the resource configuration method may also include:

[0111] If the resource content is a background type resource, configure the background content of the digital human based on the resource content and re-render the image; or,

[0112] When the resource content is clothing type, the clothing content of the digital human is configured based on the resource content, and the screen is re-rendered.

[0113] After obtaining the resource content that needs to be replaced, determine whether it is a background resource type or a clothing type, and then dynamically change the front-end rendered screen.

[0114] Specifically, resource content for different marketing scenarios can be personalized. For example, various backgrounds with different themes such as festivals, culture, technology, information halls, and outlets can be set in advance. Digital human clothing can be adapted to different scenarios based on different target keywords, and scene-specific clothing and makeup can be matched according to target keywords to enhance the realism and technological feel of digital human interaction.

[0115] After successfully matching the target keywords, the corresponding resource content will be obtained. If the resource content is a background type resource, the background content of the digital human will be configured based on the resource content, and the image will be re-rendered. Alternatively, if the resource content is a clothing type resource, the clothing content of the digital human will be configured based on the resource content, and the image will be re-rendered.

[0116] It should be added that if the above-mentioned resource content is clothing type resource, the resource content may not only include clothing resources of digital humans, but also makeup, hairstyles, etc. that are adapted to clothing resources. This application does not impose specific restrictions on this.

[0117] In this embodiment, different resource content is matched using target keywords, and the resource configuration of the digital human is changed accordingly. When the marketing scenario changes, the question-and-answer text obtained in the corresponding scenario is different, and consequently, the target keywords extracted based on the question-and-answer text are also different. This ultimately leads to changes in the resource content configuration of the digital human in different marketing scenarios. In summary, this embodiment can realize personalized configuration of digital human backgrounds, clothing, and other resources according to different marketing scenarios, effectively meeting the diverse needs of different scenarios and greatly enhancing the visual appeal to users.

[0118] To facilitate understanding of the resource configuration method provided in the above embodiments, the following describes the method using a specific scenario embodiment. Figure 2 This is a schematic flowchart of a scenario embodiment of the resource configuration method provided in this application.

[0119] The application scenario of this example can be: in the banking marketing business scenario, it is necessary to personalize the background resources or clothing resources of the digital human according to different marketing scenarios.

[0120] In this scenario embodiment, a keyword index library containing multiple core fields can be pre-built; any one of the multiple core fields includes a core keyword, at least one approximate keyword, a semantic identifier, and resource content.

[0121] Furthermore, this scenario embodiment also includes an NLP (Natural Language Processing) parsing interface. When a question-and-answer text is input into this interface, the interface will parse the question-and-answer text into corresponding semantic identifiers and internally parsed approximate keywords.

[0122] This scenario implementation example may specifically include the following steps:

[0123] Step 1: Obtain the question and answer text and extract the target keywords from it.

[0124] Step two: Match core keywords.

[0125] Based on the target keywords mentioned above, the system matches the core keywords in the keyword index. If a core keyword is found, the system returns the resource content in the corresponding core field and changes the background type resource content or switches the clothing type resource content of the digital human based on the resource content.

[0126] Step 3: Match similar keywords.

[0127] If the target keyword cannot be matched in the core keywords, then the similar keywords in the keyword index will be matched. There will be more similar keywords than core keywords, and the matching complexity is greater than that of core keywords. If a similar keyword is found, the corresponding resource content will be returned according to the core field to which the similar keyword belongs, and the digital human resource configuration will be changed.

[0128] Step 4, NLP matching.

[0129] If no approximate keyword matching the target keyword can be found, the NLP interface will be switched. The NLP interface allows the above question and answer text to be passed in. The NLP interface will parse the question and answer content into corresponding semantic tags and internally parsed approximate keywords based on the corresponding natural language processing technology.

[0130] Thus, after obtaining the semantic identifiers corresponding to the above question and answer text, the matching semantic identifiers are found in the keyword index library based on the semantic identifiers, and the corresponding resource content is returned based on the core fields described by the semantic identifiers. Based on the resource content, the background type resource content or clothing type resource content of the digital human is changed or switched.

[0131] In addition, the parsed similar keywords are added to the corresponding core fields, making the similar keywords in the keyword index library more comprehensive. This helps to improve the probability of successful keyword matching in subsequent scenarios, thereby more fully guaranteeing the personalized configuration of target resources in different marketing scenarios and effectively meeting the diverse needs of different scenarios.

[0132] Step 5, change the background and clothing:

[0133] After a successful match with the target keywords, the corresponding resource content will be obtained. If the resource content is a background type resource, the background content of the digital human will be configured based on the resource content, and the image will be re-rendered. Alternatively, if the resource content is a clothing type resource, the clothing content of the digital human will be configured based on the resource content, and the image will be re-rendered.

