CN110909154A - Abstract generation method and device - Google Patents

Abstract generation method and device Download PDF

Info

Publication number
CN110909154A
CN110909154A CN201911141552.3A CN201911141552A CN110909154A CN 110909154 A CN110909154 A CN 110909154A CN 201911141552 A CN201911141552 A CN 201911141552A CN 110909154 A CN110909154 A CN 110909154A
Authority
CN
China
Prior art keywords
abstract
template
keyword
content
entity category
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911141552.3A
Other languages
Chinese (zh)
Inventor
兰枫
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing ByteDance Network Technology Co Ltd
Original Assignee
Beijing ByteDance Network Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beijing ByteDance Network Technology Co Ltd filed Critical Beijing ByteDance Network Technology Co Ltd
Priority to CN201911141552.3A priority Critical patent/CN110909154A/en
Publication of CN110909154A publication Critical patent/CN110909154A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/34Browsing; Visualisation therefor
    • G06F16/345Summarisation for human users
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33Querying
    • G06F16/332Query formulation
    • G06F16/3329Natural language query formulation or dialogue systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The embodiment of the disclosure provides a method and a device for generating an abstract, wherein the method comprises the following steps: receiving at least one keyword of a first object input by a user; acquiring a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is greater than a first threshold; generating a summary of the first object based on the at least one keyword and the summary template. The embodiment of the disclosure does not need to manually write the abstract text, thereby reducing the burden of the user.

