Text display method and device, electronic equipment and computer readable medium

Abstract

The embodiment of the invention discloses a text display method and device, electronic equipment and a computer readable medium. A specific embodiment of the method comprises the steps of obtaining an input topic and a material file; performing semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information; performing information summarization on the first semantic split information and the second semantic split information to generate semantic split summarized information; determining a topic index content set; generating topic popularity information; obtaining a topic evaluation standard file; generating a topic flow effect text; performing text content verification on the topic flow effect text to generate a verification result; and carrying out multi-modal display on the topic flow effect text. The embodiment can be suitable for different topic flow platforms, and the topic flow text corresponding to the input topic is accurately and efficiently generated under the condition that the current computing resources are fully utilized, so that the flow effect corresponding to the input topic is effectively displayed in a multi-mode manner.

Description

Technical Field

[0001] Embodiments of the present disclosure relate to the field of computer technology, and in particular to a text display method, device, electronic device, and computer-readable medium. Background Art

[0002] At present, with the continuous development of society, discussions on various types of topics are becoming more and more extensive. How to predict the topic circulation effect for the generated topics has become one of the development directions in various fields. For the prediction of topic circulation effect, the usual method is: for the target topic circulation platform, a targeted topic circulation effect prediction model (for example, a sentiment analysis model with a predetermined network structure) is trained to achieve the prediction of topic circulation effect.

[0003] However, when using the above method to predict the topic flow effect, the following technical problems often occur:

[0004] In cross-platform and cross-cultural scenarios, the use of fixed algorithm models has the problem of poor adaptability in predicting the effect of topic flow, resulting in inaccurate prediction of topic flow effects. In addition, when using neural networks to predict the effect of topic flow, there is often unreasonable scheduling of computing resources, resulting in large delays and freezes in the generation of topic flow effects.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the inventive concept and therefore it may contain information that does not form the prior art that is already known in this country to a person of ordinary skill in the art. Summary of the invention

[0006] The content of this disclosure is used to introduce concepts in a brief form, which will be described in detail in the detailed implementation section below. The content of this disclosure is not intended to identify the key features or essential features of the technical solution claimed for protection, nor is it intended to limit the scope of the technical solution claimed for protection.

[0007] Some embodiments of the present disclosure propose text display methods, devices, electronic devices, and computer-readable media to solve one or more of the technical problems mentioned in the above background technology section.

[0008] In a first aspect, some embodiments of the present disclosure provide a text display method, including: obtaining an input topic input by a target user on a target input page and a material file corresponding to the input topic, wherein the material file includes: a material file in image form, a material file in video form, and a material file in text form; performing semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information; aggregating the first semantic splitting information and the second semantic splitting information to generate semantic splitting summary information; according to the semantic splitting summary information, using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor, determining the topic indicator content under a predetermined topic parameter indicator set to obtain a topic indicator content set, wherein the topic indicator content set It is a content set for each topic circulation platform; based on the above topic indicator content set, the topic popularity information for the above input topic is generated; a topic review standard file is obtained, wherein the above topic review standard file includes: at least one standard information corresponding to at least one topic review direction; based on the above topic popularity information and the above topic review standard file, a pre-trained large language model is used to generate a topic circulation effect text in a target format corresponding to the above input topic, wherein the model type corresponding to the above large language model is the model type selected by the above target user in the topic circulation effect text generation page; the text content of the above topic circulation effect text is verified to generate a verification result; in response to determining that the above verification result indicates that the text content in the topic text is correct, the above topic circulation effect text is multimodally displayed on the target terminal.

[0009] In a second aspect, some embodiments of the present disclosure provide a text display device, including: a first acquisition unit, configured to acquire an input topic input by a target user on a target input page and a material file corresponding to the input topic, wherein the material file includes: a material file in image form, a material file in video form, and a material file in text form; a semantic splitting unit, configured to perform semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information; an information aggregation unit, configured to aggregate the first semantic splitting information and the second semantic splitting information to generate semantic splitting aggregation information; a determination unit, configured to determine the topic indicator content under a predetermined topic parameter indicator set based on the semantic splitting aggregation information and using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor to obtain a topic indicator content set, wherein the topic indicator content set is for each A content set of a topic circulation platform; a first generation unit, configured to generate topic popularity information for the above-mentioned input topic according to the above-mentioned topic indicator content set; a second acquisition unit, configured to acquire a topic review standard file, wherein the above-mentioned topic review standard file includes: at least one standard information corresponding to at least one topic review direction; the second generation unit, configured to generate a topic circulation effect text in a target format corresponding to the above-mentioned input topic according to the above-mentioned topic popularity information and the above-mentioned topic review standard file, using a pre-trained large language model, wherein the model type corresponding to the above-mentioned large language model is the model type selected by the above-mentioned target user in the topic circulation effect text generation page; a verification unit, configured to perform text content verification on the above-mentioned topic circulation effect text to generate a verification result; a display unit, configured to perform multimodal display of the above-mentioned topic circulation effect text on a target terminal in response to determining that the above-mentioned verification result indicates that the text content in the topic text is correct.

[0010] In a third aspect, some embodiments of the present disclosure provide an electronic device comprising: one or more processors; a storage device on which one or more programs are stored, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in any implementation manner in the first aspect.

[0011] In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium having a computer program stored thereon, wherein when the program is executed by a processor, the method described in any implementation manner in the first aspect is implemented.

[0012] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the text display method of some embodiments of the present disclosure, it is applicable to different topic circulation platforms, and the topic circulation text corresponding to the input topic is accurately and efficiently generated under the condition of making full use of the current computing resources, so as to effectively display the corresponding circulation effect of the input topic in a multi-modal manner. Specifically, the reason why the corresponding circulation effect of the relevant input topic is not accurate enough and the computing resource allocation is unreasonable is that in the cross-platform and cross-cultural scenarios, the fixed algorithm model is used, and there is a problem of poor adaptability in predicting the topic circulation effect, resulting in the prediction of the topic circulation effect is not accurate enough. In addition, based on the use of neural networks, the prediction of the topic circulation effect often has the situation of unreasonable scheduling of computing resources, resulting in large delays and freezes in the generation of the topic circulation effect. Based on this, the text display method of some embodiments of the present disclosure first obtains the input topic input by the target user on the target input page and the material file corresponding to the above input topic, so as to facilitate the subsequent prediction and display of the topic circulation effect for the input topic. Among them, the above material files include: material files in the form of images, material files in the form of videos, and material files in the form of text. In addition, by inputting on the target input page, it is convenient to upload the topic-related information of the target user. Then, the above-mentioned input topic and the above-mentioned material file are semantically split to obtain the first semantic splitting information and the second semantic splitting information, so as to facilitate the semantic extraction of the input topic and the material file at the semantic granularity. At the same time, it is also convenient to determine the index content set corresponding to the topic parameter index set in the subsequent determination. Secondly, according to the above-mentioned semantic splitting summary information, the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor can be used to accurately determine the topic index content under the predetermined topic parameter index set to obtain the topic index content set. Among them, the above-mentioned topic index content set is a content set for each topic circulation platform. The obtained topic index content set is used to generate topic popularity information in the subsequent generation. Moreover, by considering the amount of available computing resources corresponding to the central processing unit and the amount of available computing resources corresponding to the image processor, the corresponding computing resources can be prepared in advance for the generation of the topic index content set, so as to avoid the generation of the topic circulation effect with large delays and jams.