[0134] In this scenario embodiment, keyword matching based on relevance similarity is performed when matching target keywords. First, the target keyword is extracted from the question-and-answer text and matched against core keywords in the keyword index. If no matching core keyword is found, similar keywords in the keyword index are matched. If no similar keyword is found, the question-and-answer text is parsed using an NLP interface to obtain semantic identifiers and similar keywords. Based on these semantic identifiers, a matching semantic identifier is found in the keyword index, and the corresponding resource content is returned based on the core field of the semantic identifier, and the digital human's resource configuration is changed. Then, the parsed similar keywords are added to the keyword index for easy retrieval in future searches.

[0135] The resource configuration scheme provided in this scenario embodiment can personalize the configuration of digital human backgrounds, clothing, and other resources according to different marketing scenarios, effectively meeting the diverse needs of different scenarios and greatly enhancing the visual appeal to users. Furthermore, it also provides more scenario options during the digital human business processing, fully meeting the business service needs of enterprise mobile banking platforms.

[0136] Based on the resource allocation method provided in the above embodiments, this application also provides a resource allocation device corresponding to the above resource allocation method, which will be described below. Figure 3 A detailed introduction to the resource allocation device is provided.

[0137] Figure 3 A schematic diagram of the structure of a resource configuration device provided in an embodiment of this application is shown. Figure 3 The resource allocation device 300 shown includes:

[0138] The first acquisition module 310 is used to acquire target keywords in the question and answer text;

[0139] The first query module 320 is used to query the core keywords in the keyword index library that match the target keyword; wherein, the keyword index library stores multiple core fields; any one of the multiple core fields includes the core keyword, at least one approximate keyword, and resource content;

[0140] The first configuration module 330 is used to configure resources for the target object according to the resource content in the first core field when the first core keyword in the first core field matches the target keyword; the multiple core fields include the first core field;

[0141] The second query module 340 is used to query similar keywords that match the target keyword in the keyword index library when the core keyword that matches the target keyword cannot be found.

[0142] The second configuration module 350 is used to configure resources for the target object based on the resource content in the second core field when any approximate keyword in the second core field matches the target keyword.

[0143] The resource configuration device of this application embodiment obtains target keywords from question-and-answer text by setting corresponding functional modules; queries core keywords that match the target keywords in a keyword index library; if the first core keyword in the first core field of the keyword index library matches the target keyword, resource configuration is performed on the target object according to the resource content in the first core field; if no core keyword matching the target keyword can be found, similar keywords matching the target keyword are queried in the keyword index library; if any similar keyword in the second core field matches the target keyword, resource configuration is performed on the target object according to the resource content in the second core field. The resource configuration device provided in this application embodiment can extract target keywords from question-and-answer text to match core keywords in a keyword index library, and can further match similar keywords to expand the matching range when no core keyword can be matched, thereby configuring resources on the target object according to the resource content corresponding to the queried core keywords or similar keywords. This enables personalized configuration of target object resources in different marketing scenarios, effectively meeting the diverse needs of different scenarios.

[0144] In some possible implementations, considering the possibility that the keyword index library does not contain similar keywords matching the target keywords, furthermore, to fully ensure the personalized configuration of target object resources in different marketing scenarios and effectively meet the diverse needs of different scenarios, any one of the multiple core fields may also include a semantic identifier; the resource configuration device 300 may also include:

[0145] The first parsing module can be used to parse the question and answer text into the first semantic identifier based on natural language processing technology when no similar keywords matching the target keyword can be found.

[0146] The third query module can be used to query the keyword index library for word meaning identifiers that match the first word meaning identifier;

[0147] The third configuration module can be used to configure resources for the target object based on the resource content in the third core field, provided that the second semantic identifier, which may be included in the third core field, matches the first semantic identifier.

[0148] In some possible implementations, when no approximate keywords matching the target keyword can be found, in order to more fully guarantee the personalized configuration of target resources in different marketing scenarios and effectively meet the diverse needs of different scenarios, the resource configuration device 300 may further include:

[0149] The second parsing module can be used to parse the question-and-answer text based on natural language processing technology to obtain N target approximate keywords corresponding to the first semantic identifier, where N is a positive integer;

[0150] The first addition module can be used to add M target similar keywords to the third core field when the second semantic identifier that can be included in the third core field matches the first semantic identifier;

[0151] Among them, the M target approximation keywords can be approximation keywords that do not overlap with at least one approximation keyword in the third core field from among the N target approximation keywords, and M is a positive integer.

[0152] In some possible implementations, considering that there may still be cases where the keyword index library does not contain the aforementioned second semantic identifier, further, in order to expand and optimize the keyword index library so as to fully ensure the personalized configuration of target object resources in different marketing scenarios, thereby effectively meeting the diverse needs of different scenarios, the resource configuration device 300 may further include:

[0153] The first storage module can be used to store the question-and-answer text and / or the first word meaning identifier when a word meaning identifier that matches the first word meaning identifier cannot be found in the keyword index library;

[0154] The first generation module can be used to generate a fourth core field if the number of times the question-and-answer text and / or the first word meaning identifier is stored within a preset time period is greater than a preset threshold. The fourth core field may include core keywords determined based on the question-and-answer text and / or the first word meaning identifier, at least one approximate keyword, and resource content.

[0155] The second storage module can be used to store the fourth core field in the keyword index library.

[0156] In some possible implementations, to more rationally construct the aforementioned keyword index library and facilitate keyword matching queries in subsequent steps, the resource configuration device 300 may further include:

[0157] The first determining module can be used to determine at least one approximate keyword and resource content corresponding to each of the multiple core keywords based on multiple core keywords.

[0158] The second generation module can be used to generate multiple core fields based on multiple core keywords, at least one approximate keyword corresponding to each of the multiple core keywords, and resource content.

[0159] The third storage module can be used to store multiple core fields in the keyword index library.

[0160] In some possible implementations, considering more specific banking business scenarios and actual marketing needs, the target object can be a digital human; the resource content can be divided into background type resources and clothing type resources; the resource configuration device 300 may further include:

[0161] The fourth configuration module can be used to configure the background content of the digital human based on the resource content when the resource content is a background type resource, and then re-render the image; or,

[0162] The fifth configuration module can be used to configure the clothing content of the digital human based on the resource content when the resource content is clothing type, and then re-render the screen.

[0163] Based on the resource configuration method provided in the above embodiments of this application, a resource configuration method provided in this application is described below. Please refer to... Figure 4 , Figure 4 This is a schematic diagram of the structure of a resource configuration device provided in an embodiment of this application.

[0164] The resource allocation device may include a processor 401 and a memory 402 storing computer program instructions.

[0165] Specifically, the processor 401 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0166] Memory 402 may include mass storage for data or instructions. For example, and not limitingly, memory 402 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 402 may include removable or non-removable (or fixed) media. Where appropriate, memory 402 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 402 is non-volatile solid-state memory.

[0167] Memory may include read-only memory (ROM), random access memory (RAM), disk storage media devices, optical storage media devices, flash memory devices, and electrical, optical, or other physical / tangible memory storage devices. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the methods according to one aspect of this disclosure.

[0168] The processor 401 implements any of the resource allocation methods described in the above embodiments by reading and executing computer program instructions stored in the memory 402.

[0169] In one example, the data resource configuration device may further include a communication interface 403 and a bus 410. Wherein, as... Figure 4 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0170] The communication interface 403 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0171] Bus 410 includes hardware, software, or both, that couples components of a resource configuration device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 410 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, any suitable bus or interconnect is contemplated herein.

[0172] The resource configuration device executes the resource configuration method in the embodiments of this application, thereby achieving... Figure 1 The resource allocation method described.

[0173] Furthermore, in conjunction with the resource allocation methods in the above embodiments, this application embodiment can provide a computer storage medium for implementation. This computer storage medium stores computer program instructions; when these computer program instructions are executed by a processor, they implement any of the resource allocation methods in the above embodiments.

[0174] Based on the resource configuration method in the above embodiments, this application provides a computer program product. When the instructions in the computer program product are executed by the processor of an electronic device, the electronic device performs the resource configuration method provided in any of the above embodiments of this application.

[0175] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0176] The functional blocks shown in the above-described structural diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0177] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0178] The aspects of this disclosure have been described above with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It should be understood that each block in 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, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that these instructions, executable via the processor of the computer or other programmable data processing apparatus, enable the implementation of the functions / actions specified in one or more blocks of the flowchart illustrations and / or block diagrams. Such a processor can be, but is not limited to, a general-purpose processor, a special-purpose processor, a special application processor, or a field-programmable logic circuit. It is also understood that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can also be implemented by special-purpose hardware performing the specified functions or actions, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0179] The above description is merely a specific implementation of this application. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, modules, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. It should be understood that the protection scope of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the protection scope of this application.

Claims

1. A resource configuration method, characterized by, The method comprises: acquiring a target keyword in a question and answer text; querying a core keyword matching the target keyword in a keyword index library; wherein the keyword index library stores a plurality of core fields; each of the plurality of core fields comprises a core keyword, at least one approximate keyword, and resource content; in a case where a first core keyword in a first core field matches the target keyword, configuring resources for a target object according to resource content in the first core field; the plurality of core fields comprise the first core field; in a case where the core keyword matching the target keyword cannot be queried, querying an approximate keyword matching the target keyword in the keyword index library; in a case where any approximate keyword in a second core field matches the target keyword, configuring resources for the target object according to resource content in the second core field; each of the plurality of core fields further comprises a word sense identifier; the method further comprises: in a case where the approximate keyword matching the target keyword cannot be queried, parsing the question and answer text into a first word sense identifier based on a natural language processing technology; querying a word sense identifier matching the first word sense identifier in the keyword index library; in a case where a second word sense identifier included in a third core field matches the first word sense identifier, configuring resources for the target object according to resource content in the third core field; in a case where the approximate keyword matching the target keyword cannot be queried, the method further comprises: parsing the question and answer text based on a natural language processing technology to obtain N target approximate keywords corresponding to the first word sense identifier, N being a positive integer; in a case where a second word sense identifier included in a third core field matches the first word sense identifier, adding M target approximate keywords to the third core field; wherein the M target approximate keywords are approximate keywords in the N target approximate keywords that do not coincide with at least one approximate keyword in the third core field, M being a positive integer.

2. The method of claim 1, wherein, The method further comprises: in a case where the word sense identifier matching the first word sense identifier in the keyword index library cannot be queried, storing the question and answer text or / and the first word sense identifier; if the storage times of the question and answer text or / and the first word sense identifier within a preset time period are greater than a preset threshold, generating a fourth core field; the fourth core field comprises a core keyword, at least one approximate keyword, and resource content determined based on the question and answer text or / and the first word sense identifier; storing the fourth core field to the keyword index library.

3. The method according to claim 1 or 2, characterized in that, Before the querying of the core keyword matching the target keyword in the keyword index library, the method further comprises: based on a plurality of core keywords, respectively determining at least one approximate keyword and resource content corresponding to each core keyword in the plurality of core keywords; generating the plurality of core fields based on the plurality of core keywords, at least one approximate keyword corresponding to each core keyword in the plurality of core keywords, and resource content; storing the plurality of core fields to the keyword index library.

4. The method according to claim 1 or 2, characterized in that, The target object is a digital human; the resource content is divided into background type resource and clothing type resource; the method further comprises: In the case that the resource content is a background type resource, configuring background content of the digital human based on the resource content, and re-rendering the picture; or, In the case that the resource content is a clothing type resource, configuring clothing content of the digital human based on the resource content, and re-rendering the picture.

5. A resource configuration apparatus, characterized by comprising: The device comprises: A first acquisition module for acquiring a target keyword in a question and answer text; A first query module for querying a core keyword matching the target keyword in a keyword index library; wherein the keyword index library stores a plurality of core fields; any core field in the plurality of core fields includes a core keyword, at least one approximate keyword, and resource content; A first configuration module for, in the case that a first core keyword in a first core field matches the target keyword, configuring a target object with resource content in the first core field; the plurality of core fields include the first core field; A second query module for, in the case that the target keyword matching core keyword cannot be queried, querying an approximate keyword matching the target keyword in the keyword index library; A second configuration module for, in the case that any approximate keyword in a second core field matches the target keyword, configuring the target object with resource content in the second core field; Any core field in the plurality of core fields further includes a word sense identifier; the device further comprises: A first analysis module for, in the case that the target keyword matching approximate keyword cannot be queried, analyzing the question and answer text into a first word sense identifier based on natural language processing technology; A third query module for querying a word sense identifier matching the first word sense identifier in the keyword index library; A third configuration module for, in the case that a second word sense identifier included in a third core field matches the first word sense identifier, configuring the target object with resource content in the third core field; In the case that the target keyword matching approximate keyword cannot be queried, the device further comprises: A second analysis module for analyzing the question and answer text based on natural language processing technology to obtain N target approximate keywords corresponding to the first word sense identifier, N being a positive integer; A first adding module for, in the case that a second word sense identifier included in a third core field matches the first word sense identifier, adding M target approximate keywords to the third core field; Wherein the M target approximate keywords are approximate keywords in the N target approximate keywords that do not coincide with at least one approximate keyword in the third core field, M being a positive integer.

6. A resource configuration device, comprising: The device comprises a processor and a memory storing computer program instructions; The processor executes the computer program instructions to implement the resource configuration method in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, The computer program instructions are stored on the computer readable storage medium and are executed by the processor to implement the resource configuration method in any one of claims 1-4.

8. A computer program product, characterised in that, The instructions in the computer program product are executed by the processor of the electronic device, and the electronic device executes the resource configuration method in any one of claims 1-4.

Images