Level generation method and device of virtual scene, electronic equipment, computer readable storage medium and computer program product

Abstract

The application provides a level generation method and device of a virtual scene, electronic equipment, a computer readable storage medium and a computer program product; the method comprises the following steps: screening at least one level word meeting a first generation requirement from a plurality of candidate words; performing splitting processing on each level word to obtain a plurality of split words of each level word; performing path exploration processing based on the plurality of split words of each level word to obtain a level of the virtual scene; and performing difficulty evaluation processing based on the level to obtain a difficulty evaluation result of the level. Through the application, the level with controllable difficulty can be automatically generated, and the generation efficiency and quality of the level of the virtual scene can be effectively improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computer, and particularly relates to a virtual scene level generation method and device, electronic equipment, computer readable storage medium and computer program product. BACKGROUND

[0002] In the related art, virtual scene levels on the market are mainly generated in an artificial editing manner. This manner is inefficient and the quality of the generated levels is uneven, which is difficult to provide players with continuous freshness and challenge. With the increasing demand of players and the complication of game content, the virtual scene levels generated by artificial editing are often insufficient, and it is difficult to generate levels on a large scale while ensuring their playability and challenge. SUMMARY

[0003] The embodiments of the present application provide a virtual scene level generation method and device, electronic equipment, computer readable storage medium and computer program product, which can automatically generate levels with controllable difficulty, and can effectively improve the generation efficiency and quality of virtual scene levels.

[0004] The technical solutions of the embodiments of the present application are as follows:

[0005] The embodiments of the present application provide a virtual scene level generation method, which comprises the following steps:

[0006] screening at least one level word meeting a first generation requirement from a plurality of candidate words;

[0007] splitting each level word to obtain a plurality of split words of each level word;

[0008] performing path exploration processing based on the plurality of split words of each level word to obtain a level of the virtual scene;

[0009] performing difficulty evaluation processing based on the level to obtain a difficulty evaluation result of the level.

[0010] The embodiments of the present application provide a virtual scene level generation device, which comprises the following steps:

[0011] a screening module configured to screen at least one level word meeting a first generation requirement from a plurality of candidate words;

[0012] a splitting module configured to split each level word to obtain a plurality of split words of each level word;

[0013] an exploration module configured to perform path exploration processing based on the plurality of split words of each level word to obtain a level of the virtual scene;

[0014] The evaluation module is configured to perform a difficulty evaluation process based on the level to obtain a difficulty evaluation result of the level.

[0015] In the scheme, the screening module is further configured to extract a difficulty value and a theme from the first generated demand, and screen at least one level word from a plurality of candidate words according to the difficulty value and the theme.

[0016] In the scheme, the screening module is further configured to determine a word length positively correlated with the difficulty value, a word frequency negatively correlated with the difficulty value, and a word type corresponding to the difficulty value, wherein different word types correspond to different splitting numbers; screen a plurality of first words from the plurality of candidate words according to the word length, the word frequency, and the word type; and screen a level word according to the theme from the plurality of first words.

[0017] In the scheme, the screening module is further configured to screen a second word according to the theme from the plurality of first words, and delete a word of a risk type from the plurality of second words to obtain the level word.

[0018] In the scheme, the splitting module is further configured to identify a word type to which each level word belongs, and perform the following processing for each word type: based on a splitting rule corresponding to the word type, split a level word corresponding to the word type to obtain a plurality of split words of the level word.

[0019] In the scheme, the splitting module is further configured to perform the following processing for each level word: perform classification processing on characters in the level word to obtain a predicted position type of each character; and when the predicted position type represents that the character is a boundary character, split the level word by taking the character as a boundary to obtain a plurality of split words of the level word.

[0020] In the scheme, the classification processing is achieved by calling a word classification model, and the splitting module is further configured to perform classification processing on each character sample in a word sample by using the word classification model to obtain a predicted position type of each character sample; determine a first loss based on an actual position type of each character sample in each word sample and the predicted position type of each character sample in each word sample; and train the word classification model based on the first loss to obtain a trained word classification model.

[0021] In the solution, the exploration module is further configured to perform path exploration on the initial arrangement of the split words in the level to determine a word elimination path of the level; when the word elimination path of the level is unique, take the initial arrangement of the split words in the level as the level of the virtual scene; when the word elimination path of the level is not unique, adjust the initial arrangement of the split words in the level, and repeatedly perform the path exploration based on the adjusted arrangement until the word elimination path of the level is unique, and take the adjusted arrangement as the level of the virtual scene.

[0022] In the solution, the evaluation module is further configured to determine a first difficulty value of the level based on the word frequency of the level words; determine a second difficulty value of the level based on the number of the split words in the level; determine a third difficulty value of the level based on the number of the level words in the level; determine a fourth difficulty value of the level based on the position placement relationship of the split words of each level word; and fuse the first difficulty value, the second difficulty value, the third difficulty value and the fourth difficulty value, and take the fusion result as the difficulty evaluation result of the level.

[0023] An electronic device is provided in an embodiment of the present application, and the electronic device comprises:

[0024] A memory is configured to store computer executable instructions or computer programs.

[0025] A processor is configured to execute the computer executable instructions or computer programs stored in the memory to implement the level generation method of the virtual scene provided in the embodiments of the present application.

[0026] A computer readable storage medium is provided in an embodiment of the present application, and the computer readable storage medium stores computer programs or computer executable instructions, and is configured to be executed by a processor to implement the level generation method of the virtual scene provided in the embodiments of the present application.

[0027] A computer program product is provided in an embodiment of the present application, and the computer program product comprises computer programs or computer executable instructions, and the computer programs or computer executable instructions are executed by a processor to implement the level generation method of the virtual scene provided in the embodiments of the present application.

[0028] The embodiments of the present application have the following beneficial effects:

[0029] First, the checkpoint words meeting the first generation requirement are screened from a plurality of candidate words, so that the screened checkpoint words can meet the first generation requirement; then, each checkpoint word is split to obtain a plurality of split words of the checkpoint word, thereby providing a minimum operable unit for subsequent path exploration; subsequently, path exploration processing is performed according to the plurality of split words of each checkpoint word, to obtain a checkpoint of the virtual scene, thereby enhancing the logical depth and strategy of the game; finally, difficulty evaluation is performed on the checkpoint to obtain a difficulty evaluation result of the checkpoint, thereby ensuring the balance and challenge of the game experience. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 is an architecture schematic diagram of a checkpoint generation system 100 of a virtual scene provided by an embodiment of the present application;

[0031] Figure 2 is a structure schematic diagram of an electronic device 500 provided by an embodiment of the present application;

[0032] Figure 3 is a flow schematic diagram of a checkpoint generation method of a virtual scene provided by an embodiment of the present application;

[0033] Figure 4 is a flow schematic diagram of a checkpoint generation method of a virtual scene provided by an embodiment of the present application;

[0034] Figure 5 is a flow schematic diagram of a checkpoint generation method of a virtual scene provided by an embodiment of the present application;

[0035] Figure 6 is a word elimination process schematic diagram of a month theme checkpoint provided by an embodiment of the present application;

[0036] Figure 7 is an interface schematic diagram of a shining theme checkpoint provided by an embodiment of the present application. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings, and the described embodiments should not be regarded as limiting the present application. All other embodiments obtained by a person of ordinary skill in the art without making any creative effort fall within the scope of protection of the present application.

[0038] In the following description, “some embodiments” are described, which describe a subset of all possible embodiments, but it can be understood that “some embodiments” can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict.

[0039] It can be understood that in the embodiments of the present application, related data such as user information is involved, and when the embodiments of the present application are applied to specific products or technologies, user permission or consent needs to be obtained, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards.

[0040] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined target, and can be implemented entirely or partially by using software, hardware (such as a processing circuit or a memory) or a combination thereof. Similarly, one processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be a part of an overall module or unit that includes the functions of the module or unit.

[0041] In the following description, the term "first\second\..." is only to distinguish similar objects, and does not represent a specific order of the objects. It can be understood that "first\second\..." can be interchanged in a specific order or sequence as allowed, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0042] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application and are not intended to limit the present application.

[0043] Before the embodiments of the present application are further described in detail, the terms and phrases involved in the embodiments of the present application are explained, and the terms and phrases involved in the embodiments of the present application are applicable to the following explanations.

[0044] 1) Word splitting: word splitting refers to using morphological analysis techniques to decompose complex words into multiple roots, affixes or common letter combinations, thereby forming split word combinations, and each split word combination is embedded in a Tile block, and the standard size of the block is 2x2.

[0045] 2) Stacking: stacking refers to forming a Tile stacking area based on split word combinations for player interaction, which can optimize space utilization and improve the interactive experience of players. The blocked blocks display the back, hide all letters, and increase the challenge of the game.

[0046] 3) Level theme: the level theme is a preset theme in the level design, which can be used to guide word selection and ensure that all words are related to the theme, which can increase the interest and educational significance of the game.

[0047] 4) Two-segment or three-segment words: In a level, a word split into two segments is called a two-segment word, and a word split into three segments is called a three-segment word.

[0048] 5) Word elimination: When the player clicks to split the word combination to build a word that meets the theme of the level, the corresponding Tile block is removed, and the hidden block is revealed.

[0049] 6) Monte Carlo simulation: Monte Carlo simulation is a randomized statistical method applied to optimization algorithms. Through a large number of random nodes and path explorations, combined with dynamic weight adjustment, the optimal solution is quickly approached, and a highly reliable difficulty assessment is provided.

[0050] In related technologies, virtual scene levels on the market are mainly generated by manual editing. This method is inefficient and the quality of the generated levels is uneven, making it difficult to provide players with continuous freshness and challenge. With the increasing demand of players and the complexity of game content, virtual scene levels generated by manual editing often seem inadequate, making it difficult to generate levels on a large scale while ensuring their playability and challenge.

[0051] Based on this, the embodiments of the present application provide a virtual scene level generation method, device, electronic device, computer readable storage medium and computer program product, which can automatically generate levels with controllable difficulty, effectively improving the generation efficiency and quality of virtual scene levels. The following describes an exemplary application of the electronic device provided by the embodiments of the present application. The electronic device provided by the embodiments of the present application can be implemented as a notebook computer, a tablet computer, a desktop computer, a set-top box, a smart phone, a smart speaker, a smart watch, a smart television, a vehicle-mounted terminal, and various types of terminal devices. It can also be implemented as a server.

[0052] Referring to Figure 1 , Figure 1 is an architecture diagram of a virtual scene level generation system provided by the embodiments of the present application, to realize the application of supporting the generation of levels of a virtual scene, as shown in Figure 1 The virtual scene level generation system 100 includes a server 200, a network 300, a terminal 400, and a database 600. The database 600 is connected to the server 200, and the terminal 400 is connected to the server 200 through the network 300. The network 300 can be a local area network or a wide area network, or a combination of the two.

[0053] In some embodiments, a first generation requirement is input through a terminal 400 and transmitted to a server 200 via a network 300. The server 200 filters multiple candidate words in a database 600 based on the first generation requirement to obtain at least one level word. Subsequently, the server 200 generates a level of a virtual scene based on the at least one level word. After that, the server 200 sends the level of the virtual scene to the terminal 400 via the network 300 and displays it on the client of the terminal 400.

[0054] The virtual scene level generation method provided in this application can be applied to various scenarios that require virtual scene level generation, such as the following scenarios:

[0055] (1) In a children’s literacy app, assuming that the first generation requirement is limited to nouns learned before the second grade of primary school, the server selects multiple words learned before the second grade from multiple candidate words stored in the database as level words, such as “apple”, “schoolbag”, “sun”, etc., and generates corresponding levels based on the selected level words, thereby enhancing the fun and educational significance of the game.

[0056] (2) In the English word elimination mobile game, assuming that the first generation requirement is limited to TOEFL core vocabulary with a length of 5-7 letters, the server selects multiple TOEFL core words with a length of 5-7 letters from the English words stored in the database as level words, such as "planet", "author", "vivid", etc., and generates corresponding elimination levels based on the selected level words, thereby enhancing the fun and educational significance of the game.

[0057] Example, Figure 1 The server 200 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms. The terminal 400 can be a smartphone, tablet, laptop, desktop computer, smart speaker, smartwatch, in-vehicle terminal, etc., but is not limited to these. The terminal 400 and server 200 can be directly or indirectly connected via wired or wireless communication, which is not limited in this embodiment.

[0058] The structure of the electronic device provided in the embodiments of this application will be further described below. Taking the electronic device as a terminal as an example, see... Figure 2 , Figure 2FIG. 1 is a structural schematic diagram of an electronic device 500 according to an embodiment of the present application, Figure 2 The electronic device 500 shown includes at least one processor 510, a memory 550, at least one network interface 520, and a user interface 530. The various components in the electronic device 500 are coupled together by a bus system 540. It can be appreciated that the bus system 540 is used for communicating data between the components. The bus system 540 includes a data bus, a power bus, a control bus, and a state signal bus, but for clarity's sake, only the data bus is shown in Figure 2 FIG. 1.

[0059] The processor 510 can be an integrated circuit chip that has the processing capability of signals, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor.

[0060] The user interface 530 includes one or more output devices 531 that enable presentation of media content, including one or more speakers and / or one or more visual display screens. The user interface 530 also includes one or more input devices 532 that facilitate user input, such as a keyboard, a mouse, a microphone, a touch screen display, a camera, other input buttons and controls.

[0061] The memory 550 can be removable, non-removable, or a combination thereof. Exemplary hardware devices include solid-state memory, hard drives, optical drives, etc. The memory 550 optionally includes one or more storage devices physically located in proximity to the processor 510.

[0062] The memory 550 includes volatile memory or non-volatile memory, and can include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), and the volatile memory can be random access memory (RAM). The memory 550 described in the embodiments of the present application is intended to include any suitable type of memory.

[0063] In some embodiments, the memory 550 is capable of storing data to support various operations, examples of which include programs, modules, and data structures or a subset or superset thereof, which are described below.

[0064] The operating system 551 includes system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, and the like, for implementing various basic services and processing hardware-based tasks;

[0065] The network communication module 552 is configured to reach other computing devices via one or more (wired or wireless) network interfaces 520, exemplary network interfaces 520 including Bluetooth, Wireless Fidelity (WiFi), Universal Serial Bus (USB), and the like;

[0066] The presentation module 553 is configured to enable presentation of information via one or more output devices 531 (e.g., a display screen, a speaker, and the like) associated with the user interface 530 (e.g., a user interface for operating peripheral devices and displaying content and information);

[0067] The input processing module 554 is configured to detect and interpret one or more user inputs or interactions from one or more input devices 532.

[0068] In some embodiments, the apparatus provided by the embodiments of the present application can be implemented in software, Figure 2 A level generation apparatus 555 of a virtual scene stored in the memory 550 is shown, which can be in the form of software such as programs and plug-ins, and includes the following software modules: a screening module 5551, a splitting module 5552, an exploration module 5553, and an evaluation module 5554, which are logical and thus can be combined or further split according to the implemented functions.

[0069] The level generation method of a virtual scene provided by the embodiments of the present application will be described in detail below in conjunction with exemplary applications and implementations of a terminal provided by the embodiments of the present application.

[0070] Referring to Figure 3 , Figure 3 is a flowchart of the level generation method of a virtual scene provided by the embodiments of the present application, which will be described in conjunction with the steps shown. Figure 3

[0071] It should be noted that Figure 3 The method shown can be executed by various forms of computer programs run by a terminal, and is not limited to a client, for example, can also be an operating system, a software module, a script, and an applet, and the like as described above, thus the following examples of a client should not be considered as a limitation on the embodiments of the present application. In addition, for convenience of description, the terminal and the client run by the terminal are not specifically distinguished in the following.

[0072] ​In step 101, at least one level word that meets the first generation requirement is screened out from multiple candidate words.

[0073] As an example, assume that the multiple candidate words include bat, emerald, apple, sun, kitten, etc., and the first generation requirement is common nouns for those below grade two. After screening according to the first generation requirement, at least one level noun that meets the first generation requirement includes apple, sun, kitten, etc.; assume that the multiple candidate words include tiger, lion, panda, elephant, sun, kitten, puppy, apple, etc., and the first generation requirement is nouns related to animals. After screening according to the first generation requirement, at least one level noun that meets the first generation requirement includes tiger, lion, elephant, kitten, puppy, etc.

[0074] In some embodiments, refer to Figure 4 , Figure 4 which is a schematic flowchart of a method for generating levels of a virtual scene provided by an embodiment of the present application. As shown in Figure 4 shown, Figure 3 step 101 shown can be implemented through Figure 4 steps 1011 to 1012 shown, and will be described in conjunction with Figure 4 the steps shown.

[0075] In step 1011, a difficulty value and a theme are extracted from the first generation requirement. <>

[0076] It should be noted that the difficulty value is used to characterize the difficulty level of the levels of the generated virtual scene. It can be measured by the time required for a user to complete the levels of the virtual scene. The longer the time required to complete a level, the greater the difficulty value of the level, that is, the more difficult the level of the generated virtual scene; it can also be measured by the number of specified words. For example, if the specified words include words with greater difficulty such as taotie and maodie, then the more the number of specified words included in the levels of the generated virtual scene, the greater the difficulty value of the level, that is, the more difficult the level of the generated virtual scene. It can also be measured according to the usage frequency level of words. For example, the usage frequencies of daily nouns such as raining and sunny are relatively high, and the usage frequency of proper nouns such as mosquito larvae is relatively low. Then, the higher the average usage frequency of the nouns included in the levels of the generated virtual scene, the smaller the difficulty value of the level, that is, the simpler the level of the generated virtual scene. The theme is used to characterize the type of level words in the levels of the generated virtual scene. For example, if the theme is animals, then the level words in the levels of the virtual scene are words related to animals such as lion and tiger.

[0077] As an example, when the difficulty value and the theme are included in the content of the first generation requirement, they can be directly extracted from the content of the first generation requirement, for example, the first generation requirement is a generation difficulty value of 3, and the theme is a word related to fruits, and the extracted difficulty value is 3 and the theme is a word related to fruits; when the content of the first generation requirement does not include the difficulty value and the theme, the difficulty value and the theme corresponding to the first generation request can be obtained through mapping processing according to the content of the first generation request, for example, the first generation requirement is a common noun below the second grade, and through the preset mapping relationship, the difficulty value corresponding to the first generation requirement is obtained. 2, the theme is a word related to fruits or animals composed of 2-3 characters.

[0078] In step 1012, at least one level word meeting the difficulty value and the theme is selected from a plurality of candidate words.

[0079] As an example, taking the English word screening scenario as an example, assuming that the difficulty value is 3, that is, screening words composed of 5-7 letters, and the theme is TOEFL core words, the candidate words include "tiktok", "deepfake", "planet", "author", "vivid", etc. Through screening, the level words meeting the difficulty value and the theme include "planet", "author", "vivid", etc.

[0080] In some embodiments, referring to Figure 5 , Figure 5 is a flowchart of a level generation method of a virtual scene provided by the embodiments of the present application, as shown in Figure 5 , the step 1012 shown in the figure can be implemented by the steps 10121 to 10124 shown in Figure 4 , which will be described in combination with the steps shown in Figure 5 Figure 5

[0081] In step 10121, the word length positively correlated with the difficulty value and the word frequency negatively correlated with the difficulty value are determined.

[0082] It should be noted that the word frequency is used to represent the frequency of the word, that is, the higher the word frequency, the more common the word, and the lower the difficulty value corresponding to the word. The lower the word frequency, the more obscure the word, and the higher the difficulty value corresponding to the word, for example, the word frequency of "sun" is high, and the difficulty value corresponding to "sun" is low, and the word frequency of "photosynthesis" is low, and the difficulty value corresponding to "photosynthesis" is high.

[0083] As an example, assuming that the mapping relationship between the difficulty value x and the word length L is ​​When the extracted difficulty value is 3, the word length of 6, i.e., containing 6 English words, can be obtained according to the mapping relationship; assuming that the mapping relationship of the difficulty value x and the word frequency w is When the extracted difficulty value is 2, the word frequency of 0.05, i.e., the English word appearing 5 times in every 100 words, can be obtained according to the mapping relationship.

[0084] In step 10122, the word type corresponding to the difficulty value is determined.

[0085] Here, different word types correspond to different splitting numbers.

[0086] As an example, assuming that the word splitting number of the word type A corresponding to the difficulty value of 1 is 2, the word splitting number of the word type B corresponding to the difficulty value of 2 is 3, and the word splitting number of the word type C corresponding to the difficulty value of 3 is 4, when the extracted difficulty value is 2, the corresponding word type B can be obtained.

[0087] In step 10123, a plurality of first words meeting the word length, the word frequency, and the word type are screened from a plurality of candidate words.

[0088] As an example, assuming that the word length of the candidate word A is 6, the word frequency is 0.03, and the splitting number corresponding to the word type is 3, the word length of the candidate word B is 7, the word frequency is 0.05, and the splitting number corresponding to the word type is 2, the word length of the candidate word C is 5, the word frequency is 0.04, and the splitting number corresponding to the word type is 2, the word length of the candidate word D is 8, the word frequency is 0.1, and the splitting number corresponding to the word type is 2; the screening range of the word length is 5-7, the screening range of the word frequency is 0.02-0.06, and the splitting number corresponding to the word type is 2-3, and through screening, the candidate words A, B, and C meet the screening requirements, i.e., the first words A, B, and C are obtained.

[0089] In step 10124, a level word meeting the theme is screened from the plurality of first words.

[0090] Continuing the above example, assuming that the theme of the first word A is animal, the theme of the first word B is animal, and the theme of the first word C is fruit, and the screening range of the theme is animal, the level words meeting the theme can be obtained as A and B.

[0091] In some embodiments, the step 10124 described above can also be implemented by: screening the second words from the plurality of first words according to the theme; and deleting the words of the risk type from the plurality of second words to obtain the checkpoint words. In this way, by deleting the second words that do not meet the theme and the words of the risk type respectively through double screening, the checkpoint words obtained can meet the theme relevance and safety requirements at the same time, which can reduce the cost of manual review and significantly reduce the compliance risk of the checkpoint.

[0092] For example, assuming that the theme of the first word A is animal, the theme of the first word B is animal, and the theme of the first word C is fruit, and the screening range corresponding to the theme is animal, the second words that meet the theme can be obtained as A and B. However, the second word B belongs to the risk type, so the second word B is deleted, and the checkpoint word A is obtained, that is, the final checkpoint word A that meets the theme is obtained.

[0093] In this way, by quantifying the word length, word frequency, and word type of the screening range corresponding to the difficulty value, the candidate words can be accurately screened to obtain the words that meet the screening range, and then the checkpoint words that meet the theme type are obtained by screening the words that meet the screening range according to the theme type. By using a multi-layer filtering mechanism, efficient and accurate checkpoint word screening is realized, which can ensure the screening quality of the checkpoint words.

[0094] In this way, the difficulty of the checkpoint to be generated can be scientifically determined according to the difficulty value and the theme generated according to the generation requirement, and then the complexity of the checkpoint can be accurately controlled by screening the checkpoint words that meet the difficulty value and the theme from the plurality of candidate words, so as to ensure that the difficulty value of the generated checkpoint is consistent with the difficulty value corresponding to the generation requirement.

[0095] In step 102, each of the checkpoint words is split to obtain a plurality of split words of each of the checkpoint words.

[0096] It should be noted that the method of splitting the checkpoint words includes various methods, which can be based on syllables, word roots and affixes, common spelling words, or uniform segmentation, and the specific method can be determined according to actual conditions, which is not limited here.

[0097] As an example, taking splitting based on syllables of a word as an example, "magical" can be split into "ma", "gi", and "cal" according to syllables; taking splitting based on root affixes of a word as an example, "unbreakable" can be split into "un", "break", and "able" according to root affixes; taking splitting according to common spelling words as an example, "station" can be split into "sta" and "tion" according to common spelling words; taking uniform segmentation as an example, "potion" can be split into "po", "ti", and "on" according to uniform segmentation.

[0098] In some embodiments, the step 102 described above can also be implemented in the following manner: identifying a word type to which each of the checkpoint words belongs; and performing the following processing for each of the word types: performing split processing on the checkpoint words corresponding to the word type based on the split rule corresponding to the word type, to obtain a plurality of split words of the checkpoint words. In this way, different split processing is performed for different word types, which can improve the rationality of the split processing, and by using the method of first identifying the word type and then performing split processing according to the type, the word can be avoided from being violently disassembled, and the computing cost can be reduced.

[0099] As an example, assuming that the word type to which the checkpoint word "artificial intelligence" belongs is "composite technology neologism", and the split rule corresponding to the word type "composite technology neologism" is "split into'modifier' + 'central noun'", according to the split rule, the checkpoint word "artificial intelligence" is split into two split words "artificial" and "intelligence"; assuming that the word type to which the checkpoint word "Beijing University" belongs is "special agency name", and the split rule corresponding to the word type "special agency name" is "split into'regional proper name' + 'common name'", according to the split rule, the checkpoint word "Beijing University" is split into two split words "Beijing" and "University".

[0100] In some embodiments, the step 102 described above can also be implemented in the following manner: performing the following processing for each of the checkpoint words: performing classification processing on the characters in the checkpoint word to obtain a predicted position type of each of the characters; and when the predicted position type represents that the character belongs to a boundary character, splitting the checkpoint word with the character as a boundary to obtain a plurality of split words of the checkpoint word. In this way, by character boundary prediction, the internal structure of a compound word can be accurately captured, and the rule dependence of word split processing is converted into data dependence, so that accurate split of the word is realized.

[0101] It should be noted that the boundary character refers to the boundary of the word for splitting processing, for example, "Xue" in the word "deep learning" is a boundary character, and the word "deep" and "learning" can be obtained by splitting the word "deep learning" as a boundary with the character "Xue".

[0102] As an example, assuming that the checkpoint word is "anti-quantum encryption algorithm", the classification processing is performed on each character in the checkpoint word, and the predicted position type of each character can be obtained, that is, "anti" is a non-boundary character, "quantum" is a non-boundary character, "sub" is a non-boundary character, "add" is a boundary character, "secret" is a boundary character, "algorithm" is a boundary character, and "law" is a non-boundary character. According to the predicted position type of each character, the checkpoint word can be split to obtain the split words of the checkpoint word as "anti-quantum", "encryption" and "algorithm".

[0103] In some embodiments, the classification processing is implemented by calling a word classification model. Before performing the above classification processing on the characters in the checkpoint word to obtain the predicted position type of each character, the following processing can be performed: performing classification processing on each character sample in a word sample by the word classification model to obtain the predicted position type of each character sample; determining a first loss based on the actual position type of each character sample in each word sample and the predicted position type of each character sample in each word sample; and training the word classification model based on the first loss to obtain a trained word classification model. In this way, the position type of the character in the checkpoint word is accurately predicted by the model training method, and the checkpoint word is automatically split based on the position type of the character, which can effectively reduce the labor cost and improve the splitting efficiency of the checkpoint word.

[0104] It should be noted that the first loss can be calculated by a specific loss function, such as a cross-entropy loss function, a mean square error loss function, an average absolute error loss function, etc.

[0105] As an example, taking the cross-entropy loss function as an example, the classification processing is performed on each character sample by the word classification model to obtain the probability distribution of the predicted position type of each character sample, the first loss is calculated by formula (1), and the parameters of the word classification model are updated according to the first loss to obtain a trained word classification model.

[0106] (1)

[0107] wherein, represents the actual position type of the i-th character sample, represents the probability distribution of the predicted position type of the i-th character sample.

[0108] In actual application, when the first loss reaches a parameter threshold, the first loss is back-propagated in the word classification model, and the model parameters of the word classification model are updated in the process of back-propagation.

[0109] Here, the back-propagation is described. The character sample is input to the input layer of the word classification model, passes through the hidden layer, and finally reaches the output layer and outputs the result, which is the forward propagation process of the word classification model. Since there is an error between the output result of the word classification model and the actual result, the error between the result and the actual value is calculated, and the error is back-propagated from the output layer to the hidden layer until it propagates to the input layer. In the process of back-propagation, the value of the model parameter is adjusted according to the error; the above process is iterated continuously until convergence.

[0110] Taking the loss function (1) of the word classification model in the embodiment of the present application as an example, the server determines the first loss based on the loss function, back-propagates the first loss from the output layer in the first model, and back-propagates the first loss layer by layer. When the first loss reaches each layer, the gradient (that is, the partial derivative of the loss function with respect to the parameters of each layer) is solved in combination with the conducted first loss, and the parameters of each layer are updated to the corresponding gradient value.

[0111] In step 103, path exploration processing is performed based on the plurality of split words of each level word of the virtual scene to obtain the level of the virtual scene.

[0112] It should be noted that the path exploration processing can be implemented by specific recursive backtracking algorithm, or can be implemented by large language model reasoning, which is not limited here.

[0113] In some embodiments, the step 103 described above can also be implemented in the following way: performing path exploration on the initial arrangement and layout of the split words in the level to determine the word elimination path of the level; when the word elimination path of the level is unique, taking the initial arrangement and layout of the split words in the level as the level of the virtual scene; when the word elimination path of the level is not unique, adjusting the initial arrangement and layout of the split words in the level, and repeatedly performing the path exploration based on the adjusted arrangement and layout until the word elimination path of the level is unique, and taking the adjusted arrangement and layout as the level of the virtual scene. In this way, by adjusting the arrangement and layout of the split words in the level through path exploration, it can be ensured that the generated level is solvable and the solution of each word elimination is unique, thereby enhancing the logical depth and strategy of the generated level.

[0114] As an example, first, for the initial arrangement of split words in the current level, all split words that can be combined are found in the initial arrangement, and a set of split words that can be combined is selected according to a preset weight (for example, according to the length of the combined word) for elimination processing, to obtain an eliminated arrangement; subsequently, all split words that can be combined are found in the eliminated arrangement; if there is at least one split word that can be combined, a set of split words that can be combined is selected according to a preset weight (for example, according to the length of the combined word) for elimination processing, to obtain an eliminated arrangement, which is iterated until there is no split word that can be combined in the eliminated arrangement, and the current elimination path is recorded as a successful path, and a level of a virtual scene is generated according to the initial arrangement of split words corresponding to the current elimination path; if there is no split word that can be combined in the eliminated arrangement, the current elimination path is recorded as a failed path, and the initial arrangement of split words in the current level is adjusted, and the above path exploration process is performed on the adjusted arrangement, until a successful path is found or all arrangements of split words are traversed.

[0115] In step 104, difficulty evaluation processing is performed based on the level to obtain a difficulty evaluation result of the level.

[0116] It should be noted that the difficulty evaluation result of the level can include a difficulty value of the level and an estimated completion time of the level, or can only include the difficulty value of the level; the automatic trial play can be realized by Monte Carlo simulation, and the level difficulty evaluation result can be obtained by analyzing the dynamic difficulty curve; or the difficulty evaluation result of the level can be obtained by statistically analyzing different influencing factors in the level according to a preset rule.

[0117] In some embodiments, step 104 described above can also be realized in the following manner: determining a first difficulty value of the level based on the word frequency of the level words; determining a second difficulty value of the level based on the number of split words in the level; determining a third difficulty value of the level based on the number of level words in the level; determining a fourth difficulty value of the level based on the position placement relationship of the split words of each level word; and fusing the first difficulty value, the second difficulty value, the third difficulty value and the fourth difficulty value, and taking the fusion result as the difficulty evaluation result of the level. In this way, the word frequency, the number of split words, the number of level words and the position placement relationship of the split words are used to realize accurate evaluation of the level difficulty from multiple dimensions, which can ensure the balance and challenge of the game experience.

[0118] It should be noted that the fusion of different difficulty values can be realized by weighted summation.

[0119] As an example, we can count the word frequency of each level's words contained in the current level. Calculate the average word frequency of the words in the current level. Average word frequency of words in the level Perform mapping processing to obtain the first difficulty value of the level. ; Count the number of split words included in the current level The number of words to be split Perform mapping processing to obtain the second difficulty value of the level. ; Count the number of words in the current level. The number of words in the level By performing mapping processing, the third difficulty value of the level is obtained. Based on the positional arrangement of multiple split words corresponding to each word in the current level, the number of level words arranged in sequence can be counted separately. The number of level words that are not arranged in sequence The number of words in the level arranged in order The number of level words that are not arranged in sequence After performing mapping processing, the fourth difficulty value of the level is obtained. Assume the weight of the first difficulty value is... The weight of the second difficulty value is The weight of the third difficulty level is The weight of the fourth difficulty level is Set the first difficulty value Second difficulty level The third difficulty level And the fourth difficulty level By performing weighted fusion, the overall difficulty value of the level is obtained. for This allows us to obtain the difficulty assessment results.

[0120] In some embodiments, automated playthroughs can be achieved through Monte Carlo simulations, a dynamic difficulty curve can be constructed based on the data obtained from the simulation playthroughs, and the dynamic difficulty curve can be analyzed to obtain an evaluation result of the level difficulty.

[0121] As an example, first, a player virtual pool is constructed on the server side, including 500,000 virtual player profiles, the profiles including vocabulary levels of the players (for example, basic, intermediate, and advanced), and the profiles are stored in the cache in the form of numbers for the scheduler to randomly draw; then, 100,000 copies of the word elimination level to be tested are cloned; each instance maintains the same layout, and all instances are mounted to distributed computing nodes, and the nodes take tasks through a queue, then the scheduler draws virtual players from the profile pool in a random order, and injects the profile information of the virtual players into the current level instance, and the node drives the instance to run to an end condition (passing or failing); after all the trials end, a difficulty curve is drawn according to all operation records of the same level; finally, according to the passing rate of the "basic vocabulary" profile in all virtual players, the level difficulty value is determined, for example, if the passing rate is less than 60%, the level is marked as "difficult", if the passing rate is 60% to 80%, the level is marked as "intermediate", and if the passing rate is higher than 80%, the level is marked as "easy", and the median time consumption of all successful passing records is taken as the estimated completion time of the level; the level difficulty value and the estimated completion time are combined to obtain the difficulty evaluation result of the level.

[0122] Below, an exemplary application of the embodiment of the present application in an actual application scenario will be described. The exemplary application describes the specific implementation process of the level generation method of the virtual scene in the word game level generation scenario.

[0123] Word games are a major branch of leisure games. In the era of rapid development of information technology, traditional game level design methods mainly rely on manual editing. This method is not only inefficient, but also the quality of the generated levels is uneven, and it is difficult to provide players with continuous freshness and challenge. With the increase in player demand and the complexity of game content, levels designed by manual editing often seem inadequate, and it is difficult to generate levels on a large scale while ensuring their playability and challenge.

[0124] Based on this, the present application provides a level generation method of a virtual scene, which can automatically generate game levels with clear solution paths and controllable difficulty based on natural language processing and artificial intelligence methods, thereby improving the efficiency and quality of level generation and further improving the gaming experience of players.

[0125] In some embodiments, firstly, based on a large-scale corpus, relevant words are selected and preprocessed, sensitive words are screened out, and words corresponding to the theme content of the level and meeting the difficulty requirements are selected according to the preset difficulty value, mainly from the two dimensions of word length and word frequency, here, a sensitive word table is maintained in the database, which includes a plurality of sensitive words, and the sensitive words are classified and graded, and when the sensitive words are screened out, the words are filtered according to the sensitive word table, to ensure the educational nature and fun of the game; then, according to the preset level difficulty value, the specified number of words of the level, the use frequency level (COCA value) filtering range of the words, the number of two-section words and three-section words, and different combination ways of two-section words and three-section words are obtained by mapping, so as to accurately control the difficulty and complexity of the level; then, according to the number of letter combinations after word splitting and the stacking complexity, the level map is generated, here, the letter combinations after word splitting are fixed in 1 to 3 ways, mainly according to the root, affix and common letter combinations for automatic splitting and storage, and in the level map generation process, the arrangement mode of axial symmetry, central symmetry or combined symmetry can be set, so as to ensure the playability and aesthetics of the level; then, the words are selected and split through morphological analysis, and the split words are filled into the level structure to ensure the diversity and interest of the game; then, with the help of artificial intelligence algorithm, the word elimination path is explored and adjusted to ensure the completeness of the solution path of the level and the uniqueness of the solution of each word combination, which can greatly improve the challenge and interest of the game; finally, Monte Carlo simulation is used for large-scale automatic simulation to analyze the dynamic difficulty curve of the level, and the difficulty and expected completion time of the level are finally evaluated to ensure the balance and challenge of the game experience.

[0126] In some embodiments, the level generation system mainly includes a theme selection module, a lexical analysis module, a path exploration module, and an automated evaluation module. In the theme selection module, words that meet the theme content are extracted and screened from the corpus, and sensitive words are excluded to ensure that all selected words meet the requirements of the theme content; in the lexical analysis module, the selected words are split using a lexical analysis algorithm, and the split words are filled into the level structure to ensure the randomness and challenge of the game; in the path exploration module, path exploration and adjustment are performed through an artificial intelligence (AI) algorithm. Here, path exploration refers to randomly attempting all possible solutions (i.e., possible word combinations) for the current situation of the original level with weights, and recursively executing the process to ensure that the level is solvable and that each word has a unique solution, thereby enhancing the logical depth and strategy of the game; in the automated evaluation module, difficulty evaluation and adjustment are performed using Monte Carlo simulation, and the main influencing factors include the frequency of words that can be combined, the number of split words, the number of words that can be eliminated, and the placement position of split words. Among them, the higher the frequency of words, the lower the difficulty of the level; the fewer the number of split words, the lower the difficulty of the level; the more the number of words that can be eliminated, the lower the difficulty of the level; the more the number of sequentially arranged split words, the lower the difficulty of the level.

[0127] In some embodiments, referring to Figure 6 , Figure 6 is a word elimination process diagram of a month theme level provided by the embodiments of the present application, as shown in Figure 6 , the initial layout of the level includes two word blocks "AP" and "RIL". After "APRIL" is found by clicking the two word blocks, more word blocks are displayed, and the current level is completed by sequentially clicking the word blocks to eliminate them until there is no word block position in the level.

[0128] In some embodiments, referring to Figure 7 , Figure 7 is an interface diagram of a sparkling theme level provided by the embodiments of the present application, as shown in Figure 7 , the initial layout of the level includes six word blocks "DIA", "RL", "PEA", "CRYS", "MOND", and "TAL". By clicking the word block "DIA" and the word block "MOND", "DIAMOND" can be eliminated, by clicking the word block "PEA" and the word block "RL", "PEARL" can be eliminated, and by clicking the word block "CRYS" and the word block "TAL", "CRYSTAL" can be eliminated, thereby completing the word block layout of the current page. After that, more word blocks will be displayed, and the current level is completed by sequentially clicking the word blocks to eliminate them until there is no word block position in the level.

[0129] In summary, the application provides a level generation method for a virtual scene, which can automate level generation and difficulty assessment, significantly improve the efficiency and quality of game level design, ensure the aesthetic appearance of the level shape, reduce the dependence of game level design on manual editing by combining artificial intelligence and natural language processing technology, and provide players with a progressive and varied gaming experience.

[0130] The following continues to illustrate an exemplary structure of the implementation of the level generation apparatus 555 for a virtual scene provided by the embodiments of the application as a software module. In some embodiments, as shown in Figure 2 The software module stored in the level generation apparatus 555 for a virtual scene in the memory 550 can include a screening module 5551, a splitting module 5552, an exploration module 5553, and an assessment module 5554.

[0131] The screening module 5551 is configured to screen at least one level word that meets a first generation requirement from a plurality of candidate words. The splitting module 5552 is configured to perform splitting processing on each level word to obtain a plurality of split words of each level word. The exploration module 5553 is configured to perform path exploration processing based on the plurality of split words of each level word to obtain a level of the virtual scene. The assessment module 5554 is configured to perform difficulty assessment processing based on the level to obtain a difficulty assessment result of the level.

[0132] In some embodiments, the screening module 5551 is further configured to extract a difficulty value and a theme from the first generation requirement, and screen at least one level word that meets the difficulty value and the theme from a plurality of candidate words.

[0133] In some embodiments, the screening module 5551 is further configured to determine a word length positively correlated with the difficulty value and a word frequency negatively correlated with the difficulty value, determine a word type corresponding to the difficulty value, wherein different word types correspond to different splitting numbers, screen a plurality of first words that meet the word length, the word frequency, and the word type from a plurality of candidate words, and screen a level word that meets the theme from the plurality of first words.

[0134] In some embodiments, the screening module 5551 is further configured to screen a second word that meets the theme from the plurality of first words, and delete a word that meets a risk type from the plurality of second words to obtain the level word.

[0135] In some embodiments, the splitting module 5552 is further configured to identify a type of each of the stage words; and perform the following processing for each of the types of words: splitting the stage words corresponding to the type of words according to a splitting rule corresponding to the type of words, to obtain a plurality of split words of the stage words.

[0136] In some embodiments, the splitting module 5552 is further configured to perform the following processing for each of the stage words: classifying characters in the stage words to obtain a predicted position type of each of the characters; and splitting the stage words according to the characters when the predicted position type represents that the character is a boundary character, to obtain a plurality of split words of the stage words.

[0137] In some embodiments, the classification processing is implemented by invoking a word classification model, and the splitting module 5552 is further configured to classify each character sample in a word sample by the word classification model to obtain a predicted position type of each of the character samples; determine a first loss based on an actual position type of each of the character samples in each of the word samples and the predicted position type of each of the character samples in each of the word samples; and train the word classification model based on the first loss to obtain a trained word classification model.

[0138] In some embodiments, the exploration module 5553 is further configured to perform path exploration on an initial arrangement of the split words in the stage to determine a word elimination path of the stage; and when the word elimination path of the stage is unique, take the initial arrangement of the split words in the stage as the stage of the virtual scene; when the word elimination path of the stage is not unique, adjust the initial arrangement of the split words in the stage, and repeatedly perform the path exploration based on the adjusted arrangement until the word elimination path of the stage is unique, and take the adjusted arrangement as the stage of the virtual scene.

[0139] In some embodiments, the evaluation module 5554 is further configured to determine a first difficulty value of the stage based on a word frequency of the stage words; determine a second difficulty value of the stage based on a number of the split words in the stage; determine a third difficulty value of the stage based on a number of the stage words in the stage; determine a fourth difficulty value of the stage based on a position arrangement relationship of the split words of each of the stage words; and fuse the first difficulty value, the second difficulty value, the third difficulty value, and the fourth difficulty value, and take a fusion result as a difficulty evaluation result of the stage.

[0140] It should be noted that the description of the device of the embodiments of the present application is similar to the description of the above-mentioned method embodiments, has similar beneficial effects as the method embodiments, and therefore will not be described again. For the technical details not described in the level generation device of the virtual scene provided by the embodiments of the present application, they can be understood according to the description of any figure. Figure 3 、 Figure 4 、 Figure 5 .

[0141] The embodiments of the present application provide a computer program product, which includes a computer program or computer executable instructions stored in a computer readable storage medium. The processor of the electronic device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the electronic device executes the level generation method of the virtual scene provided by the embodiments of the present application.

[0142] The embodiments of the present application provide a computer readable storage medium, which stores computer executable instructions or computer programs. When the computer executable instructions or computer programs are executed by the processor, the processor will execute the level generation method of the virtual scene provided by the embodiments of the present application, for example, the level generation method of the virtual scene as shown in Figure 3 、 Figure 4 、 Figure 5 .

[0143] In some embodiments, the computer readable storage medium can be a ferroelectric memory (FRAM), a ROM, a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM) memory, etc. It can also be various devices including one or any combination of the above memories.

[0144] In some embodiments, the computer-executable instructions can be in the form of programs, procedures, modules, scripts, or code, and can be written in any programming language, including compiled or interpreted languages, or declarative or procedural languages; and can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0145] By way of example, a computer-executable instruction can, but need not, correspond to a file in a file system. A computer-executable instruction can be stored in one or more files, in a single file, or in multiple files that, in combination, store the instruction. A computer-executable instruction can be, for example, code kept in storage as a computer-readable file; code kept in storage as a computer-readable file, stored on a magnetic disk, magnetic tape, optical medium, or other storage medium, kept as a computer-readable file, or loaded into a computer's memory for execution; or code kept in storage as a computer-readable file, stored on a magnetic disk, magnetic tape, optical medium, or other storage medium, kept as a computer-readable file, or loaded into a computer's memory for execution.

[0146] By way of example, a computer-executable instruction can be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed among multiple sites and interconnected through a communication network.

[0147] To sum up, by the embodiments of the present application, first, the checkpoint words meeting the first generation requirement are screened from the multiple candidate words, which can ensure that the screened checkpoint words can meet the first generation requirement; then, each checkpoint word is split to obtain multiple split words of the checkpoint word, to provide the minimum operable unit for subsequent path exploration; thereafter, path exploration processing is performed according to the multiple split words of each checkpoint word, to ensure that the checkpoint is solvable and the path is unique, to obtain the checkpoint of the virtual scene, thereby enhancing the logical depth and strategy of the game; finally, difficulty evaluation is performed on the checkpoint to obtain the difficulty evaluation result of the checkpoint, thereby ensuring the balance and challenge of the game experience

[0148] The above merely describes the embodiments of the present application, but is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, and improvement within the spirit and scope of the present application shall be included in the protection scope of the present application.

Claims

1. A method for generating levels in a virtual scene, characterized in that, The method includes: Select at least one level word that meets the first generation requirement from multiple candidate words; Each of the level words is split into multiple split words for each level word; Explore the initial arrangement of the split words in the level to determine the word elimination path of the level; When the word elimination path of the level is unique, the initial arrangement of the split words in the level is taken as the level of the virtual scene; when the word elimination path of the level is not unique, the initial arrangement of the split words in the level is adjusted, and the path exploration is repeated based on the adjusted arrangement until the word elimination path of the level is unique, and the adjusted arrangement is taken as the level of the virtual scene. Based on the level, a difficulty assessment process is performed to obtain the difficulty assessment result of the level; wherein, the difficulty assessment process includes: constructing an archive pool on the server side, wherein virtual player profiles representing vocabulary level are stored in a cache in the archive pool in the form of numbers; cloning the word elimination level to be tested into multiple level instances with the same layout and mounting them to distributed computing nodes; randomly extracting the virtual player profiles from the archive pool and injecting them into the current level instance, driving the current level instance to run until the termination condition is met; Based on all operation records for the same level, the level difficulty value is determined by the pass rate of the basic vocabulary level among all virtual players, and the median time taken by all successful pass records is rounded down as the estimated completion time of the level.

2. The method according to claim 1, characterized in that, The step of selecting at least one level word that meets the first generation requirement from multiple candidate words includes: Extract the difficulty value and theme from the first generated requirement; Select at least one level word from multiple candidate words that matches the difficulty value and the theme.

3. The method according to claim 2, characterized in that, The step of selecting at least one level word from multiple candidate words that matches the difficulty value and the theme includes: Determine the word lengths that are positively correlated with the difficulty value and the word frequencies that are negatively correlated with the difficulty value; Determine the word type corresponding to the difficulty value, wherein different word types correspond to different numbers of splits; Select multiple first words that meet the specified word length, word frequency, and word type from a pool of candidate words; Select the level words that match the theme from the plurality of first words.

4. The method according to claim 3, characterized in that, The step of selecting checkpoint words that match the theme from the plurality of first words includes: Select second words that match the theme from the plurality of first words; The checkpoint words are obtained by deleting words that match the risk type from the plurality of second words.

5. The method according to claim 1, characterized in that, The step of splitting each level term into multiple sub-terms for each level term includes: Identify the word type to which each of the level's words belongs; For each of the word types, the following processing is performed: based on the splitting rules corresponding to the word type, the level words corresponding to the word type are split to obtain multiple split words of the level words.

6. The method according to claim 1, characterized in that, The step of splitting each level term into multiple sub-terms for each level term includes: For each of the aforementioned level words, the following processing is performed: The characters in the checkpoint words are classified to obtain the predicted position type of each character; When the predicted position type indicates that the character belongs to the boundary character, the level word is split using the character as the boundary to obtain multiple split words of the level word.

7. The method according to claim 6, characterized in that, The classification process is implemented by calling a word classification model. Before classifying the characters in the level words to obtain the predicted position type of each character, the method further includes: The word classification model is used to classify each character sample in the word sample to obtain the predicted position type of each character sample; A first loss is determined based on the actual position type of each character sample in each word sample and the predicted position type of each character sample in each word sample; The word classification model is trained based on the first loss to obtain the trained word classification model.

8. The method according to claim 1, characterized in that, The process of performing a difficulty assessment based on the level to obtain the difficulty assessment result of the level also includes: Based on the word frequency of the words in the level, the first difficulty value of the level is determined; The second difficulty value of the level is determined based on the number of multiple split words in the level; The third difficulty value of the level is determined based on the number of level words in the level. The fourth difficulty value of the level is determined based on the positional arrangement of multiple split words in each level's words. The first difficulty value, the second difficulty value, the third difficulty value, and the fourth difficulty value are merged, and the fusion result is used as the difficulty evaluation result of the level.

9. A virtual scene level generation device, characterized in that, The device includes: The filtering module is used to select at least one level word from multiple candidate words that meets the first generation requirement; The splitting module is used to split each of the level words to obtain multiple split words for each level word; The exploration module is used to explore the initial arrangement of the split words in the level and determine the word elimination path of the level; When the word elimination path of the level is unique, the initial arrangement of the split words in the level is taken as the level of the virtual scene; when the word elimination path of the level is not unique, the initial arrangement of the split words in the level is adjusted, and the path exploration is repeated based on the adjusted arrangement until the word elimination path of the level is unique, and the adjusted arrangement is taken as the level of the virtual scene. An evaluation module is used to perform difficulty evaluation processing based on the level to obtain the difficulty evaluation result of the level; wherein, the difficulty evaluation processing includes: building an archive pool on the server side, wherein virtual player profiles representing vocabulary level are stored in a cache in the archive pool in the form of numbers; cloning the word elimination level to be tested into multiple level instances with the same layout and mounting them to distributed computing nodes; randomly extracting the virtual player profiles from the archive pool and injecting them into the current level instance, driving the current level instance to run until the termination condition is met; Based on all operation records for the same level, the level difficulty value is determined by the pass rate of the basic vocabulary level among all virtual players, and the median time taken by all successful pass records is rounded down as the estimated completion time of the level.

10. An electronic device, characterized in that, include: Memory, used to store executable instructions; A processor, when executing executable instructions stored in the memory, implements the level generation method for a virtual scene as described in any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that, It stores executable instructions for use by a processor to implement the level generation method of the virtual scene as described in any one of claims 1 to 8.

12. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by the processor, they implement the level generation method for the virtual scene according to any one of claims 1 to 8.

Images