Description

Abstract generation method and device
Technical Field
The embodiment of the disclosure relates to the technical field of internet, in particular to a method and a device for generating an abstract.
Background
With the improvement of the performance of mobile terminals, more and more users use the mobile terminals to perform operations such as information reading and commodity purchase.
In order to quickly display information data such as texts or videos, a short abstract needs to be generated for the information data, so that a user can obtain basic content or key points of the information data through the abstract, and preview of the user is facilitated. At present, the generation mode of the abstract is mainly to write an abstract text by a user himself and then upload the abstract text so as to display the abstract text on a terminal interface. When a user needs to show information frequently, frequent abstract writing causes a large burden on the user.
Disclosure of Invention
The embodiment of the disclosure provides a method and a device for generating an abstract, which are used for solving the problem that the existing scheme generates the abstract and generates a large burden on a user.
In a first aspect, an embodiment of the present disclosure provides a digest generation method, including:
receiving at least one keyword of a first object input by a user;
acquiring a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is greater than a first threshold;
generating a summary of the first object based on the at least one keyword and the summary template.
In one possible implementation, generating the summary of the first object based on the at least one keyword and the summary template includes:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
acquiring entity categories corresponding to each content to be filled in the abstract template;
and generating the abstract of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled.
In a possible implementation manner, generating the summary of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled includes:
determining the corresponding relation between the at least one keyword and the at least one content to be filled according to the entity category corresponding to each keyword and the entity category corresponding to each content to be filled;
and filling the at least one keyword into the content to be filled in the abstract template according to the corresponding relation to obtain the abstract of the first object.
In a possible implementation manner, acquiring the abstract template corresponding to the first object based on the at least one keyword includes:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
determining the object type of the first object according to the entity type corresponding to each keyword;
and acquiring a summary template corresponding to the first object based on the object type of the first object, wherein the object type of the first object is the same as the object type corresponding to the summary template.
In a possible implementation manner, before obtaining the abstract template corresponding to the first object based on the at least one keyword, the method further includes:
acquiring a preset abstract, wherein the click rate of the preset abstract is greater than a first threshold value;
and generating the abstract template based on the preset abstract.
In a possible implementation manner, generating the summary template based on the preset summary includes:
inputting the preset abstract into a named entity recognition model to obtain the entity category of each vocabulary in the preset abstract output by the named entity recognition model;
and generating the abstract template according to the entity category of each vocabulary in the preset abstract.
In a possible implementation manner, generating the abstract template according to the entity category of each vocabulary in the preset abstract includes:
acquiring a first entity category set and a second entity category set;
determining a first vocabulary in the preset abstract as the template content, wherein the entity category of the first vocabulary is located in the first entity category set;
and deleting a second vocabulary in the preset abstract, and determining the content of the position of the second vocabulary as the content to be filled, wherein the entity category of the second vocabulary is located in the second entity category set.
In one possible implementation, the number of generated digests of the first object is greater than 1, the method further comprising:
displaying at least two abstracts of the first object based on the weight values of the abstract templates corresponding to the at least two abstracts;
receiving a target abstract selected by a user from the at least two abstracts;
determining the target summary as a summary of the first object.
In a possible implementation manner, after determining the target digest as the digest of the first object, the method further includes:
and updating the weight values of the abstract templates corresponding to the at least two abstract according to the target abstract.
In a second aspect, an embodiment of the present disclosure provides a digest generation apparatus, including:
the receiving module is used for receiving at least one keyword of a first object input by a user;
the obtaining module is used for obtaining a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is larger than a first threshold value;
and the processing module is used for generating the abstract of the first object based on the at least one keyword and the abstract template.
In a possible implementation manner, the processing module is specifically configured to:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
acquiring entity categories corresponding to each content to be filled in the abstract template;
and generating the abstract of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled.
In a possible implementation manner, the processing module is specifically configured to:
determining the corresponding relation between the at least one keyword and the at least one content to be filled according to the entity category corresponding to each keyword and the entity category corresponding to each content to be filled;
and filling the at least one keyword into the content to be filled in the abstract template according to the corresponding relation to obtain the abstract of the first object.
In a possible implementation manner, the obtaining module is specifically configured to:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
determining the object type of the first object according to the entity type corresponding to each keyword;
and acquiring a summary template corresponding to the first object based on the object type of the first object, wherein the object type of the first object is the same as the object type corresponding to the summary template.
In a possible implementation manner, the obtaining module is further configured to:
acquiring a preset abstract, wherein the click rate of the preset abstract is greater than a first threshold value;
and generating the abstract template based on the preset abstract.
In a possible implementation manner, the obtaining module is further configured to:
inputting the preset abstract into a named entity recognition model to obtain the entity category of each vocabulary in the preset abstract output by the named entity recognition model;
and generating the abstract template according to the entity category of each vocabulary in the preset abstract.
In a possible implementation manner, the obtaining module is further configured to:
acquiring a first entity category set and a second entity category set;
determining a first vocabulary in the preset abstract as the template content, wherein the entity category of the first vocabulary is located in the first entity category set;
and deleting a second vocabulary in the preset abstract, and determining the content of the position of the second vocabulary as the content to be filled, wherein the entity category of the second vocabulary is located in the second entity category set.
In a possible implementation manner, the processing module is further configured to, when the number of the generated digests of the first object is greater than 1:
displaying at least two abstracts of the first object based on the weight values of the abstract templates corresponding to the at least two abstracts;
receiving a target abstract selected by a user from the at least two abstracts;
determining the target summary as a summary of the first object.
In a possible implementation manner, the processing module is further configured to, after determining the target digest as the digest of the first object:
and updating the weight values of the abstract templates corresponding to the at least two abstract according to the target abstract.
In a third aspect, an embodiment of the present disclosure provides an electronic device, including: at least one processor and memory;
the memory stores computer-executable instructions;
the at least one processor executing the memory-stored computer-executable instructions causes the at least one processor to perform the summary generation method of any one of the first aspects.
In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium, where computer-executable instructions are stored, and when a processor executes the computer-executable instructions, the digest generation method according to any one of the first aspect is implemented.
The method and the device for generating the abstract provided by the embodiment of the disclosure firstly receive at least one keyword of a first object input by a user, then obtain an abstract template corresponding to the first object based on the at least one keyword, and finally generate the abstract of the first object based on the at least one keyword and the abstract template. When a user needs to obtain an abstract of a first object, only one or a plurality of key words related to the first object need to be input, the scheme provided by the embodiment of the disclosure can obtain a corresponding abstract template according to the key words input by the user, the abstract template comprises template content and content to be filled, then the key words input by the user are added to the content to be filled of the abstract template to obtain the abstract of the first object, so that the corresponding abstract is generated on the premise of inputting fewer key words without manually writing an abstract text, and the burden of the user is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present disclosure, and for those skilled in the art, other drawings can be obtained according to the drawings without inventive exercise.
Fig. 1 is a schematic view of an application scenario provided in the embodiment of the present disclosure;
fig. 2 is a schematic flowchart of a digest generation method according to an embodiment of the present disclosure;
fig. 3 is a schematic diagram of summary generation provided by the embodiment of the present disclosure;
fig. 4 is a flowchart illustrating a summary generation method according to another embodiment of the disclosure;
fig. 5 is a schematic flowchart of acquiring a summary template according to an embodiment of the present disclosure;
fig. 6 is a schematic diagram of generating a summary template according to an embodiment of the disclosure;
FIG. 7 is a schematic diagram of summary template filling provided by an embodiment of the present disclosure;
FIG. 8 is a keyword filling diagram provided by an embodiment of the present disclosure;
FIG. 9 is a schematic diagram illustrating multi-summary generation provided by an embodiment of the present disclosure;
fig. 10 is a schematic structural diagram of a summary generation apparatus according to an embodiment of the present disclosure;
fig. 11 is a schematic structural diagram of an electronic device according to an embodiment of the present disclosure.
Detailed Description
To make the objects, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions of the embodiments of the present disclosure will be described clearly and completely with reference to the drawings in the embodiments of the present disclosure, and it is obvious that the described embodiments are some, but not all embodiments of the present disclosure. All other embodiments, which can be derived by a person skilled in the art from the embodiments disclosed herein without making any creative effort, shall fall within the protection scope of the present disclosure.
Fig. 1 is a schematic view of an application scenario provided by the embodiment of the present disclosure, as shown in fig. 1, a user wishes to display related information on an information platform. For example, a user may place an advertisement on an information platform, and may publish an article, a video, etc. on the information platform.
Since the user needs to display the relevant information on the information platform, and the relevant information may need to occupy a larger space, a shorter summary is usually generated according to the relevant information displayed by the user and is published on the information platform. When other users browse the information platform, the abstracts of all the information are acquired firstly, and the information which needs to be viewed is selected according to the abstracts of all the information.
Therefore, the abstract is usually the key for deciding whether other users click and check related information, so that a brief and attractive abstract is generated, and is particularly important for publishing and popularizing the related information.
The existing scheme generates the abstract mode, and the user who needs to release the relevant information usually writes the abstract text, and the writing of the abstract text is a large burden for the user. It is to be realized that in the terminal device of the example of fig. 1, keywords such as "shirt", "cool" and "298" related to information desired to be presented are input, and then a summary of the presented information is obtained, wherein the generated summary may have one or more, thereby providing a user with a choice of the most desired summary text, and the example summary generated in fig. 1 is "cool without others |)! Short style printed shirt sold only 298 Yuan! ".
It should be noted that the application scenario shown in fig. 1 is only an example, and the summary generation method provided in the embodiment of the present disclosure may also be applied to many other scenarios, for example, besides being applied to an information platform, a social platform, a search platform, and the like, and the device for executing the operation steps may be the terminal device illustrated in fig. 1, and may also be other possible electronic devices, which are not all listed in the embodiment of the present disclosure.
Hereinafter, the technical means shown in the present disclosure will be described in detail by specific examples. It should be noted that the following embodiments may be combined with each other, and the description of the same or similar contents in different embodiments is not repeated.
Fig. 2 is a schematic flow chart of a digest generation method provided in the embodiment of the present disclosure, and as shown in fig. 2, the method includes:
s21, receiving at least one keyword of the first object input by the user;
the execution subject of the embodiment of the present disclosure is an electronic device, or a text processing apparatus provided in an electronic device. Alternatively, the text processing means may be implemented by software, or by a combination of software and hardware. Optionally, the electronic device may be a computer, a server, a mobile phone, or the like.
The first object in the embodiment of the present disclosure is a related object of information to be displayed, for example, in the advertisement delivery field, if a commodity advertised by an advertisement is a shirt, the first object is a shirt; in the video field, assuming that the video is related to travel promotion, the first object is a travel related object, such as a travel destination, a travel, etc.
When a user needs to show the related information of the first object and generate the abstract, at least one keyword can be input, and the keyword can express key features of the first object or key points which the user wants to embody in the abstract. For example, in the advertisement delivery field, when the first object is a commodity, the keyword of the first object may be a selling point, discount offer, price, and the like of the commodity.
It is understood that the keywords may be words in different languages, in the embodiment of the present disclosure, the words may be in the form of words, may also be in the form of numbers, and may also include characters, and further, the keywords may also be represented in the form of a combination of two or more of the words, the numbers, the characters, and the like. For example, if the user wishes to display the price of a good in the summary, which is $32, a keyword relating to the price may be entered: and $ 32.
S22, acquiring a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is greater than a first threshold value;
and after obtaining at least one keyword, acquiring a summary template corresponding to the first object according to the at least one keyword. There may be multiple abstract templates, and different abstract templates correspond to different abstract. After the keywords of the abstract are determined, the abstract template corresponding to the keywords can be obtained according to the keywords. This process will be described below with reference to fig. 3.
Fig. 3 is a schematic diagram of summary generation provided in the embodiment of the present disclosure, and as shown in fig. 3, a corresponding summary template is generated according to a preset summary. In the embodiment of the disclosure, the preset abstract is an already published abstract, and the click rate of the preset abstract is greater than a first threshold. When the click rate of the already published summary is higher, it indicates that the summary is viewed by many people, so the probability that the summary is a more attractive summary is higher. Therefore, when the preset abstract is obtained, the embodiment of the disclosure only obtains the abstract with the click rate greater than the first threshold. The first threshold is a positive number, and the obtaining of the first threshold may be determined according to actual needs, which is not particularly limited in the embodiments of the present disclosure.
Fig. 3 illustrates two preset abstracts, which are abstracts for clothing type goods, namely a preset abstract one and a preset abstract two, and the contents of the two preset abstracts are as follows:
presetting abstract one, namely 'one piece of windbreaker to make your temperament distinctive'.
The summary I is preset, namely, the first summary, the first choice for men, the classic of the achievement suit and the selling of the new season suit are only 999 yuan.
And after the two preset abstracts are obtained, obtaining two corresponding abstract templates according to the two preset abstracts.
For example, a first abstract template is obtained for the first preset abstract, and a second abstract template is obtained for the second preset abstract, where the two abstract templates are as follows:
abstract template one{ trade name }Let your temperament be distinctive ".
Summary template 1,' first choice and achievement for men{ trade name }Classic and New season{ trade name }Sell only{ commodity price }Yuan ".
The abstract template comprises template contents and at least one content to be filled, wherein the template contents need to be reserved, and the content to be filled needs to be filled according to keywords input by a user. And in the subsequently generated abstract, the retained template content and the content to be filled in by adopting the keywords input by the user are included. For example, in the abstract template one, the template content is "one piece ____, which makes your quality distinctive", and the content to be filled is the name of the product; in the summary template two, the template content is "men's first choice, achievement ____ classic, new season ____ only sells ____ yuan", and the content to be filled includes the commodity name and the commodity price.
The first abstract template and the second abstract template are both abstract templates, but the two templates are different. In the summary template one, the main salient is the name of the commodity, i.e. the subject of the summary, and the content to be filled is also the name of the commodity. In the abstract template two, the name of the commodity and the price of the commodity are required to be filled in. Therefore, when the keywords input by the user include only the commodity name, the first abstract template corresponding to the keywords can be obtained, when the keywords input by the user include the commodity name and the commodity price, the second abstract template corresponding to the two keywords can be obtained, when the keywords input by the user include other types of keywords besides the commodity name and the commodity price, the first abstract template and the second abstract template do not meet the requirements of the user, and other abstract templates meeting the requirements of the user need to be searched according to the keywords input by the user, and the like.
For example, when the keyword input by the user is "sweater", which is one of the categories of clothing, i.e., names of goods, the summary template one illustrated in fig. 3 can be acquired according to the keyword "sweater". When the keywords input by the user are "sweater" and "200", the "200" is a number, which is an indication of the price of the commodity, and the second abstract template illustrated in fig. 3 can be obtained according to the keywords "sweater" and "200". When the keywords input by the user are sweater and goodness, the goodness is description of the commodity and is a selling point of the commodity, and the contents to be filled aiming at the selling point of the commodity are not contained in the abstract template I and the abstract template II, so that the abstract template I and the abstract template II do not meet the requirement of the user at the moment, and other abstract templates are required to be obtained to fill the commodity name and the selling point of the commodity.
S23, generating the abstract of the first object based on the at least one keyword and the abstract template.
After obtaining at least one keyword and a summary template, obtaining a summary of the first object according to the keyword and the summary template. The abstract template is obtained according to the keywords, so that the obtained abstract template corresponds to the keywords, and the keywords can be filled in the positions to be filled in the abstract template according to the corresponding relation between the keywords and the abstract template. For example, when the user inputs the keyword "sweater", the "sweater" is filled in the place to be filled corresponding to the commodity name of the abstract template one, and the abstract at this time is obtained as follows: a piece of wind coat makes your temperament distinctive. When the user inputs the keywords 'sweater' and '200', filling the 'sweater' into the position to be filled corresponding to the commodity name of the abstract template II, and filling the '200' into the position to be filled corresponding to the commodity price of the abstract template II to obtain the abstract at the moment: the man prefers to sell the ready-made sweater classically, and the new-season sweater only sells 200 yuan.
If there are multiple templates for the input keyword, multiple abstracts are generated correspondingly, and the user can select one of the abstracts as the first object according to the preference of the user and display the abstract.
The abstract generating method provided by the embodiment of the disclosure includes the steps of firstly receiving at least one keyword of a first object input by a user, then obtaining an abstract template corresponding to the first object based on the at least one keyword, and finally generating an abstract of the first object based on the at least one keyword and the abstract template. When a user needs to obtain an abstract of a first object, only one or a plurality of key words related to the first object need to be input, the scheme provided by the embodiment of the disclosure can obtain a corresponding abstract template according to the key words input by the user, the abstract template comprises template content and content to be filled, then the key words input by the user are added to the content to be filled of the abstract template to obtain the abstract of the first object, so that the corresponding abstract is generated on the premise of inputting fewer key words without manually writing an abstract text, and the burden of the user is reduced.
On the basis of the above-described embodiments, the technical solutions of the present disclosure will be described in detail below with specific embodiments. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.
Fig. 4 is a flowchart illustrating a digest generation method according to another embodiment of the present disclosure, as shown in fig. 4, including:
s41, receiving at least one keyword of the first object input by the user.
The specific implementation of S41 is similar to S21 in fig. 2, and is not described here again.
S42, obtaining an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of the object indicated by the keyword.
In the embodiment of the present disclosure, the entity category represents a category of a keyword, and before generating the abstract according to the keyword, a generation operation needs to be performed according to the entity category of the keyword. Still taking the advertisement field as an example, when the input keyword is "shirt", the entity category of "shirt" is goods or goods name, when the input keyword is "$ 32", the entity category of "$ 32" is price, when the input keyword is "cool", the entity category of "cool" is goods selling point or goods advantage, when the input keyword is "pure cotton", the entity category of "pure cotton" is goods material, and the like.
Since the keywords are expected to be embodied in the abstract by the user, the keywords can be set at corresponding positions of the abstract template according to the entity type of the keywords.
S43, determining the object type of the first object according to the entity type corresponding to each keyword;
after the entity categories of the keywords are obtained, the object category of the first object needs to be determined according to the entity category of each keyword. For example, if the keyword is "shirt", the entity category of the keyword is product, and the object category of the first object can be further specified as clothing product and the related information of the first object can be further specified from the keyword "shirt" whose entity category is product.
S44, acquiring the abstract template corresponding to the first object based on the object type of the first object, wherein the object type of the first object is the same as the object type corresponding to the abstract template.
Before obtaining the summary template corresponding to the first object, a plurality of summary templates are obtained first, and the process will be described with reference to fig. 5.
Fig. 5 is a schematic flowchart of a process for obtaining a summary template according to an embodiment of the present disclosure, as shown in fig. 5, including:
s51, acquiring a preset abstract, wherein the click rate of the preset abstract is greater than a first threshold value;
on various platforms or web pages, there are various summaries. Searching advertisement abstracts released in the advertisement field on a platform or a webpage when the abstracts of the advertisement field are needed; when the abstract of the video field is needed, searching the abstract of the video field on a platform or a webpage; when the abstract of the education training field is required, the abstract of the education training field is searched on a platform or a web page, and the like.
Because the information is complicated, only the summary with the click rate greater than the first threshold is selected as the preset summary in the embodiment of the present disclosure, because the click rate is intuitive reflection and evaluation of the user on the summary content to some extent. Compared with the summary of low click rate, the summary of high click rate has a higher possibility of being more excellent in the aspects of popularity, attractiveness to the user and the like, so that only the summary of which the click rate is greater than the first threshold needs to be acquired.
And S52, generating the abstract template based on the preset abstract.
Specifically, the preset abstract is input into the named entity recognition model, and the entity category of each vocabulary in the preset abstract output by the named entity recognition model is obtained.
And after the entity category of each vocabulary is obtained, generating a summary template according to the entity category of each vocabulary in the preset summary. For example, a first entity category set and a second entity category set are obtained first, then a first vocabulary in a preset abstract is determined as template content, a second vocabulary in the preset abstract is deleted, and content of the position of the second vocabulary is determined as content to be filled, wherein the entity category of the first vocabulary is located in the first entity category set, and the entity category of the second vocabulary is located in the second entity category set.
Named entity recognition refers to a subtask of information extraction, which is used to locate and classify named entities in a text into predefined categories, such as people, organizations, locations, time, quantity, and the like. The named entity recognition model can be trained by adopting a machine learning method, in the embodiment of the disclosure, a preset abstract needs to be input into the named entity recognition model, and the entity category of each vocabulary output by the named entity recognition model is obtained.
Before training the named entity recognition model, predefined categories are determined, and the embodiment of the disclosure mainly includes two large categories, namely a first entity category and a second entity category, wherein words in the first entity category need to be retained in the abstract template, and words in the second entity category need to be deleted to serve as contents to be filled in the abstract template. The division of the vocabulary in the first entity category and the second entity category may be partially different for different domains.
Taking the advertisement delivery field as an example, when the abstract for the clothing needs to be generated in the advertisement delivery field, the second entity category may include the contents of the commodity name, the commodity price, the commodity selling point, and the like, and the rest of the contents may be the contents of the first entity category. Therefore, when the user uses the abstract template, the user only needs to input keywords related to the commodity name, the commodity price, the commodity selling point and the like.
And then, reserving the contents belonging to the first vocabulary in the abstract template, and filling the keywords into the corresponding contents to be filled to obtain the abstract of the first object.
Fig. 6 is a schematic diagram of generating a summary template according to an embodiment of the present disclosure, and as shown in fig. 6, the input preset summary is: the man prefers to have a classic achievement shirt and only sell 298 Yuan for the new season shirt.
And after the preset abstract is input into the named entity recognition model, the entity category of each vocabulary in the preset abstract output by the model is obtained, and a first entity category set and a second entity category set are obtained. In FIG. 6, words in the first set of entity categories are represented by boxes and words in the second set of entity categories are represented by underlining. The words in the summary belonging to the first set of entity categories include: the first choice, achievement, classics, new season and sale only of men, and the vocabularies belonging to the second entity category set comprise: shirt, 298 yuan.
Then, the vocabulary in the preset abstract belonging to the first entity category set is used as the fixed content in the abstract template generated by the preset abstract, the relative positions of the vocabularies are unchanged, the vocabulary belonging to the second entity category set is deleted, namely, the shirt and the 298 yuan in the preset abstract are deleted, and the deleted vacant positions are the content to be filled and the corresponding vocabularies need to be filled.
After the above operations are performed, the abstract template generated by the preset abstract can be obtained:
the first choice for men is the achievement ____ classic, which is only ____ yuan sold in the new season ____.
It should be noted that, in the embodiment of the present disclosure, after the object type of the first object is obtained, the abstract template corresponding to the first object is further obtained, and the object type of the first object is the same as the object type corresponding to the abstract template. For example, if the object type of the first object is a clothing type product, the object type of the obtained abstract template is the object type of the clothing type product, and if the object type of the first object is a food type product, the object type of the obtained abstract template is the object type of the electronic type product. This process will be described below with reference to fig. 7.
Fig. 7 is a schematic diagram of filling a summary template provided in the embodiment of the present disclosure, as shown in fig. 7, including a preset summary of clothing:
the man prefers to have a classic achievement shirt and only sell 298 Yuan for the new season shirt.
The abstract template generated by the preset abstract is as follows:
the first choice for men is the achievement ____ classic, which is only ____ yuan sold in the new season ____.
The abstract template is an abstract template for clothing, and can only be used for clothing commodities. If applied to other ranges, an inappropriate digest may result. For example, if the product that the user needs to publicize is milk, after inputting the keywords "milk" and "20", the obtained abstract is:
the male prefers to sell the achievement milk classics and the new-season milk is only sold 20 yuan.
The above generated abstract is obviously not suitable for propaganda of milk. In the embodiment of the disclosure, after a corresponding abstract template is generated according to a preset abstract, the generated abstract template can be marked to indicate the application range of the abstract template.
S45, obtaining an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of the object indicated by the keyword.
For example, the keyword is "shirt", the entity category corresponding to the keyword is name of goods, "cool" and the entity category corresponding to the keyword is "point of sale of goods", etc.
And S46, acquiring the entity category corresponding to each content to be filled in the abstract template.
When the keyword is filled into the content to be filled in the abstract template, the entity category corresponding to the content to be filled in the abstract template also needs to be acquired, so that the category of the keyword and the entity category corresponding to the content to be filled are correspondingly filled.
And S47, generating the abstract of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled.
Specifically, according to the entity category corresponding to each keyword and the entity category corresponding to each content to be filled, the corresponding relationship between at least one keyword and at least one content to be filled is determined, and then, according to the corresponding relationship, at least one keyword is filled into the content to be filled in the abstract template, so that the abstract of the first object is obtained.
Fig. 8 is a keyword filling schematic diagram provided in the embodiment of the present disclosure, and as shown in fig. 8, the keyword filling schematic diagram includes a summary template generated by presetting a summary:
not the other ____! Short stamp ____ is sold only at ____.
In the abstract template, the entity category of the main content to be filled includes a commodity selling point, a commodity name and a commodity price, and the total three keywords input by the user are "shirt", "cool" and "298 yuan", respectively, of which the entity category corresponding to "shirt" is the commodity name, corresponding to the second content to be filled of the abstract template, and should be filled to the second content to be filled of the abstract template, the entity category corresponding to "cool" is the commodity selling point, corresponding to the first content to be filled of the abstract template, and should be filled to the first content to be filled of the abstract template, the entity category corresponding to "298 yuan" is the commodity price, corresponding to the third content to be filled of the abstract template, and should be filled to the third content to be filled of the abstract template, and so on.
According to the filling, the obtained abstract is as follows:
nothing else is a cooling speed! The short style printed shirt is sold only 298 Yuan.
The filling of the keywords is only an example, and in practice, the entity types of the keywords are not limited to the above examples, and the embodiments of the disclosure are not described in detail.
In addition to the above method for generating a summary, the present disclosure also provides a method for generating a summary, where a keyword input by a user is first obtained, and then the keyword input by the user is input into a named entity recognition model, so as to obtain the keyword and a tag corresponding to the keyword, where the tag corresponding to the keyword reflects an entity type of the keyword, and a training process of the named entity recognition model is similar to that in the above embodiments, and is not repeated here.
After obtaining the keywords and the labels corresponding to the keywords, inputting the keywords and the labels into a text generation model to obtain a generated abstract, wherein the text generation model is obtained by training according to a training sample set. The training sample set comprises preset abstracts, keywords and labels corresponding to the keywords. The keywords and the corresponding abstracts comprise millions of parallel training corpora to form a huge training sample set, and the training sample set is trained until the model converges by constructing a deep neural network, such as a seq2seq text generation network, so as to obtain a text generation model.
In the process of generating the abstract, a plurality of abstract templates can be obtained according to the input keywords, and at the moment, the plurality of abstract templates need to be displayed for the user to select. Fig. 9 is a schematic view illustrating summary generation provided by the embodiment of the present disclosure, and as shown in fig. 9, an embodiment of the present disclosure may adopt a scheme that at least two summaries of a first object are displayed based on weight values of summary templates corresponding to the at least two summaries, then a target summary selected by a user in the at least two summaries is received, and the target summary is determined as the summary of the first object.
Specifically, the click rate of each abstract template is labeled behind each abstract template, the abstract templates are sequentially ordered according to the click rate of each abstract template, and when a certain abstract template selected by a user generates a target abstract, the click rate of the abstract template is updated, and the weight value of each abstract template is updated.
In fig. 9, after the keyword "shirt" is input, five filled abstracts are generated, which are:
abstract one, "a shirt, make your quality distinctive";
abstract II, printed shirt, fine quality, classic taste;
thirdly, the short-money printed shirt is soft in texture and rich in elasticity, and earning is achieved when winning;
fourth, a short shirt, simple and plain, and free cleaning without deformation;
fifth abstract, the new-season shirt is pushed out from the heavyweight, the materials are selected and superior, and the shirt is suitable for four seasons.
After the templates of the five summaries are labeled with the corresponding click rates respectively, as shown in fig. 9, the click times of the five summaries are 980 times, 1006 times, 5029 times, 3298 times and 4200 times in sequence.
As can be seen, the click rate of summary three is the highest, indicating that the most users are viewed. At this time, the click rate of the summaries can be provided for the user for reference, and the user can arbitrarily select one of the summaries as the selected target summary. For example, when the user selects abstract four, the number of clicks of abstract four will be updated to 3299.
The abstract generating method provided by the embodiment of the disclosure includes the steps of firstly receiving at least one keyword of a first object input by a user, then obtaining an abstract template corresponding to the first object based on the at least one keyword, and finally generating an abstract of the first object based on the at least one keyword and the abstract template. When a user needs to obtain an abstract of a first object, only one or a plurality of key words related to the first object need to be input, the scheme provided by the embodiment of the disclosure can obtain a corresponding abstract template according to the key words input by the user, the abstract template comprises template content and content to be filled, then the key words input by the user are added to the content to be filled of the abstract template to obtain the abstract of the first object, so that the corresponding abstract is generated on the premise of inputting fewer key words without manually writing an abstract text, and the burden of the user is reduced.
Fig. 10 is a schematic structural diagram of the summary generation apparatus provided in the embodiment of the present disclosure, as shown in fig. 10, including a receiving module 101, an obtaining module 102, and a processing module 103, where:
the receiving module 101 is configured to receive at least one keyword of a first object input by a user;
the obtaining module 102 is configured to obtain, based on the at least one keyword, a summary template corresponding to the first object, where the summary template includes template content and at least one content to be filled, the summary template is generated according to a preset summary, and a click rate of the preset summary is greater than a first threshold;
the processing module 103 is configured to generate a summary of the first object based on the at least one keyword and the summary template.
In a possible implementation manner, the processing module 103 is specifically configured to:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
acquiring entity categories corresponding to each content to be filled in the abstract template;
and generating the abstract of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled.
In a possible implementation manner, the processing module 103 is specifically configured to:
determining the corresponding relation between the at least one keyword and the at least one content to be filled according to the entity category corresponding to each keyword and the entity category corresponding to each content to be filled;
and filling the at least one keyword into the content to be filled in the abstract template according to the corresponding relation to obtain the abstract of the first object.
In a possible implementation manner, the obtaining module 102 is specifically configured to:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
determining the object type of the first object according to the entity type corresponding to each keyword;
and acquiring a summary template corresponding to the first object based on the object type of the first object, wherein the object type of the first object is the same as the object type corresponding to the summary template.
In a possible implementation manner, the obtaining module 102 is further configured to:
acquiring a preset abstract, wherein the click rate of the preset abstract is greater than a first threshold value;
and generating the abstract template based on the preset abstract.
In a possible implementation manner, the obtaining module 102 is further configured to:
inputting the preset abstract into a named entity recognition model to obtain the entity category of each vocabulary in the preset abstract output by the named entity recognition model;
and generating the abstract template according to the entity category of each vocabulary in the preset abstract.
In a possible implementation manner, the obtaining module 102 is further configured to:
acquiring a first entity category set and a second entity category set;
determining a first vocabulary in the preset abstract as the template content, wherein the entity category of the first vocabulary is located in the first entity category set;
and deleting a second vocabulary in the preset abstract, and determining the content of the position of the second vocabulary as the content to be filled, wherein the entity category of the second vocabulary is located in the second entity category set.
In a possible implementation manner, the processing module 103 is further configured to, when the number of generated digests of the first object is greater than 1:
displaying at least two abstracts of the first object based on the weight values of the abstract templates corresponding to the at least two abstracts;
receiving a target abstract selected by a user from the at least two abstracts;
determining the target summary as a summary of the first object.
In a possible implementation manner, the processing module 103 is further configured to, after determining the target digest as the digest of the first object:
and updating the weight values of the abstract templates corresponding to the at least two abstract according to the target abstract.
The apparatus provided in the embodiment of the present disclosure may be configured to implement the technical solution of the method embodiment, and the implementation principle and the technical effect are similar, which are not described herein again.
Fig. 11 is a schematic structural diagram of an electronic device provided in an embodiment of the present disclosure, and referring to fig. 11, a schematic structural diagram of an electronic device 110 suitable for implementing an embodiment of the present disclosure is shown, where the electronic device 110 may be a terminal device or a server. Among them, the terminal Device may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a Digital broadcast receiver, a Personal Digital Assistant (PDA), a tablet computer (PAD), a Portable Multimedia Player (PMP), a car terminal (e.g., car navigation terminal), etc., and a fixed terminal such as a Digital TV, a desktop computer, etc. The electronic device shown in fig. 11 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 11, the electronic device 110 may include a processing device (e.g., a central processing unit, a graphics processor, etc.) 111, which may perform various suitable actions and processes according to a program stored in a Read Only Memory (ROM) 112 or a program loaded from a storage device 118 into a Random Access Memory (RAM) 113. In the RAM 113, various programs and data necessary for the operation of the electronic apparatus 110 are also stored. The processing device 111, the ROM112, and the RAM 113 are connected to each other through a bus 114. An input/output (I/O) interface 115 is also connected to bus 114.
Generally, the following devices may be connected to the I/O interface 115: input devices 116 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 117 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage devices 118 including, for example, magnetic tape, hard disk, etc.; and a communication device 119. The communication means 119 may allow the electronic device 110 to communicate wirelessly or by wire with other devices to exchange data. While fig. 11 illustrates an electronic device 110 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication device 119, or installed from the storage device 118, or installed from the ROM 112. The computer program performs the above-described functions defined in the digest generation method of the embodiment of the present disclosure when executed by the processing device 111.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, 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 disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to perform the methods shown in the above embodiments.
Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of Network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The units described in the embodiments of the present disclosure may be implemented by software or hardware.
Finally, it should be noted that: the above embodiments are only used for illustrating the technical solutions of the embodiments of the present disclosure, and not for limiting the same; although embodiments of the present disclosure have been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; such modifications and substitutions do not depart from the spirit and scope of the embodiments of the present disclosure.