[0013] Then, according to the above-mentioned topic indicator content set, the topic popularity information for the above-mentioned input topic can be accurately generated. Here, by accurately generating the topic popularity information, the corresponding circulation heat of the input topic can be characterized. Then, the topic review standard file is obtained, so as to facilitate the subsequent evaluation of the topic circulation for at least one topic review direction, so as to obtain a more accurate topic circulation effect text in the future. Among them, the above-mentioned topic review standard file includes: at least one standard information corresponding to at least one topic review direction. Furthermore, according to the above-mentioned topic popularity information and the above-mentioned topic review standard file, the pre-trained large language model can be used to accurately generate the topic circulation effect text in the target format corresponding to the above-mentioned input topic, wherein the model type corresponding to the above-mentioned large language model is the model type selected by the above-mentioned target user in the topic circulation effect text generation page. The topic circulation effect text generated here can fully express the circulation effect of the input topic under each topic review direction, making the circulation effect of the input topic more explicit. Finally, in response to determining that the text content in the topic text represented by the above-mentioned verification result is correct, the above-mentioned topic circulation effect text is displayed on the target terminal. In summary, by pre-setting a topic parameter indicator set, the popularity prediction of an input topic can be accurately realized, and by using at least one standard information corresponding to at least one topic comment direction, a topic flow effect text with more explicit topic flow can be accurately generated. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The above and other features, advantages and aspects of the embodiments of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings. Throughout the accompanying drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that components and elements are not necessarily drawn to scale.

[0015] Figure 1 is a flow chart of some embodiments of the text display method according to the present disclosure;

[0016] Figure 2 is a schematic structural diagram of some embodiments of the text display device according to the present disclosure;

[0017] Figure 3 It is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. DETAILED DESCRIPTION

[0018] Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although certain embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be construed as being limited to the embodiments set forth herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes and are not intended to limit the scope of protection of the present disclosure.

[0019] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0020] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0021] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0022] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only used for illustrative purposes and are not used to limit the scope of these messages or information.

[0023] The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0024] refer to Figure 1 , shows a process 100 of some embodiments of the text display method according to the present disclosure. The text display method comprises the following steps:

[0025] Step 101: obtaining an input topic input by a target user on a target input page and a material file corresponding to the input topic.

[0026] In some embodiments, the execution subject (for example, an electronic device) of the above-mentioned text display method can obtain the input topic input by the target user on the target input page and the material file corresponding to the above-mentioned input topic through a wired connection method or a wireless connection method. Among them, the target user can be a user who initiates the topic. The input topic can be a topic initiated by the user on the corresponding topic circulation platform. The topic circulation platform can be a platform that supports topic discussion at a certain scale. For example, the topic circulation platform can be a news event discussion platform, or it can be a public opinion video discussion platform. For example, the input topic can be "Is the weather in Beijing, Tianjin and Hebei too hot today?". The material file can be a file related to the topic content between the input topic. The material file can be a file in various modal forms. For example, each modal form can include: text mode, image mode, video mode and audio mode. For example, the material file can be a support file of the topic viewpoint corresponding to the input topic, or it can be a flowchart of the topic content corresponding to the input topic, or it can be an entertaining picture of the topic content corresponding to the input topic. The specific file content of the specific material file will not be repeated one by one. Among them, the target input page can be a page that supports the target user to input topics and material files. In practice, the target input page supports multimodal input. For example, if the input topic is in audio format, the audio input to the target input page can be supported. The audio is then converted to text via the target input page.

[0027] Step 102: semantically split the input topic and the material file to obtain first semantic split information and second semantic split information.

[0028] In some embodiments, the execution subject may perform semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information. The first semantic splitting information may be information after the input topic is subjected to topic semantic splitting. The material file may be information after the input topic is subjected to topic semantic splitting. In practice, the first semantic splitting information may include: name information, place name information, organization information, entity relationship information, event type information, event time information, emotion information, and semantic role information. Similarly, the second semantic splitting information includes the same type of information as the first semantic splitting information.

[0029] Step 103: Aggregate the first semantic split information and the second semantic split information to generate semantic split summary information.

[0030] In some embodiments, the execution subject may aggregate the first semantic split information and the second semantic split information to generate semantic split aggregate information, wherein the semantic split aggregate information may be information after the semantic information is aggregated.

[0031] As an example, the execution entity may fuse the first semantic split information and the second semantic split information under the same information type to generate each fused information under each information type as the semantic split summary information.

[0032] Step 104, based on the semantic splitting summary information, using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor, determine the topic indicator content under the predetermined topic parameter indicator set to obtain a topic indicator content set.

[0033] In some embodiments, the execution subject may determine the topic indicator content under the predetermined topic parameter indicator set based on the semantic splitting summary information, using the available computing resources corresponding to the central processing unit (CPU) and / or the available computing resources corresponding to the graphics processing unit (GPU), to obtain a topic indicator content set. The topic indicator content set is a content set for each topic circulation platform. The predetermined topic parameter indicator may be a pre-set parameter indicator related to topic circulation. The topic indicator content may be the indicator content corresponding to the predetermined topic parameter indicator in the current topic scenario. The platforms in each topic circulation platform are different. The available computing resources may be available online computing resources.

[0034] In some optional implementations of some embodiments, the above-mentioned determining the topic indicator content under the predetermined topic parameter indicator set based on the above-mentioned semantic splitting summary information using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor to obtain the topic indicator content set may include the following steps:

[0035] In the first step, based on the amount of information corresponding to the semantic split summary information and the amount of calculation corresponding to each predetermined topic parameter indicator in the predetermined topic parameter indicator set, the estimated calculation resources corresponding to the topic indicator content set are generated. The estimated calculation resources can represent the calculation resources required to generate the topic indicator content set. The estimated calculation resources can be determined based on the total amount of calculation corresponding to the estimated topic parameter indicator set.

[0036] In the second step, in response to determining that the available computing resources corresponding to the above-mentioned central processing unit are greater than the above-mentioned estimated computing resources, based on the above-mentioned semantic splitting summary information, the available computing resources corresponding to the central processing unit are used to determine the topic indicator content under the predetermined topic parameter indicator set to obtain the topic indicator content set.

[0037] In the third step, in response to determining that the available computing resources corresponding to the above-mentioned central processing unit are less than the above-mentioned estimated computing resources and the available computing resources corresponding to the target image processor are greater than the above-mentioned estimated computing amount, according to the above-mentioned semantic splitting summary information, using the available computing resources corresponding to the above-mentioned target image processor, determine the topic indicator content under the predetermined topic parameter indicator set to obtain the topic indicator content set.

[0038] In the fourth step, in response to determining that the available computing resources corresponding to the above-mentioned central processing unit are less than the above-mentioned estimated computing resources and the available computing resources corresponding to any image processor are less than the above-mentioned estimated computing amount, based on the above-mentioned semantic splitting summary information, utilizing the available computing resources corresponding to each distributed image processor, determine the topic indicator content under the predetermined topic parameter indicator set to obtain the topic indicator content set.

[0039] In some optional implementations of some embodiments, the semantic split summary information includes: behavior noun information, event noun information, person noun information, organization noun information, media noun information and place name noun information. The predetermined topic parameter index set includes: topic promotion average slope index, topic subject word-of-mouth index, first topic flow index, topic subject popularity index, second topic flow index, estimated topic flow index, communication meta-rating index, topic subject attitude index. The topic promotion average slope index can characterize the trend of topic flow change. The larger the index value corresponding to the topic promotion average slope index, the larger the trend of corresponding topic flow change. In practice, topic flow can be topic propagation. The topic subject word-of-mouth index can characterize the word-of-mouth situation of the topic issuing subject. The larger the index value corresponding to the topic subject word-of-mouth index, the better the word-of-mouth situation of the topic issuing subject. In practice, the word-of-mouth situation of the topic subject can be determined by obtaining the published comments and reply information corresponding to the topic subject. The first topic flow index can characterize the median of the topic flow in the topic propagation environment. The topic propagation environment can be various environments for topic propagation. For example, various platforms for topic propagation. The topic subject popularity index can represent the popularity of the topic issuing subject. The larger the corresponding indicator value of the topic subject popularity index, the higher the popularity of the topic issuing subject. The second topic flow index can represent the total topic flow in the topic propagation environment. That is, the second topic flow index can be the total flow of each topic in the topic propagation environment. The estimated topic flow index can represent the total flow of the evaluated topic in each time period. The evaluated topic can be a topic predicted and evaluated by the topic flow effect. Each time period can be a pre-set time period. The specific setting method will not be repeated. The communication meta-scoring index can represent the scoring of the communication meta-vocabulary. The topic subject attitude index can represent the public's attitude towards the topic issuing subject. The higher the corresponding indicator value of the topic subject attitude index, the better the public's attitude towards the topic issuing object.

[0040] In practice, the communication meta-vocabulary can be the various vocabularies after classification and grading. The communication meta-score can be determined based on the category and level corresponding to the communication meta-vocabulary. Each communication meta-vocabulary has a corresponding vocabulary category. Specifically, the LLM (Large Language Model) model can be used to implement the classification and grading of vocabulary. For example, the communication meta-vocabulary can be the vocabulary of the corresponding level under category 1, or the vocabulary of the corresponding level under category 2, and so on. The specific categories include but are not limited to category 1 and category 2. Specifically, category 1 can be a category related to "survival and safety risks". There are 5 levels of vocabulary classification under category 1. The first level under category 1 can be various vocabulary related to "culture and philosophy field". For example, various vocabulary related to "culture and philosophy field" can include but are not limited to at least one of the following: traditional customs, etiquette, festivals, spiritual sustenance, philosophy of life, cultural integration, philosophical enlightenment, moral exploration. The second level under category 1 can be various vocabulary related to "social and environmental background". For example, various vocabulary related to "social and environmental background" can include but are not limited to at least one of the following: community support, family, friendship, environmental adaptation, climate change, social stability, environmental protection, ecological restoration. The third level under category 1 may be various words related to "survival strategies and risk management". For example, various words related to "survival strategies and risk management" may include but are not limited to at least one of the following: resource acquisition, wealth, income, risk management, self-defense, emergency preparedness, investment success, safety measures, job hunting, career development, workplace skills. The fourth level under category 1 may be various words related to "survival-related items and behaviors". For example, various words related to "survival-related items and behaviors" may include but are not limited to at least one of the following: food, water, shelter, natural disasters, war, accidents, poisons, chemicals, difficulties, rescue, escape, and escape. The fifth level under category 1 may be various words related to "survival and safety risks". For example, various words related to "survival and safety risks" may include but are not limited to at least one of the following: survival, safety, health, death, danger, fatality, extinction, extinction, survival, and recovery. Specifically, category 2 may be a category related to "reproduction". There are 5 levels of vocabulary classification under category 2. The first level under category 2 may be various words related to "human evolution and environmental factors". For example, the words related to "human evolution and environmental factors" may include but are not limited to at least one of the following: ecosystem, environmental protection, climate change, ecological balance, sustainable development. The second level under Category 2 may be the words related to "reproductive rights and social and cultural influences". For example, the words related to "reproductive rights and social and cultural influences" may include but are not limited to at least one of the following: reproductive rights, stars, idols, second child, social role. The third level under Category 2 may be the words related to "reproductive health and rights".For example, the various words related to "reproductive health and rights" may include but are not limited to at least one of the following: reproductive health, maternal and child health, reproductive rights, fertility policy, reproductive technology. The fourth level under Category 2 may be the various words related to "reproductive behavior". For example, the various words related to "reproductive behavior" may include but are not limited to at least one of the following: pregnancy, childbirth, reproduction, marriage, personal relationship, family harmony, successful children's education, family. The fifth level under Category 2 may be the various words related to "sexual behavior". For example, the various words related to "sexual behavior" may include but are not limited to at least one of the following: long legs, genetics, sexual behavior, muscle, charm.

[0041] Here, the communication meta-score can be a value obtained by multiplying the corresponding level of the communication meta-word by the corresponding category coefficient of the category to which it belongs. That is, each category has a corresponding category coefficient. For example, the category coefficient corresponding to category 1 is the value "1". The category coefficient corresponding to category 2 is the value "1.2".

[0042] Optionally, the above-mentioned semantic splitting and summarizing information is used to determine the topic indicator content under the predetermined topic parameter indicator set using the available computing resources corresponding to the central processing unit, and obtain the topic indicator content set, including:

[0043] Using the available computing resources corresponding to the CPU, the following determination steps are performed:

[0044] The first step is to determine the average slope of topic improvement corresponding to the above-mentioned average slope indicator of topic improvement.

[0045] The second step is to determine the topic subject reputation value corresponding to the above topic subject reputation indicator based on the above semantic splitting and summary information.

[0046] As an example, first, the execution subject can use the APIs of various social platforms and search platforms or network worms to crawl the topic subject in the semantic split summary information, and obtain the comments and replies to the communication subject (i.e., the target user) under the corresponding topic. Then, use LLM to evaluate the semantic color of each sentence, distinguish between positive and negative sentences, and subtract the number of negative sentences from the number of positive sentences to obtain the word-of-mouth value.

[0047] The third step is to determine the median of the topic flow corresponding to the first topic flow indicator, wherein the first topic flow indicator corresponds to each topic at the current time.

[0048] As an example, first, use the APIs of various social platforms and search platforms or crawl with web worms to obtain the traffic trends of the topics on the list. Then, record the topic traffic values ​​every hour to obtain a traffic value sequence. Then, select the median from the traffic value sequence as the topic traffic median.

[0049] The fourth step is to determine the topic subject popularity corresponding to the above topic subject popularity index based on the above semantic splitting and summary information.

[0050] As an example, first, the number of likes and fans of the articles published by the subject of the topic subject in the semantic split summary information is captured by using the API of each social platform and search platform or a network worm. Then, based on the number of likes and fans, the popularity association table is used to determine the topic subject popularity corresponding to the above topic subject popularity index.

[0051] The fifth step is to determine the total amount of topic traffic corresponding to the second topic traffic indicator. The second topic traffic indicator corresponds to each topic at the current time. The topics may be topics on the popularity list of each platform.

[0052] As an example, the total traffic value of the topics on the list is obtained by using the APIs of various social platforms and search platforms or crawling with web worms, and the total traffic value of the topics is obtained by summing them up.

[0053] Step 6: According to the semantic splitting summary information, determine the total estimated topic traffic volume set corresponding to the estimated topic traffic index, wherein the estimated topic traffic index corresponds to the input topic.

[0054] As an example, the topic information in the semantic split summary information is captured using APIs or web worms of various social platforms and search platforms, the total value of the evaluated traffic is obtained, and the sum is calculated to obtain the total traffic of the evaluated topic.

[0055] The seventh step is to determine the communication meta-score corresponding to the above communication meta-score indicators based on the above semantic split summary information.

[0056] As an example, the execution subject may determine a classification and grading table corresponding to the communication meta-vocabulary. Then, determine the vocabulary associated with the communication meta-vocabulary in the semantic split summary information to obtain at least one vocabulary. Next, find out at least one score corresponding to the at least one vocabulary from the classification and grading table. Finally, add up the scores in the at least one score to generate a communication meta-score.

[0057] The eighth step is to determine the topic subject attitude value corresponding to the above topic subject attitude indicator.

[0058] As an example, first, the APIs or web worms of various social platforms and search platforms are used to crawl the comment content set of articles published by the subject of the topic subject in the semantic split summary information. Then, sentiment semantic analysis is performed on each comment content in the comment content set to generate a sentiment semantic analysis information set. Based on the above sentiment semantic analysis information set, the attitude tendency value of the above topic subject is determined as the topic subject attitude value.

[0059] In the ninth step, the average slope of topic improvement, the reputation value of the topic subject, the median of the topic traffic, the popularity of the topic subject, the total amount of topic traffic, the estimated total amount of topic traffic, the communication meta-score and the attitude value of the topic subject are determined as the topic indicator content set.

[0060] Optionally, the determining, according to the semantic splitting summary information, the topic improvement average slope corresponding to the topic improvement average slope indicator may include the following steps:

[0061] The first step is to determine the maximum value of topic traffic, the starting value of topic traffic, the ending value of topic traffic, the first time point corresponding to the maximum value of topic traffic, the second time point corresponding to the starting value of topic traffic, and the third time point corresponding to the ending value of topic traffic based on the above semantic splitting summary information. The maximum value of topic traffic may be the maximum value of topic traffic of each topic on the list. The starting value of topic traffic may be the starting value of topic traffic of each topic on the list. The ending value of topic traffic may be the extinction point value of topic traffic of each topic on the list. The first time point may be the time point when the maximum value of topic traffic appears. The second time point may be the time point when the starting value of topic traffic appears. The third time point may be the time point when the ending value of topic traffic appears.

[0062] The second step is to determine the time difference between the first time point and the second time point as the first time difference.

[0063] The third step is to subtract the starting value of the topic traffic from the maximum value of the topic traffic to obtain a second subtraction value.

[0064] The fourth step is to divide the second subtraction value by the first time difference to generate a third division value.

[0065] The fifth step is to subtract the above topic traffic end value from the above topic traffic maximum value to obtain a third subtraction value.

[0066] Step 6: Determine the time difference between the first time point and the third time point as the second time difference.

[0067] The seventh step is to divide the third subtraction value by the second time difference to obtain a fourth division value.

[0068] In the eighth step, the third division value is added to the fourth division value to obtain a third added value.

[0069] In the ninth step, the third added value is divided by the third predetermined value to generate a fifth divided value as the topic improvement average slope.

[0070] Optionally, the above-mentioned determination of the topic subject attitude value corresponding to the above-mentioned topic subject attitude indicator may include the following steps:

[0071] The first step is to obtain a topic reply information set corresponding to the input topic, wherein the topic reply information may be information replied to the input topic.

[0072] The second step is to determine the image set corresponding to the topic reply information set, wherein the image set may be each image that appears in the topic reply information set.

[0073] In the third step, for each image in the above image set, perform the following first generation step:

[0074] Sub-step 1: determining the image occurrence frequency corresponding to the above image, wherein the image occurrence frequency may be the number of times the image appears.

[0075] Sub-step 2: determining at least one topic reply text message corresponding to the above image, wherein the at least one topic reply text message may be each reply text replying together with the corresponding image.

[0076] Sub-step 3: determining the text emotion information corresponding to each topic reply text information in the at least one topic reply text information, and obtaining at least one text emotion information.

[0077] As an example, the execution entity may use a sentiment analysis model to determine text sentiment information corresponding to each topic reply text message in the at least one topic reply text message, and obtain at least one text sentiment information.

[0078] Sub-step 4: setting image emotion tendency information for the above image based on the above at least one text emotion information.

[0079] As an example, first, the execution subject may obtain the number of information on each emotional tendency in at least one text emotional information, and then select the emotional tendency with the highest number of corresponding information on each emotional tendency as the image emotional tendency information.

[0080] The fourth step is to determine the topic reply text information set corresponding to the topic reply information set, wherein the topic reply information in the topic reply information set corresponds one to one with the topic reply text information in the topic reply text information set.

[0081] Step 5: For each topic reply information in the above topic reply information set, execute the following second generation step:

[0082] Sub-step 1: determine the topic reply text information corresponding to the above topic reply information as the target topic reply text information.

[0083] Sub-step 2: determining the text sentiment information corresponding to the above target topic reply text information as the target text sentiment information.

[0084] Sub-step 3: performing sentiment vocabulary extraction processing on the above target topic reply text information to generate at least one sentiment word.

[0085] Sub-step 4, determining at least one word sentiment tendency information corresponding to the at least one sentiment word. Among them, there is a one-to-one correspondence between the sentiment word in the at least one sentiment word and the word sentiment tendency information in the at least one word sentiment tendency information. The word sentiment tendency information can characterize the sentiment tendency of the vocabulary. For example, the word sentiment tendency information can be but not limited to at least one of the following: negative tendency information, positive tendency information, neutral tendency information. Each type of word sentiment tendency information can be set with a corresponding tendency value. For example, the tendency value corresponding to the negative tendency information can be the value "2". The tendency value corresponding to the positive tendency information can be the value "1". The tendency value corresponding to the neutral tendency information can be the value "0.5". Sub-step 5, determining the image sentiment tendency information corresponding to the above-mentioned target topic reply information as the target image sentiment tendency information.

[0086] Sub-step 6, generating emotional tendency summary information corresponding to the above-mentioned topic reply information based on the above-mentioned target text emotional information, the above-mentioned at least one word emotional information and the above-mentioned target image emotional tendency information.

[0087] The sixth step is to generate the above topic subject attitude value based on the obtained sentiment tendency summary information set.

[0088] As an example, first, each sentiment tendency summary information in the sentiment tendency summary information set is digitized to generate a sentiment value to obtain a sentiment value set. Then, each sentiment value in the sentiment value set is added to generate a topic subject attitude value.

[0089] The above "optionally" content is one of the inventive points of the present disclosure, which solves the technical problem mentioned in the background technology that "the generated topic subject attitude value is not accurate enough". Based on this, the present disclosure can accurately generate the topic subject attitude value by performing multi-modal precise analysis through the image content and text content corresponding to the topic reply.

[0090] Step 105: Generate topic popularity information for the input topic based on the topic indicator content set.

[0091] In some embodiments, the execution entity may generate topic popularity information for the input topic based on the topic indicator content set. The topic popularity information may represent the popularity (i.e., the degree of circulation) of the input topic. In practice, the topic popularity information may be the topic heat corresponding to the input topic. The topic popularity information may be information in numerical form. The larger the value, the higher the topic heat representing the input topic.

[0092] Optionally, the generating of topic popularity information for the input topic according to the topic indicator content set may include the following steps:

[0093] In the first step, the estimated total topic traffic amounts in the above-mentioned estimated total topic traffic amount set are added together to obtain a first added value.

[0094] In the second step, the median of the topic traffic is subtracted from the first added value to generate a first subtracted value.

[0095] The third step is to multiply the first subtraction value by the average slope of topic improvement to generate a first multiplication value.

[0096] The fourth step is to perform a negation process on the first multiplication value to generate a negated multiplication value.

[0097] The fifth step is to use the multiplied value after negation as the exponent and the first predetermined value as the base to generate a first exponent value.

[0098] The sixth step is to add the first index value to the second predetermined value to obtain a second added value.

[0099] Step 7: Divide the total topic traffic by the second added value to obtain the first divided value.

[0100] In the eighth step, the communication meta-score is divided by the third predetermined value to obtain a second division value.

[0101] In the ninth step, the second division value is used as an exponent and the first predetermined value is used as a base to generate a second exponent value.

[0102] The tenth step is to multiply the second division value by the second exponent value to obtain a second multiplication value.

[0103] The eleventh step is to multiply the topic subject popularity by the topic subject attitude value to obtain a third multiplication value.

[0104] The twelfth step is to perform absolute value processing on the reputation value of the above topic subject to generate the absolute value of the reputation.

[0105] Step 13: multiply the third multiplication value by the absolute value of the word-of-mouth to generate topic popularity information.

[0106] Step 106, obtaining the topic review standard file.

[0107] In some embodiments, the execution entity may obtain a topic review standard file. The topic review standard file includes: at least one standard information corresponding to at least one topic review direction. The at least one topic review direction may be a pre-set topic review direction. The topic review direction may be a review angle for reviewing a topic. There is a one-to-one correspondence between the standard information in at least one standard information and the topic review direction in at least one topic review direction. The standard information may be a judgment rule for the topic review direction.

[0108] Step 107, based on the topic popularity information and the topic review standard file, using a pre-trained large language model, generate a topic circulation effect text in a target format corresponding to the input topic.

[0109] In some embodiments, the execution subject may generate a topic circulation effect text in a target format corresponding to the input topic using a pre-trained large language model based on the topic popularity information and the topic comment standard file. In practice, the large language model may be an LLM model. The model type corresponding to the large language model is selected by the target user in the topic circulation effect text generation page. That is, in practice, the large language model may select the model structure in the topic circulation effect text generation page.

[0110] In some optional implementations of some embodiments, the above-mentioned topic review standard files include: topic popularity review standard source file, target user review standard source file, heat review standard source file, word-of-mouth assignment review standard source file, word-of-mouth evaluation review standard source file, and place name heat review standard source file. The topic popularity review standard source file may be a source file for standard review of topic popularity. The target user review standard source file may be a source file for standard review of user-related situations. The heat review standard source file may be a source file for standard review of topic heat. The word-of-mouth assignment review standard source file may be a standard source file for how to assign word-of-mouth. The place name heat review standard source file may be a source file for standard review of place name heat.

[0111] Optionally, the step of generating the topic circulation effect text in the target format corresponding to the input topic using the pre-trained large language model according to the topic popularity information and the topic review standard file may include the following steps:

[0112] The first step is to combine the source file contents of the above-mentioned topic popularity evaluation standard source file, the above-mentioned target user evaluation standard source file, the above-mentioned heat evaluation standard source file, the above-mentioned word-of-mouth assignment evaluation standard source file, the above-mentioned word-of-mouth evaluation standard source file and the above-mentioned place name heat evaluation standard source file to generate a combined source file.

[0113] In the second step, the combined source file and the topic popularity information are input into the large language model to generate the topic circulation effect text.

[0114] Step 108, verify the text content of the above topic circulation effect text to generate a verification result.

[0115] In some embodiments, the execution entity may perform text content verification on the topic circulation effect text to generate a verification result.

[0116] In some optional implementations of some embodiments, the text content verification of the topic circulation effect text to generate a verification result may include the following steps:

[0117] In the first step, for each topic review direction in the at least one topic review direction, the following third generation step is performed:

[0118] Sub-step 1: using at least one pre-trained sentiment analysis neural network model, determine the first topic review content corresponding to the above topic review direction.

[0119] Sub-step 2: determining the second topic review content in the topic flow effect text corresponding to the topic review direction.

[0120] Sub-step 3: determining the content difference between the first topic review content and the second topic review content.

[0121] The second step is to generate the verification result according to the obtained at least one content difference.

[0122] Step 109, in response to determining that the verification result indicates that the text content in the topic text is correct, the topic circulation effect text is multimodally displayed on the target terminal.

[0123] In some embodiments, in response to determining that the verification result indicates that the text content in the topic text is correct, the execution subject may perform multimodal display of the topic flow effect text on the target terminal. The multimodal display may include: text mode display and image mode display. In the case of image mode display, the topic flow effect text needs to be converted into an image.

[0124] The above-mentioned embodiments of the present disclosure have the following beneficial effects: through the text display method of some embodiments of the present disclosure, it is applicable to different topic circulation platforms, and the topic circulation text corresponding to the input topic is accurately and efficiently generated under the condition of making full use of the current computing resources, so as to effectively display the corresponding circulation effect of the input topic in a multi-modal manner. Specifically, the reason why the corresponding circulation effect of the relevant input topic is not accurate enough and the computing resource allocation is unreasonable is that in the cross-platform and cross-cultural scenarios, the fixed algorithm model is used, and there is a problem of poor adaptability in predicting the topic circulation effect, resulting in the prediction of the topic circulation effect is not accurate enough. In addition, based on the use of neural networks, the prediction of the topic circulation effect often has the situation of unreasonable scheduling of computing resources, resulting in large delays and freezes in the generation of the topic circulation effect. Based on this, the text display method of some embodiments of the present disclosure first obtains the input topic input by the target user on the target input page and the material file corresponding to the above input topic, so as to facilitate the subsequent prediction and display of the topic circulation effect for the input topic. Among them, the above material files include: material files in the form of images, material files in the form of videos, and material files in the form of text. In addition, by inputting on the target input page, it is convenient to upload the topic-related information of the target user. Then, the above-mentioned input topic and the above-mentioned material file are semantically split to obtain the first semantic splitting information and the second semantic splitting information, so as to facilitate the semantic extraction of the input topic and the material file at the semantic granularity. At the same time, it is also convenient to determine the index content set corresponding to the topic parameter index set in the subsequent determination. Secondly, according to the above-mentioned semantic splitting summary information, the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor can be used to accurately determine the topic index content under the predetermined topic parameter index set to obtain the topic index content set. Among them, the above-mentioned topic index content set is a content set for each topic circulation platform. The obtained topic index content set is used to generate topic popularity information in the subsequent generation. Moreover, by considering the amount of available computing resources corresponding to the central processing unit and the amount of available computing resources corresponding to the image processor, the corresponding computing resources can be prepared in advance for the generation of the topic index content set, so as to avoid the generation of the topic circulation effect with large delays and jams.

[0125] Then, according to the above-mentioned topic indicator content set, the topic popularity information for the above-mentioned input topic can be accurately generated. Here, by accurately generating the topic popularity information, the corresponding circulation heat of the input topic can be characterized. Then, the topic review standard file is obtained, so as to facilitate the subsequent evaluation of the topic circulation for at least one topic review direction, so as to obtain a more accurate topic circulation effect text in the future. Among them, the above-mentioned topic review standard file includes: at least one standard information corresponding to at least one topic review direction. Furthermore, according to the above-mentioned topic popularity information and the above-mentioned topic review standard file, the pre-trained large language model can be used to accurately generate the topic circulation effect text in the target format corresponding to the above-mentioned input topic, wherein the model type corresponding to the above-mentioned large language model is the model type selected by the above-mentioned target user in the topic circulation effect text generation page. The topic circulation effect text generated here can fully express the circulation effect of the input topic under each topic review direction, making the circulation effect of the input topic more explicit. Finally, in response to determining that the text content in the topic text represented by the above-mentioned verification result is correct, the above-mentioned topic circulation effect text is displayed on the target terminal. In summary, by pre-setting a topic parameter indicator set, the popularity prediction of an input topic can be accurately realized, and by using at least one standard information corresponding to at least one topic comment direction, a topic flow effect text with more explicit topic flow can be accurately generated.

[0126] Further references Figure 2 As an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of a text display device, and these device embodiments are Figure 1 Corresponding to the method embodiments shown, the text display device can be specifically applied to various electronic devices.

[0127] like Figure 2As shown, a text display device 200 includes: a first acquisition unit 201, a semantic splitting unit 202, an information aggregation unit 203, a determination unit 204, a first generation unit 205, a second acquisition unit 206, a second generation unit 207, a verification unit 208 and a display unit 209. Among them, the first acquisition unit 201 is configured to obtain the input topic input by the target user on the target input page and the material file corresponding to the above input topic, wherein the above material file includes: material file in image form, material file in video form and material file in text form; the semantic splitting unit 202 is configured to perform semantic splitting on the above input topic and the above material file to obtain first semantic splitting information and second semantic splitting information; the information aggregation unit 203 is configured to aggregate the above first semantic splitting information and the above second semantic splitting information to generate semantic splitting aggregation information; the determination unit 204 is configured to determine the topic indicator content under the predetermined topic parameter indicator set according to the above semantic splitting aggregation information, using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor, and obtain a topic indicator content set, wherein the above topic indicator content set is a content set for each topic circulation platform; the first generation unit 204 is configured to perform semantic splitting on the above input topic and the above material file to obtain first semantic splitting information and second semantic splitting information; the information aggregation unit 203 is configured to aggregate the above first semantic splitting information and the above second semantic splitting information to generate semantic splitting aggregation information; the determination unit 204 is configured to determine the topic indicator content under the predetermined topic parameter indicator set according to the above semantic splitting aggregation information, using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor, and obtain a topic indicator content set, wherein the above topic indicator content set is a content set for each topic circulation platform; The first generation unit 205 is configured to generate topic popularity information for the input topic according to the topic indicator content set; the second acquisition unit 206 is configured to acquire a topic review standard file, wherein the topic review standard file includes: at least one standard information corresponding to at least one topic review direction; the second generation unit 207 is configured to generate a topic circulation effect text in a target format corresponding to the input topic according to the topic popularity information and the topic review standard file using a pre-trained large language model, wherein the large language model corresponds to a model type selected by the target user in the topic circulation effect text generation page; the verification unit 208 is configured to perform text content verification on the topic circulation effect text to generate a verification result; the display unit 209 is configured to perform multimodal display on the target terminal in response to determining that the verification result indicates that the text content in the topic text is correct.

[0128] It is understandable that the units recorded in the text display device 200 and the reference Figure 1 Therefore, the operations, features and beneficial effects described above for the method are also applicable to the text display device 200 and the units included therein, and will not be described in detail here.

[0129] Reference below Figure 3 , which shows a structural schematic diagram of an electronic device (eg, an electronic device) 300 suitable for implementing some embodiments of the present disclosure. Figure 3The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0130] like Figure 3 As shown, the electronic device 300 may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0131] Typically, the following devices may be connected to the I / O interface 305: input devices 306 including, for example, a touch screen, a touch pad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 307 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 308 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 309. The communication devices 309 may allow the electronic device 300 to communicate with other devices wirelessly or by wire to exchange data. Although Figure 3 The electronic device 300 is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown. More or fewer devices may be implemented or possessed instead. Figure 3 Each block shown in the figure may represent one device, or may represent multiple devices as required.

[0132] In particular, according to some embodiments of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, some embodiments of the present disclosure include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In some such embodiments, the computer program can be downloaded and installed from the network through the communication device 309, or installed from the storage device 308, or installed from the ROM 302. When the computer program is executed by the processing device 301, the above-mentioned functions defined in the method of some embodiments of the present disclosure are executed.

[0133] It should be noted that the computer-readable medium in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In some embodiments of the present disclosure, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In some embodiments of the present disclosure, the computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, in which a computer-readable program code is carried. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer readable signal medium may also be any computer readable medium other than a computer readable storage medium, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer readable medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0134] In some embodiments, the client and the server may communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), an internet (e.g., the Internet), and a peer-to-peer network (e.g., an ad hoc peer-to-peer network), as well as any currently known or future developed network.

[0135] The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or it may exist independently without being installed in the electronic device. The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the electronic device, the electronic device: obtains the input topic input by the target user on the target input page and the material file corresponding to the above-mentioned input topic, wherein the above-mentioned material file includes: material file in image form, material file in video form and material file in text form; performs semantic splitting on the above-mentioned input topic and the above-mentioned material file to obtain first semantic splitting information and second semantic splitting information; summarizes the above-mentioned first semantic splitting information and the above-mentioned second semantic splitting information to generate semantic splitting summary information; according to the above-mentioned semantic splitting summary information, using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor, determines the topic indicator content under the predetermined topic parameter indicator set to obtain a topic indicator content set, wherein The above-mentioned topic indicator content set is a content set for each topic circulation platform; based on the above-mentioned topic indicator content set, generate the topic popularity information for the above-mentioned input topic; obtain the topic review standard file, wherein the above-mentioned topic review standard file includes: at least one standard information corresponding to at least one topic review direction; based on the above-mentioned topic popularity information and the above-mentioned topic review standard file, use the pre-trained large language model to generate the topic circulation effect text in the target format corresponding to the above-mentioned input topic, wherein the model type corresponding to the above-mentioned large language model is the model type selected by the above-mentioned target user in the topic circulation effect text generation page; perform text content verification on the above-mentioned topic circulation effect text to generate a verification result; in response to determining that the above-mentioned verification result represents that the text content in the topic text is correct, the above-mentioned topic circulation effect text is multimodally displayed on the target terminal.

[0136] Computer program code for performing the operations of some embodiments of the present disclosure may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0137] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some implementations as replacements, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0138] The units described in some embodiments of the present disclosure may be implemented by software or by hardware. The described units may also be provided in a processor, for example, may be described as: a processor including a first acquisition unit, a semantic splitting unit, an information aggregation unit, a determination unit, a first generation unit, a second acquisition unit, a second generation unit, a verification unit and a display unit. Among them, the names of these units do not constitute a limitation on the unit itself in certain cases, for example, the first acquisition unit may also be described as "a unit for acquiring input topics corresponding to the target user and material files corresponding to the above-mentioned input topics".

[0139] The functions described above herein may be performed at least in part by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0140] The above descriptions are only some preferred embodiments of the present disclosure and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solutions formed by a specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features are replaced with the technical features with similar functions disclosed in the embodiments of the present disclosure (but not limited to) and the technical solutions formed.

Claims

1. A text display method, comprising: Acquire an input topic input by a target user on a target input page and a material file corresponding to the input topic, wherein the material file includes: a material file in the form of an image, a material file in the form of a video, and a material file in the form of a text; Performing semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information; Aggregating the first semantic splitting information and the second semantic splitting information to generate semantic splitting summary information; According to the semantic split summary information, using available computing resources corresponding to the central processing unit and / or available computing resources corresponding to the image processor, determining the topic index content under the predetermined topic parameter index set, and obtaining a topic index content set, wherein the topic index content set is a content set for each topic circulation platform; Generating topic popularity information for the input topic according to the topic indicator content set; Acquire a topic review standard file, wherein the topic review standard file includes: at least one standard information corresponding to at least one topic review direction; According to the topic popularity information and the topic comment standard file, using a pre-trained large language model, generating a topic circulation effect text in a target format corresponding to the input topic, wherein the large language model corresponds to a model type that the target user clicks and selects in a topic circulation effect text generation page; Performing text content verification on the topic circulation effect text to generate a verification result; In response to determining that the verification result indicates that the text content in the topic text is correct, the topic circulation effect text is multimodally displayed on the target terminal.

2. The method according to claim 1, wherein: The method of determining the topic indicator content under a predetermined topic parameter indicator set by using the available computing resources corresponding to the central processing unit and / or the available computing resources corresponding to the image processor according to the semantic splitting summary information to obtain a topic indicator content set includes: Generate an estimated computing resource corresponding to the topic indicator content set according to the information amount corresponding to the semantic split summary information and the computing amount corresponding to each predetermined topic parameter indicator in the predetermined topic parameter indicator set; In response to determining that the available computing resources corresponding to the central processing unit are greater than the estimated computing resources, determining the topic indicator content under the predetermined topic parameter indicator set using the available computing resources corresponding to the central processing unit according to the semantic split summary information, and obtaining a topic indicator content set; In response to determining that the available computing resources corresponding to the central processing unit are less than the estimated computing resources and the available computing resources corresponding to the target image processor are greater than the estimated computing amount, according to the semantic splitting summary information, using the available computing resources corresponding to the target image processor, determining the topic indicator content under the predetermined topic parameter indicator set, and obtaining a topic indicator content set; In response to determining that the available computing resources corresponding to the central processing unit are less than the estimated computing resources and the available computing resources corresponding to any image processor are less than the estimated computing amount, based on the semantic splitting summary information and utilizing the available computing resources corresponding to each distributed image processor, the topic indicator content under the predetermined topic parameter indicator set is determined to obtain a topic indicator content set.

3. The method according to claim 2, wherein: The semantic split summary information includes: behavior noun information, event noun information, person noun information, organization noun information, media noun information and place name noun information, and the predetermined topic parameter index set includes: topic promotion average slope index, topic subject word-of-mouth index, first topic flow index, topic subject popularity index, second topic flow index, estimated topic flow index, communication meta-score index, topic subject attitude index; and The method of determining the topic index content under the predetermined topic parameter index set by using the available computing resources corresponding to the central processing unit according to the semantic splitting summary information to obtain the topic index content set includes: Using the available computing resources corresponding to the CPU, the following determination steps are performed: Determine the topic improvement average slope corresponding to the topic improvement average slope indicator; Determine the topic subject word-of-mouth value corresponding to the topic subject word-of-mouth indicator according to the semantic split summary information; Determine a median of topic traffic corresponding to the first topic traffic indicator, wherein the first topic traffic indicator corresponds to each topic at the current time; Determine the topic subject popularity corresponding to the topic subject popularity index according to the semantic split summary information; Determine the total amount of topic traffic corresponding to the second topic traffic indicator, wherein the second topic traffic indicator corresponds to each topic at the current time; Determine, according to the semantic split summary information, a total set of estimated topic traffic corresponding to the estimated topic traffic indicator, wherein the estimated topic traffic indicator corresponds to the input topic; Determining a communication meta-score corresponding to the communication meta-score indicator according to the semantic split summary information; Determine a topic subject attitude value corresponding to the topic subject attitude indicator; The average slope of topic improvement, the topic subject word-of-mouth value, the median of topic traffic, the topic subject popularity, the total topic traffic, the estimated total topic traffic, the communication meta-score and the topic subject attitude value are determined as the topic indicator content set.

4. The method according to claim 3, wherein: The step of generating topic popularity information for the input topic according to the topic indicator content set includes: Adding the estimated topic traffic totals of each of the estimated topic traffic totals to obtain a first added value; Subtracting the median of the topic traffic from the first added value to generate a first subtracted value; Multiplying the first subtraction value by the average slope of topic promotion to generate a first multiplication value; Performing a negation process on the first multiplication value to generate a negated multiplication value; Using the multiplied value after negation as an exponent and the first predetermined value as a base to generate a first exponent value; Adding the first index value to a second predetermined value to obtain a second added value; Divide the total topic traffic by the second added value to obtain a first divided value; Dividing the communication element score by a third predetermined value to obtain a second division value; Using the second divided value as an exponent and the first predetermined value as a base, generating a second exponent value; multiplying the second divided value by the second exponent value to obtain a second multiplied value; Multiplying the topic subject popularity by the topic subject attitude value to obtain a third multiplied value; Performing absolute value processing on the topic subject word-of-mouth value to generate an absolute word-of-mouth value; The third multiplication value is multiplied by the absolute value of the word-of-mouth to generate topic popularity information.

5. The method according to claim 3, wherein: The determining, according to the semantic split summary information, the topic improvement average slope corresponding to the topic improvement average slope indicator comprises: According to the semantic split summary information, determine the maximum value of topic traffic, the starting value of topic traffic, the ending value of topic traffic, the first time point corresponding to the maximum value of topic traffic, the second time point corresponding to the starting value of topic traffic, and the third time point corresponding to the ending value of topic traffic; determining a time difference between the first time point and the second time point as a first time difference; Subtract the topic flow starting value from the topic flow maximum value to obtain a second subtraction value; dividing the second subtracted value by the first time difference to generate a third divided value; Subtract the topic flow ending value from the topic flow maximum value to obtain a third subtraction value; determining a time difference between the first time point and the third time point as a second time difference; Dividing the third subtraction value by the second time difference to obtain a fourth division value; Adding the third division value and the fourth division value to obtain a third added value; The third added value is divided by a third predetermined value to generate a fifth divided value as the topic improvement average slope.

6. The method according to claim 1, wherein: The topic review standard files include: topic popularity review standard source files, target user review standard source files, popularity review standard source files, word-of-mouth evaluation review standard source files, word-of-mouth evaluation review standard source files, and place name popularity review standard source files; and The method of generating a topic circulation effect text in a target format corresponding to the input topic using a pre-trained large language model according to the topic popularity information and the topic comment standard file includes: Combining the content of the topic popularity evaluation standard source file, the target user evaluation standard source file, the popularity evaluation standard source file, the word-of-mouth value evaluation standard source file, the word-of-mouth evaluation standard source file, and the place name popularity evaluation standard source file to generate a combined source file; The combined source file and the topic popularity information are input into the large language model to generate the topic circulation effect text.

7. The method according to claim 1, wherein: The text content verification of the topic circulation effect text to generate a verification result includes: For each topic review direction in the at least one topic review direction, the following third generation step is performed: Determine the first topic review content corresponding to the topic review direction by using at least one pre-trained sentiment analysis neural network model; Determining a second topic review content in the topic flow effect text corresponding to the topic review direction; Determining a content difference between the first topic review content and the second topic review content; The verification result is generated according to the obtained at least one content difference.

8. A text display device, comprising: A first acquisition unit is configured to acquire an input topic input by a target user on a target input page and a material file corresponding to the input topic, wherein the material file includes: a material file in the form of an image, a material file in the form of a video, and a material file in the form of a text; A semantic splitting unit is configured to perform semantic splitting on the input topic and the material file to obtain first semantic splitting information and second semantic splitting information; an information summarizing unit, configured to summarize the first semantic splitting information and the second semantic splitting information to generate semantic splitting summary information; A determination unit is configured to determine the topic index content under a predetermined topic parameter index set based on the semantic splitting summary information by using available computing resources corresponding to the central processing unit and / or available computing resources corresponding to the image processor, and obtain a topic index content set, wherein the topic index content set is a content set for each topic circulation platform; A first generating unit is configured to generate topic popularity information for the input topic according to the topic indicator content set; A second acquisition unit is configured to acquire a topic review standard file, wherein the topic review standard file includes: at least one standard information corresponding to at least one topic review direction; The second generating unit is configured to generate a topic circulation effect text in a target format corresponding to the input topic using a pre-trained large language model according to the topic popularity information and the topic comment standard file, wherein the large language model corresponds to a model type selected by the target user in a topic circulation effect text generation page; A verification unit is configured to perform text content verification on the topic circulation effect text to generate a verification result; The display unit is configured to, in response to determining that the verification result represents that the text content in the topic text is correct, display the topic circulation effect text in a multimodal manner on the target terminal.

9. An electronic device, comprising: one or more processors; a storage device having one or more programs stored thereon, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 7.

10. A computer readable medium having a computer program stored thereon, wherein: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images