CN108460131B - Classification label processing method and device - Google Patents
Classification label processing method and device Download PDFInfo
- Publication number
- CN108460131B CN108460131B CN201810175718.2A CN201810175718A CN108460131B CN 108460131 B CN108460131 B CN 108460131B CN 201810175718 A CN201810175718 A CN 201810175718A CN 108460131 B CN108460131 B CN 108460131B
- Authority
- CN
- China
- Prior art keywords
- target classification
- label
- determining
- classification
- tag
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/954—Navigation, e.g. using categorised browsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Accounting & Taxation (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a method and a device for processing classified labels, wherein the method comprises the following steps: obtaining a plurality of source classification labels; creating a plurality of label categories, wherein each label category corresponds to at least one source classification label; determining at least one theme element according to at least two of the label categories; and determining each theme element as a target classification label through a preset natural language model. In the embodiment of the invention, the theme elements at least correspond to two label categories, and are further processed to determine the theme elements as the target classification labels according with the language habits, so that the target classification labels according to the embodiment of the invention can provide accurate recommendation for users.
Description
Technical Field
The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for processing a classification tag.
Background
In the user interfaces of e-commerce websites, video websites, news websites, etc., classification tags are usually set in advance for the contents of goods, videos, news, etc., so that when a user clicks on each classification tag in the user interface of the website, the corresponding contents of goods, videos, or news can be browsed.
In the prior art, when a classification tag is set in each website, only a rough classification tag is usually set, for example, a classification tag for clothing, food, electric appliances and the like is set in an e-commerce website, a classification tag for category, place of production, language and the like is set in a video website, and a classification tag for entertainment, military and the like is set in a news website.
However, in the process of studying the above technical solutions, the skilled person finds that the above technical solutions have the following disadvantages: the rough classification label cannot accurately reflect a category of contents, for example, an inland classification label in a video website, as long as an inland video is generated, videos related to love, thriller and war all appear in the inland classification label, and an accurate search basis cannot be provided for a user.
Disclosure of Invention
In view of the above, embodiments of the present invention are proposed to provide a sort label processing method and apparatus that overcome or at least partially solve the above problems.
According to a first aspect of the present invention, there is provided a classification tag processing method, the method comprising:
obtaining a plurality of source classification labels;
creating a plurality of label categories, wherein each label category corresponds to at least one source classification label;
determining at least one theme element according to at least two of the label categories;
and determining each theme element as a target classification label through a preset natural language model.
According to a second aspect of the present invention, there is provided a sorting label processing apparatus, the apparatus comprising:
the source classification label acquisition module is used for acquiring a plurality of source classification labels;
a tag category creating module, configured to create a plurality of tag categories, where each tag category corresponds to at least one source classification tag;
the theme element determining module is used for determining at least one theme element according to at least two label categories;
and the target classification label determining module is used for determining each topic element as a target classification label through a preset natural language model.
The embodiment of the invention has the following advantages:
according to the embodiment of the invention, after a plurality of source classification labels are obtained, a plurality of label categories are created according to the source video labels, and at least one theme element is determined according to at least two label categories, each theme element at least corresponds to two label categories, and the source video labels in each label category have specific contents, so that each theme element can be determined to be the target classification label according with language habits through a preset natural language model according to the contents of the source video labels in the label categories corresponding to each theme element. In the embodiment of the invention, the subject element at least corresponds to two label categories, so compared with the mode of classifying the objects to be classified only according to one source classification label in one classification in the prior art, the classification method of the embodiment of the invention is more accurate, meanwhile, as the subject element is further processed, the subject element is determined as the target classification label according with the language habit, the search input habit of the user can be better met, and when the user inputs related search content, the content similarity between the input search content and the target classification label is greatly improved, so that accurate recommendation can be provided for the user according to the target classification label of the embodiment of the invention.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The drawings are only for purposes of illustrating the preferred embodiments and are not to be construed as limiting the invention. Also, like reference numerals are used to refer to like parts throughout the drawings. In the drawings:
fig. 1 is a flowchart of a method for processing a category label according to an embodiment of the present invention;
FIG. 2 is a diagram showing a source sort label according to an embodiment of the present invention;
fig. 3 is a specific flowchart of a method for processing a category label according to an embodiment of the present invention;
FIG. 4 is a diagram illustrating a target classification label according to an embodiment of the present invention;
fig. 5 is a block diagram of a sorting label processing apparatus according to an embodiment of the present invention;
fig. 6 is a specific block diagram of a sorting label processing apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
It should be understood that the specific embodiments described herein are merely illustrative of the invention, but do not limit the invention to only some, but not all embodiments.
