A method and device for generating article title

Abstract

The present application discloses a method for generating an article title. The method can be executed by a first device, which can be a terminal device or a server. Specifically: the first device can obtain a target text, and then determine a target style, which can be the style of a title generated for the target text. Then, the first device can output a stylized title of the target style corresponding to the target text. The stylized title mentioned here refers to a title with a certain style. The stylized title has a relatively large appeal. It can be seen that, by using this solution, an attractive article title can be determined for the target text, thereby improving the efficiency of determining an attractive article title.

Description

Technical Field

[0001] The present application relates to the field of data processing, and in particular to a method and device for generating article titles. Background Art

[0002] Users can attract the attention of other users by writing and publishing articles on the network platform. For an article, the first thing users look at is the article title. The attractiveness of the article title can affect the number of readers of the article to a certain extent.

[0003] Currently, users need to think a lot to determine an attractive article title, which is inefficient. Therefore, a solution is needed to solve the above problem. Summary of the invention

[0004] The technical problem to be solved by the present application is how to improve the efficiency of generating attractive article titles, and to provide a method and device for generating article titles.

[0005] In a first aspect, an embodiment of the present application provides a method for generating an article title, the method comprising:

[0006] Get the target text;

[0007] Determine the target style;

[0008] Outputting the styled title of the target text corresponding to the target style.

[0009] Optionally, the method further includes:

[0010] The target text is input into a machine learning model to obtain a stylized title of the target style corresponding to the target text. The machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0011] Optionally, determining the target style includes:

[0012] Output at least one title style;

[0013] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0014] Optionally, determining the target style includes:

[0015] The target style is determined according to the reading group of the target text.

[0016] Optionally, determining the target style includes:

[0017] A preset style is determined as the target style.

[0018] Optionally, the method further includes:

[0019] Obtaining a training text and a stylized title of the training text;

[0020] The machine learning model is trained based on the training text and the stylized title of the training text.

[0021] Optionally, the preset style includes one or more of the following:

[0022] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0023] Optionally, the step of inputting the target text into a machine learning model to obtain a stylized title of the target text corresponding to the target style includes:

[0024] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0025] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the method further includes:

[0026] In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

[0027] In a second aspect, an embodiment of the present application provides a device for generating an article title, the device comprising:

[0028] A first acquisition unit, used for acquiring a target text;

[0029] A determination unit, used to determine a target style;

[0030] The first output unit is used to output the stylized title of the target text corresponding to the target style.

[0031] Optionally, the device further comprises:

[0032] A processing unit is used to input the target text into a machine learning model to obtain a stylized title of the target style corresponding to the target text, and the machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0033] Optionally, the determining unit is used to:

[0034] Output at least one title style;

[0035] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0036] Optionally, the determining unit is used to:

[0037] The target style is determined according to the reading group of the target text.

[0038] Optionally, the determining unit is used to:

[0039] A preset style is determined as the target style.

[0040] Optionally, the device further comprises:

[0041] A second acquisition unit, used to acquire a training text and a stylized title of the training text;

[0042] A training unit is used to train the machine learning model based on the training text and the stylized title of the training text.

[0043] Optionally, the preset style includes one or more of the following:

[0044] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0045] Optionally, the processing unit is used to:

[0046] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0047] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the device further includes:

[0048] A generating unit is used to generate an article including the first title and the target text in response to a selection operation triggered by a user on a first title among the multiple titles.

[0049] In a third aspect, an embodiment of the present application provides a device for generating an article title, comprising a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors, wherein the one or more programs include instructions for performing the following operations:

[0050] Get the target text;

[0051] Determine the target style;

[0052] Outputting the styled title of the target text corresponding to the target style.

[0053] Optionally, the operation further includes:

[0054] The target text is input into a machine learning model to obtain a stylized title of the target style corresponding to the target text. The machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0055] Optionally, determining the target style includes:

[0056] Output at least one title style;

[0057] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0058] Optionally, determining the target style includes:

[0059] The target style is determined according to the reading group of the target text.

[0060] Optionally, determining the target style includes:

[0061] A preset style is determined as the target style.

[0062] Optionally, the operation further includes:

[0063] Obtaining a training text and a stylized title of the training text;

[0064] The machine learning model is trained based on the training text and the stylized title of the training text.

[0065] Optionally, the preset style includes one or more of the following:

[0066] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0067] Optionally, the step of inputting the target text into a machine learning model to obtain a stylized title of the target text corresponding to the target style includes:

[0068] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0069] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the operation further includes:

[0070] In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

[0071] In a fourth aspect, an embodiment of the present application provides a computer-readable medium having instructions stored thereon, which, when executed by one or more processors, enables the device to execute any of the methods described in the first aspect above.

[0072] Compared with the prior art, the embodiments of the present application have the following advantages:

[0073] An embodiment of the present application provides a method for generating an article title. The method can be executed by a first device, which can be a terminal device or a server. Specifically: the first device can obtain a target text, and then determine a target style, which can be the style of a title generated for the target text. Then, the first device can output a stylized title of the target style corresponding to the target text. The stylized title mentioned here refers to a title with a certain style. The stylized title has a relatively large appeal. It can be seen that, using this solution, an attractive article title can be determined for the target text, thereby improving the efficiency of determining an attractive article title. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0075] Figure 1 A flowchart of a method for generating an article title provided in an embodiment of the present application;

[0076] Figure 2 A schematic diagram of the structure of a device for generating article titles provided in an embodiment of the present application;

[0077] Figure 3 A schematic diagram of the structure of the client provided in the embodiment of the present application;

[0078] Figure 4 A schematic diagram of the structure of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0079] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0080] The inventor of the present application has found through research that the attractiveness of an article title can affect the number of readers of the article to a certain extent. At present, most article titles are written by users themselves. However, in general, article titles written by users themselves are not very attractive. To write an attractive article title, users need to think a lot, which is inefficient.

[0081] In order to solve the above problems, the embodiments of the present application provide a method and device for generating an article title.

[0082] Various non-limiting implementations of the present application are described in detail below in conjunction with the accompanying drawings.

[0083] Exemplary Methods

[0084] See also Figure 1 , which is a flow chart of a method for generating article titles provided in an embodiment of the present application.

[0085] The method provided in the embodiment of the present application can be executed by a first device, for example. The first device mentioned here includes but is not limited to a terminal device and a server. The terminal device mentioned here can be a mobile terminal such as a smart phone or a tablet computer, or a terminal device such as a desktop computer.

[0086] It should be noted that the solution provided in the embodiment of the present application can be applied to a network platform that provides an article writing function. For example, for a certain web page, a user can write an article on the web page.

[0087] In this embodiment, the method may be implemented, for example, through the following steps S101-S103.

[0088] S101: Obtain target text.

[0089] In an embodiment of the present application, the target text may be input by a user on the Internet. In one example, for a network platform that provides an article writing function, the corresponding web page includes a text input area, and the target text may be text input by a user in the text input area. The text mentioned here includes, but is not limited to, text in any language such as Chinese text, English text, Korean text, and Japanese text. In another example, the target text may be obtained by recognizing an image including text, for example, the image may be recognized using optical character recognition (OCR) technology to obtain the target text.

[0090] S102: Determine the target style.

[0091] In the embodiment of the present application, there are multiple ways to determine the target style in specific implementation. Three possible ways are introduced below.

[0092] In one example, the first device may output at least one title style. The user may select one or more styles from the at least one title style. In an embodiment of the present application, if the user triggers a selection operation for one or more of the title styles, the title style selected by the user is determined as the target style. For example, the first device outputs a data style, a substitution style, an exclamation style, a question style, a suspense style, a contrast style, a pain point style, a label style, a scenario style, and a hot style, and the user triggers a selection operation for the question style, then the question style is determined as the target style. If the user triggers a selection operation for the question style and the pain point style, the question style and the pain point style are determined as the target style. For data style, substitution style, exclamation style, question style, suspense style, contrast style, pain point style, label style, scenario style, and hot style, please refer to the relevant description section below, which will not be described in detail here.

[0093] In another example, considering that different articles correspond to different reading groups, and different reading groups are accustomed to different titles. Therefore, in order to increase the reading volume of the target text, the target style can be determined according to the reading group of the target text. For example, the reading group of the target text is female, and females prefer immersive article titles. Therefore, the target style can be determined to be an immersive style. Among them:

[0094] The target text's reading group can be determined based on the network platform that provides the article writing function, and the present application embodiment does not make specific limitations. For example, for a romance novel writing platform, its reading group is mostly female. Therefore, if the target text is published on a romance novel writing platform, the target text's reading group can be female. The target text's reading group can also be input by the user, and the present application embodiment does not make specific limitations.

[0095] In another example, a preset style can be determined as the target style, wherein the preset style can include one or more of the aforementioned data style, substitution style, exclamation style, question style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0096] S103: Outputting a stylized title of the target text corresponding to the target style.

[0097] In the embodiment of the present application, after executing S101 and S102, it can be determined that the target text corresponds to the stylized title of the target style, and further, the stylized title of the target text corresponding to the target style can be output.

[0098] In one example, the target text can be input into a machine learning model to obtain a stylized title of the target text corresponding to the target style, and the machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style. It can be understood that the title of the preset style is relatively attractive.

[0099] Regarding the preset style, in some embodiments, an attractive title style can be determined in advance through research, and the attractive title style can be determined as the preset style.

[0100] The embodiments of the present application do not specifically limit the preset style. In some embodiments, considering that the data style, substitution style, exclamation style, question style, suspense style, contrast style, pain point style, label style, scenario style and hot style are all relatively attractive, the preset style can be one or more of the above styles. Next, the data style, substitution style, exclamation style, question style, suspense style, contrast style, pain point style, label style, scenario style and hot style are introduced respectively.

[0101] Data-style refers to the inclusion of data in the article title, where the data mentioned here can be numbers, for example. The title of a data-style article can be, for example: "After interviewing 3,000 college students, we summarized 7 ways to chase goddesses"; another data-style article title can be, for example: "When you are not feeling well, your feet will have 4 changes."

[0102] The immersive style refers to the use of first-person or second-person descriptions in the title of an article, which creates a sense of immersion for the reader. For example, the immersive style title of an article can be: "You look like a failure for not getting an offer"; another immersive style title can be: "The day I received the offer, I posted a circle of friends and received 29 blessing calls"; another immersive style title can be: "The strongest May Day program in history, have you ever seen a tiger kept as a pet?"

[0103] The exclamatory style refers to the use of an exclamatory tone in the title of an article. For example, an exclamatory style article title could be: "What a pity! Another domestic mobile phone giant has completely collapsed!"; another exclamatory style article title could be: "The ugly makeup in these five TV dramas absolutely amazed me!".

[0104] The questioning style refers to the title of the article being described in a questioning tone. The questioning tone can include interrogative and rhetorical questions. For example, the title of an interrogative style article can be: "How can fresh graduates adjust their mindset and find a job they like?"; the title of a rhetorical style article can be: "Can't you be more tolerant of your child when he is young? No!".

[0105] Suspense style refers to the title of the article leaving suspense. For example, the title of a suspense style article can be: "I am shocked! There is such a magical profession in the world."; Another title of a suspense style article can be: "Why do we sometimes forget what we want to say? I never expected that the culprit was actually..."

[0106] The contrast style refers to the presence of contrast in the article title. Among them, the contrast mentioned here can include at least three situations. One of the situations is numerical contrast, including the title of the contrast style article of numerical contrast, for example, it can be: "The difference between a monthly salary of 3,000 and a monthly salary of 30,000". Another situation is contradictory contrast, including the title of the contrast style article of contradictory contrast, for example, it can be: "Annual salary of 600,000, but not a piece of clothing worth more than 100 yuan: saving money is the highest level of self-discipline". Another situation is contrary to common sense, including the title of the contrast style article that goes against common sense, for example, it can be: "The son who "would rather raise a dog than his parents" has won the support of everyone".

[0107] Pain point style refers to the existence of problem pain points in the article title. For example, the title of a pain point style article can be: "90% of startups die at this step. How to get the first batch of customers in the startup stage?"; another pain point style article title can be: "Children cry all day long. How should parents manage their emotions?"

[0108] The tag style refers to the inclusion of tags in the article title. The tags mentioned here can be used to identify groups, places, or things. For example, the title of an article with a tag style can be: "What advantages do Taurus people have that other constellations don't have?"; another example of a title of an article with a tag style can be "Do "working people" not deserve a sweet weekend?".

[0109] The scene-based style refers to the inclusion of specific scenes in the article title. For example, the scene-based style article title can be: "A 1994 girl with a monthly salary of 4,600 yuan saved up for two houses: How terrible is it to be addicted to saving money?"; Another scene-based style article title can be: "Bai Baihe turned out to have such a good figure. The moment she bent down and pulled her skirt, everyone in the audience stared at her."

[0110] The hot-selling style refers to the inclusion of hot content in the article title. The hot content mentioned here includes but is not limited to: hot events, hot topics or celebrities. The title of an article in the hot-selling style can be, for example: "28-year-old Li fell ill and cried during an interview: Adult life is much sadder than you think"; the title of an article in the hot-selling style can be, for example: "Even Huo has begun to "learn" from Wu, so why don't you learn these legal common sense in love quickly?" In this embodiment of the application, the machine learning model can be pre-trained. In other words, before executing S101-S103 above, the machine learning model can also be pre-trained. The embodiment of the application does not specifically limit the machine learning model. The machine learning model can be, for example, a convolutional neural network (CNN) model, or a deep neural network (DNN) model. The embodiment of the application does not make specific restrictions. Regarding the training part of the machine learning model, please refer to the relevant description part below, which will not be described in detail here.

[0111] As described above, the solution provided by the embodiment of the present application can be applied to a network platform that provides an article writing function. For this case, in one example, after the user edits the target text on the web page, a title generation operation can be triggered on the web page. For example, the web page may include a "title generation" control, and the user triggers the title generation operation by clicking on the area where the control is located. After the user triggers the title generation operation, the first device can input the target text into a machine learning model, thereby obtaining a stylized title corresponding to the target style of the target text.

[0112] After the first device obtains the stylized title of the target style corresponding to the target text, it can output the stylized title of the target style corresponding to the target text. The output of the stylized title of the target style corresponding to the target text mentioned here can be, for example, displaying the stylized title of the target text corresponding to the target style on the web page for editing the target text. In one example, the web page can include a title recommendation area, which is used to display the stylized title recommended for the target text. The stylized title of the target style corresponding to the target text can be displayed in the title recommendation area.

[0113] It is understandable that after the target text is input into the machine learning model, the stylized title output by the machine learning model may include one or more titles, that is, outputting titles of one or more styles among the above data style, substitution style, exclamation style, question style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0114] It is understandable that:

[0115] If the target style is a preset style, the multiple titles output by the machine learning model are stylized titles corresponding to each of the multiple title styles included in the target style.

[0116] If the target style is a subset of the preset style, for example, the target style is selected by the user, or the target style is determined according to the target text's reading group. In addition to outputting the stylized title of the target text corresponding to the target style, the first device also outputs stylized titles of other styles corresponding to the target text. In this case, the stylized title of the target text corresponding to the target style can be displayed preferentially. For example, the multiple titles are displayed in the form of a list, and the stylized title of the target text corresponding to the target style is located at the first item in the list.

[0117] In an embodiment of the present application, if the machine learning model outputs multiple titles. Then after the first device outputs the multiple titles, the user can select one of the multiple titles as the final article title to be used. In this case, if the user selects the first title among the multiple titles, the first device can generate an article including the first title and the target text in response to the selection operation triggered by the user for the first title. Wherein: generating the article including the first title and the target text may be, for example, automatically inputting the first title in the title input area of ​​the web page. In an embodiment of the present application, the user triggers a selection operation for the first title, for example, the user may click or double-click the first title displayed on the web page.

[0118] The method for generating article titles provided in the embodiments of the present application is now described by way of example:

[0119] After the user completes editing the target text in the text input area on the web page, he clicks the "title generation" control on the web page, and the first device obtains the stylized title of the target text according to the target text and the machine learning model, and displays the stylized title of the target text in the title recommendation area on the web page. The stylized title may include the following titles, for example:

[0120] 1. The new version of "The Deer and the Cauldron" scored 2.7 points. Zhang's exaggerated acting successfully dissuaded the crowd from watching the show.

[0121] 2. The new version of "The Deer and the Cauldron" scored 2.7! Zhang's exaggerated acting successfully dissuaded the audience from watching the film.

[0122] 3. The new version of "The Deer and the Cauldron" scored 2.7 points. Zhang Moumou was ridiculed for playing Wei Xiaobao?

[0123] 4. The new version of "The Deer and the Cauldron" scored 2.7! Zhang played the "crazy monkey", which is more realistic than the real eunuch?

[0124] 5. What do you think of Zhang's version of The Deer and the Cauldron? Douban's score hits a record low

[0125] The first title mentioned above, for example, may be a stylized title corresponding to the target style.

[0126] When the user clicks on the first title "The new version of "The Deer and the Cauldron" scores 2.7! Zhang XX plays the "crazy monkey", which is more realistic than the real eunuch?", the title input area of ​​the web page is automatically filled with the first title "The new version of "The Deer and the Cauldron" scores 2.7! Zhang XX plays the "crazy monkey", which is more realistic than the real eunuch?".

[0127] From the above description, it can be seen that by using this solution, the user does not need to go through a lot of thinking to determine the title of the target text. The machine learning model can be used to determine an attractive article title for the target text, thereby improving the efficiency of determining an attractive article title.

[0128] Next, the method of training the machine learning model is introduced.

[0129] In an embodiment of the present application, the machine learning model can be trained through the following steps AB.

[0130] Step A: Get the training text and the stylized title of the training text.

[0131] Step B: training the machine learning model based on the training text and the stylized title of the training text.

[0132] In the embodiment of the present application, the training text may be, for example, an article that has been published on various network platforms. Therefore, the training text may be obtained from various network platforms.

[0133] In some examples, the stylized title of the training text may be the title of the training text that has been published. In this case, the training text and the stylized title of the training text may be obtained from a network platform.

[0134] In other examples, considering that the title of the training text published on the network platform may not be a stylized title. For this method, the training text can be manually annotated with a stylized title. In other words, the training title can also be manually annotated, that is, after obtaining the training text, the training title is manually annotated with a stylized title. In this case, the training text can be obtained from the network platform, and the stylized title of the training text input by the user can be received.

[0135] It should be noted that, for a training text, the corresponding stylized title may include one or more titles. In this way, the trained machine learning model can determine one or more styles of titles for the target text.

[0136] Regarding step B, it should be noted that a training text and its corresponding stylized title can be used as a set of training samples. In other words, multiple groups of training samples can be used to train the machine learning model. The process of training a machine learning model is actually a process of determining the model parameters through multiple iterative calculations. When training a machine learning model, the multiple groups of training samples can be divided into a training set and a validation set. The training set is used to train the model parameters, and the validation set is used to verify the accuracy of the model. When the accuracy of the trained model verified by the validation set meets certain conditions (for example, determining that the accuracy of the stylized title is higher than 85%), the iteration can be ended, and the corresponding model at the end of the iteration is used as the machine learning model.

[0137] Exemplary Devices

[0138] Based on the method provided in the above embodiment, the embodiment of the present application further provides a device, which is described below in conjunction with the accompanying drawings.

[0139] See also Figure 2 , which is a schematic diagram of the structure of a device for generating article titles provided in an embodiment of the present application. The device 200 may specifically include: a first acquisition unit 201, a determination unit 202 and a first output unit 203.

[0140] A first acquisition unit 201 is used to acquire a target text;

[0141] A determination unit 202, configured to determine a target style;

[0142] The first output unit 203 is used to output the stylized title of the target text corresponding to the target style.

[0143] Optionally, the device further comprises:

[0144] A processing unit is used to input the target text into a machine learning model to obtain a stylized title of the target style corresponding to the target text, and the machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0145] Optionally, the determining unit 202 is configured to:

[0146] Output at least one title style;

[0147] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0148] Optionally, the determining unit 202 is configured to:

[0149] The target style is determined according to the reading group of the target text.

[0150] Optionally, the determining unit 202 is configured to:

[0151] A preset style is determined as the target style.

[0152] Optionally, the device further comprises:

[0153] A second acquisition unit, used to acquire a training text and a stylized title of the training text;

[0154] A training unit is used to train the machine learning model based on the training text and the stylized title of the training text.

[0155] Optionally, the preset style includes one or more of the following:

[0156] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0157] Optionally, the processing unit is used to:

[0158] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0159] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the device further includes:

[0160] A generating unit is used to generate an article including the first title and the target text in response to a selection operation triggered by a user on a first title among the multiple titles.

[0161] Since the device 200 is a device corresponding to the method provided in the above method embodiment, the specific implementation of each unit of the device 200 is based on the same concept as the above method embodiment. Therefore, for the specific implementation of each unit of the device 200, reference can be made to the description part of the above method embodiment, which will not be repeated here.

[0162] The method provided in the embodiment of the present application can be executed by a client or a server. The client and server that execute the above method are described below respectively.

[0163] Figure 31 shows a block diagram of a client 300. For example, the client 300 may be a mobile phone, a computer, a digital broadcast terminal, a messaging device, a game console, a tablet device, a medical device, a fitness device, a personal digital assistant, etc.

[0164] Reference Figure 3 The client 300 may include one or more of the following components: a processing component 302 , a memory 304 , a power component 306 , a multimedia component 308 , an audio component 310 , an input / output (I / O) interface 33 , a sensor component 314 , and a communication component 316 .

[0165] The processing component 302 generally controls the overall operation of the client 300, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 302 may include one or more processors 320 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 302 may include one or more modules to facilitate the interaction between the processing component 302 and other components. For example, the processing component 302 may include a multimedia module to facilitate the interaction between the multimedia component 308 and the processing component 302.

[0166] The memory 304 is configured to store various types of data to support operations on the client 300. Examples of such data include instructions for any application or method operating on the client 300, contact data, phone book data, messages, pictures, videos, etc. The memory 304 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0167] The power supply component 306 provides power to the various components of the client 300. The power supply component 306 may include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power to the client 300.

[0168] The multimedia component 308 includes a screen that provides an output interface between the client 300 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 308 includes a front camera and / or a rear camera. When the client 300 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and the rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0169] The audio component 310 is configured to output and / or input audio signals. For example, the audio component 310 includes a microphone (MIC), and when the client 300 is in an operation mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 304 or sent via the communication component 316. In some embodiments, the audio component 310 also includes a speaker for outputting audio signals.

[0170] The I / O interface provides an interface between the processing component 302 and the peripheral interface module, which may be a keyboard, a click wheel, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0171] The sensor assembly 314 includes one or more sensors for providing various aspects of status assessment for the client 300. For example, the sensor assembly 314 can detect the open / closed state of the device 300, the relative positioning of components, such as the display and keypad of the client 300, and the sensor assembly 314 can also detect the position change of the client 300 or a component of the client 300, the presence or absence of user contact with the client 300, the orientation or acceleration / deceleration of the client 300, and the temperature change of the client 300. The sensor assembly 314 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 314 may also include an optical sensor, such as a CMOS or CCD image sensor, for use in imaging applications. In some embodiments, the sensor assembly 314 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0172] The communication component 316 is configured to facilitate wired or wireless communication between the client 300 and other devices. The client 300 can access a wireless network based on a communication standard, such as WiFi, 2G or 3G, or a combination thereof. In an exemplary embodiment, the communication component 316 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 316 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0173] In an exemplary embodiment, the client 300 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the following methods:

[0174] Get the target text;

[0175] Determine the target style;

[0176] Outputting the styled title of the target text corresponding to the target style.

[0177] Optionally, the method further includes:

[0178] The target text is input into a machine learning model to obtain a stylized title of the target style corresponding to the target text. The machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0179] Optionally, determining the target style includes:

[0180] Output at least one title style;

[0181] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0182] Optionally, determining the target style includes:

[0183] The target style is determined according to the reading group of the target text.

[0184] Optionally, determining the target style includes:

[0185] A preset style is determined as the target style.

[0186] Optionally, the method further includes:

[0187] Obtaining a training text and a stylized title of the training text;

[0188] The machine learning model is trained based on the training text and the stylized title of the training text.

[0189] Optionally, the preset style includes one or more of the following:

[0190] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0191] Optionally, the step of inputting the target text into a machine learning model to obtain a stylized title of the target text corresponding to the target style includes:

[0192] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0193] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the method further includes:

[0194] In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

[0195] Figure 4 4 is a schematic diagram of the structure of the server in the embodiment of the present application. The server 400 may have relatively large differences due to different configurations or performances, and may include one or more central processing units (CPU) 422 (for example, one or more processors) and memory 432, and one or more storage media 430 (for example, one or more mass storage devices) storing application programs 442 or data 444. Among them, the memory 432 and the storage medium 430 can be short-term storage or permanent storage. The program stored in the storage medium 430 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations on the server. Furthermore, the central processing unit 422 can be configured to communicate with the storage medium 430 to execute a series of instruction operations in the storage medium 430 on the server 400.

[0196] Furthermore, the CPU 422 may execute the following method:

[0197] Get the target text;

[0198] Determine the target style;

[0199] Outputting the styled title of the target text corresponding to the target style.

[0200] Optionally, the method further includes:

[0201] The target text is input into a machine learning model to obtain a stylized title of the target style corresponding to the target text. The machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0202] Optionally, determining the target style includes:

[0203] Output at least one title style;

[0204] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0205] Optionally, determining the target style includes:

[0206] The target style is determined according to the reading group of the target text.

[0207] Optionally, determining the target style includes:

[0208] A preset style is determined as the target style.

[0209] Optionally, the method further includes:

[0210] Obtaining a training text and a stylized title of the training text;

[0211] The machine learning model is trained based on the training text and the stylized title of the training text.

[0212] Optionally, the preset style includes one or more of the following:

[0213] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0214] Optionally, the step of inputting the target text into a machine learning model to obtain a stylized title of the target text corresponding to the target style includes:

[0215] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0216] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the method further includes:

[0217] In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

[0218] The server 400 may also include one or more power supplies 426, one or more wired or wireless network interfaces 450, one or more input and output interfaces 456, one or more keyboards 456, and / or one or more operating systems 441, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM, etc.

[0219] The present application also provides a computer-readable medium having instructions stored thereon, which, when executed by one or more processors, causes the device to perform the following method:

[0220] Get the target text;

[0221] Determine the target style;

[0222] Outputting the styled title of the target text corresponding to the target style.

[0223] Optionally, the method further includes:

[0224] The target text is input into a machine learning model to obtain a stylized title of the target style corresponding to the target text. The machine learning model is used to generate a stylized title for the article based on the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style.

[0225] Optionally, determining the target style includes:

[0226] Output at least one title style;

[0227] In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

[0228] Optionally, determining the target style includes:

[0229] The target style is determined according to the reading group of the target text.

[0230] Optionally, determining the target style includes:

[0231] A preset style is determined as the target style.

[0232] Optionally, the method further includes:

[0233] Obtaining a training text and a stylized title of the training text;

[0234] The machine learning model is trained based on the training text and the stylized title of the training text.

[0235] Optionally, the preset style includes one or more of the following:

[0236] Data style, substitution style, exclamatory style, questioning style, suspense style, contrast style, pain point style, label style, scenario style and hype style.

[0237] Optionally, the step of inputting the target text into a machine learning model to obtain a stylized title of the target text corresponding to the target style includes:

[0238] In response to a title generation operation triggered by a user, the target text is input into a machine learning model to obtain a stylized title of the target text corresponding to the target style.

[0239] Optionally, if multiple titles are obtained after the target text is input into the machine learning model, the method further includes:

[0240] In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

[0241] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0242] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

[0243] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for generating an article title, characterized in that: Applied to a network platform providing an article writing function, the method comprises: Edit the target text on the web page corresponding to the network platform; Determining a target style, including: determining a reading group of the target text according to the network platform; determining the target style according to the reading group of the target text, with different titles corresponding to different reading groups; In response to the triggered title generation operation, the target text is input into the machine learning model to obtain a stylized title of the target style corresponding to the target text, and the machine learning model is used to generate a stylized title for the article according to the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style. The title generation operation is to trigger the title generation operation by clicking on the area where the title generation control is located, and the preset style includes one or more of the following: data style, substitution style, suspense style, contrast style, pain point style, label style, scenario style and hot style; Outputting a stylized title of the target text corresponding to the target style, including: displaying the stylized title of the target style corresponding to the target text on a web page for editing the target text, the web page including a title recommendation area, the title recommendation area being used to display the stylized title recommended for the target text, wherein, if the target style is the preset style, the multiple titles output are stylized titles corresponding to each of the multiple title styles included in the target style, and if the target style is a subset of the preset style, in addition to outputting the stylized title of the target text corresponding to the target style, stylized titles of other styles corresponding to the target text are also output, and the multiple titles are displayed in the form of a list, and the stylized title of the target text corresponding to the target style is located at the first item in the list; In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

2. The method according to claim 1, characterized in that: The determining of the target style further includes: Output at least one title style; In response to a selection operation triggered by a user on one or more of the at least one title style, the title style selected by the user is determined as the target style.

3. The method according to claim 1, characterized in that The determining of the target style further includes: A preset style is determined as the target style.

4. The method according to claim 1, characterized in that: The method further comprises: Obtaining a training text and a stylized title of the training text; The machine learning model is trained based on the training text and the stylized title of the training text.

5. A device for generating article titles, characterized in that: A network platform for providing an article writing function, the device comprising: A first acquisition unit is used to edit the target text on the web page corresponding to the network platform; A determination unit is used to determine a target style, including: determining a reading group of the target text according to the network platform; determining the target style according to the reading group of the target text, and different reading groups correspond to different titles; A processing unit, configured to, in response to a triggered title generation operation, input the target text into a machine learning model to obtain a stylized title of the target style corresponding to the target text, wherein the machine learning model is configured to generate a stylized title for the article according to the article text, wherein the stylized title is a title with a preset style, wherein the preset style includes the target style, and the title generation operation is triggered by clicking on an area where a title generation control is located, wherein the preset style includes one or more of the following: data style, substitution style, suspense style, contrast style, pain point style, label style, scenario style, and hype style; A first output unit is used to output a stylized title of the target text corresponding to the target style, which includes: displaying the stylized title of the target style corresponding to the target text on a web page for editing the target text, the web page including a title recommendation area, the title recommendation area being used to display the stylized title recommended for the target text, wherein if the target style is the preset style, the multiple titles output are stylized titles corresponding to each of the multiple title styles included in the target style, and if the target style is a subset of the preset style, in addition to outputting the stylized title of the target text corresponding to the target style, stylized titles of other styles corresponding to the target text are also output, and the multiple titles are displayed in the form of a list, and the stylized title of the target text corresponding to the target style is located at the first item in the list; A generating unit is used to generate an article including the first title and the target text in response to a selection operation triggered by a user on a first title among the multiple titles.

6. The device according to claim 5, characterized in that The determining unit is used for: Output at least one title style; In response to a selection operation triggered by a user for one or more of the at least one title style, determining the target style based on the title style selected by the user; or, A preset style is determined as the target style.

7. The device according to claim 6, characterized in that The device also includes: A second acquisition unit, used to acquire a training text and a stylized title of the training text; A training unit is used to train the machine learning model based on the training text and the stylized title of the training text.

8. A device for generating article titles, characterized in that: A network platform for providing an article writing function includes a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by one or more processors. The one or more programs include instructions for performing the following operations: Edit the target text on the web page corresponding to the network platform; Determining a target style, including: determining a reading group of the target text according to the network platform; determining the target style according to the reading group of the target text, with different titles corresponding to different reading groups; In response to the triggered title generation operation, the target text is input into the machine learning model to obtain a stylized title of the target style corresponding to the target text, and the machine learning model is used to generate a stylized title for the article according to the article text, wherein the stylized title is a title with a preset style, and the preset style includes the target style. The title generation operation is to trigger the title generation operation by clicking on the area where the title generation control is located, and the preset style includes one or more of the following: data style, substitution style, suspense style, contrast style, pain point style, label style, scenario style and hot style; Outputting a stylized title of the target text corresponding to the target style, including: displaying the stylized title of the target style corresponding to the target text on a web page for editing the target text, the web page including a title recommendation area, the title recommendation area being used to display the stylized title recommended for the target text, wherein, if the target style is the preset style, the multiple titles output are stylized titles corresponding to each of the multiple title styles included in the target style, and if the target style is a subset of the preset style, in addition to outputting the stylized title of the target text corresponding to the target style, stylized titles of other styles corresponding to the target text are also output, and the multiple titles are displayed in the form of a list, and the stylized title of the target text corresponding to the target style is located at the first item in the list; In response to a selection operation triggered by a user on a first title among the multiple titles, an article including the first title and the target text is generated.

9. A computer-readable medium having instructions stored thereon, which, when executed by one or more processors, causes an apparatus to perform the method of any one of claims 1 to 4.

Images