A virtual character perfecting method, device and medium based on user preferences

Abstract

The application discloses a virtual character perfecting method and device based on user preferences and a medium, and relates to the field of artificial intelligence.The method comprises the following steps: obtaining user search records;obtaining user preferences according to the user search records; and determining the data of a virtual character in a virtual character database according to the user preferences to perfect the virtual character.The application can automatically modify the virtual character according to the user preferences, and the user does not need to manually change the virtual character, thereby improving the user's experience.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence, and in particular to a virtual character perfecting method based on user preferences, a device and a medium. BACKGROUND

[0002] In recent years, with the continuous development of mobile terminal technologies such as mobile phones, mobile terminals have become more and more intelligent, and the functions provided by mobile terminals have become more and more numerous. Virtual characters are one of the important functions provided by mobile terminals, can accompany people, make up for people's social needs, and are loved by more and more people. Existing virtual characters such as mobile phone pets are too abstract, and the user-specificity and pertinence are not strong enough, and the user experience is not enough. SUMMARY

[0003] The present application aims to at least solve one of the technical problems in the related art. To this end, the present application provides a virtual character perfecting method based on user preferences, a device and a medium, which improves user-specificity and pertinence to improve user experience.

[0004] In a first aspect, the embodiments of the present application provide a virtual character perfecting method based on user preferences, comprising:

[0005] obtaining user search records;

[0006] obtaining user preferences from the user search records;

[0007] determining virtual character data in a virtual character database according to the user preferences to perfect virtual characters.

[0008] Optionally, in an embodiment of the present application, the obtaining user preferences from the user search records comprises:

[0009] extracting user keywords from the user search records; the user keywords include category keywords and other keywords;

[0010] classifying the user search records according to the category keywords to obtain classification categories to which the user search records belong; machine learning;

[0011] determining effective keywords in the other keywords according to the classification categories to which the user search records belong;

[0012] the user preferences include the classification categories to which the user search records belong, the category keywords and the effective keywords.

[0013] Optionally, in an embodiment of the present application, the determining virtual character data in a virtual character database according to the user preferences comprises:

[0014] determining whether there is a keyword similar to the user preference in the database of the virtual role according to the user preference;

[0015] If yes, the keyword in the table is determined and the keyword is input into the virtual role.

[0016] Optionally, in an embodiment of the present application, the classification category to which the user search record belongs includes a user preference dress and a user life content.

[0017] Optionally, in an embodiment of the present application, the determining whether there is a keyword similar to the user preference in the database of the virtual role according to the user preference includes: determining a table of the user preference in the database of the virtual role according to the classification category to which the user search record in the user preference belongs.

[0018] If the valid keyword exists, it is determined whether there is a keyword in the table with a similarity greater than a similarity threshold to the valid keyword; if yes, the keyword with the highest similarity is determined; otherwise, the process is ended.

[0019] If the valid keyword does not exist, it is determined whether there is a keyword in the table with a similarity greater than a similarity threshold to the category keyword; if yes, the keyword with the highest similarity is determined; otherwise, the process is ended.

[0020] Optionally, in an embodiment of the present application, if it is determined that there is no keyword similar to the user preference in the database of the virtual role according to the user preference, it is determined whether the classification category to which the user search record belongs is the user life content.

[0021] If yes, the user search record is sent to a web server with which a connection is established.

[0022] According to the user search record, the web server performs a search and sends a search result after the search is performed.

[0023] The search result sent by the web server is received.

[0024] The search result is input into the virtual role.

[0025] Optionally, in an embodiment of the present application, if it is determined that there is no keyword similar to the user preference in the database of the virtual role according to the user preference, it is determined that the classification category to which the user search record belongs is the user preference dress, and the virtual role remains unchanged.

[0026] The embodiments of the present application have at least the following beneficial effects: the user preference is obtained according to the user search record, and whether to modify and perfect the virtual role is determined according to the user preference, the virtual role can be automatically modified according to the user preference, the user does not need to manually change, the user's specificity and pertinence are improved, and the user's experience is improved.

[0027] In a second aspect, the embodiments of the present application also provide an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the virtual role perfecting method as described above when executing the computer program.

[0028] In a third aspect, the embodiments of the present application also provide a computer readable storage medium, which stores computer executable instructions, and the computer executable instructions are used to make a computer execute the virtual role perfecting method as described above. BRIEF DESCRIPTION OF DRAWINGS

[0029] The accompanying drawings are used to provide a further understanding of the technical scheme of the present application, and constitute a part of the specification, and are used to explain the technical scheme of the present application together with the embodiments of the present application, and do not constitute a limitation on the technical scheme of the present application.

[0030] Figure 1 is a flowchart of a virtual role perfecting method based on user preference provided by one embodiment of the present application;

[0031] Figure 2 is a schematic diagram of an electronic device provided by another embodiment of the present application. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below in combination with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, and do not limit the present application.

[0033] It should be noted that although the functional modules are divided in the system schematic diagram, and the logical order is shown in the flowchart, in some cases, the steps shown or described in the modules in the system or the order in the flowchart can be executed. The terms "first", "second", etc. in the specification, claims and above-described drawings are used to distinguish similar objects, and do not necessarily describe a specific order or sequence.

[0034] Reference Figure 1 In an embodiment of the present application, a virtual role perfecting method based on user preference comprises:

[0035] Step S100, obtaining a user search record;

[0036] Step S200, obtaining user preference according to user search record;

[0037] Step S300, determining data of virtual role in database of virtual role according to user preference, to perfect virtual role.

[0038] It should be noted that the data of the virtual role includes the appearance of the virtual role and the content of the virtual role reminding the user, and the database of the virtual role is a database storing all data of the virtual role; the attributes of the virtual role are determined by the data of the virtual role, and the attributes of the virtual role are determined by calling some data in the database of the virtual role.

[0039] It should be further noted that when the virtual role is established, the data has been given an initial value, and in the process of user use, the life condition of the user will change, therefore, the embodiment obtains user preference according to user search record, and determines in real time whether the virtual role needs to be modified and perfected; for example, the search record of the user today is about red, and the relevant data of the virtual role is determined to be red by the method of the embodiment; the search record of the user tomorrow is about green, and the relevant data of the virtual role is modified to be green by the method of the embodiment.

[0040] In the embodiment, the user preference is obtained according to the user search record, and whether the virtual role needs to be modified and perfected is determined through the user preference, so that the virtual role can be automatically modified according to the user preference, the user does not need to manually change, the specificity and pertinence of the user are improved, and the experience of the user is improved.

[0041] In an embodiment of the application, in step S200, obtaining user preference according to user search record includes:

[0042] extracting user key words from the user search record; the user key words include category key words and other key words;

[0043] classifying the user search record according to the category key words, to obtain a classification category to which the user search record belongs;

[0044] determining effective key words from the other key words according to the classification category to which the user search record belongs;

[0045] The user preference includes the classification category to which the user search record belongs, the category key words and the effective key words.

[0046] It should be noted that the number of category keywords is less than or equal to 2, and other keywords are keywords extracted in addition to category keywords, such as: a user searches for "where is the cheapest apple in June", the extracted keywords include "June", "apple", "where" and "cheap", the category keyword is "where", and the other keywords include "June", "apple" and "cheap", and the effective keywords include "apple" and "cheap", so the invalid keyword "June" is excluded, and the word irrelevant to the meaning expressed by the user's actual search is excluded.

[0047] In an embodiment of the present application, extracting user keywords from the user search record specifically includes: inputting the user search record into the trained keyword extraction model, and the trained keyword extraction model outputs category keywords and other keywords respectively. In an embodiment, the keyword extraction model is trained using the SVM method.

[0048] In an embodiment of the present application, the user search record is classified according to the category keyword to obtain the classification category to which the user search record belongs, including:

[0049] The category keyword is searched in the set category keyword and classification category association table to determine the classification category corresponding to the category keyword.

[0050] It should be noted that the category keyword and classification category association table has been established when the virtual role is established; if the category keyword is not found in the association table, the process ends, and the category keyword is considered invalid.

[0051] In an embodiment of the present application, the classification category to which the user search record belongs includes user preferred dressing and user life content.

[0052] For the category keyword identified by the user search record, the category keyword is divided into two categories, i.e. the classification category to which the user search record belongs; for example: in a category keyword and classification category association table, "cold", "apple" and "wash hands" correspond to the category of user life content, and "red clothes" and "yellow hat" correspond to the user preferred dressing; and when the user searches for "what should I pay attention to when I have a cold", the category keyword "cold" is obtained through the trained keyword extraction model, and the user search record is obtained as the category of user life content through the category keyword and classification category association table; when the user searches for "shopping software", there is no category keyword about "shopping software" in the category keyword and classification category association table, so the search record is considered invalid and is not classified.

[0053] In an embodiment of the present application, determining the effective keyword from the other keywords according to the classification category to which the user search record belongs comprises: performing text correlation analysis on the classification category to which the user search record belongs and the other keywords, and determining the other keyword with a confidence degree greater than a confidence degree threshold as the effective keyword.

[0054] The keywords related to the classification category are screened out, for example, the user searches "apple and pear I", the user keywords "I", "apple" and "pear" are identified, the classification keyword is "apple", and the effective keyword "pear" is obtained through text correlation analysis, and "I" is not related to "apple" and "pear".

[0055] In an embodiment of the present application, determining the data of the virtual role in the database of the virtual role comprises:

[0056] According to the user preference, it is determined whether there is a keyword similar to the user preference in the database of the virtual role.

[0057] If yes, the keyword in the table is determined and the keyword is input into the virtual role.

[0058] In an embodiment of the present application, according to the user preference, it is determined whether there is a keyword similar to the user preference in the database of the virtual role.

[0059] According to the classification category to which the user search record belongs in the user preference, the table of the user preference in the database of the virtual role is determined, and the classification of the table of the user preference in the database of the virtual role has been set.

[0060] If there is an effective keyword, it is determined whether there is a keyword with a similarity greater than a similarity threshold to the effective keyword in the table, if yes, the keyword with the highest similarity is determined, otherwise, the process is ended.

[0061] If there is no effective keyword, it is determined whether there is a keyword with a similarity greater than a similarity threshold to the classification keyword in the table, if yes, the keyword with the highest similarity is determined, otherwise, the process is ended.

[0062] It should be noted that the search record is short, the user keyword extracted is only one and is determined as the classification keyword, and the similarity comparison is performed on the classification keyword; the similarity threshold is a value set by a person skilled in the art according to the actual situation.

[0063] In an embodiment of the present application, the TF-IDF algorithm is used for similarity comparison, and the similarity threshold is 0.6.

[0064] For example, the user search record is "red shirt", the classification category to which the user search record belongs is user preferred dressing, and in the table of the user search record in the database of the virtual role, the table determined as the user preferred dressing includes a "clothes" table and a "trousers" table, the "clothes" table includes a "red dotted shirt" and a "yellow striped shirt", and the "trousers" table includes a "red dotted trousers" and a "yellow striped trousers", wherein the similarity of the "red dotted shirt", the "yellow striped shirt" and the "red dotted trousers" to the "red shirt" is greater than the similarity threshold, and the similarity of the "red dotted shirt" to the "red shirt" is the highest, and thus the "red dotted shirt" is determined as the keyword.

[0065] In an embodiment of the present application, according to the user preference, if it is determined in the database of the virtual role that there is no keyword similar to the user preference, it is determined whether the classification category to which the user search record belongs is user life content;

[0066] If yes, the user search record is sent to the connected Web server;

[0067] According to the user search record, the Web server performs a search and sends the search result after obtaining the search result;

[0068] The search result sent by the Web server is received;

[0069] The search result is input into the virtual role.

[0070] For the user life content, if there is no keyword related to the user life content in the database, a search is performed on the Web server, and the related search result is sent after obtaining the search result; for example, the user searches "what should be paid attention to for a cold", the keywords of the user are "cold" and "attention", there is no keyword related to the keywords of the user in the database, "what should be paid attention to for a cold" is sent to the connected Web server, the Web server performs a search, the first search result "drink more hot water and pay attention to rest" is taken as the search result, and the search result is sent; the search result is input into the virtual role, and the virtual role reminds the user of "drink more hot water and pay attention to rest".

[0071] In an embodiment of the present application, if it is determined in the database of the virtual role that there is no keyword similar to the user preference, it is determined that the classification category to which the user search record belongs is user preferred dressing, and the virtual role remains unchanged.

[0072] It should be noted that, since the dressing of the virtual role needs to be modeled and constructed, the user demand cannot be met by simple network search, and the dressing in the database established when the virtual role is established can only be called to modify.

[0073] In addition, with reference toFigure 2 One embodiment of the present application also provides an electronic device, comprising a memory 11, a processor 12, and a computer program stored in the memory 11 and executable on the processor 12.

[0074] The processor 12 and the memory 11 can be connected by a bus or other means.

[0075] A non-transitory software program and instructions required for implementing the above-mentioned embodiment of the virtual character perfecting method based on user preferences are stored in the memory 11, and when executed by the processor 12, the virtual character perfecting method based on user preferences in the above-mentioned embodiment is executed.

[0076] The embodiment of the present application also provides a storage medium storing computer executable instructions for executing the above-mentioned virtual character perfecting method based on user preferences.

[0077] In one embodiment, the storage medium stores computer executable instructions, which are executed by one or more control processors.

[0078] The above-described embodiments are merely illustrative, and units described as separate components can or can not be physically separate, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected according to actual needs to achieve the purpose of the embodiment.

[0079] As will be appreciated by one of ordinary skill in the art, all or some steps, systems of the above-disclosed methods can be implemented as software, firmware, hardware, or any suitable combination thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application- specific integrated circuit. Such software can be distributed on computer readable media, which can comprise computer storage media (or non-transitory media), and communication media (or transitory media). As is well known to those of ordinary skill in the art, the term computer storage media includes both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data. Computer storage media include, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Further, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules, or other data in a modulated data signal, such as carrier waves or other transport mechanisms, and includes any information delivery media.

[0080] Embodiments of the present application are described herein with reference to the drawings, including embodiments presently preferred by the inventors for the practice of the present application. Variations on described embodiments can become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the scope of the application to include all such modifications and equivalents resorting to the claims appended hereto as presently stated and as subsequently amended. The scope of the application covers any combination of the above-described elements in all possible variations thereof unless otherwise explicitly indicated herein or otherwise clearly contradicted by context.

Claims

1. A method for perfecting a virtual character based on user preferences, the method comprising: receiving a user profile; receiving a user preference; and generating a virtual character based on the user profile and the user preference. The method comprises: obtaining a user search record; obtaining a user preference according to the user search record; determining a virtual role data in a virtual role database according to the user preference to perfect the virtual role; wherein the obtaining the user preference according to the user search record comprises: inputting the user search record into a trained keyword extraction model, the trained keyword extraction model outputs category keywords and other keywords respectively; finding the category keywords in a category keyword and category association table, determining a category corresponding to the category keywords as a category to which the user search record belongs; the category to which the user search record belongs comprises a user preference dressing and a user life content; performing a text association analysis on the category to which the user search record belongs and the other keywords, and determining an effective keyword with a confidence greater than a confidence threshold; the user preference comprises the category to which the user search record belongs, the category keywords and the effective keyword; wherein the determining the virtual role data in the virtual role database according to the user preference comprises: determining whether there is a keyword similar to the user preference in the virtual role database according to the user preference; if yes, determining the keyword in the table and inputting the keyword into the virtual role; wherein the determining whether there is a keyword similar to the user preference in the virtual role database according to the user preference comprises: determining a table of the user preference in the virtual role database according to the category to which the user search record belongs in the user preference; if the effective keyword exists, determining whether there is a keyword with a similarity greater than a similarity threshold to the effective keyword in the table; if yes, determining the keyword with the highest similarity; otherwise, ending the process; if the effective keyword does not exist, determining whether there is a keyword with a similarity greater than a similarity threshold to the category keyword in the table; if yes, determining the keyword with the highest similarity; otherwise, ending the process; wherein, if it is determined in the virtual role database that there is no keyword similar to the user preference, it is determined whether the category to which the user search record belongs is the user life content; if yes, sending the user search record to a web server connected; the web server performs a search according to the user search record, and sends a search result after obtaining the search result; receiving the search result sent by the web server; inputting the search result into the virtual role; if it is determined in the virtual role database that there is no keyword similar to the user preference, it is determined that the category to which the user search record belongs is the user preference dressing, and the virtual role remains unchanged.

2. An electronic device comprising: A memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the method of claim 1.

3. 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 of claim 1.

Images