Example one
Referring to fig. 1, a flow diagram of a method of sorting labels is shown.
It can be understood that the embodiment of the present invention may be applied to a server side, where the server side may be a WEB server (World Wide WEB) or other types of servers, and the embodiment of the present invention is not particularly limited thereto.
In specific application, the server can process the source classification labels in the database under the condition that the access amount of users is less at night, so that the processing efficiency is improved, and the load of the server during daytime running is reduced; the server may also perform source classification tag processing operation according to an actual application scenario, and the embodiment of the present invention does not limit a specific implementation time for performing the classification tag processing method of the present invention.
The method specifically comprises the following steps:
step 101: a plurality of source classification tags is obtained.
In the embodiment of the present invention, the server may obtain at least one source classification tag from a database, a cache, a local disk, and the like.
In a specific application, the source classification tag may be a classification tag that is set in a user interface such as each website and web page and summarizes presentation of related content. For example, as shown in fig. 2, "chinese," "usa," "japan," "comedy," "tragedy," "action," "2018," "2017," "2016" and the like in the user interface of a video website may be considered as the source classification label corresponding to the video website. It should be understood that, in websites such as news, e-commerce, novels, etc., corresponding source classification tags are also set according to actual situations, the principle is similar to the above source classification tags, and is not illustrated one by one here, and the following steps of the embodiment of the present invention are also illustrated by only using the source classification tags of the video websites for movies, and the embodiment is only for the purpose of clearly explaining the present invention, but not for limiting the present invention.
Step 102: creating a plurality of label categories, wherein each of the label categories corresponds to at least one of the source classification labels.
In the embodiment of the present invention, after the source classification tags are obtained, the tag categories may be created according to the specific content of each source classification tag, and all the tag categories include source classification tags belonging to the same type.
For example, assuming that the obtained source classification tags are "chinese," "usa," "japan," "comedy," "tragedy," "love," "2018," "2017," and "2016," a tag category "country," "genre," "year" may be created according to the content of the source classification tags, specifically, the tag category "country" corresponds to the source classification tag "chinese," "usa," "japan," the tag category "genre" corresponds to the source classification tag "comedy," "tragedy," and "action," and the tag category "year" corresponds to the source classification tag "2018," "2017," and "2016.
Step 103: and determining at least one theme element according to at least two label categories.
In the embodiment of the invention, the theme elements can be determined in a mode of combining at least two created label categories.
For example, after creating the tag categories "country", "style", "year", the way of combining at least two of the three tag categories may be: "country + style", "style + year", "country + style + year"; the theme element may be at least one of "country + style", "style + year", "country + style + year".
Step 104: and determining each theme element as a target classification label through a preset natural language model.
In the embodiment of the invention, the preset natural language model is a model trained according to the natural language habit of people, and the training steps can be as follows:
firstly, analyzing the attributes of the label categories according to the content, semantics and the like of the label categories, and establishing a classification table comprising the corresponding relation between the label categories and the attributes of the label categories. For example, the label category "country" may correspond to "regional attributes", the label category "genre" may correspond to "content attributes", the label category "emotion" may correspond to "emotional attributes", and so on.
Secondly, setting a prefix and/or a suffix of the source classification label corresponding to the classification label according to the attribute of the label category; for example, "region attribute" or "content attribute" is usually a noun description, and "about" or "based" may be added in front of and "behind" in a source classification tag corresponding to any one of the tag categories "country" or "style"; the "emotional attribute" is usually an adjective, and "joy" can be added in front of, followed by "and the like.
Thirdly, setting the number of the label categories which can be contained in the natural language model, the positions of the label categories to be added with prefixes and suffixes and serving as suffix words of the natural language model categories by combining different application scenes; for example, 2 or 3 tag categories may be set, the first occurring tag category is added with a prefix and/or a suffix, and the later occurring tag categories are not added with a prefix and/or a suffix; for the natural language model of the movie classification, suffix words are collectively set as movies.
In specific application, when the tag category included in the theme element is "country + style", the source classification tag corresponding to the tag category "country" is "usa", and the source classification tag corresponding to the tag category "style" is "comedy", so that a "comedy movie about usa" and a "U.S movie about comedy" can be obtained; when the theme element contains the tag category of "country + genre + emotion", assuming that the source classification tag corresponding to the tag category "emotion" is "impairment", it is possible to obtain "impairment-oriented american comedy movie", "impairment-oriented comedy american movie", "impairment-oriented comedy movie" and the like.
And fourthly, training a large number of label categories with different attributes and different numbers in the natural language model to determine a final natural language model according with the language habits of people. For example, if the "traumatic American comedy movie" determined in the third step often conforms to the language habit of people, the natural language model when the subject element contains "Country + Style + Emotion" can be determined as: the first place sets "emotion" and adds prefix and/or suffix, the second place sets "country", the third place sets "genre", and finally adds movie.
It can be understood that a person skilled in the art can train a suitable natural language model according to an actual application scenario, and the embodiment of the present invention does not limit a specific form of the natural language model.
After the main element is obtained, the label categories contained in the main element can be matched with the classification table of the natural language model to obtain the attributes of each label category, and then the attributes are set at the corresponding positions of the natural language model according to the attributes to obtain the target classification labels meeting the natural language habits of people.
It is understood that in news and E-commerce websites, object classification labels like "funny news about entertainment", "fresh Japanese women's clothes" and the like can be obtained through corresponding natural language models. A person skilled in the art can train a corresponding natural language model according to an actual application scenario to determine a corresponding target classification label, and the embodiment of the present invention does not specifically limit a specific form of the natural language model.
In summary, in the embodiments of the present invention, after a plurality of source classification tags are obtained, a plurality of tag categories are created according to source video tags, and at least one topic element is determined according to at least two tag categories, so that each topic element at least corresponds to two tag categories, and a source video tag in each tag category has specific content, and therefore, according to the content of a source video tag in a tag category corresponding to each topic element, each topic element can be determined as a target classification tag conforming to a language habit through a preset natural language model. In the embodiment of the invention, the subject element at least corresponds to two label categories, so compared with the mode of classifying the objects to be classified only according to one source classification label in one classification in the prior art, the classification method of the embodiment of the invention is more accurate, meanwhile, as the subject element is further processed, the subject element is determined as the target classification label according with the language habit, the search input habit of the user can be better met, and when the user inputs related search content, the content similarity between the input search content and the target classification label is greatly improved, so that accurate recommendation can be provided for the user according to the target classification label of the embodiment of the invention.
Example two
Referring to fig. 3, a specific flowchart of a method for processing a classified tag is shown, which specifically includes the following steps:
step 201: a plurality of source classification tags is obtained.
Step 202: creating a plurality of label categories, wherein each of the label categories corresponds to at least one of the source classification labels.
Step 203: and determining at least one theme element according to at least two label categories.
As a preferable solution of the embodiment of the present invention, step 203 includes:
substep A1: and performing a combination operation on at least two label categories.
Substep A2: and respectively determining a theme element according to each combination operation.
In a specific application, the combining operation performed on at least two of the tag categories may be: enumerating all combination modes in which at least two label categories can be combined, wherein each combination mode corresponds to one theme element, and determining all possible theme elements.
For example, after creating the tag categories "country", "style", "year", the way of combining at least two of the three tag categories may be: "country + style", "style + year", "country + style + year"; the theme elements may be "country + style", "style + year", "country + style + year".
In the embodiment of the invention, a mode of enumerating at least two label category combinations is adopted, and the determined subject elements cover a comprehensive label category combination mode, so that the method and the device can adapt to different application scenes.
As a preferable solution of the embodiment of the present invention, step 203 includes:
substep B1: and determining each label category into at least two label category groups according to the semantics and the content of each label category.
In the embodiment of the present invention, after the tag category is created, the tag category needs to be further processed, and the tag categories with similar contents are determined as a plurality of tag category groups.
For example, when creating the tag category, there may be a tag category "marital" or "campus", and although "marital" is for a married or divorced movie and "campus" is for a student love, unmarried movie, both of which generally belong to the category of love, the tag category "marital" or "campus" may be set in the tag category group of "love".
Substep B2: and performing a combination operation on at least two label category groups.
Substep B3: and respectively determining a theme element according to each combination operation.
In a specific application, the combining operation performed on at least two of the tag category groups may be: enumerating all possible combinations of at least two tag category groups, wherein each combination corresponds to one theme element, and determining all possible theme elements.
In the embodiment of the invention, after the tag category is created, the tag category with similar content is determined as a tag category group according to the content of the tag category, and then the subject element is determined through the tag category group, compared with the substeps of determining the subject element directly through the tag category in steps A1 and A2, the number of the tag category group in the embodiment of the invention is obviously less than that of the tag category, so that the operation of combining when determining the subject element is correspondingly less, and the efficiency of determining the subject element is improved.
Step 204: and determining each theme element as a target classification label through a preset natural language model.
Step 205: and determining the target classification object corresponding to each target classification label according to the matching relation between the source classification label of each object to be classified and each target classification label.
In the embodiment of the invention, after the target classification label is determined, the target classification label corresponds to the active classification label, so that the source classification label and the target classification label contained in each object to be classified can be matched, and the matched object to be classified is arranged under the corresponding target classification label to be used as the target classification object corresponding to the target classification label.
As a preferred solution of the embodiment of the present invention, the target classification label and the target classification object corresponding to the target classification label may be directly displayed on a user interface. When a user browses a user interface, the user can see the target classification tags conforming to natural language habits and the corresponding target classification objects, and directly click the interested target classification tags or target classification objects without inputting '… movie related to …' into a search box for further search, so that the technical effects of providing accurate recommendation tags for the user and facilitating user operation are achieved.
As another preferred solution of the embodiment of the present invention, when the target classification label and the target classification object corresponding to the target classification label are displayed, the operations of steps 206 and 208 may also be performed.
Step 206: and determining the weight of each target classification label according to the user click rate of each target classification label and/or the user click rate of the target classification object corresponding to each target classification label.
In a specific application, the interests of users are very different, and are embodied on a user interface, and the user can be a target classification label or/and/or a target classification object click rate corresponding to the target classification label, the user often clicks the target classification label of interest of the user, and/or the target classification object corresponding to the target classification label, after counting a plurality of samples of the target classification labels clicked by the users, and/or the target classification object corresponding to the target classification label, the user click rate of each target classification label and/or the user click rate of the target classification object corresponding to each target classification label can be determined according to the samples, and then the user click rate of the target classification label can be determined as the weight of the target classification label, or the user click rate of the target classification object corresponding to the target classification label can be determined as the weight of the target classification label, or comprehensively analyzing the user click rate of the target classification label and the user click rate of the target classification object corresponding to the target classification label to determine the weight of the target classification label.
Step 207: and sequencing the target classification labels according to the weight of each target classification label.
In specific application, the target classification labels with larger weights are usually the target classification labels which are interested by most users, so that the target classification labels are sorted in an ascending or descending manner according to the weight of each target classification label.
Step 208: and displaying the target classification labels meeting the first preset condition and the target classification objects corresponding to the target classification labels meeting the first preset condition on a user interface according to the sequencing result of the target classification labels.
In the embodiment of the present invention, the target classification tags that satisfy the first preset condition may be: the last target classification tags corresponding to the ascending sorting or the positive target classification tags corresponding to the descending sorting may be determined according to the number of target classification tags to be displayed on the user interface, which is not specifically limited in the embodiment of the present invention.
As a preferable solution of the embodiment of the present invention, before step 208, the method may further include:
substep C1: and determining the weight of the target classification object in each target classification label according to the user click rate of the target classification object in each target classification label.
In a specific application, the hobbies of the public are always influenced by popular elements, for example, a large amount of clicks are always obtained when a movie which is most recently shown or publicized, and in order to cater to the interest wind direction of the public, the weight determination needs to be performed on the target classification object in the target classification tag, so that a user can directly see the target classification object with a high click rate in a user interface.
Substep C2: and sequencing the target classification objects in each target classification label according to the weight of the target classification objects in each target classification label.
Then step 208 may be: displaying the target classification labels meeting the first preset condition and the target classification objects meeting the second preset condition on a user interface according to the sequencing result of the target classification objects in the target classification labels; the target classification objects meeting the second preset condition are as follows: and in the target classification objects corresponding to the target classification labels meeting the first preset condition, the target classification objects meeting the second preset condition.
In the embodiment of the present invention, the target classification object that satisfies the second preset condition may be: in a certain target classification label, several target classification objects which are counted backwards and correspond to the ascending sorting, or several target classification objects which are counted forwards and correspond to the descending sorting, the specific number may be determined according to the number of the target classification objects which need to be displayed on the user interface, and the embodiment of the present invention is not particularly limited to this.
For example, as shown in fig. 4, not only the display positions of two target classification tags on the user interface are determined according to the weights of the target classification tags "about growing comedy movies" and "grave crime movies", but also the target classification objects displayed under each target classification tag are determined according to the weights of the target classification objects corresponding to the target classification tags, for example, "plains Buddha" and "fake espionage" are sequentially set according to the weights under the target classification tag "about growing comedy movies".
It can be understood that, in the step 206-.
In the embodiment of the invention, the target classification label determined according to the user interest and the target classification object corresponding to the target classification label are displayed in the user interface, so that the recommendation label which best meets the user requirement can be provided.
Step 209: and when a search request of a user is received, searching in the target classification label according to a search identifier carried by the search request.
Step 210: and returning the target classification object corresponding to the searched target classification label to the user.
In the embodiment of the invention, the search request of the user is often a sentence which accords with natural semantics, so that when the search request of the user is received, the search identifier carried by the search request is searched in the target classification label, the corresponding target classification label can be matched quickly, and the target classification object corresponding to the searched target classification label is returned to the user, thereby greatly improving the efficiency of searching for the user.
In summary, in the embodiments of the present invention, after a plurality of source classification tags are obtained, a multi-tag category is created according to a source video tag, and at least one theme element is determined according to at least two tag categories, where each theme element corresponds to at least two tag categories, and a source video tag in each tag category has specific content, so that each theme element can be determined as a target classification tag that conforms to a language habit through a preset natural language model according to the content of the source video tag in the tag category corresponding to each theme element. In the embodiment of the invention, the subject element at least corresponds to two label categories, so compared with the mode of classifying the objects to be classified only according to one source classification label in one classification in the prior art, the classification method of the embodiment of the invention is more accurate, meanwhile, as the subject element is further processed, the subject element is determined as the target classification label according with the language habit, the search input habit of the user can be better met, and when the user inputs related search content, the content similarity between the input search content and the target classification label is greatly improved, so that accurate recommendation can be provided for the user according to the target classification label of the embodiment of the invention.
It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.
EXAMPLE III
Referring to fig. 5, a block diagram of a sorting label processing apparatus is shown, which may specifically include:
a source classification tag obtaining module 310, configured to obtain a plurality of source classification tags;
a tag category creating module 320, configured to create a plurality of tag categories, where each tag category corresponds to at least one source classification tag.
A theme element determining module 330, configured to determine at least one theme element according to at least two of the tag categories.
And the target classification label determining module 340 is configured to determine each topic element as a target classification label through a preset natural language model.
Preferably, referring to fig. 6, on the basis of fig. 5, the classification label processing apparatus further includes:
and a target classification object determining module 350, configured to determine a target classification object corresponding to each target classification label according to a matching relationship between the source classification label of each object to be classified and each target classification label.
The target classification label weight determining module 360 is configured to determine a weight of each target classification label according to a user click rate of each target classification label and/or a user click rate of a target classification object corresponding to each target classification label.
A first sorting module 370, configured to sort the target classification labels according to the weight of each target classification label.
And the target classification object weight determining module 380 is configured to determine the weight of the target classification object in each target classification label according to the user click rate of the target classification object in each target classification label.
The second sorting module 390 is configured to sort the target classification objects in each target classification label according to the weight of the target classification object in each target classification label.
A display module 400, configured to display the target classification label and the target classification object corresponding to the target classification label on a user interface.
The searching module 410 is configured to search in the target classification tag according to a search identifier carried in a search request when the search request of a user is received.
And a returning module 420, configured to return the target classification object corresponding to the found target classification tag to the user.
The theme element determination module 330 may include:
and the first combination sub-module is used for carrying out combination operation on at least two label categories.
And the first theme element determining submodule is used for determining a theme element according to each combination operation.
The theme element determination module 330 may further include:
and the tag category group determining submodule is used for determining each tag category into at least two tag category groups according to the semantics and the content of each tag category.
And the second combination sub-module is used for carrying out combination operation on at least two label category groups.
And the second theme element determining submodule is used for determining a theme element according to each combination operation.
The display module 400 may include:
and the first display sub-module is used for displaying the target classification tags meeting the first preset condition and the target classification objects corresponding to the target classification tags meeting the first preset condition on a user interface according to the sequencing result of the target classification tags.
The display module 400 may further include:
the second display sub-module is used for displaying the target classification labels meeting the first preset condition and the target classification objects meeting the second preset condition on a user interface according to the sequencing result of the target classification objects in the target classification labels; the target classification objects meeting the second preset condition are as follows: and in the target classification objects corresponding to the target classification labels meeting the first preset condition, the target classification objects meeting the second preset condition.
In the embodiment of the present invention, after the source classification tag obtaining module 310 obtains a plurality of source classification tags, a plurality of tag categories are created by the tag category creating module 320 according to the source video tags, and then the subject element determining module 330 determines at least one subject element according to at least two tag categories, so that each subject element at least corresponds to two tag categories, and the source video tags in each tag category have specific contents, and therefore, according to the contents of the source video tags in the tag categories corresponding to each subject element, the target classification tag determining module 340 may determine each subject element as a target classification tag conforming to the language habit through a preset natural language model. In the embodiment of the invention, the subject element at least corresponds to two label categories, so compared with the mode of classifying the objects to be classified only according to one source classification label in one classification in the prior art, the classification method of the embodiment of the invention is more accurate, meanwhile, as the subject element is further processed, the subject element is determined as the target classification label according with the language habit, the search input habit of the user can be better met, and when the user inputs related search content, the content similarity between the input search content and the target classification label is greatly improved, so that accurate recommendation can be provided for the user according to the target classification label of the embodiment of the invention.
For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.
The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
In a typical configuration, the computer device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. The memory may include forms of volatile memory in a computer readable medium, Random Access Memory (RAM) and/or non-volatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of a computer-readable medium. Computer-readable media, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include non-transitory computer readable media (fransitory media), such as modulated data signals and carrier waves.
Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable classification terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable classification terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable classification terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable classification terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.
The above detailed description is provided for a sorting label processing method and a sorting label processing apparatus, and the principle and the implementation of the present invention are explained by applying specific examples, and the description of the above examples is only used to help understanding the method and the core idea of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (12)
1. A method for processing a classification tag, the method comprising:
obtaining a plurality of source classification labels, wherein the source classification labels are labels for summarizing related contents of a website;
creating a plurality of label categories, wherein each label category corresponds to at least one source classification label;
determining at least one theme element according to at least two of the label categories;
determining each topic element as a target classification label through a preset natural language model;
before each topic element is determined as a target classification tag through a preset natural language model, the method further includes:
analyzing the attribute of the label category, and establishing a classification table corresponding to the label category and the attribute of the label category;
setting a prefix and/or a suffix of the source classification label according to the attribute of the label category;
setting the number of the label categories contained in the natural language model, the positions of the label categories needing to be added with prefixes and suffixes, and suffix words serving as the natural language model categories;
and training the label categories with different attributes and different numbers in the natural language model to determine the final natural language model.
2. The method of claim 1, wherein the step of determining at least one subject element from at least two of the tag categories comprises:
performing a combination operation on at least two of the label categories;
and respectively determining a theme element according to each combination operation.
3. The method of claim 1, wherein said step of determining at least one subject element from at least two of said tag categories comprises:
determining each label category into at least two label category groups according to the semantics and the content of each label category;
performing a combination operation on at least two of the tag category groups;
and respectively determining a theme element according to each combination operation.
4. The method of claim 1, further comprising:
determining a target classification object corresponding to each target classification label according to the matching relation between the source classification label of each object to be classified and each target classification label;
and displaying the target classification label and a target classification object corresponding to the target classification label on a user interface.
5. The method of claim 4, further comprising:
determining the weight of each target classification label according to the user click rate of each target classification label and/or the user click rate of a target classification object corresponding to each target classification label;
sorting the target classification labels according to the weight of each target classification label;
the step of displaying the target classification label and the target classification object corresponding to the target classification label on a user interface includes:
and displaying the target classification labels meeting the first preset condition and the target classification objects corresponding to the target classification labels meeting the first preset condition on a user interface according to the sequencing result of the target classification labels.
6. The method of claim 5, further comprising:
determining the weight of the target classification object in each target classification label according to the user click rate of the target classification object in each target classification label;
sorting the target classification objects in each target classification label according to the weight of the target classification objects in each target classification label;
the step of displaying the target classification label and the target classification object corresponding to the target classification label on a user interface further includes:
displaying the target classification labels meeting the first preset condition and the target classification objects meeting the second preset condition on a user interface according to the sequencing result of the target classification objects in the target classification labels;
the target classification objects meeting the second preset condition are as follows: and in the target classification objects corresponding to the target classification labels meeting the first preset condition, the target classification objects meeting the second preset condition.
7. A sorting label processing apparatus, characterized in that the apparatus comprises:
the source classification label acquisition module is used for acquiring a plurality of source classification labels, wherein the source classification labels are labels for summarizing relevant contents of a website;
a tag category creating module, configured to create a plurality of tag categories, where each tag category corresponds to at least one source classification tag;
the theme element determining module is used for determining at least one theme element according to at least two label categories;
the target classification label determining module is used for determining each topic element as a target classification label through a preset natural language model;
the target classification label determination module is further configured to:
analyzing the attribute of the label category, and establishing a classification table corresponding to the label category and the attribute of the label category;
setting a prefix and/or a suffix of the source classification label according to the attribute of the label category;
setting the number of the label categories contained in the natural language model, the positions of the label categories needing to be added with prefixes and suffixes, and suffix words serving as the natural language model categories;
and training the label categories with different attributes and different numbers in the natural language model to determine the final natural language model.
8. The apparatus of claim 7, wherein the theme element determination module comprises:
the first combination sub-module is used for carrying out combination operation on at least two label categories;
and the first theme element determining submodule is used for determining a theme element according to each combination operation.
9. The apparatus of claim 7, wherein the theme element determination module comprises:
the tag category group determining submodule is used for determining each tag category into at least two tag category groups according to the semantics and the content of each tag category;
the second combination sub-module is used for carrying out combination operation on at least two label category groups;
and the second theme element determining submodule is used for determining a theme element according to each combination operation.
10. The apparatus of claim 7, further comprising:
the target classification object determining module is used for determining a target classification object corresponding to each target classification label according to the matching relation between the source classification label of each object to be classified and each target classification label;
and the display module is used for displaying the target classification label and the target classification object corresponding to the target classification label on a user interface.
11. The apparatus of claim 10, further comprising:
the target classification label weight determining module is used for determining the weight of each target classification label according to the user click rate of each target classification label and/or the user click rate of a target classification object corresponding to each target classification label;
the first sequencing module is used for sequencing the target classification labels according to the weight of each target classification label;
the display module comprises:
and the first display sub-module is used for displaying the target classification tags meeting the first preset condition and the target classification objects corresponding to the target classification tags meeting the first preset condition on a user interface according to the sequencing result of the target classification tags.
12. The apparatus of claim 11, further comprising:
the target classification object weight determining module is used for determining the weight of the target classification object in each target classification label according to the user click rate of the target classification object in each target classification label;
the second sorting module is used for sorting the target classification objects in each target classification label according to the weight of the target classification objects in each target classification label;
the display module comprises:
the second display sub-module is used for displaying the target classification labels meeting the first preset condition and the target classification objects meeting the second preset condition on a user interface according to the sequencing result of the target classification objects in the target classification labels; the target classification objects meeting the second preset condition are as follows: and in the target classification objects corresponding to the target classification labels meeting the first preset condition, the target classification objects meeting the second preset condition.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810175718.2A CN108460131B (en) | 2018-03-02 | 2018-03-02 | Classification label processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810175718.2A CN108460131B (en) | 2018-03-02 | 2018-03-02 | Classification label processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108460131A CN108460131A (en) | 2018-08-28 |
| CN108460131B true CN108460131B (en) | 2021-11-26 |
Family
ID=63217676
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810175718.2A Active CN108460131B (en) | 2018-03-02 | 2018-03-02 | Classification label processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108460131B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109587568A (en) * | 2018-11-01 | 2019-04-05 | 北京奇艺世纪科技有限公司 | Video broadcasting method, device, computer readable storage medium |
| CN109992346A (en) * | 2019-04-08 | 2019-07-09 | 香港乐蜜有限公司 | Methods of exhibiting, device and the equipment of virtual objects |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678335A (en) * | 2012-09-05 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Method and device for identifying commodity with labels and method for commodity navigation |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8447588B2 (en) * | 2008-12-18 | 2013-05-21 | Palo Alto Research Center Incorporated | Region-matching transducers for natural language processing |
| CN102760142A (en) * | 2011-04-29 | 2012-10-31 | 北京百度网讯科技有限公司 | Method and device for extracting subject label in search result aiming at searching query |
| CN105335386B (en) * | 2014-07-01 | 2018-10-16 | 阿里巴巴集团控股有限公司 | A kind of method and device that navigation tag is provided |
| CN106874279B (en) * | 2015-12-11 | 2021-01-15 | 腾讯科技（深圳）有限公司 | Method and device for generating application category label |
| CN107463552A (en) * | 2017-07-20 | 2017-12-12 | 北京奇艺世纪科技有限公司 | A kind of method and apparatus for generating video subject title |
-
2018
- 2018-03-02 CN CN201810175718.2A patent/CN108460131B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103678335A (en) * | 2012-09-05 | 2014-03-26 | 阿里巴巴集团控股有限公司 | Method and device for identifying commodity with labels and method for commodity navigation |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108460131A (en) | 2018-08-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108694223B (en) | User portrait database construction method and device | |
| US9514405B2 (en) | Scoring concept terms using a deep network | |
| JP6855595B2 (en) | Using machine learning to recommend live stream content | |
| CN104573054B (en) | A kind of information-pushing method and equipment | |
| CN108228758B (en) | Text classification method and device | |
| US11176142B2 (en) | Method of data query based on evaluation and device | |
| CN106649316B (en) | Video pushing method and device | |
| US20160048754A1 (en) | Classifying resources using a deep network | |
| US10902077B2 (en) | Search result aggregation method and apparatus based on artificial intelligence and search engine | |
| CN111178970B (en) | Advertisement putting method and device, electronic equipment and computer readable storage medium | |
| CN107784010B (en) | Method and equipment for determining popularity information of news theme | |
| US20170193075A1 (en) | System and method for aggregating, classifying and enriching social media posts made by monitored author sources | |
| CN110019943B (en) | Video recommendation method and device, electronic equipment and storage medium | |
| US10503803B2 (en) | Animated snippets for search results | |
| CN111159341B (en) | Information recommendation method and device based on user investment and financial management preference | |
| CN109255037B (en) | Method and apparatus for outputting information | |
| CN111400586A (en) | Group display method, terminal, server, system and storage medium | |
| CN108763369B (en) | Video searching method and device | |
| US20100169318A1 (en) | Contextual representations from data streams | |
| CN111552884A (en) | Method and apparatus for content recommendation | |
| CN108460131B (en) | Classification label processing method and device | |
| CN103425767B (en) | A kind of determination method and system pointing out data | |
| US20200073925A1 (en) | Method and system for generating a website from collected content | |
| KR102560610B1 (en) | Reference video data recommend method for video creation and apparatus performing thereof | |
| CN116738060A (en) | Content generation method and device and electronic equipment |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |