Method and apparatus for commenting on an interactive account

Abstract

The present disclosure relates to an interactive account comment method and device. The method comprises: obtaining user post content; obtaining a first target generative language model corresponding to the user post content; wherein different target generative language models correspond to different accounts, different accounts correspond to different attributes, and the attributes include: roles, topics, text types, and / or comment methods; generating comment information corresponding to the user post content according to the first target generative language model; and outputting the comment information after the user post content through the account corresponding to the first target generative language model. In the present disclosure, based on the content of the generative language model, the user's needs and preferences can be better understood, personalized communication can be generated, the user experience can be improved, and based on the generative language model, more natural and coherent text content can be generated, so that the user feels communicating with a human.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of information processing, and particularly relates to an interactive account comment method and device. BACKGROUND

[0002] In a community APP, companies often use automated accounts to post, comment or announce official activities, etc., to increase user interaction and social interaction, and improve user stickiness and participation.

[0003] Currently, a set of predefined text templates are constructed for different types of comments or topics. These text templates can contain placeholders to facilitate filling in relevant information when actually generating posts or replies.

[0004] However, this approach uses pre-defined text templates, and the generated content is prone to repetition and lacks novelty, i.e., the diversity of generated text is poor. SUMMARY

[0005] To overcome the problems in the related art, the embodiments of the present disclosure provide an interactive account comment method and device. The technical solution is as follows:

[0006] According to a first aspect of the embodiments of the present disclosure, an interactive account comment method is provided, comprising:

[0007] obtaining user post content;

[0008] obtaining a first target generative language model corresponding to the user post content; wherein different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes including: role, topic, text type, and / or comment mode;

[0009] generating comment information corresponding to the user post content according to the first target generative language model;

[0010] outputting the comment information after the user post content through the account corresponding to the first target generative language model.

[0011] In one embodiment, the obtaining of the first target generative language model corresponding to the user post content comprises: obtaining a category corresponding to the user post content.

[0012] obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, the first preset mapping relationship table including a corresponding relationship between the category and the first target generative language model.

[0013] In an embodiment, the obtaining the first target generative language model corresponding to the user post content comprises: receiving a guide keyword, the guide keyword indicating the first target generative language model corresponding to the user post content.

[0014] The second preset mapping relationship table comprises a corresponding relationship between the guide keyword and the first target generative language model.

[0015] In an embodiment, the method further comprises:

[0016] Obtaining a training sample;

[0017] Training a generative language model according to the training sample to obtain a target generative language model.

[0018] In an embodiment, the method further comprises:

[0019] Receiving a post keyword;

[0020] Obtaining a second target generative language model corresponding to the post keyword;

[0021] Generating post content according to the target generative language model;

[0022] Publishing a new post according to the post content.

[0023] According to a second aspect of the embodiments of the present disclosure, an interactive account comment device is provided, comprising:

[0024] A first obtaining module is configured to obtain user post content.

[0025] A second obtaining module is configured to obtain a first target generative language model corresponding to the user post content; different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes comprising: a role, a topic, a text type, and / or a comment mode.

[0026] A first generating module is configured to generate comment information corresponding to the user post content according to the first target generative language model.

[0027] An output module is configured to output the comment information after the user post content through an account corresponding to the first target generative language model.

[0028] In an embodiment, the second obtaining module comprises:

[0029] A first obtaining submodule is configured to obtain a category corresponding to the user post content.

[0030] a second obtaining sub-module, configured to obtain a first target generative language model corresponding to the category according to a first preset mapping relationship table, wherein the first preset mapping relationship table comprises a corresponding relationship between the category and the first target generative language model.

[0031] In an embodiment, the second obtaining module comprises:

[0032] a receiving sub-module, configured to receive a guide keyword, wherein the guide keyword indicates the first target generative language model corresponding to the post content of the user;

[0033] a third obtaining sub-module, configured to obtain a first target generative language model corresponding to the guide keyword according to a second preset mapping relationship table, wherein the second preset mapping relationship table comprises a corresponding relationship between the guide keyword and the first target generative language model.

[0034] In an embodiment, the apparatus further comprises:

[0035] a third obtaining module, configured to obtain a training sample;

[0036] a training module, configured to train a generative language model according to the training sample to obtain a target generative language model.

[0037] In an embodiment, the apparatus further comprises:

[0038] a receiving module, configured to receive a post keyword;

[0039] a fourth obtaining module, configured to obtain a second target generative language model corresponding to the post keyword;

[0040] a second generating module, configured to generate post content according to the target generative language model;

[0041] a post module, configured to post a new post according to the post content.

[0042] According to a third aspect of the embodiments of the present disclosure, an interactive account comment apparatus is provided, comprising:

[0043] a processor;

[0044] a memory for storing processor-executable instructions;

[0045] wherein the processor is configured to:

[0046] obtain post content of a user;

[0047] obtain a first target generative language model corresponding to the user post content; different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes including: a role, a topic, a text type, and / or a comment manner;

[0048] generate, according to the first target generative language model, comment information corresponding to the user post content;

[0049] output, by an account corresponding to the first target generative language model, the comment information after the user post content.

[0050] According to a fourth aspect of the embodiments of the present disclosure, a computer readable storage medium is provided, and the computer readable storage medium stores computer instructions. The computer instructions are executed by a processor to implement the steps of the method in any one of the first aspect.

[0051] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated into the specification and constitute part of the specification, illustrate embodiments consistent with the present disclosure and, together with the specification, serve to explain the principles of the present disclosure.

[0053] Figure 1 is a flowchart of an interactive account comment method according to an exemplary embodiment.

[0054] Figure 2 is a block diagram of an interactive account comment device according to an exemplary embodiment.

[0055] Figure 3 is a block diagram of an interactive account comment device 80 according to an exemplary embodiment. DETAILED DESCRIPTION

[0056] The exemplary embodiments will be described in detail hereinafter with reference to the accompanying drawings. In the following description, the same numbers refer to the same or similar elements throughout the drawings, unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present disclosure. Instead, they only represent examples of devices and methods consistent with some aspects of the present disclosure, as detailed in the appended claims.

[0057] In community APPs, companies often use automated accounts to post, comment, or announce official activities, increase user interaction and social interaction, and improve user stickiness and participation. Common methods are as follows:

[0058] 1. Template construction: Construct a set of predefined text templates for different types of comments or topics. These text templates can contain placeholders for filling in relevant information when generating actual posts or replies.

[0059] The following mainly takes replies as an example, and the way of posting and other text operations is similar.

[0060] 2. Keyword extraction: When a user makes a comment or asks a question in the APP, extract keywords and part-of-speech information from the user's input (for comments) or relevant news information (for posts).

[0061] 3. Template matching: Select a suitable template from the predefined reply templates based on the extracted keyword information. Filtering can be done by matching keywords, part-of-speech, etc.

[0062] 4. Fill in the content: Fill in the placeholders in the selected reply template with the extracted keyword information to generate complete reply content.

[0063] For example: The original text is "How many points does this cotton doll get?" The keyword extraction is "cotton doll"; the template matching is the cute template "Wow, this xxx, so cute!"; the filled keyword generates an automatic comment: "Wow, this cotton doll, so cute!"

[0064] 5. Different automated account output text: reply or post.

[0065] This method has many obvious shortcomings:

[0066] 1. Poor diversity of generated text: Since the text templates used are predefined, the generated content is prone to repetition and lacks novelty.

[0067] 2. Lack of personalization: Template-based text generation may not fully capture the characteristics and potential intent of the original content, resulting in ordinary text that may not meet highly personalized needs.

[0068] 3. Limited naturalness: Since the text is generated based on predefined templates, it may appear stiff or unnatural. Compared to neural network-based methods, this approach is more difficult to generate language that is coherent and fluent.

[0069] 4. Difficult to expand and maintain: When new text types need to be added or text content needs to be updated, templates need to be manually modified and added, which is costly to maintain.

[0070] To solve the above technical problems, the present disclosure provides an interactive account comment method and device.

[0071] Figure 1 This is a flowchart illustrating an interactive account commenting method according to an exemplary embodiment, such as... Figure 1 As shown, the process includes the following steps S101-S104:

[0072] In step S101, the user's post content is obtained;

[0073] In step S102, a first target generative language model corresponding to the user's post content is obtained; wherein, different target generative language models correspond to different accounts, and different accounts correspond to different attributes, including: role, topic, text type, and / or, comment method;

[0074] The characters can have different personas, for example, using A's language style to comment on this post, or using B's language style to comment on this post.

[0075] Comment styles can be varied, including: a lighthearted and humorous style, a serious and earnest style, a Xiaohongshu (Little Red Book) style, a Weibo (Weibo) style, a government report style, etc.

[0076] Text types include classical Chinese, poetry, song lyrics, and novels. Topics can generate related content associated with a user's post, such as associating the post with the vastness of the universe or a holiday to match the festive atmosphere.

[0077] Generative language models, such as ChatGPT, are artificial intelligence models that can automatically generate text sequences that conform to grammatical and semantic rules. Typically, these models are based on neural network structures, pre-trained using historical text data, and can predict the next word or character given a context, continuously generating new text.

[0078] This invention is based on a generative language model to comment on user posts or generate new posts.

[0079] Different generative language models correspond to different automated accounts, and different automated accounts have different language styles, topics, roles, text types and other attributes.

[0080] In one embodiment, step S102 includes the following sub-steps A1-A2:

[0081] A1. Obtain the category corresponding to the user's post content;

[0082] For example, the category of a user's post can be determined by the corresponding tag in the app when the user posts.

[0083] The keywords in the post content of the user can also be extracted, and then the corresponding category of the post content of the user corresponding to the keywords is queried through a database. The database can prestore the corresponding relationship between the keywords and the categories.

[0084] For example, it can be determined through the keywords that the category corresponding to the post content of the user is serious. Here, serious means the category corresponding to the post content of the user.

[0085] A2, obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, the first

[0086] The preset mapping relationship table includes the corresponding relationship between the category and the first target generative language model.

[0087] Continuing with the above example, when it is determined through the keywords that the category corresponding to the post content of the user is serious, the first target generative language model corresponding to serious can be obtained through the first preset mapping relationship table. Here, the comment information generated by the first target generative language model corresponding to serious is serious.

[0088] In an embodiment, in order to improve the diversity of comments, a plurality of first target generative language models described above can be obtained, and each first target generative language model can generate comment information corresponding to the post content of the user. Then, the comment information is outputted in a preset order by the account corresponding to each first target generative language model after the post content of the user, for example, it can be randomly outputted, or it can be sequentially outputted. The order of outputting the comments is not limited in the present disclosure.

[0089] For example, the content of the post is: I saw XX's cotton doll, and I was very excited.

[0090] If the content of the post of the user is obtained in the background, and the category corresponding to the post content of the user obtained is the Little Red Book style, then the first target generative language model corresponding to the Little Red Book style is obtained in the background, and the following content is outputted to the first target generative language model: "Please make a comment in the Little Red Book style based on the following text: I saw XX's cotton doll, and I was very excited." At this time, the comment information generated by the first target generative language model corresponding to the Little Red Book style is: "OMG! XX's cotton doll is simply a masterpiece in the world. Come and take a close-up photo and check in a wave!"

[0091] If the user post content is obtained in the background, and the category corresponding to the obtained user post content is serious, the background will obtain the first target generative language model corresponding to the serious, and then output the following content to the first target generative language model: "Please combine the following text and post a comment in a serious style: 'I saw XX's cotton doll and was very excited.' At this time, the comment information generated by the first target generative language model corresponding to the serious style is: "XX represents fashion and beauty, her cotton doll is not only a response to fans, but also a symbol of a positive attitude towards life and the transmission of positive energy!"

[0092] In another embodiment, step S102 includes the following sub-steps B1-B2:

[0093] B1, receive a guide keyword, the guide keyword indicating a first target generative language model corresponding to the user post content;

[0094] B2, obtain the first target generative language model corresponding to the guide keyword through a second preset mapping relationship table, the second preset mapping relationship table including the correspondence between the guide keyword and the first target generative language model.

[0095] Among them, the target generative language model can directly call a public generative language model, or train a generative language model. In training, the training sample can be obtained first, and then the target generative language model is obtained by training the generative language model through the obtained training sample.

[0096] In one embodiment, in order to improve the diversity of comments, a plurality of first target generative language models described above can be obtained, and then each first target generative language model generates comment information corresponding to the user post content, and then the comment information is output according to a preset order through the account corresponding to each first target generative language model after the user post content, for example: it can be randomly output, or it can be output in turn, and the present disclosure does not limit the order of outputting the comments.

[0097] For example: the content of the post is: "I saw XX's cotton doll and was very excited."

[0098] If the guide keyword is: "Please comment on this post: 'I saw XX's cotton doll and was very excited' in the language styles of A / B / C / D / E / F respectively."

[0099] The system will call the first target generative language model corresponding to the language style of A to generate the following comment: "The emotion presented in this moment will gradually dissipate over time, leaving only a calm and peacefulness", and the system will call the first target generative language model corresponding to the language style of C to generate the following comment: "XX's cotton doll is like a dream, and the only real surprise is the one based on personal experience" and the like.

[0100] Notably, the above guiding keywords are "Please use the language styles of A / B / C / D / E / F to comment on this post: "I saw XX's cotton doll and was very excited". The guiding keywords here include multiple language styles. Of course, the guiding keywords can also output one language style at a time, for example: the guiding keywords are: "Please use the language style of A to comment on this post: "I saw XX's cotton doll and was very excited".

[0101] Based on different guiding keywords, the first target generative language model can generate diverse text for posting or commenting to avoid repetition and monotony.

[0102] In step S103, the comment information corresponding to the user post content is generated according to the first target generative language model;

[0103] In step S104, the comment information is outputted after the user post content by the account corresponding to the first target generative language model.

[0104] The present disclosure not only can post comments, but also can directly post, specifically, in an embodiment, the above method further includes the following sub-steps C1-C4:

[0105] C1, receiving a post keyword;

[0106] C2, obtaining a corresponding second target generative language model according to the post keyword;

[0107] C3, generating post content according to the target generative language model;

[0108] C4, issuing a new post according to the post content.

[0109] The maintenance personnel of the APP can input the post keyword, and then the system can call the corresponding second target generative language model according to the post keyword, then generate the post content corresponding to the post keyword based on the second target generative language model, and issue a new post based on the obtained post content.

[0110] The scheme in the present disclosure can also be applied to social media, online Q&A, customer service, etc.

[0111] The content based on the generative language model in the present disclosure can better understand the needs and preferences of users, generate personalized communication, improve user experience, and based on the generative language model, more natural and coherent text content can be generated, so that users feel communicating with humans. Moreover, since predefined text templates are not used, the generated content is less likely to be repetitive and novel, i.e., the diversity of generated text is improved.

[0112] The generative language model used in the present disclosure can be conveniently used for transfer learning and fine-tuning without manual modification of templates, facilitating multi-field and multi-scenario applications.

[0113] The following is an apparatus embodiment of the present disclosure, which can be used to execute the method embodiments of the present disclosure.

[0114] Figure 2 is a block diagram of an interactive account comment device according to an exemplary embodiment. The device can be implemented as part or all of an electronic device through software, hardware, or a combination of both.

[0115] As shown in Figure 2 , the interactive account comment device includes:

[0116] The first acquisition module 11 is configured to acquire user post content.

[0117] The second acquisition module 12 is configured to acquire a first target generative language model corresponding to the user post content. Different target generative language models correspond to different accounts, and different accounts correspond to different attributes, including roles, topics, text types, and / or comment methods.

[0118] The first generation module 13 is configured to generate comment information corresponding to the user post content according to the first target generative language model.

[0119] The output module 14 is configured to output the comment information after the user post content through the account corresponding to the first target generative language model.

[0120] In one embodiment, the second acquisition module 12 includes:

[0121] The first acquisition sub-module is configured to acquire a category corresponding to the user post content.

[0122] The second acquisition sub-module is configured to acquire a first target generative language model corresponding to the category through a first preset mapping relationship table. The first preset mapping relationship table includes a corresponding relationship between the category and the first target generative language model.

[0123] In an embodiment, the second obtaining module 12 comprises:

[0124] The receiving sub-module is configured to receive a guide keyword, the guide keyword indicating the first target generative language model corresponding to the user post content.

[0125] The third obtaining sub-module is configured to obtain, through a second preset mapping relationship table, a first target generative language model corresponding to the guide keyword, the second preset mapping relationship table including a correspondence between the guide keyword and the first target generative language model.

[0126] In an embodiment, the apparatus further comprises:

[0127] The third obtaining module is configured to obtain a training sample.

[0128] The training module is configured to train a generative language model according to the training sample to obtain a target generative language model.

[0129] In an embodiment, the apparatus further comprises:

[0130] The receiving module is configured to receive a post keyword.

[0131] The fourth obtaining module is configured to obtain, according to the post keyword, a second target generative language model corresponding thereto.

[0132] The second generating module is configured to generate post content according to the target generative language model.

[0133] The post module is configured to post a new post according to the post content.

[0134] According to a third aspect of the embodiments of the present disclosure, an interactive account comment apparatus is provided, comprising:

[0135] A processor;

[0136] A memory for storing processor-executable instructions;

[0137] The processor is configured to:

[0138] Obtain user post content;

[0139] Obtain a first target generative language model corresponding to the user post content; different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes including roles, topics, text types, and / or comment manners;

[0140] Generate comment information corresponding to the user post content according to the first target generative language model;

[0141] The comment information is output by an account corresponding to the first target generative language model after the user posts the content.

[0142] The processor can be further configured to:

[0143] In one embodiment, the processor is configured to obtain the first target generative language model corresponding to the user post content, including: obtaining a category corresponding to the user post content.

[0144] The first target generative language model corresponding to the category is obtained through a first preset mapping relationship table, and the first preset mapping relationship table includes a corresponding relationship between the category and the first target generative language model.

[0145] In one embodiment, the processor is configured to obtain the first target generative language model corresponding to the user post content, including: receiving a guide keyword, the guide keyword indicating the first target generative language model corresponding to the user post content.

[0146] In one embodiment, the processor is configured to obtain the first target generative language model corresponding to the user post content, including: receiving a guide keyword, the guide keyword indicating the first target generative language model corresponding to the user post content.

[0147] The first target generative language model corresponding to the guide keyword is obtained through a second preset mapping relationship table, and the second preset mapping relationship table includes a corresponding relationship between the guide keyword and the first target generative language model.

[0148] In one embodiment, the processor is further configured to:

[0149] Obtain a training sample;

[0150] The target generative language model is obtained by training the generative language model according to the training sample.

[0151] In one embodiment, the processor is further configured to:

[0152] Receive a post keyword;

[0153] Generate post content according to the target generative language model;

[0154] Post a new post according to the post content.

[0155] As to the apparatus in the above embodiments, the specific manners in which various modules perform operations have been described in detail in the embodiments of the method, and will not be described here in detail.

[0156] Figure 3is a block diagram of an apparatus 80 for interactive account commenting, in accordance with an example embodiment, which is suitable for use in a terminal device. For example, the apparatus 80 can be a mobile telephone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, and the like.

[0157] The apparatus 80 can include one or more of the following components: a processing component 802, a memory 804, a power supply component 806, a multimedia component 808, an audio component 810, an input / output (I / O) interface 812, a sensor component 814, and a communication component 816.

[0158] The processing component 802 usually controls overall operations of the apparatus 80, such as operations associated with displaying, making phone calls, data communications, camera operations, and recording operations. The processing component 802 can include one or more processors 820 to execute instructions to complete all or part of steps of the above-described methods. In addition, the processing component 802 can include one or more modules to facilitate interaction between the processing component 802 and other components. For example, the processing component 802 can include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0159] The memory 804 is configured to store various types of data to support operations of the apparatus 80. Examples of these data include instructions to perform any of the applications or methods on the apparatus 80, contact data, phonebook data, messages, pictures, videos, and so on. The memory 804 can be implemented by any type of volatile or non-volatile storage devices or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage devices, flash memory, magnetic disks, or optical disks.

[0160] The power supply component 806 supplies electrical power for the various components of the apparatus 80. The power supply component 806 can include a power supply management system, one or more power sources, and other components associated with generating, managing, and distributing power for the apparatus 80.

[0161] The multimedia component 808 includes a screen providing an output interface between the device 80 and a user. In some embodiments, the screen can include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen can be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense a touch, a slide and a gesture on the touch panel. The touch sensor can not only sense a boundary of a touching or sliding action, but also detect duration and pressure related to the touching or sliding action. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 80 is in an operation mode, such as a camera mode or a video mode, the front camera and / or the rear camera can receive external multimedia data.

[0162] The video mode, the front camera and / or the rear camera can receive external multimedia data. Each of the front camera and the rear camera can be a fixed optical lens system or have a focal length and optical zoom capability.

[0163]

[0164] The audio component 810 is configured to output and / or input an audio signal. For example, the audio component 810 includes a microphone (MIC) to receive an external audio signal when the device 80 is in an operation mode, such as a call mode, a recording mode and a voice recognition mode. The received audio signal can be further stored in the memory 804 or transmitted via the communication component 816. In some embodiments, the audio component 810 further includes a speaker for outputting an audio signal.

[0165] The I / O interface 812 provides an interface between the processing component 802 and peripheral interface modules, such as a keypad, a click wheel, buttons and the like. The buttons can include, but are not limited to, a home button, a volume button, a start button and a lock button.

[0166] The sensor component 814 includes one or more sensors to provide various state assessments for the device 80. For example, the sensor component 814 can detect an open / closed state of the device 80, relative positioning of components, such as a display and a keypad of the device 80, a change in position of the device 80 or a component of the device 80, presence or absence of user contact with the device 80, an orientation or acceleration / deceleration of the device 80, and a temperature change of the device 80. The sensor component 814 can include a proximity sensor configured to detect presence of a nearby object without any physical touch. The sensor component 814 can further include a light sensor, such as a CMOS or CCD image sensor, for use in an imaging application. In some embodiments, the sensor component 814 can further include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor or a temperature sensor.​

[0167] The communication component 816 is configured to facilitate wired or wireless communication between the device 80 and other devices. The device 80 can access a wireless network based on a communication standard, such as WiFi, 2G, or 3G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near-field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.

[0168] In an exemplary embodiment, the device 80 can be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, micro-controllers, microprocessors, or other electronic components, for performing the above-described methods.

[0169] In an exemplary embodiment, a non-transitory computer-readable storage medium, such as the memory 804 including instructions, is also provided, which can be executed by the processor 820 of the device 80 to complete the above-described methods. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, and an optical data storage device, etc.

[0170] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a device 80, enables the device 80 to perform the above-described interactive account comment method, the method comprising:

[0171] Obtaining user post content;

[0172] Obtaining a first target generative language model corresponding to the user post content; wherein different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes including: role, topic, text type, and / or comment mode;

[0173] Generating comment information corresponding to the user post content according to the first target generative language model;

[0174] Outputting the comment information after the user post content through the account corresponding to the first target generative language model.

[0175] In one embodiment, the first target generative language model corresponding to the user post content is obtained by:

[0176] The method comprises the following steps:

[0177] Obtaining a category corresponding to the post content of the user;

[0178] Obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, wherein the first preset mapping relationship table comprises a corresponding relationship between the category and the first target generative language model.

[0179] In one embodiment, the method further comprises:

[0180] Obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, wherein the first preset mapping relationship table comprises a corresponding relationship between the category and the first target generative language model.

[0181] In one embodiment, the method further comprises:

[0182] Obtaining a training sample;

[0183] Training a generative language model according to the training sample to obtain a target generative language model.

[0184] In one embodiment, the method further comprises:

[0185] Receiving a post keyword;

[0186] Obtaining a second target generative language model corresponding to the post keyword;

[0187] Generating post content according to the target generative language model;

[0188] Publishing a new post according to the post content.

[0189] Other embodiments of the present disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the aspects disclosed herein. This application is intended to cover any variations, uses or adaptations of the present disclosure that are deemed to fall within the general principles of the present disclosure and include commonly known or customary practice in the art. The specification and examples are to be regarded as illustrative only and the true scope and spirit of the present disclosure is indicated by the following claims.

[0190] It should be understood that the present disclosure is not limited to the precise construction that has been described above and shown in the accompanying drawings, and that various modifications and changes can be made by those skilled in the art without departing from the scope of the present disclosure. The scope of the present disclosure is limited only by the appended claims.

Claims

1. An interactive account review method, characterized by, The method comprises: obtaining user post content; obtaining a first target generative language model corresponding to the user post content, comprising: obtaining a category corresponding to the user post content; obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, wherein the first preset mapping relationship table comprises a corresponding relationship between the category and the first target generative language model, and further comprising: receiving a guide keyword, wherein the guide keyword indicates the first target generative language model corresponding to the user post content; obtaining a first target generative language model corresponding to the guide keyword through a second preset mapping relationship table, wherein the second preset mapping relationship table comprises a corresponding relationship between the guide keyword and the first target generative language model; wherein different target generative language models correspond to different accounts, and different accounts correspond to different attributes, wherein the attributes comprise: role, topic, text type, and / or comment mode; generating comment information corresponding to the user post content according to the first target generative language model; outputting the comment information after the user post content through the account corresponding to the first target generative language model.

2. The method of claim 1, wherein, The method further comprises: obtaining a training sample; training a generative language model according to the training sample to obtain a target generative language model.

3. The method of claim 2, wherein, The method further comprises: receiving a post keyword; obtaining a corresponding second target generative language model according to the post keyword; generating post content according to the target generative language model; posting a new post according to the post content.

4. An interactive account review apparatus, characterized by, The method comprises: a first obtaining module, configured to obtain user post content; a second obtaining module, configured to obtain a first target generative language model corresponding to the user post content; wherein different target generative language models correspond to different accounts, and different accounts correspond to different attributes, wherein the attributes comprise: role, topic, text type, and / or comment mode; a first generating module, configured to generate comment information corresponding to the user post content according to the first target generative language model; an output module, configured to output the comment information after the user post content through the account corresponding to the first target generative language model; The second obtaining module comprises: a first obtaining submodule, configured to obtain a category corresponding to the user post content; a second obtaining submodule, configured to obtain a first target generative language model corresponding to the category through a first preset mapping relationship table, wherein the first preset mapping relationship table comprises a corresponding relationship between the category and the first target generative language model; a receiving submodule, configured to receive a guide keyword, wherein the guide keyword indicates the first target generative language model corresponding to the user post content; a third obtaining submodule, configured to obtain a first target generative language model corresponding to the guide keyword through a second preset mapping relationship table, wherein the second preset mapping relationship table comprises a corresponding relationship between the guide keyword and the first target generative language model.

5. An interactive account review apparatus, characterized by, The method comprises: a processor; a memory for storing processor-executable instructions; The processor is configured to: obtain user post content; obtain a first target generative language model corresponding to the user post content, including: obtaining a category corresponding to the user post content; obtaining a first target generative language model corresponding to the category through a first preset mapping relationship table, the first preset mapping relationship table including a corresponding relationship between the category and the first target generative language model; further including: receiving a guide keyword, the guide keyword indicating the first target generative language model corresponding to the user post content; obtaining a first target generative language model corresponding to the guide keyword through a second preset mapping relationship table, the second preset mapping relationship table including a corresponding relationship between the guide keyword and the first target generative language model; wherein different target generative language models correspond to different accounts, and different accounts correspond to different attributes, the attributes including: a role, a topic, a text type, and / or a comment mode; generating comment information corresponding to the user post content according to the first target generative language model; outputting the comment information through an account corresponding to the first target generative language model after the user post content.

6. A computer readable storage medium having stored thereon computer instructions, wherein, The instruction is executed by the processor to implement the steps of the method of any one of claims 1 to 3.

Images