Claims (12)

1. A method for generating a summary, comprising:
receiving at least one keyword of a first object input by a user;
acquiring a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is greater than a first threshold;
generating a summary of the first object based on the at least one keyword and the summary template.
2. The method of claim 1, wherein generating the summary of the first object based on the at least one keyword and the summary template comprises:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
acquiring entity categories corresponding to each content to be filled in the abstract template;
and generating the abstract of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled.
3. The method of claim 2, wherein generating the summary of the first object based on the entity category corresponding to each keyword and the entity category corresponding to each content to be filled comprises:
determining the corresponding relation between the at least one keyword and the at least one content to be filled according to the entity category corresponding to each keyword and the entity category corresponding to each content to be filled;
and filling the at least one keyword into the content to be filled in the abstract template according to the corresponding relation to obtain the abstract of the first object.
4. The method according to any one of claims 1 to 3, wherein obtaining the abstract template corresponding to the first object based on the at least one keyword comprises:
acquiring an entity category corresponding to each keyword, wherein the entity category is used for indicating the category of an object indicated by the keyword;
determining the object type of the first object according to the entity type corresponding to each keyword;
and acquiring a summary template corresponding to the first object based on the object type of the first object, wherein the object type of the first object is the same as the object type corresponding to the summary template.
5. The method according to any one of claims 1 to 4, wherein before obtaining the summary template corresponding to the first object based on the at least one keyword, the method further comprises:
acquiring a preset abstract, wherein the click rate of the preset abstract is greater than a first threshold value;
and generating the abstract template based on the preset abstract.
6. The method of claim 5, wherein generating the summary template based on the predetermined summary comprises:
inputting the preset abstract into a named entity recognition model to obtain the entity category of each vocabulary in the preset abstract output by the named entity recognition model;
and generating the abstract template according to the entity category of each vocabulary in the preset abstract.
7. The method of claim 6, wherein generating the abstract template according to the entity category of each vocabulary in the preset abstract comprises:
acquiring a first entity category set and a second entity category set;
determining a first vocabulary in the preset abstract as the template content, wherein the entity category of the first vocabulary is located in the first entity category set;
and deleting a second vocabulary in the preset abstract, and determining the content of the position of the second vocabulary as the content to be filled, wherein the entity category of the second vocabulary is located in the second entity category set.
8. The method according to any one of claims 1-7, wherein the number of generated digests of the first object is greater than 1, the method further comprising:
displaying at least two abstracts of the first object based on the weight values of the abstract templates corresponding to the at least two abstracts;
receiving a target abstract selected by a user from the at least two abstracts;
determining the target summary as a summary of the first object.
9. The method of claim 8, wherein determining the target digest as the digest of the first object further comprises:
and updating the weight values of the abstract templates corresponding to the at least two abstract according to the target abstract.
10. An apparatus for generating a summary, comprising:
the receiving module is used for receiving at least one keyword of a first object input by a user;
the obtaining module is used for obtaining a summary template corresponding to the first object based on the at least one keyword, wherein the summary template comprises template content and at least one content to be filled, the summary template is generated according to a preset summary, and the click rate of the preset summary is larger than a first threshold value;
and the processing module is used for generating the abstract of the first object based on the at least one keyword and the abstract template.
11. An electronic device, comprising: at least one processor and memory;
the memory stores computer-executable instructions;
the at least one processor executing the memory-stored computer-executable instructions cause the at least one processor to perform the summary generation method of any of claims 1 to 9.
12. A computer-readable storage medium having stored thereon computer-executable instructions which, when executed by a processor, implement the summary generation method of any one of claims 1 to 9.
CN201911141552.3A 2019-11-20 2019-11-20 Abstract generation method and device Pending CN110909154A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911141552.3A CN110909154A (en) 2019-11-20 2019-11-20 Abstract generation method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911141552.3A CN110909154A (en) 2019-11-20 2019-11-20 Abstract generation method and device

Publications (1)

Publication Number Publication Date
CN110909154A true CN110909154A (en) 2020-03-24

Family

ID=69818393

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911141552.3A Pending CN110909154A (en) 2019-11-20 2019-11-20 Abstract generation method and device

Country Status (1)

Country Link
CN (1) CN110909154A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112819567A (en) * 2021-01-21 2021-05-18 万象春天实业集团(武汉)有限公司 Management method and device based on intelligent bidding platform
TWI764566B (en) * 2021-02-09 2022-05-11 中國信託商業銀行股份有限公司 Method and system for generating subject matter of electronic publicity

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090041642A (en) * 2007-10-24 2009-04-29 에스케이 텔레콤주식회사 System and method for producing template-based user created content, and apparatus to the same
CN110059309A (en) * 2018-01-18 2019-07-26 北京京东尚科信息技术有限公司 The generation method and device of information object title
CN110134931A (en) * 2019-05-14 2019-08-16 北京字节跳动网络技术有限公司 Media title generation method, device, electronic equipment and readable medium

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20090041642A (en) * 2007-10-24 2009-04-29 에스케이 텔레콤주식회사 System and method for producing template-based user created content, and apparatus to the same
CN110059309A (en) * 2018-01-18 2019-07-26 北京京东尚科信息技术有限公司 The generation method and device of information object title
CN110134931A (en) * 2019-05-14 2019-08-16 北京字节跳动网络技术有限公司 Media title generation method, device, electronic equipment and readable medium

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112819567A (en) * 2021-01-21 2021-05-18 万象春天实业集团(武汉)有限公司 Management method and device based on intelligent bidding platform
CN112819567B (en) * 2021-01-21 2024-04-30 万象春天实业集团(武汉)有限公司 Intelligent bidding platform-based management method and device
TWI764566B (en) * 2021-02-09 2022-05-11 中國信託商業銀行股份有限公司 Method and system for generating subject matter of electronic publicity

Similar Documents

Publication Publication Date Title
US11776018B2 (en) Universal ad creative
US10620804B2 (en) Optimizing layout of interactive electronic content based on content type and subject matter
CN110134931B (en) Medium title generation method, medium title generation device, electronic equipment and readable medium
CN109740085B (en) Page content display method, device, equipment and storage medium
WO2020003296A1 (en) System, device, and method of automatic construction of digital advertisements
US9613268B2 (en) Processing of images during assessment of suitability of books for conversion to audio format
US20130080928A1 (en) Embeddable context sensitive chat system
US9830388B2 (en) Modular search object framework
CN113111026A (en) Gallery of messages with shared interests
US20120030014A1 (en) Automatic Ad Creative Generation
CN113760158A (en) Target object display method, object association method, device, medium and equipment
WO2020215977A1 (en) System, method and device for displaying information
EP3905177A1 (en) Recommending that an entity in an online system create content describing an item associated with a topic having at least a threshold value of a performance metric and to add a tag describing the item to the content
WO2023241321A1 (en) Recommendation method and apparatus, and device, storage medium and computer program product
CN110827058A (en) Multimedia promotion resource insertion method, equipment and computer readable medium
CN112100558A (en) Method, device, equipment and storage medium for object recommendation
CN110909154A (en) Abstract generation method and device
CN113159870B (en) Push information display method and device and computer equipment
WO2020147714A1 (en) Deliverable entity pushing method and device
CN113077319A (en) Dynamic recommendation method and device for micro detail page
CN111930459A (en) Content display method and device, electronic equipment and computer readable medium
CN113763009A (en) Picture processing method, picture skipping method, device, equipment and medium
CN111400464B (en) Text generation method, device, server and storage medium
CN115017200B (en) Method and device for sorting search results, electronic equipment and storage medium
US20230195774A1 (en) Systems and methods for generating interactable elements in text strings relating to media assets

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination