Description file generation method and related equipment

Abstract

The invention discloses a description file generation method and related equipment. The description file generation method comprises the steps of obtaining an associated portrait of a target object and a theme range of user generation content to be generated by the target object; determining a plurality of knowledge points in a knowledge reserve graph of the target object based on the associated portrait and the theme range; obtaining a reply of the target object for the associated question of each knowledge point, and obtaining a plurality of reply messages in one-to-one correspondence with the plurality of knowledge points; and generating a description file based on the multiple knowledge points and the multiple pieces of reply information. Therefore, the description file which is matched with the theme range of the user generated content and contains the personalized features of the target object and the understanding of the target object can be automatically generated, and the target object is assisted in creation of the user generated content.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a description file generation method and related equipment. Background Art

[0002] The production of user-generated content, such as pictures, audio, and video, is a relatively low-output task. The main reason is that in the creative planning phase at the beginning of production, creators need to consider which knowledge points to use to create the user-generated content based on the theme requirements of the user-generated content to be created, and whether the user-generated content created using these knowledge points will be promoted and paid attention to.

[0003] The above creative planning often requires a lot of time and energy from the creators. Currently, in order to produce user-generated content that meets the theme requirements and has high content quality, it is often necessary to maintain a dedicated team for creative planning, or spend a lot of time and energy on creation, resulting in high production costs for user-generated content. Summary of the Invention

[0004] The present application provides a description file generation method and related equipment, which can automatically generate a description file that is compatible with the subject scope of user-generated content and contains the personalized characteristics of the target object and the target object's understanding, assisting the target object in creating user-generated content and reducing the production cost of user-generated content.

[0005] In a first aspect, a description file generation method provided in an embodiment of the present application includes:

[0006] Obtaining a related profile of a target object and a subject range of user-generated content to be generated by the target object;

[0007] Based on the association profile and the subject scope, determining a plurality of knowledge points in the knowledge reserve map of the target object;

[0008] Obtaining responses from the target subject to questions related to each knowledge point, and obtaining a plurality of response information, wherein the plurality of response information corresponds one-to-one to the plurality of knowledge points;

[0009] A description file is generated based on the multiple knowledge points and the multiple reply information.

[0010] It can be seen that this method automatically determines multiple knowledge points that are consistent with the associated portrait and the subject range of the user-generated content to be generated by the target object in the knowledge reserve map of the target object through the associated portrait of the target object and the subject range, thereby realizing the automatic selection of creative themes and content. Then, the target object's response to the associated questions for each knowledge point is obtained, and then a description file is generated based on multiple knowledge points and multiple reply information to assist the target object in creating user-generated content. Thus, by generating questions associated with knowledge points and interacting with the target object, the target object can be induced to think about the relevant knowledge points and give corresponding reply information, and then the subsequently generated description file contains the target object's personalized understanding of the relevant knowledge points, so that the user-generated content generated based on the description file can be more in line with the target object and have higher quality.

[0011] In a second aspect, a description file generation device provided in an embodiment of the present application includes:

[0012] An acquisition module, configured to acquire a related portrait of a target object and a subject range of user-generated content to be generated by the target object;

[0013] a determination module, configured to determine a plurality of knowledge points in the knowledge reserve map of the target object based on the association portrait and the subject scope;

[0014] The acquisition module is further configured to acquire the target subject's responses to the associated questions of each knowledge point, thereby obtaining a plurality of response information, wherein the plurality of response information corresponds one-to-one to the plurality of knowledge points;

[0015] A generation module is used to generate a description file based on the multiple knowledge points and the multiple reply information.

[0016] In a third aspect, an electronic device provided in an embodiment of the present application comprises a processor, a memory and a computer program or instructions stored on the memory, wherein the processor executes the computer program or instructions to implement the steps in the method designed in the first aspect above.

[0017] In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, wherein the computer program or instructions are stored. When the computer program or instructions are executed, the steps of the method according to the first aspect are implemented. For example, the computer program or instructions are executed by a processor.

[0018] In a fifth aspect, a computer program product provided in accordance with an embodiment of the present application includes a computer program or instructions, wherein when the computer program or instructions are executed, the steps of the method according to the first aspect are performed. For example, the computer program or instructions are executed by a processor.

[0019] The beneficial effects brought about by the technical solutions of the second to fifth aspects can be referred to the technical effects brought about by the technical solution of the first aspect, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the implementation methods of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the implementation methods or the description of the prior art. Obviously, the drawings described below are only some implementation methods recorded in one or more of the present specifications. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0021] Figure 1 is a schematic diagram of a description file generation system proposed in an embodiment of the present application;

[0022] Figure 2 A flowchart of a description file generation method proposed in an embodiment of the present application;

[0023] Figure 3 This is a schematic diagram of a knowledge reserve map proposed in an embodiment of the present application;

[0024] Figure 4 A schematic diagram of an embodiment of the present application for fusing a knowledge point with its corresponding reply information to obtain the fused knowledge point;

[0025] Figure 5 This is a functional module block diagram of a description file generation device proposed in an embodiment of the present application;

[0026] Figure 6 It is a structural diagram of an electronic device proposed in an embodiment of the present application. DETAILED DESCRIPTION

[0027] It should be understood that the terms "first," "second," and the like in the embodiments of this application are used to distinguish between different objects, rather than to describe a specific order. Furthermore, the terms "including," "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, software, product, or device that includes a series of steps or units is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units inherent to the process, method, product, or device.

[0028] The phrase "embodiment" in the embodiments of this application means that the specific features, structures, or characteristics described in conjunction with the embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive with other embodiments. It is understood explicitly and implicitly by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0029] The “and / or” in the embodiments of the present application describes the association relationship between associated objects, indicating that three types of relationships may exist.

[0030] In the implementation manner of the present application, the symbol “ / ” can indicate that the preceding and following associated objects are in an “or” relationship.

[0031] "At least one item" or similar expressions in the embodiments of this application refers to any combination of these items, including any combination of single items or multiple items, and refers to one or more, and multiple refers to two or more.

[0032] In the embodiments of this application, "higher than" can be used to express the same concept as "greater than", "lower than" can be used to express the same concept as "less than", "not lower than" can be used to express the same concept as "higher than or equal to" or "greater than or equal to", and "not higher than" can be used to express the same concept as "lower than or equal to" or "less than or equal to". In this application, for the same solution, "equal to" can be used in conjunction with "less than" or with "greater than", but not with "less than" and "greater than" at the same time. When "equal to" is used in conjunction with "less than", the technical solution adopted by "less than" is applicable. When "equal to" is used in conjunction with "greater than", the technical solution adopted by "greater than" is applicable.

[0033] In the embodiments of the present application, the terms "of", "corresponding / relevant", "corresponding", "indicated", "associated", etc. may be used interchangeably.

[0034] In the embodiments of this application, the terms “associated with,” “corresponding to,” “is,” “of,” “for,” “belong to,” “as,” and “regarded as” may sometimes be used interchangeably.

[0035] The "connection" in the embodiments of the present application refers to various connection methods such as direct connection or indirect connection to achieve communication between devices, and there is no limitation on this.

[0036] In the embodiments of the present application, the “network” and the “system” may be expressed as the same concept, and the communication system is the communication network.

[0037] The following describes the relevant contents, concepts, meanings, technical issues, technical solutions and beneficial effects involved in the implementation of this application.

[0038] First, see Figure 1 , Figure 1 This is a schematic diagram of a description file generation system proposed in the embodiment of this application. Figure 1 As shown, the generation system may include a terminal device, a generation device and a database.

[0039] In this embodiment, the target object can submit a request to generate user-generated content on the terminal device. After receiving the generation request, the generation device parses the request to obtain the target object's associated portrait and the subject range of the user-generated content to be generated. Then, the generation device obtains the target object's knowledge reserve map from the database, and then determines multiple knowledge points in the knowledge reserve map based on the associated portrait and the subject range. Then, the generation device generates corresponding associated questions for each knowledge point, and sends these associated questions to the terminal device, which displays them to the target object through the terminal device to obtain the target object's response to the associated questions for each knowledge point, and obtain multiple response information. Finally, the generation device generates a description file based on the multiple knowledge points and multiple response information, and sends it to the terminal device.

[0040] It is understandable that Figure 1 The forms and quantities of the terminal devices, generating devices and databases shown in the figure are for example only and do not constitute a limitation on the implementation methods of the present application.

[0041] In this embodiment, the terminal device and the generating device may be a device with a communication function, which may be referred to as a terminal, user equipment (UE), mobile station (MS), mobile terminal (MT), access terminal device, vehicle-mounted terminal device, industrial control terminal device, UE unit, UE station, mobile station, remote station, remote terminal device, mobile device, UE terminal device, wireless communication device, UE agent or UE apparatus, etc. The terminal device and the generating device may be fixed or mobile. It should be noted that the terminal device and the generating device may support at least one wireless communication technology, such as LTE, new radio (NR), wideband code division multiple access (WCDMA), etc. For example, the terminal device and the generating device can be a mobile phone, a tablet computer, a desktop computer, a laptop computer, an all-in-one computer, an in-vehicle terminal, a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a wireless terminal in self-driving, a wireless terminal in remote medical surgery, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), a handheld device with wireless communication function, a computing device or other processing device connected to a wireless modem, a wearable device, a terminal device in a future mobile communication network, or a terminal device in a future evolved public mobile land network (PLMN), etc. In some embodiments of the present application, the terminal device and the generating device may also be devices with transceiver functions, such as a chip system, wherein the chip system may include a chip and may also include other discrete devices.

[0042] In this embodiment, the terminal device and the generating device can also be a server, for example, it can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms. This application does not make specific limitations on this.

[0043] In this embodiment, the database may also be a server or a memory that provides data storage services, such as: read-only memory (ROM) or other types of static storage devices that can store static information and instructions, random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compact disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and can be accessed by a computer. This application also does not make specific limitations on this.

[0044] In this embodiment, the database may also be a storage unit integrated into the terminal device and / or the generating device, and may be a component of the terminal device and / or the generating device.

[0045] See Figure 2 , Figure 2 A flow chart of a description file generation method proposed in the embodiment of this application, which is applied to Figure 1 The generation system shown in FIG. 1 can be implemented by the generation device in the generation system. Figure 2 As shown, the method includes:

[0046] S201: Obtain an associated portrait of a target object and a subject range of user-generated content to be generated by the target object.

[0047] In this embodiment, the target object may refer to the creator of user-generated content, and user-generated content (UGC) may refer to content originally created by the target object, and its form may include: articles, pictures, audio, video, etc.

[0048] In this embodiment, the associated portrait and the subject scope of the UGC can be specified by the target object, or determined by the creative activity in which the target object participates. Specifically, the creative requirements of the creative activity can be obtained, from which keywords representing the object-oriented nature of the activity and the creative scope can be extracted. The user portrait of the object-oriented nature can then be obtained as the associated portrait, and the subject scope of the UGC can be determined using the keywords of the creative scope.

[0049] For example, if the creative activity's creation requirements are as follows:

[0050] This creative event is aimed at enthusiasts with a keen interest in technology, particularly those interested in artificial intelligence and large-scale model applications. It aims to inspire creators to explore the potential and applications of large-scale models in various fields and, through the combination of creativity and technology, demonstrate how large-scale model technology impacts our lives and work.

[0051] After extracting and analyzing keywords, we can find that the target audience of this activity is technology enthusiasts, and the creative scope is artificial intelligence and large-scale model applications. Therefore, we can obtain the following related portraits and UGC topic ranges:

[0052] Related portrait: User portrait of technology enthusiasts.

[0053] The subject range of UGC: artificial intelligence and large-scale model applications.

[0054] In this embodiment, the associated portrait may also refer to the portraits of other users associated with the target object. Exemplarily, the associated portrait may also refer to the user portrait of the target object itself, and the group portrait of users who follow the target object on the UGC publishing platform. Alternatively, the associated portrait may also be a comprehensive portrait determined by the object-oriented user portrait determined by the creative requirements of the creative activity, the user portrait of the target object itself, and the group portrait of users who follow the target object on the UGC publishing platform. For example, a weight can be set for each portrait, and then these portraits can be fused based on their corresponding weights to obtain the fused comprehensive portrait as the associated portrait.

[0055] S202: Based on the association portrait and the subject scope, multiple knowledge points are determined in the knowledge reserve map of the target object.

[0056] In this embodiment, the target user's knowledge reserve map refers to a map that includes the topics and knowledge points contained in the materials that the target user has learned through reading, viewing, etc. Specifically, the topics and knowledge points of the target user's learning videos, reading notes, and reading documents can be extracted and counted, and then the target user's knowledge reserve map can be constructed based on the extracted and counted topics and knowledge points.

[0057] For example, Figure 3 As shown, Figure 3 A schematic diagram of a knowledge reserve map proposed for an embodiment of the present application. In the map, each topic and each knowledge point can be regarded as an independent node. There are lines between topic nodes, between topic nodes and knowledge point nodes, and between knowledge point nodes and knowledge point nodes that represent the relationship between the two. For example: if knowledge point 1 in the map is the superordinate knowledge of knowledge point 2, then there is a line between the node of knowledge point 1 and the node of knowledge point 2 that represents the superordinate relationship between the two; or, if knowledge point 1 belongs to the knowledge point under topic 1, or topic 1 includes knowledge point 1, then there is also a line between the node of topic 1 and the node of knowledge point 1 that represents the inclusion relationship between the two.

[0058] In this embodiment, a weight is also set for each knowledge point node, and the weight is used to characterize the target object's understanding of the knowledge point corresponding to the knowledge point node. Specifically, the number of occurrences of the knowledge point corresponding to the knowledge point node in the historical browsing data of the target object can be determined, and then the third score of the knowledge point corresponding to the knowledge point node can be determined based on the number of occurrences. Then, in the self-assessment table that records the score of the target object's familiarity with each knowledge point in the knowledge reserve map, the score of the knowledge point corresponding to the knowledge point node is determined to obtain a fourth score. Finally, the weight of the knowledge point corresponding to the knowledge point node is determined based on the third score and the fourth score.

[0059] Exemplarily, the third scores corresponding to different ranges of the number of occurrences can be pre-set, for example: 1 point for 1-5 occurrences; 2 points for 6-10 occurrences; 3 points for 11-15 occurrences; 4 points for 16-20 occurrences; and 5 points for 21 or more occurrences. Subsequently, if the number of occurrences of the knowledge point corresponding to the knowledge point node in the historical browsing data is 14, then the corresponding third score is 3 points. The self-assessment form can be displayed to the target object through the terminal device, and then the target object can fill in the self-assessment form on the terminal device and score each knowledge point based on their familiarity with each knowledge point. After obtaining the third score and fourth score corresponding to the knowledge point corresponding to the knowledge point node, the third score and the fourth score can be weighted and summed according to the preset weight to obtain the weight of the knowledge point corresponding to the knowledge point node.

[0060] Therefore, through the knowledge reserve map, we can clearly know the topics and knowledge points that the target object understands, the relationship between these topics and knowledge points, and the target object's understanding of each knowledge point.

[0061] In this embodiment, the most appropriate first topic can be determined in the target object's graph based on the associated profile and the topic scope, serving as the UGC creation topic. Specifically, feature extraction can first be performed on the associated profile to obtain a first feature vector, and feature extraction can also be performed on the topic scope to obtain a second feature vector. This feature extraction method can be any commonly used feature extraction method in the field, such as feature extraction through word embedding, feature extraction through principal component analysis, feature extraction through deep learning neural networks, etc., which are not limited in this application. Then, a first similarity is determined between the feature vector of each topic in the knowledge reserve graph and the first feature vector, as well as a second similarity between the feature vector of each topic and the second feature vector. The feature vector of each topic in the knowledge reserve graph can be pre-extracted using a method for extracting the first feature vector and the second feature vector and stored in the node corresponding to the topic. This similarity can be represented by calculating the cosine similarity, Euclidean distance, Manhattan distance, etc. between the two feature vectors, which are not limited in this application. Finally, based on the first and second similarities corresponding to each topic, the first topic is determined in the knowledge reserve graph. For example, the first similarity and the second similarity can be weighted and summed based on preset weights to obtain the third similarity corresponding to each topic, and then the topic with the third highest similarity in the knowledge reserve map is used as the first topic.

[0062] For example, this embodiment provides a method for determining the matching degree between each topic in the knowledge reserve graph and the associated portrait and topic scope based on a pre-trained language model (such as RoBERTa), as follows:

[0063] (1) Combine the association profile, topic scope, and each topic in the knowledge reserve map to obtain the corresponding text input RoBERTa model;

[0064] Specifically, the data format of the input text is:

[0065] Related Portraits [SEP] Subject Area [SEP] Subject;

[0066] For example, if the associated profile is a user profile of a technology enthusiast, the subject area is artificial intelligence and large model applications, and the topics include: traditional classification models and artificial intelligence neural networks, then the following two input texts can be obtained:

[0067] User profile of technology enthusiasts [SEP] Artificial intelligence and large model applications [SEP] Traditional classification models;

[0068] User profile of technology enthusiasts [SEP] Artificial intelligence and large model applications [SEP] Artificial intelligence neural networks.

[0069] (2) The RoBERTa model performs word segmentation on each input text and performs word embedding on the words obtained after word segmentation to obtain the word vectors of each word corresponding to each text. Then, the RoBERTa model concatenates the word vectors corresponding to each text into the feature vectors of each text according to the order of the words corresponding to each word vector in each text, and then obtains the first feature vector of the association portrait, the second feature vector of the topic range, and the feature vector of the topic in the text;

[0070] (3) Obtain the following two parameters through the hidden layer of the RoBERTa model:

[0071] a) a first similarity between the feature vector of the topic in the text and the first feature vector of the associated portrait;

[0072] b) a second similarity between the feature vector of the topic in the text and the second feature vector of the topic scope;

[0073] (4) Encoding the first similarity threshold and the second similarity threshold set manually respectively and passing them into the hidden layer of the RoBERTa model as the reference vector layer. The first similarity threshold corresponds to the first similarity, and the second similarity threshold corresponds to the second similarity;

[0074] (5) Subtracting the first similarity from the first similarity threshold to obtain a first difference; subtracting the second similarity from the second similarity threshold to obtain a second difference; outputting the first difference and the second difference;

[0075] (6) performing softmax processing on the first difference and the second difference to obtain a processing result;

[0076] Specifically, the first difference and the second difference can be converted into their corresponding probabilities through Softmax processing, and the formula can be expressed by formula ①:

[0077]

[0078] Among them, P(Δ i ) represents the probability corresponding to the i-th difference. Since there are only two differences in this application, the value of i is 1 or 2. e is a natural constant.

[0079] Then, based on their respective corresponding weights, the two probabilities are weighted and summed to obtain the processing result.

[0080] (7) Determine the matching degree of each topic with the associated portrait and the topic scope based on the processing results;

[0081] Specifically, a series of classification threshold ranges can be set, for example: [0, 0.25), [0.25, 0.75), and [0.75, 1], which correspond to different matching values, for example: [0, 0.25) corresponds to a matching value of 1; [0.25, 0.75) corresponds to a matching value of 2; and [0.75, 1] ​​corresponds to a matching value of 3. Thus, according to which range the value of the processing result in step (6) falls into, the matching value of each topic and the associated portrait and the topic range can be determined.

[0082] The output format can be:

[0083] Related Profile [SEP] Subject Range [SEP] Subject Matching Value;

[0084] For example, if the matching values ​​of the two input texts are 1 and 2 respectively, the following output can be obtained:

[0085] User profile of technology enthusiasts [SEP] Artificial intelligence and large model applications [SEP] Traditional classification model 1;

[0086] User profile of technology enthusiasts [SEP] Artificial intelligence and large model applications [SEP] Artificial intelligence neural network 2.

[0087] Then, based on this model, we can obtain the matching degree between each topic in the knowledge reserve and the associated portrait and topic range, and then we can use the topic with the highest matching degree as the first topic.

[0088] Therefore, the first topic is determined based on the associated portrait of the target object and the topic range of the UGC. Among them, the associated portrait of the target object includes the user portrait or group portrait of the masses to which the UGC is directed after its release, which contains information that characterizes the preferences of such groups for UGC content. Then, based on the associated object, the topics that the masses to which the UGC is directed after its release can be screened out to ensure the promotion and attention of the UGC after its release. The topic range plays a screening role, further screening the topics determined by the associated objects to ensure that the final determined topic falls within the topic range. Then, through the joint screening of the associated portrait and the topic range, the first topic that meets the topic requirements and ensures the promotion and attention of the UGC after its release can be obtained.

[0089] In this embodiment, after obtaining the first theme, multiple knowledge points can be determined as the main content of the UGC based on the first theme and the knowledge reserve map. Specifically, first, multiple fifth knowledge points that have a connection relationship with the first theme in the knowledge reserve map can be obtained. From the above content, it can be seen that in the knowledge reserve map, there are lines representing the relationship between the theme nodes and theme nodes, between the theme nodes and knowledge point nodes, and between the knowledge point nodes and knowledge point nodes. If there is a line between the two, it means that there is a connection relationship between the two nodes at both ends of the line. Based on this, by searching for multiple knowledge point nodes that have a connection relationship with the nodes corresponding to the first theme in the knowledge reserve map, the multiple knowledge points corresponding to the multiple knowledge point nodes are used as the multiple fifth knowledge points.

[0090] Then, a first score representing the target subject's familiarity with each fifth knowledge point can be determined. Specifically, the first score can be determined by statistically analyzing the target subject's understanding of the knowledge point and all knowledge points associated with the knowledge point. Exemplarily, in the knowledge reserve map, all knowledge points connected to the fifth knowledge point are obtained to obtain a plurality of sixth knowledge points, which are the knowledge points associated with the fifth knowledge point. Then, a second score for the fifth knowledge point and a second score for each of the sixth knowledge points corresponding to the fifth knowledge point are determined. This second score represents the target subject's understanding of the knowledge point. In the knowledge reserve map, the target subject's understanding of the knowledge point is already represented by the weight of each knowledge point. Therefore, the weight of the fifth knowledge point and each of the sixth knowledge points corresponding to the fifth knowledge point can be directly obtained from the knowledge reserve map to serve as the second score for the fifth knowledge point and the second score for each of the sixth knowledge points corresponding to the fifth knowledge point. Finally, the sum of the second score for the fifth knowledge point and the second score for each of the sixth knowledge points corresponding to the fifth knowledge point is used as the first score for the fifth knowledge point.

[0091] Finally, the fifth knowledge points whose first scores are greater than or equal to the first threshold among the multiple fifth knowledge points can be obtained to obtain multiple knowledge points. Thus, for each of the selected knowledge points, the target object is sufficiently familiar with it and its corresponding knowledge system. In subsequent interactions with the target object, a more personalized understanding of the target object can be obtained, and this understanding will not include misunderstandings caused by unfamiliarity with the relevant knowledge points, thereby ensuring the quality of the ultimately generated UGC.

[0092] In this embodiment, a method for determining multiple knowledge points is also provided. Specifically, the similarity between the feature vector of each knowledge point in the knowledge reserve map and the feature vector of the first theme can be determined to obtain multiple third similarities. Then, the knowledge points in the knowledge reserve map whose third similarity is greater than or equal to the second threshold are obtained to obtain multiple seventh knowledge points. Then, the scores that characterize the target object's familiarity with each seventh knowledge point are determined to obtain multiple fifth scores, and then the seventh knowledge points whose fifth scores are greater than or equal to the second threshold among the multiple seventh knowledge points are used as multiple knowledge points. Thus, the scope of the knowledge points that are initially screened is expanded from the knowledge points that have a connection relationship with the first theme to all knowledge points in the knowledge map, so that the comprehensiveness and diversity of the multiple knowledge points determined are higher.

[0093] S203: Obtain the target object's response to the related questions of each knowledge point, and obtain multiple response information.

[0094] In this embodiment, the multiple reply messages correspond to multiple knowledge points in a one-to-one manner. Specifically, the associated questions are some questions related to the knowledge point. For example, if a knowledge point is sparrow, its associated questions can be as follows:

[0095] 1. What is the main habitat of sparrows?

[0096] 2. What are the characteristics of a sparrow’s body shape and feather color?

[0097] 3.What are the main food sources for sparrows?

[0098] 4. What sounds do sparrows make? What purposes do they usually serve?

[0099] 5. What are the characteristics of sparrows’ lifestyle and behavior in urban environments?

[0100] In this embodiment, by presenting these related questions to the target object, the target object's response information to these questions can be obtained.

[0101] S204: Generate a description file of the user-generated content based on the multiple knowledge points and the multiple reply information.

[0102] In this embodiment, first, multiple knowledge points can be grouped to obtain multiple knowledge groups. Specifically, if a first knowledge point and a second knowledge point have a first relationship in the knowledge reserve map, the first knowledge point and the reply information corresponding to the first knowledge point are fused to obtain a third knowledge point corresponding to the first knowledge point. The second knowledge point and the reply information corresponding to the second knowledge point are fused to obtain a fourth knowledge point corresponding to the second knowledge point. The third knowledge point and the fourth knowledge point are then grouped into the same knowledge group. In this way, multiple knowledge points are traversed and grouped to obtain multiple knowledge groups.

[0103] Specifically, as we've seen from the above discussion of the knowledge reserve map, the relationships between knowledge points in the knowledge reserve map can include: general-specific relationships, causal relationships, and progressive relationships. The nodes of two knowledge points with a relationship will be connected in the knowledge reserve map, and the relationship between them will be labeled. Therefore, by consulting the knowledge reserve map, you can quickly determine the relationship between any two knowledge points.

[0104] In this embodiment, the first relationship can be any one of the above-mentioned total-specific relationship, causal relationship and progressive relationship. Taking the first relationship as a total-specific relationship as an example, when there is a total-specific relationship between the first knowledge point and the second knowledge point in the knowledge reserve map, the first knowledge point and the reply information corresponding to the first knowledge point are fused to obtain the third knowledge point corresponding to the first knowledge point, and the second knowledge point and the reply information corresponding to the second knowledge point are fused to obtain the fourth knowledge point corresponding to the second knowledge point, and the third knowledge point and the fourth knowledge point are divided into the same knowledge group.

[0105] For example, there are nine knowledge points, among which knowledge points A, C, and E are main knowledge points, knowledge points D and G are sub-knowledge points under knowledge point A, knowledge points B, F, and I are sub-knowledge points under knowledge point C, and knowledge point H is a sub-knowledge point under knowledge point E. Then, the response information corresponding to knowledge point A and knowledge point A are combined to obtain knowledge point A1; the response information corresponding to knowledge point D and knowledge point D are combined to obtain knowledge point D1; the response information corresponding to knowledge point G and knowledge point G are combined to obtain knowledge point G1; knowledge points A1, D1, and G1 are divided into knowledge group 1. The response information corresponding to knowledge point C and knowledge point C are combined to obtain knowledge point C1; the response information corresponding to knowledge point B and knowledge point B are combined to obtain knowledge point B1; the response information corresponding to knowledge point F and knowledge point F are combined to obtain knowledge point F1; the response information corresponding to knowledge point I and knowledge point I are combined to obtain knowledge point I1; and knowledge points C1, C1, F1, and I1 are divided into knowledge group 2. Fusion of knowledge point E and its corresponding reply information yields knowledge point E1; fusion of knowledge point H and its corresponding reply information yields knowledge point H1; and division of knowledge point E1 and knowledge H1 into knowledge group 3. Finally, the following three knowledge groups are obtained:

[0106] Knowledge group 1 [knowledge point A1, knowledge point D1, knowledge point G1];

[0107] Knowledge group 2 [knowledge point C1, knowledge point B1, knowledge point F1, knowledge point I1];

[0108] Knowledge Group 3 [Knowledge Point E1, Knowledge Point H1].

[0109] In this embodiment, if Figure 4 As shown, when integrating knowledge points and their corresponding reply information, corresponding supplementary content can also be generated for each knowledge point. This supplementary content can be the definition of the corresponding knowledge point, either official or popular; or it can be the extension of the corresponding knowledge point, such as what is the superordinate concept of the knowledge point, how it differs from other concepts under the superordinate concept, and some positive and negative examples.

[0110] For example, taking the above-mentioned knowledge point of sparrow as an example, the supplementary content may include the following:

[0111] definition:

[0112] Official definition: Sparrows are birds in the family Passeridae, order Passeriformes. They are typically small, with brown or gray plumage, ranging from 10 to 20 cm in length, and short, pointed beaks adapted for pecking seeds and insects. There are often significant sexual differences; for example, male sparrows tend to have brighter plumage. They are widely distributed worldwide, with most species living in open environments such as cities, rural areas, and grasslands.

[0113] Popular definition: Sparrows are small, common birds, especially those found in both urban and rural areas. Their clear and frequent calls make them commonplace in both urban and rural environments.

[0114] Superordinate concept: bird.

[0115] The difference between sparrows and crows is that they are two distinct bird species. Sparrows are small, typically brown or gray in color, feed primarily on seeds and insects, are highly social, and are found in both urban and rural areas. Crows, on the other hand, are large, all-black, and omnivorous, consuming a diet that includes carrion and garbage. They are highly intelligent and have complex behaviors, and are found in both forests and open areas.

[0116] In this embodiment, after obtaining supplementary content, the generated supplementary content can be modified based on the multiple responses corresponding to that knowledge point. For example, if the supplementary information states, "Sparrow feathers are mostly brown or gray," and the target subject, in the response to related question 2, states that sparrow feathers can be reddish-brown in addition to brown or gray, then based on this response, the supplementary information "Sparrow feathers are mostly brown or gray" can be modified to "Sparrow feathers are mostly brown, gray, or reddish-brown." Information that is present in the multiple responses but not in the supplementary information can be added to the supplementary information.

[0117] In this embodiment, the supplementary information that has been corrected and added can be added to the corresponding knowledge point to complete the fusion of the knowledge point and its corresponding reply information.

[0118] Then, the knowledge points in each knowledge group can be arranged according to the relationship order corresponding to the first relationship to obtain the knowledge sequence group corresponding to each knowledge group. Specifically, each knowledge sequence group includes at least one knowledge sequence. When the first relationship is a total-score relationship, the relationship order should be the total-score order, that is, the total knowledge point is ranked first, and the sub-knowledge points are arranged in sequence behind the total knowledge point; when the first relationship is a cause-and-effect relationship, the relationship order should be the cause-and-effect order, that is, the knowledge point representing the cause is ranked first, and the knowledge point representing the effect is arranged in sequence behind the total knowledge point; when the first relationship is a progressive relationship, the relationship order should be the progressive order. Taking the three knowledge groups in the above example as an example, after arranging them in the order of total score, the following three knowledge sequence groups can be obtained:

[0119] Knowledge sequence group 1:

[0120] Knowledge sequence 1: knowledge point A1, knowledge point D1, knowledge point G1;

[0121] Knowledge sequence 2: knowledge point A1, knowledge point G1, knowledge point D1;

[0122] Knowledge sequence group 2:

[0123] Knowledge sequence 1: knowledge point C1, knowledge point B1, knowledge point F1, knowledge point I1;

[0124] Knowledge sequence 2: knowledge point C1, knowledge point B1, knowledge point I1, knowledge point F1;

[0125] Knowledge sequence 3: knowledge point C1, knowledge point F1, knowledge point B1, knowledge point I1;

[0126] Knowledge sequence 4: knowledge point C1, knowledge point F1, knowledge point I1, knowledge point B1;

[0127] Knowledge sequence 5: knowledge point C1, knowledge point I1, knowledge point B1, knowledge point F1;

[0128] Knowledge sequence 6: knowledge point C1, knowledge point I1, knowledge point F1, knowledge point B1;

[0129] Knowledge sequence group 3:

[0130] Knowledge sequence 1: Knowledge point E1, knowledge point H1.

[0131] Then, multiple splicing processes can be performed based on the multiple knowledge sequence groups to obtain multiple first knowledge sequences. In each splicing process, a knowledge sequence is selected from each knowledge sequence group for splicing to obtain a first knowledge sequence, and any two first knowledge sequences among the multiple first knowledge sequences are different.

[0132] For example, taking the three knowledge sequence groups obtained above as an example, each time a knowledge sequence is selected from knowledge sequence group 1, knowledge sequence group 2, and knowledge sequence group 3 for splicing. For example, in a certain splicing, the selected knowledge sequences are as follows:

[0133] Knowledge sequence group 1:

[0134] Knowledge sequence 1: knowledge point A1, knowledge point D1, knowledge point G1;

[0135] Knowledge sequence group 2:

[0136] Knowledge sequence 1: knowledge point C1, knowledge point B1, knowledge point F1, knowledge point I1;

[0137] Knowledge sequence group 3:

[0138] Knowledge sequence 1: Knowledge point E1, knowledge point H1.

[0139] The first knowledge sequence obtained by this splicing is: knowledge point A1, knowledge point D1, knowledge point G1, knowledge point C1, knowledge point B1, knowledge point F1, knowledge point I1, knowledge point E1, knowledge point H1. Finally, a description file can be generated based on multiple first knowledge sequences. Specifically, the distance between each two adjacent knowledge points in each first knowledge sequence can be determined to obtain multiple first distances corresponding to each first knowledge sequence. For example, the first knowledge sequence is as follows:

[0140] Knowledge point A1, knowledge point D1, knowledge point G1, knowledge point C1, knowledge point B1, knowledge point F1, knowledge point I1, knowledge point E1, knowledge point H1.

[0141] Then, eight first distances are calculated respectively between knowledge point A1 and knowledge point D1, between knowledge point D1 and knowledge point G1, between knowledge point G1 and knowledge point C1, between knowledge point C1 and knowledge point B1, between knowledge point B1 and knowledge point F1, between knowledge point F1 and knowledge point I1, between knowledge point I1 and knowledge point E1, and between knowledge point E1 and knowledge point H1. The first distance can be the cosine similarity, Euclidean distance, etc. between the feature vectors of two knowledge points, which will not be repeated here in this application.

[0142] After obtaining the multiple first distances corresponding to each first knowledge sequence, the knowledge distance of each first knowledge sequence can be determined based on the multiple first distances corresponding to each first knowledge sequence. For example, the sum or average of the multiple first distances corresponding to each first knowledge sequence can be used as the knowledge distance of each first knowledge sequence. Then, the first knowledge sequence with the smallest knowledge distance is used as the description file. Thus, the relative distance between each knowledge point in the first knowledge sequence is the shortest, and the viewers of the UGC generated based on the description file generated for the first knowledge sequence will not be carried by the content shared in the UGC to jump between knowledge points, that is, the switching between each knowledge point will be smoother due to the shorter distance between them, which will help viewers better understand the content shared in the UGC.

[0143] In this embodiment, after obtaining the description file, the generation device can feed the description file back to the terminal device and display it to the target object. The target object can then create UGC based on the knowledge points, supplementary content of each knowledge point, and the arrangement order of each knowledge point in the description file to obtain a UGC file. Of course, the generation device can also input the description file into a corresponding UGC generation device, such as a Wensheng picture, Wensheng video, etc., to generate a corresponding UGC file, and then feed the UGC file back to the terminal device for display to the target object.

[0144] In this embodiment, when a UGC file is generated by a UGC generation device, the device can also evaluate the UGC file after receiving it. If the evaluation score is greater than a threshold, the UGC is fed back to the target object and the UGC file is published. Otherwise, a new UGC file is regenerated based on the UGC generation device and evaluated until a qualified UGC file is generated.

[0145] The assessment may include at least one of the following:

[0146] Content relevance assessment, continuity assessment, theme consistency assessment, emotional progression assessment, form's auxiliary assessment to content, content comprehensiveness assessment and thought-provoking assessment.

[0147] The content relevance assessment refers to whether all the shared content in the UGC file is related. Specifically, the subsection shared content generated based on each second knowledge point in the description file can be input into the pre-trained language model to obtain the vector of the subsection shared content. Then, the similarity between the vectors of each subsection shared content is calculated. If the similarity between the vectors of each subsection shared content is higher than the threshold, the content relevance assessment is determined to be qualified.

[0148] Continuity assessment refers to whether the correlation between the shared content of adjacent sections meets the standards. Specifically, the similarity between the shared content of adjacent sections is calculated. If the similarity between the shared content of adjacent sections is higher than the threshold, the continuity assessment is determined to be qualified.

[0149] The topic consistency assessment refers to whether the correlation between the topic scope and the content shared in each section meets the standards. Specifically, the similarity between the content shared in each section and the topic scope can be calculated. If the similarity between the content shared in each section and the topic scope is higher than the threshold, the topic consistency assessment is judged to be qualified.

[0150] The emotional progressive evaluation refers to whether the emotional level of the shared content in the later sections is higher than the emotional level of the shared content in the earlier sections. Specifically, the emotional level of the shared content in each section can be determined by the emotional level determination model. If the emotional level of each section content is in a progressive relationship according to the arrangement order of each section content in the UGC file, the emotional progressive evaluation is judged to be qualified.

[0151] The assessment of the supportiveness of form to content refers to whether the inclusion of form in a UGC document has a positive impact on the content description. For example, for visual UGC documents such as images, text, and videos, whether important knowledge points are highlighted, repeated, or emphasized. If all important knowledge points in the UGC document are prominently presented, the assessment of the supportiveness of form to content is considered qualified.

[0152] The content comprehensiveness assessment refers to whether the area occupied by the knowledge points in the UGC file in the knowledge graph exceeds the threshold. When the area occupied by the knowledge points in the UGC file in the knowledge graph exceeds the threshold, the content comprehensiveness assessment is considered qualified.

[0153] The thought-provoking evaluation refers to whether there are question sentences in the UGC file. The UGC file can be screened through the sentence type reasoning model. If there are question sentences, the thought-provoking evaluation is considered qualified.

[0154] In summary, this method automatically determines multiple knowledge points that conform to the associated portrait and the subject range of the user-generated content to be generated by the target object in the knowledge reserve map of the target object, thereby realizing the automatic selection of creative themes and content. Then, the target object's responses to the associated questions for each knowledge point are obtained, and then a description file is generated based on multiple knowledge points and multiple reply information to assist the target object in creating user-generated content. Thus, by generating questions associated with knowledge points and interacting with the target object, the target object can be induced to think about the relevant knowledge points and give corresponding reply information, and then the subsequently generated description file contains the target object's personalized understanding of the relevant knowledge points, so that the user-generated content generated based on the description file can be more in line with the target object and have higher quality.

[0155] The above mainly introduces the scheme of the implementation method of the present application from the perspective of the method side. It is understandable that in order to realize the above functions, the training device includes hardware structures and / or software modules corresponding to the execution of each function. Those skilled in the art should easily realize that, in combination with the units and algorithm steps of each example described in the implementation method disclosed in this article, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0156] The embodiments of the present application can divide the terminal device into functional units according to the above-mentioned method examples. For example, each functional module can be divided according to each function, or two or more functions can be integrated into one functional module. The above-mentioned integrated modules can be implemented in the form of hardware or software program modules. It should be noted that the division of modules in the embodiments of the present application is schematic and is only a logical functional division. In actual implementation, other division methods can be used.

[0157] In the case of integrated modules, Figure 5 1 is a functional module block diagram of a description file generation device proposed in an embodiment of the present application, wherein the generation device 500 includes an acquisition module 501 , a determination module 502 and a generation module 503 .

[0158] In this embodiment, the acquisition module 501, the determination module 502 and the generation module 503 may be a module for receiving and processing signals, information, etc. or determining a monitoring mechanism, and there is no specific limitation on this.

[0159] In this embodiment, the generating device 500 may further include a storage module, which is used to generate computer program codes or instructions executed by the device 500. The storage module may be a memory.

[0160] In this embodiment, the generating device 500 may be a chip or a chip module.

[0161] In this embodiment, the acquisition module 501, the determination module 502, and the generation module 503 may be integrated into a communication module, which may be a communication interface, a transceiver, a transceiver circuit, or the like.

[0162] In this embodiment, the acquisition module 501, the determination module 502 and the generation module 503 may be integrated into a processor.

[0163] It should be noted that the processor can be a baseband processor, a baseband chip, a central processing unit (CPU), a general-purpose processor, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, a transistor logic device, a hardware component or any combination thereof. It can implement or execute the various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. The processing module can also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0164] In this embodiment, the generating device 500 is used to execute any step executed by the terminal device / chip / chip module, etc. in the above method embodiment.

[0165] In specific implementation, the acquisition module 501, the determination module 502 and the generation module 503 are used to execute any step in the above method implementation, and when executing an action such as sending, other modules can be optionally called to complete the corresponding operation.

[0166] An acquisition module 501 is used to acquire a related portrait of a target object and a subject range of user-generated content to be generated by the target object;

[0167] A determination module 502 is configured to determine a plurality of knowledge points in the knowledge reserve map of the target object based on the association portrait and the subject scope;

[0168] The acquisition module 501 is further configured to acquire the target subject's responses to the questions related to each knowledge point, thereby obtaining a plurality of response information, wherein the plurality of response information corresponds one-to-one to the plurality of knowledge points;

[0169] The generating module 503 is configured to generate a description file based on the plurality of knowledge points and the plurality of reply information.

[0170] In this embodiment, in terms of generating a description file based on the multiple knowledge points and the multiple reply information, the generating module 503 is specifically configured to:

[0171] Grouping the multiple knowledge points to obtain multiple knowledge groups, wherein, if a first relationship exists between a first knowledge point and a second knowledge point in the knowledge reserve map, fusing the first knowledge point and the reply information corresponding to the first knowledge point to obtain a third knowledge point corresponding to the first knowledge point, fusing the second knowledge point and the reply information corresponding to the second knowledge point to obtain a fourth knowledge point corresponding to the second knowledge point, and grouping the third knowledge point and the fourth knowledge point into the same knowledge group, where the first knowledge point and the second knowledge point are two different knowledge points among the multiple knowledge points;

[0172] Arrange the knowledge points in each of the knowledge groups according to the arrangement order corresponding to the first relationship to obtain a knowledge sequence group corresponding to each of the knowledge groups, wherein the knowledge sequence group corresponding to each of the knowledge groups includes at least one knowledge sequence;

[0173] Performing multiple splicing processes based on the multiple knowledge sequence groups to obtain multiple first knowledge sequences, wherein in each splicing process, one knowledge sequence is selected from each knowledge sequence group for splicing to obtain a first knowledge sequence, and any two first knowledge sequences in the multiple first knowledge sequences are different;

[0174] The description file is generated based on the plurality of first knowledge sequences.

[0175] In this embodiment, in terms of generating the description file based on the multiple first knowledge sequences, the generating module 503 is specifically configured to:

[0176] determining the distance between every two adjacent knowledge points in each of the first knowledge sequences to obtain a plurality of first distances corresponding to each of the first knowledge sequences;

[0177] determining a knowledge distance of each of the first knowledge sequences based on a plurality of first distances corresponding to each of the first knowledge sequences;

[0178] The first knowledge sequence with the smallest knowledge distance is used as the description file.

[0179] In this embodiment, in determining a plurality of knowledge points in the knowledge reserve map of the target object based on the association portrait and the subject scope, the determination module 502 is specifically configured to:

[0180] Performing feature extraction on the associated portrait to obtain a first feature vector;

[0181] Performing feature extraction on the subject range to obtain a second feature vector;

[0182] Determining a first similarity between a feature vector of each topic in the knowledge reserve map and the first feature vector, and a second similarity between the feature vector of each topic and the second feature vector;

[0183] Determining a first topic in the knowledge reserve map based on the first similarity and the second similarity corresponding to each topic;

[0184] Based on the first topic and the knowledge reserve map, the multiple knowledge points are determined.

[0185] In this embodiment, in determining the plurality of knowledge points based on the first topic and the knowledge reserve map, the determination module 502 is specifically configured to:

[0186] Acquire a plurality of fifth knowledge points in the knowledge reserve map that are connected to the first topic;

[0187] determining a first score representing the target subject's familiarity with each fifth knowledge point;

[0188] Acquire the fifth knowledge points whose first scores are greater than or equal to the first threshold among the multiple fifth knowledge points to obtain the multiple knowledge points.

[0189] In this embodiment, in determining the first score representing the target subject's familiarity with each fifth knowledge point, the determination module 502 is specifically configured to:

[0190] Acquire a plurality of sixth knowledge points in the knowledge reserve map that have a connection relationship with each of the fifth knowledge points;

[0191] Determining a second score for each fifth knowledge point and a second score for each sixth knowledge point;

[0192] The first score of each fifth knowledge point is determined based on the second score of each fifth knowledge point and the second scores of the plurality of sixth knowledge points.

[0193] In this embodiment, in determining the second score of each fifth knowledge point, the determination module 502 is specifically configured to:

[0194] In the historical browsing data of the target object, determining the number of occurrences of each fifth knowledge point in the historical browsing data;

[0195] determining a third score based on the number of occurrences;

[0196] In a self-assessment form that records the target subject's score of familiarity with each knowledge point in the knowledge reserve map, determining a score for each fifth knowledge point to obtain a fourth score;

[0197] A second score for each of the fifth knowledge points is determined based on the third score and the fourth score.

[0198] In this embodiment, in determining the plurality of knowledge points based on the first topic and the knowledge reserve map, the determination module 502 is specifically configured to:

[0199] Determining the similarity between the feature vector of each knowledge point in the knowledge reserve map and the feature vector of the first topic to obtain a plurality of third similarities;

[0200] Acquire knowledge points in the knowledge reserve graph whose third similarity is greater than or equal to the second threshold, to obtain a plurality of seventh knowledge points;

[0201] determining a score representing the target subject's familiarity with each seventh knowledge point to obtain a plurality of fifth scores;

[0202] The seventh knowledge points whose fifth scores are greater than or equal to the second threshold among the multiple seventh knowledge points are taken as the multiple knowledge points.

[0203] See Figure 6 , Figure 6 6 is a schematic diagram of the structure of an electronic device proposed in an embodiment of the present application, wherein the electronic device 600 may include a processor 610, a memory 620, and a communication bus for connecting the processor 610 and the memory 620.

[0204] Optionally, the memory 620 includes but is not limited to random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or portable read-only memory (CD-ROM), and the memory 620 is used to store the program code executed by the electronic device 600 and the transmitted data.

[0205] In this embodiment, the electronic device 600 further includes a communication interface for receiving and sending data.

[0206] In this embodiment, the processor 610 may be one or more CPUs. When the processor 610 is a CPU, the CPU may be a single-core CPU or a multi-core CPU.

[0207] In this embodiment, the processor 610 can be a baseband chip, a chip, a CPU, a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof.

[0208] In a specific implementation, the processor 610 in the electronic device 600 is configured to execute the computer program or instruction 621 stored in the memory 620 to perform the following operations:

[0209] Obtaining a related profile of a target object and a subject range of user-generated content to be generated by the target object;

[0210] Based on the association profile and the subject scope, determining a plurality of knowledge points in the knowledge reserve map of the target object;

[0211] Obtaining responses from the target subject to questions related to each knowledge point, and obtaining a plurality of response information, wherein the plurality of response information corresponds one-to-one to the plurality of knowledge points;

[0212] A description file is generated based on the multiple knowledge points and the multiple reply information.

[0213] It should be noted that Figure 6 The specific implementation of each operation in the embodiment can be found in the description of the method embodiment shown above, and will not be detailed here.

[0214] The embodiments of the present application further provide a computer-readable storage medium storing a computer program or instructions, which implement the steps described in the above method embodiments when executed.

[0215] The embodiments of the present application further provide a computer program product, including a computer program or instructions, which implement the steps described in the above method embodiments when executed.

[0216] It should be noted that, for the sake of simplicity, each of the above-mentioned embodiments is expressed as a series of action combinations. Those skilled in the art should be aware that the present application is not limited by the order of the actions described, because some steps in the embodiments of the present application can be performed in other orders or simultaneously. In addition, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions, steps, modules or units involved are not necessarily required for the embodiments of the present application.

[0217] In the above embodiments, the description of each embodiment in the embodiments of the present application has different focuses. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0218] The steps of the method or algorithm described in the embodiments of the present application can be implemented in hardware or by a processor executing software instructions. The software instructions can be composed of corresponding software modules, which can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (EEPROM), registers, hard disks, mobile hard disks, read-only compact disks (CD-ROMs), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be an integral part of the processor. The processor and storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and storage medium can also exist in a terminal device or a management device as discrete components.

[0219] Those skilled in the art will appreciate that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via a wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) method. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated therein. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0220] The modules / units included in the devices and products described in the above embodiments may be software modules / units, hardware modules / units, or partly software modules / units and partly hardware modules / units. For example, for the devices and products applied to or integrated in the chip, the modules / units included therein may all be implemented in the form of hardware such as circuits, or at least part of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for the devices and products applied to or integrated in the chip module, the modules / units included therein may all be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as chip, circuit module, etc.) or different components of the chip module, or at least part of the modules / units may be It is implemented in the form of a software program, which runs on the processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits; for various devices and products applied to or integrated in the terminal equipment, the various modules / units contained therein can be implemented in the form of hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or different components in the terminal equipment, or, at least some modules / units can be implemented in the form of a software program, which runs on the processor integrated inside the terminal equipment, and the remaining (if any) modules / units can be implemented in the form of hardware such as circuits.

[0221] The specific implementation methods described above further explain in detail the purpose, technical solutions and beneficial effects of the implementation methods of the present application. It should be understood that the above description is only the specific implementation method of the implementation methods of the present application and is not intended to limit the scope of protection of the implementation methods of the present application. Any modifications, equivalent replacements, improvements, etc. made on the basis of the technical solutions of the implementation methods of the present application should be included in the scope of protection of the implementation methods of the present application.

Claims

1. A description file generation method, characterized in that: The method comprises: Obtaining a related profile of a target object and a subject range of user-generated content to be generated by the target object; Based on the association profile and the subject scope, determining a plurality of knowledge points in the knowledge reserve map of the target object; Obtaining responses from the target subject to questions related to each of the knowledge points to obtain a plurality of response messages, wherein the plurality of response messages correspond one-to-one to the plurality of knowledge points; A description file is generated based on the multiple knowledge points and the multiple reply information.

2. The method according to claim 1, characterized in that The generating of the description file based on the plurality of knowledge points and the plurality of reply information includes: Grouping the multiple knowledge points to obtain multiple knowledge groups, wherein, if a first relationship exists between a first knowledge point and a second knowledge point in the knowledge reserve map, fusing the first knowledge point and the reply information corresponding to the first knowledge point to obtain a third knowledge point corresponding to the first knowledge point, fusing the second knowledge point and the reply information corresponding to the second knowledge point to obtain a fourth knowledge point corresponding to the second knowledge point, and grouping the third knowledge point and the fourth knowledge point into the same knowledge group, where the first knowledge point and the second knowledge point are two different knowledge points among the multiple knowledge points; Arrange the knowledge points in each of the knowledge groups according to the arrangement order corresponding to the first relationship to obtain a knowledge sequence group corresponding to each of the knowledge groups, wherein the knowledge sequence group corresponding to each of the knowledge groups includes at least one knowledge sequence; Performing multiple splicing processes based on the multiple knowledge sequence groups to obtain multiple first knowledge sequences, wherein in each splicing process, one knowledge sequence is selected from each knowledge sequence group for splicing to obtain a first knowledge sequence, and any two first knowledge sequences in the multiple first knowledge sequences are different; The description file is generated based on the plurality of first knowledge sequences.

3. The method according to claim 2, characterized in that The generating the description file based on the plurality of first knowledge sequences comprises: determining the distance between every two adjacent knowledge points in each of the first knowledge sequences to obtain a plurality of first distances corresponding to each of the first knowledge sequences; determining a knowledge distance of each of the first knowledge sequences based on a plurality of first distances corresponding to each of the first knowledge sequences; The first knowledge sequence with the smallest knowledge distance is used as the description file.

4. The method according to any one of claims 1 to 3, characterized in that The determining of a plurality of knowledge points in the knowledge reserve map of the target object based on the association portrait and the subject scope includes: Performing feature extraction on the associated portrait to obtain a first feature vector; Performing feature extraction on the subject range to obtain a second feature vector; Determining a first similarity between a feature vector of each topic in the knowledge reserve map and the first feature vector, and a second similarity between the feature vector of each topic and the second feature vector; Determining a first topic in the knowledge reserve map based on the first similarity and the second similarity corresponding to each topic; Based on the first topic and the knowledge reserve map, the multiple knowledge points are determined.

5. The method according to claim 4, characterized in that The determining of the plurality of knowledge points based on the first topic and the knowledge reserve map includes: Acquire a plurality of fifth knowledge points in the knowledge reserve map that are connected to the first topic; determining a first score representing the target subject's familiarity with each fifth knowledge point; Acquire the fifth knowledge points whose first scores are greater than or equal to the first threshold among the multiple fifth knowledge points to obtain the multiple knowledge points.

6. The method according to claim 5, characterized in that The determining of a first score representing the target subject's familiarity with each fifth knowledge point includes: Acquire a plurality of sixth knowledge points in the knowledge reserve map that have a connection relationship with each of the fifth knowledge points; Determining a second score for each fifth knowledge point and a second score for each sixth knowledge point; The first score of each fifth knowledge point is determined based on the second score of each fifth knowledge point and the second scores of the plurality of sixth knowledge points.

7. The method according to claim 6, characterized in that Determining the second score of each fifth knowledge point includes: Determining the number of occurrences of each fifth knowledge point in the historical browsing data of the target object; determining a third score based on the number of occurrences; In a self-assessment form recording the target subject's score of familiarity with each knowledge point in the knowledge reserve map, determining a score for each fifth knowledge point to obtain a fourth score; A second score for each fifth knowledge point is determined based on the third score and the fourth score.

8. The method according to claim 4, characterized in that The determining of the plurality of knowledge points based on the first topic and the knowledge reserve map includes: Determining the similarity between the feature vector of each knowledge point in the knowledge reserve map and the feature vector of the first topic to obtain a plurality of third similarities; Acquire knowledge points in the knowledge reserve graph whose third similarity is greater than or equal to the second threshold, to obtain a plurality of seventh knowledge points; determining a score representing the target subject's familiarity with each seventh knowledge point to obtain a plurality of fifth scores; The seventh knowledge points whose fifth scores are greater than or equal to the second threshold among the multiple seventh knowledge points are taken as the multiple knowledge points.

9. A description file generating device, characterized in that: The device comprises: An acquisition module, configured to acquire a related portrait of a target object and a subject range of user-generated content to be generated by the target object; a determination module, configured to determine a plurality of knowledge points in the knowledge reserve map of the target object based on the association portrait and the subject scope; The acquisition module is further configured to acquire the target subject's responses to the questions related to each of the knowledge points, thereby obtaining a plurality of response messages, wherein the plurality of response messages correspond one-to-one to the plurality of knowledge points; A generation module is used to generate a description file based on the multiple knowledge points and the multiple reply information.

10. An electronic device comprising a processor, a memory, and a computer program or instruction stored in the memory, characterized in that: The processor executes the computer program or instructions to implement the steps of the method according to any one of claims 1 to 8.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program or instructions, and when the computer program or instructions are executed, the steps of the method according to any one of claims 1 to 8 are performed.

Images