CN113297493A - Method and device for identifying brand of article - Google Patents

Method and device for identifying brand of article Download PDF

Info

Publication number
CN113297493A
CN113297493A CN202110643892.7A CN202110643892A CN113297493A CN 113297493 A CN113297493 A CN 113297493A CN 202110643892 A CN202110643892 A CN 202110643892A CN 113297493 A CN113297493 A CN 113297493A
Authority
CN
China
Prior art keywords
brand
names
identified
brand name
article
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.)
Granted
Application number
CN202110643892.7A
Other languages
Chinese (zh)
Other versions
CN113297493B (en
Inventor
孔晓晴
李百川
陈第
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.)
Youmi Technology Co ltd
Original Assignee
Youmi 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 Youmi Technology Co ltd filed Critical Youmi Technology Co ltd
Priority to CN202110643892.7A priority Critical patent/CN113297493B/en
Publication of CN113297493A publication Critical patent/CN113297493A/en
Application granted granted Critical
Publication of CN113297493B publication Critical patent/CN113297493B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/55Clustering; Classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5846Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using extracted text
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/22Matching criteria, e.g. proximity measures
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00Pattern recognition
    • G06F18/20Analysing
    • G06F18/24Classification techniques

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Library & Information Science (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Machine Translation (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a brand identification method and device of an article, which can improve the extraction accuracy and efficiency of brand names in text information of the article to be identified by carrying out multimode matching analysis on the text information of the article to be identified and identification information of brands in an industry brand library corresponding to the text information of the article to be identified after identifying the category of the article to be identified, and can determine that a matched brand is a brand result of the article to be identified when the brand name matched with the brand information of the article to be identified, which is input by a user, exists in a brand name set obtained by the multimode matching analysis, so that the identification accuracy and efficiency of the brand result of the article to be identified can be improved, and whether the brand name contained in the text information of the article to be identified is the correct brand of the article to be identified can be accurately judged.

Description

Method and device for identifying brand of article
Technical Field
The invention relates to the technical field of internet, in particular to a brand identification method and device for an article.
Background
With the rapid development of the internet, the e-commerce has been developed more rapidly, and more users like to purchase desired items on the internet based on the advantages of convenience and time saving, and before purchasing the items, the users generally like to know the related information of the desired items, and to know the related information of the desired items, the brand of the desired items is determined, and in order to determine the brand of the desired items, the users often identify the picture of the desired items through their user terminals to determine the brand of the items, or identify the text information on the picture of the items to determine the brand of the items.
However, practice finds that the method of determining the brand of an article by recognizing the picture of the article often cannot accurately determine the brand of the article due to similarity of appearances of the articles; methods for determining the brand of an item by recognizing text information on a picture of the item are often due to the fact that the text information of the item often contains multiple brands, such as: the seniority hall is fast to the sun, etc., or the text information of the article can have a word of ambiguity, such as: millet 10 jin, etc. also cannot accurately determine the brand of the article. Therefore, it is important to provide a scheme for improving the identification accuracy of the brand of the article.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a method and a device for identifying a brand of an article, which can improve the identification accuracy of the brand of the article to be identified.
In order to solve the technical problem, a first aspect of the present invention discloses a method for identifying a brand of an article, the method comprising:
when the category of an article to be identified is determined, conducting multi-mode matching analysis on text information of the article to be identified and identification information of at least one brand in an industry brand library corresponding to the category of the article to be identified to obtain a first brand name set matched with the text information of the article to be identified, wherein the first brand name set comprises at least one brand name;
judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set;
and when judging that the brand name matched with the brand information of the article to be identified exists, determining the brand name matched with the brand information of the article to be identified as the brand result of the article to be identified.
In a second aspect, the present invention discloses a brand identification device for an article, the device comprising:
the matching module is used for carrying out multi-mode matching analysis on the text information of the object to be identified and the identification information of at least one brand in an industry brand library corresponding to the category of the object to be identified when the category of the object to be identified is determined, so as to obtain a first brand name set matched with the text information of the object to be identified, wherein the first brand name set comprises at least one brand name;
the judging module is used for judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set;
and the determining module is used for determining the brand name matched with the brand information of the article to be identified as the brand result of the article to be identified when the judging module judges that the brand name matched with the brand information of the article to be identified exists.
In a third aspect, the present invention discloses a brand identification device for another article, the device comprising:
a memory storing executable program code;
a processor coupled with the memory;
the processor calls the executable program code stored in the memory to execute part or all of the steps of the brand identification method of the article disclosed by the first aspect of the invention.
In a fourth aspect, the present invention discloses a computer storage medium storing computer instructions for performing some or all of the steps of the brand identification method for an article disclosed in the first aspect of the present invention when the computer instructions are invoked.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
in the embodiment of the invention, when the category of an article to be identified is determined, multi-mode matching analysis is carried out on text information of the article to be identified and identification information of at least one brand in an industry brand library corresponding to the category of the article to be identified to obtain a first brand name set matched with the text information of the article to be identified, wherein the first brand name set comprises at least one brand name; judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set; and when judging that the brand name matched with the brand information of the article to be identified exists, determining the brand name matched with the brand information of the article to be identified as a brand result of the article to be identified. Therefore, after the category of the object to be identified is identified, the multi-mode matching analysis is carried out on the text information of the object to be identified and the identification information of the brand in the industry brand library corresponding to the text information of the object to be identified, so that the extraction accuracy and efficiency of the brand name in the text information of the object to be identified can be improved, when the brand name matched with the brand information of the object to be identified, which is input by a user, exists in the brand name set obtained by the multi-mode matching analysis, the matched brand name is determined to be the brand result of the object to be identified, the identification accuracy and efficiency of the brand result of the object to be identified can be improved, and whether the brand name contained in the text information of the object to be identified is the correct brand of the object to be identified can be accurately judged.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a flow chart illustrating a method for brand identification of an article according to an embodiment of the present invention;
FIG. 2 is a flow chart illustrating another method for brand identification of an article according to an embodiment of the present disclosure;
FIG. 3 is a schematic structural diagram of a brand identification device for an article according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a brand identification device for another article according to an embodiment of the present disclosure;
FIG. 5 is a schematic structural diagram of a brand identification device for another article according to an embodiment of the present disclosure.
Detailed Description
In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The terms "first," "second," and the like in the description and claims of the present invention and in the above-described drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, apparatus, article, or article that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or article.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the invention. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The invention discloses a brand identification method and device for an article, which can perform multimode matching analysis on text information of the article to be identified and identification information of a brand in an industry brand library corresponding to the text information of the article to be identified after identifying the category of the article to be identified, can improve the extraction accuracy and efficiency of the brand name in the text information of the article to be identified, and can determine that the matched brand name is a brand result of the article to be identified when the brand name matched with the brand information of the article to be identified, which is input by a user, exists in a brand name set obtained through the multimode matching analysis. The following are detailed below.
Example one
Referring to fig. 1, fig. 1 is a flowchart illustrating a brand identification method for an article according to an embodiment of the present invention. The method described in fig. 1 may be applied to a brand identification device, such as a server, and the embodiment of the present invention is not limited thereto. As shown in fig. 1, the brand identification method of the article may include the following operations:
101. when the category of the to-be-identified object is determined, conducting multi-mode matching analysis on the text information of the to-be-identified object and the identification information of at least one brand in the industry brand library corresponding to the category of the to-be-identified object to obtain a first brand name set matched with the text information of the to-be-identified object, wherein the first brand name set comprises at least one brand name.
In the embodiment of the invention, the article to be identified comprises any commodity or non-commodity of which the brand needs to be identified. And the text information of the object to be identified comprises text information obtained by identifying the object to be identified and/or stored text information of the object to be identified. Further, the text information of the object to be identified also includes brand information of the object to be identified, which is input by a user corresponding to the object to be identified. And the text information of the item to be identified includes an item title of the item to be identified. Further, the text information of the object to be identified may include, but is not limited to, a product description (e.g., a use description) and/or store information (e.g., a store name) of the object to be identified.
In the embodiment of the present invention, optionally, the industry brand library corresponding to the category of the object to be identified is an industry brand library of an industry to which the object to be identified belongs. The identification information of each brand in the industry brand library corresponding to the category of the object to be identified comprises a main brand name, a father brand name and a son brand name of the brand, wherein the main brand name, the father brand name and the son brand name of each brand comprise at least one of a Chinese name, an English name, an alias and a brand ID. Therefore, the more contents in the industry brand library, the more the accuracy, comprehensiveness and efficiency of the brand name in the text information of the object to be identified are improved, and the accuracy and efficiency of the brand result of the object to be identified are improved. Wherein the main brand name is the final official name of the brand. And further, the name of the main brand is generally a Chinese name, such as: the Chinese name of the brand is a gale diary, the English name is a wind diary, and the gale diary is generally used as a main brand name of the brand; the father brand name is the superior brand name of the son brand name, such as: when the marbled holy leaves are subjected to the safety sun-curing, the name of the parent marbled holy leaves is the name of the child marbled Chinese herb, and the name of the main brand of the safety sun-curing is the name of the safety skin-protecting Chinese herb.
102. Judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set; when judging that the brand name matched with the brand information of the article to be identified exists, triggering to execute the step 103; when it is determined that there is no brand name matching the brand information of the article to be identified, the process may be ended.
103. And determining the brand name matched with the brand information of the article to be identified as a brand result of the article to be identified.
In the embodiment of the present invention, optionally, the brand name matched with the brand information of the object to be identified includes a brand name identical to the brand information, or a sub-brand name of the brand information, or a main brand name of the brand information.
For example, if the text title of the article to be identified is "100 ml of under-leaf sun protection and protection for skin under the beginner, then the obtained first brand name set of the article to be identified is [ beginner, skin protection for skin protection ], and if the brand information of the article to be identified input by the user is [ skin protection for skin protection ], then it is determined that [ skin protection for skin protection ] is the brand result of the article to be identified.
In the embodiment of the present invention, optionally, only the text information of one article to be recognized may be analyzed at a time, the text information of a plurality of articles to be recognized may be analyzed at the same time, and a plurality of pieces of text information of one article to be recognized may also be analyzed.
It can be seen that, by implementing the method described in fig. 1, after the category of the object to be identified is identified, the multi-mode matching analysis can be performed on the text information of the object to be identified and the identification information of the brand in the industry brand library corresponding to the text information of the object to be identified, so that the accuracy and efficiency of extracting the brand name from the text information of the object to be identified can be improved, and when the brand name matched with the brand information of the object to be identified, which is input by the user, exists in the set of brand names obtained by the multi-mode matching analysis, the matched brand name is determined to be the brand result of the object to be identified, so that the accuracy and efficiency of identifying the brand result of the object to be identified can be improved, and whether the brand name included in the text information of the object to be identified is the correct brand of the object to be identified can be accurately determined.
In an optional embodiment, before determining whether there is a brand name matching the acquired brand information of the to-be-identified item input by the user in all the brand names of the first set of brand names, the method may further include the following operations:
judging whether the brand name of the determined brand name type exists in all the brand names of the first brand name set or not according to the text information of the article to be identified;
when the first brand name set does not exist, the operation of judging whether the brand name matched with the acquired brand information of the to-be-identified object input by the user exists in all the brand names of the first brand name set is executed;
and when the name is judged to exist, deleting the name of the name type from all the names in the first name set to obtain a second name set, and executing the operation of judging whether the name matched with the acquired brand information of the object to be identified input by the user exists in all the names in the first name set, wherein the second name set is the first name set.
In this optional embodiment, optionally, the determined brand name type includes at least one of a type of the brand name included in the determined brand name, a type of a like brand name, a type of a mean brand name, and a type of a like alternative brand name. Therefore, the more the content included by the brand name types, the more the judgment accuracy, the efficiency and the comprehensiveness of the brand names which do not meet the requirements exist or do not exist in the brand name set obtained through the multi-mode matching analysis are improved, the more the judgment accuracy, the efficiency and the comprehensiveness of the brand names which do not meet the requirements are improved when the brand names which do not meet the requirements exist are judged, the more the extraction accuracy of the brand names in the text information of the object to be identified is improved, and the more the identification accuracy of the brand results of the object to be identified is improved.
Therefore, in the optional embodiment, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the object to be identified input by the user, whether unmatched brand names exist in the brand name set obtained by the multi-mode matching analysis is judged, if the unmatched brand names do not exist, the subsequent brand analysis step is directly executed, if the unmatched brand names exist, the unmatched brand names are deleted from the brand name set obtained by the multi-mode matching analysis, the situations that the brand result identification errors of the object to be identified and the identification efficiency of the brand result are reduced due to the fact that the unmatched brand names exist in the brand name set obtained by the multi-mode matching analysis but the brand result analysis is still carried out can be reduced, and the identification accuracy and efficiency of the brand result of the object to be identified can be further improved.
In this optional embodiment, as an optional embodiment mode, judging whether all the brand names in the first brand name set have the determined brand name of the brand name type according to the text information of the article to be identified includes:
judging whether a certain word in the determined word set exists in the text information of the object to be identified, and when the certain word exists, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, wherein the brand names in front of the certain word are the brand names of the determined brand name types; or,
judging whether all the brand names in the first brand name set have contained brand names according to text information of the object to be identified, and when the contained brand names are judged to exist, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, and the contained brand names are the brand names of the determined brand name types.
In this optional embodiment, optionally, the term set includes at least one of the same type, and the same type of substitution.
In this optional embodiment, when it is determined that a certain word in the determined word set does not exist in the text information of the article to be recognized, the following operation of determining whether all brand names in the first brand name set exist in the brand names included in the text information of the article to be recognized may be triggered; or, when it is determined that there is no included brand name, the above operation of determining whether there is a certain word in the determined word set in the text information of the article to be recognized may be triggered. It should be noted that a term may be understood as at least one term.
For example, the text information of the object to be identified is: the sky is fairy and flat in price to replace the Korean plum lotion, and at the moment, the sky is not the correct brand name of the object to be identified, and the sky needs to be deleted and flat in price to replace the sky in the front; the text information of the object to be identified is as follows: the gale diary glass is glazed, at the moment, two brand names of the gale diary and the gale diary can be identified, and at the moment, the contained brand name of the gale needs to be removed.
Therefore, according to the optional embodiment, when words such as translation and the like exist in the text information of the object to be identified and/or included brands exist, the determination that the unqualified brands exist in the brand name set obtained through the multi-mode matching analysis can be realized, the determination mode that the unqualified brands exist in the brand name set obtained through the multi-mode matching analysis can be enriched, and the determination probability that the unqualified brands exist in the brand name set obtained through the multi-mode matching analysis is improved.
In another optional embodiment, after the text information of the article to be recognized and the identification information of at least one brand in the industry brand library corresponding to the category of the article to be recognized are subjected to multi-mode matching analysis, so as to obtain a first brand name set matched with the text information of the article to be recognized, where the first brand name set includes at least one brand name, the method may further include the following operations:
judging whether the number of all brand names in the first brand name set is larger than a determined brand name number threshold value or not, and executing the operation of judging whether the brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names in the first brand name set or not when the number of all the brand names in the first brand name set is judged to be smaller than or equal to the brand name number threshold value;
the method may further comprise the operations of:
when the number of the brand names is judged to be larger than the threshold value, collecting appearance information of the to-be-identified article, and screening brand names matched with the appearance information of the to-be-identified article from all the brand names in the first brand name set;
and updating all the brand names of the first brand name set based on the screened brand names to obtain an updated first brand name set, and executing the operation of judging whether the brand names matched with the acquired brand information of the object to be identified input by the user exist in all the brand names of the first brand name set.
In this alternative embodiment, the appearance information of the object to be identified includes, but is not limited to, at least one of a shape of the object to be identified, a color of the object to be identified, a texture of the object to be identified, a pattern of the object to be identified, and a number of components of the object to be identified.
In this optional embodiment, the threshold value of the number of brand names may be uniform, or may be determined according to the categories of different articles to be identified, that is, different categories correspond to different threshold values of brand names.
For example, when all brand names of the first set of brand names of the item to be identified are: [ Houshengtao, sun protection, skin protection, and beauty sun protection ], and the text information of the object to be identified is: the collected appearance of an article to be identified is a red panda, and the brand name corresponding to the red panda is sunscreen lotion and sun protection lotion, so that all brand names of a first brand name set of the article to be identified are as follows: sun-screening, beautifying and protecting.
Therefore, in the optional embodiment, after the multi-mode matching analysis is performed to obtain the brand name set, whether the number of the brand names in the brand name set is large is further judged, if not, the subsequent operation of determining the brand result of the object to be identified is directly executed, if yes, the brand name matched with the object to be identified is screened according to the appearance information of the object to be identified, the accurate brand name can be extracted, the data volume of the brand name can be reduced, then the brand result of the object to be identified is determined, and the efficiency of identifying the brand result of the object to be identified can be improved.
In yet another optional embodiment, before determining whether there is a brand name matching the acquired brand information of the to-be-identified item input by the user in all the brand names of the first set of brand names, the method may further include the following operations:
judging whether the industry to which the article to be identified belongs is a certain industry in the determined industry set;
and when the text information of the article to be identified is judged to be matched with the determined content matched with the certain industry, executing the operation of judging whether the brand name matched with the acquired brand information of the article to be identified input by the user exists in all the brand names of the first brand name set.
In this alternative embodiment, all industries in the determined industry set include, but are not limited to, at least one of the digital home appliance industry, the make-up and care industry, the apparel, shoe and bag industry, the mother and infant and child industry, the food and beverage industry, and the smart home industry.
For example, the industry to which the article to be identified belongs is an article of the digital home appliance industry, and the article title of the article to be identified is: the product titles comprise peripheral products of digital products such as a mobile phone shell and a charging wire, if the corresponding single product list does not comprise the mobile phone shell and/or the charging wire, the content of the mobile phone shell and the charging wire matched with the digital household appliance industry is determined to be not matched, namely the brand result of the object to be identified is determined not to be identified, the brand result of the object to be identified is not output, and the first brand name set is returned to be empty; if the mobile phone shell and/or the charging wire exist in the corresponding single item list, but the title of the mobile phone shell and the charging wire does not contain an official product and/or a genuine product, the mobile phone shell and the charging wire are determined not to be matched with the content matched with the digital household appliance industry, namely, the brand result of the object to be identified is also determined not to be identified, namely the brand result of the object to be identified is not output, namely, the first brand name set is returned to be empty; if the title comprises an official part and/or a genuine product, determining that the content matched with the digital household appliance industry is matched with the 'mostly mobile phone shell and the charging wire', namely determining that the brand result of the object to be identified is identified, namely outputting the brand result of the object to be identified to be 'mostly'; the industry to which the to-be-identified article belongs is an article in the beauty and make-up industry, and the article title of the to-be-identified article is as follows: the lipstick number is 123cf, the brand name is '3 cf' in a single-product list corresponding to the beauty and make-up personal care industry, but the number is '12' before the '3 cf', the brand result of the object to be identified is determined not to be identified, the brand result of the object to be identified is not output, namely the first brand name set is returned to be empty, and therefore the fact that the color number and/or the model number are mistakenly identified as the brand name is conveniently prevented.
In this optional embodiment, optionally, when it is determined that the industry to which the to-be-identified item belongs is not a certain industry, the operation of determining whether a brand name matching the acquired brand information of the to-be-identified item input by the user exists in all the brand names of the first brand name set is performed.
It can be seen that, in the optional embodiment, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the to-be-identified item input by the user, it is first determined whether the industry to which the to-be-identified item belongs is some special industries, and if so, it is further automatically searched whether the content of the special industries is included in the text information of the to-be-identified item, such as: if not, the subsequent brand analysis operation is continuously executed, so that the occurrence situation that the brand result of the object to be identified is mistakenly identified because the text information of the object to be identified is used for indicating that the object to be identified is a peripheral product of a special industry or only represents the model, the color number and the like of the object can be reduced, and the identification accuracy of the brand result of the object to be identified can be further improved.
In yet another alternative embodiment, the method may further comprise the steps of:
outputting a brand result of the object to be identified;
and before outputting a brand result of the item to be identified, the method may further include the steps of:
the method comprises the steps of performing part-of-speech tagging on text information of an article to be recognized to obtain a part-of-speech tagging result of each word in the text information of the article to be recognized, determining a part-of-speech tagging result corresponding to a brand result of the article to be recognized according to the part-of-speech tagging result of each word in the text information, judging whether the part-of-speech tagging result corresponding to the brand result of the article to be recognized is the same as the part-of-speech tagging result of the determined brand result of the article to be recognized, and executing the operation of outputting the brand result of the article to be recognized when the judgment result is yes.
Therefore, before outputting the brand result of the object to be recognized, the embodiment of the invention judges whether the part-of-speech label of the brand result of the object to be recognized in the text information is consistent with the specified part-of-speech label, and if so, the part outputs the brand result of the object to be recognized, thereby reducing the occurrence of verbs and common words of brands, further improving the recognition accuracy of the brand result of the object to be recognized and improving the output accuracy of the brand result of the object to be recognized.
In yet another alternative embodiment, before outputting the brand result of the item to be identified, the method may further include the steps of:
and identifying the image of the article to be identified based on the determined image identification algorithm to obtain the brand name of the article to be identified, judging whether the identified brand name of the article to be identified is the same as the brand result of the article to be identified, and if so, executing the operation of outputting the brand result of the article to be identified.
Therefore, the embodiment of the invention judges whether the brand name obtained by identifying the image of the article to be identified is consistent with the brand result of the article to be identified before outputting the brand result of the article to be identified, and if so, outputs the brand result of the article to be identified, thereby further improving the identification accuracy of the brand result of the article to be identified and the output accuracy of the brand result of the article to be identified.
In yet another optional embodiment, before performing multi-mode matching analysis on the text information of the article to be recognized and the identification information of at least one brand in the industry brand library corresponding to the category of the article to be recognized, and obtaining a first brand name set matched with the text information of the article to be recognized, the method may further include the following operations:
and executing formatting operation on the text information of the object to be identified to obtain the text information of the formatted object to be identified, and executing multimode matching analysis on the text information of the object to be identified and the identification information of at least one brand in the industry brand library corresponding to the category of the object to be identified to obtain a first brand name set matched with the text information of the object to be identified.
In this alternative embodiment, the formatting operation includes a remove irregular characters operation and/or a text conversion operation. Wherein the irregular character includes: separating symbols such as expression bag character strings, |, @, & and the like, wherein the character conversion operation comprises all English-to-lower-case operation or all English-to-upper-case operation.
In this optional embodiment, optionally, a formatting operation is performed on the text information of the article to be recognized through a regular expression.
In this optional embodiment, optionally, after the brand information of the to-be-identified item input by the user is obtained, a formatting operation is also performed on the brand information of the to-be-identified item, where for the description of performing the formatting operation on the brand information of the to-be-identified item, reference is made to the description of performing the formatting operation on the text information of the to-be-identified item, which is not described herein again.
Therefore, the optional embodiment can obtain the text information conforming to the processing by executing the formatting operation on the text information of the article to be identified, and is beneficial to further improving the accuracy and efficiency of extracting the brand name in the text information, thereby further improving the identification accuracy and efficiency of the brand result of the article to be identified.
In yet another alternative embodiment, the method may further include the operations of:
acquiring text information of an article to be recognized, inputting the text information of the article to be recognized into the determined information type determination model for analysis, and acquiring an analysis result output by the information type determination model as the type of the article to be recognized.
For the description of the text information of the object to be identified, please refer to the description of the related contents, which is not described herein again.
In this alternative embodiment, the information category determination model comprises a model based on the fastText algorithm.
Therefore, the optional embodiment can quickly and accurately determine the category of the to-be-identified object by analyzing the text information of the to-be-identified object based on the information category determination model, so that the identification accuracy and efficiency of the brand of the to-be-identified object can be improved.
Example two
Referring to fig. 2, fig. 2 is a flowchart illustrating another method for identifying a brand of an article according to an embodiment of the present invention. The method described in fig. 2 may be applied to a brand identification device, such as a server, and the embodiment of the present invention is not limited thereto. As shown in fig. 2, the brand identification method of the article may include the following operations:
201. when the category of the to-be-identified object is determined, conducting multi-mode matching analysis on the text information of the to-be-identified object and the identification information of at least one brand in the industry brand library corresponding to the category of the to-be-identified object to obtain a first brand name set matched with the text information of the to-be-identified object, wherein the first brand name set comprises at least one brand name.
202. Judging whether all target brand names are main brand names or not, wherein all target brand names comprise all brand names of the first brand name set and acquired brand information of the object to be identified input by the user; when the judgment result is yes, step 205 may be executed; when the judgment result is no, step 203 may be performed.
203. And performing primary name conversion on the brand names which are not the primary brand names in all the target brand names to obtain a third brand name set of the to-be-identified article and/or target brand information of the to-be-identified article, wherein the third brand name set comprises at least one primary brand name.
In the embodiment of the invention, when all the brand names in the first brand name set are main brand names, the third brand name set is completely the same as the first brand name set, and when the brand information of the to-be-identified object input by the user is the main brand name, the target brand information of the to-be-identified object is completely the same as the brand information of the to-be-identified object input by the user.
For example, the text information of the object to be identified is: the safe-care sunscreen cream emulsion is characterized in that the brand of an article to be identified is identified as safe-care sunscreen, the safe-care sunscreen is an alias of the brand of the article to be identified, the official name is called safe-care skin, namely the main brand name is called safe-care skin, and therefore the alias safe-care sunscreen is required to be converted into the main name safe-care skin.
204. And judging whether the brand name matched with the target brand information of the object to be identified exists in all the main brand names in the third brand name set. When judging that the brand name matched with the brand information of the article to be identified exists, triggering to execute the step 206; when it is determined that there is no brand name matching the brand information of the article to be identified, the process may be ended.
Therefore, before analyzing the brand name set obtained by the multimode matching analysis and the brand information of the object to be identified input by the user, the embodiment of the invention judges whether the brand name set obtained by the multimode matching analysis and the brand information input by the user are both the main brand names, if so, the subsequent operation of brand analysis is continuously executed, and if not, the main name conversion is firstly carried out, so that the brand names in the brand name set obtained by the multimode matching analysis can be deduplicated, the calculation amount is reduced, and the brand result identification accuracy and efficiency of the object to be identified can be further improved.
205. Judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set; when judging that the brand name matched with the brand information of the article to be identified exists, triggering to execute the step 206; when it is determined that there is no brand name matching the brand information of the article to be identified, the process may be ended.
206. And determining the brand name matched with the brand information of the article to be identified as a brand result of the article to be identified.
In the embodiment of the present invention, for the related descriptions of step 201, step 205, and step 206, please refer to the detailed descriptions of step 101 to step 103 in the first embodiment, which is not described again in the embodiment of the present invention.
It can be seen that, by implementing the method described in fig. 2, after the category of the object to be identified is identified, the multi-mode matching analysis can be performed on the text information of the object to be identified and the identification information of the brand in the industry brand library corresponding to the text information of the object to be identified, so that the accuracy and efficiency of extracting the brand name from the text information of the object to be identified can be improved, and when the brand name matched with the brand information of the object to be identified, which is input by the user, exists in the set of brand names obtained by the multi-mode matching analysis, the matched brand name is determined to be the brand result of the object to be identified, so that the accuracy and efficiency of identifying the brand result of the object to be identified can be improved, and whether the brand name contained in the text information of the object to be identified is the correct brand of the object to be identified can be accurately determined. In addition, the brand names in the brand name set obtained through multimode matching analysis can be subjected to duplication elimination, the calculation amount is reduced, and the brand result identification accuracy and efficiency of the to-be-identified object can be further improved.
In an optional embodiment, before determining whether there is a brand name matching the acquired brand information of the to-be-identified item input by the user in all the brand names of the first set of brand names, the method may further include the following operations:
judging whether the acquired brand information of the to-be-identified object input by the user is matched with the identification information of a certain brand in the industry brand library corresponding to the category of the to-be-identified object;
and when the first brand name set is matched with the second brand name set, executing the operation of judging whether the brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set.
In this optional embodiment, optionally, when it is determined that the brand information of the to-be-identified item is not matched with the identification information of a certain brand in the industry brand library corresponding to the category of the to-be-identified item, determining that the brand result of the to-be-identified item is not recognized, and ending the process, or triggering and executing the following operation of determining the brand result of the to-be-identified item according to the number of all the brand names in the first brand name set.
Therefore, in the optional embodiment, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the to-be-identified object input by the user, whether the brand information of the to-be-identified object is matched with the identification information of a certain brand in the industry brand library corresponding to the category of the to-be-identified object is automatically judged, if yes, the user continues to perform subsequent brand analysis operation, and the identification accuracy and efficiency of the brand result of the to-be-identified object are further improved.
In another optional embodiment, the method may further comprise the operations of:
adding the brand information of the object to be identified to a first brand name set when judging that no brand name matched with the brand information of the object to be identified exists;
and judging whether the number of all the brand names in the first brand name set is equal to 1 or not, and determining all the brand names in the first brand name set as the brand results of the to-be-identified articles when judging that the number of all the brand names in the first brand name set is equal to 1.
It can be seen that, in the optional embodiment, after it is determined that no brand name matching the brand information of the to-be-identified item input by the user exists in the set of brand names obtained by the multimode matching analysis, the brand information input by the user is further added to the set of brand names obtained by the multimode matching analysis, and when it is determined that only one brand number exists in the set of brand names of the to-be-identified item, the unique brand name is directly determined as the brand result of the to-be-identified item.
In yet another alternative embodiment, after adding the brand information of the to-be-identified item to the first set of brand names and before determining whether the number of all the brand names in the first set of brand names is equal to 1, the method may further include the operations of:
judging whether all brand names in the first brand name set have brand names matched with the determined common words or not, and judging whether the brand names matched with the common words are matched with brand information of the to-be-identified articles or not when the brand names matched with the common words are judged to exist;
and when judging that the brand name matched with the common word is matched with the brand information of the object to be identified, executing the operation of judging whether the number of all the brand names in the first brand name set is equal to 1 or not.
In this alternative embodiment, common words corresponding to different industries are different, for example: the common words of the beauty care industry are 'back', 'perfect' and 'finish', etc., and the common words of the digital household appliance industry are 'line', 'shell', etc.
In this optional embodiment, optionally, the method may further include the following operations:
and when judging that no brand name matched with the common word exists, executing subsequent operation of determining the brand result of the object to be identified according to the number of all the brand names in the first brand name set.
Therefore, when the optional embodiment judges that the brand name matched with the common word exists in the brand name set of the object to be identified and the brand name is matched with the brand information input by the user, the user determines the brand result of the object to be identified according to the number of the brand names in the brand name set of the object to be identified, and the accuracy of determining the brand result of the object to be identified according to the number of the brand names can be improved.
In yet another alternative embodiment, the method may further include the operations of:
when the number of all the brand names in the first brand name set is judged to be larger than 1, all the brand names with all the brand names of the first brand name set as parent brand names are converted into corresponding child brand names, all the parent brand names are deleted from the first brand name set, and a fourth brand name set is obtained;
and calculating the number of all the sub-brand names in the fourth brand name set, and determining the brand result of the object to be identified according to the number of all the sub-brand names in the fourth brand name set.
For example, if all the brand names of the first set of brand names include begonia and sun care, the begonia is converted into sun care and deleted.
Therefore, when the number 1 of the brand names in the brand name set obtained through the multimode matching analysis is judged, the optional embodiment can enrich the determining mode of identifying the accurate brand result of the article to be identified and improve the probability of accurately identifying the brand result by converting the parent brand name in the brand name set into the child brand name, deleting the parent brand name and determining the brand result of the article to be identified according to the number of the brand names in the set.
In the foregoing optional embodiment, optionally, determining the brand result of the to-be-identified item according to the number of all the sub-brand names in the fourth set of brand names includes:
judging whether the number of all the sub-brand names in the fourth brand name set is equal to 1, and determining the sub-brand names of the fourth brand name set as the brand result of the article to be identified when the number of all the sub-brand names in the fourth brand name set is equal to 1;
the method may further comprise the steps of;
when the number of all the sub-brand names in the fourth brand name set is judged to be larger than 1, whether the brand names belonging to the determined selling platform exist in all the sub-brand names in the fourth brand name set is judged, and when the brand names of the selling platform exist, the brand names belonging to the selling platform are deleted from all the sub-brand names in the fourth brand name set, so that a fifth brand name set is obtained;
and calculating the number of all the sub-brand names in the fifth brand name set, judging whether the number of all the sub-brand names in the fifth brand name set is equal to 1, and determining the sub-brand names in the fifth brand name set as the brand result of the object to be identified when judging that the number of all the sub-brand names in the fifth brand name set is equal to 1.
It can be seen that, in the optional embodiment, after the parent brand names in the brand name set are deleted, if the number of the brand names in the brand name set is equal to 1, the brand names in the brand name set are directly determined to be the brand result of the object to be identified, and if the number of the brand names in the brand name set is still greater than 1, the brand names belonging to the selling platform in the brand name set are deleted, so that the extraction accuracy of the brand names in the text information can be further improved, the brand result of the object to be identified is determined according to the number of the brand names in the set, the determination mode of the accurate brand result of the object to be identified can be enriched, and the probability and the accuracy of the accurate brand result can be improved.
In yet another alternative embodiment, the method may further include the operations of:
when the brand names of the selling platform do not exist, or when the number of all the sub-brand names in the fifth brand name set is judged to be larger than 1, target information of each brand name in a certain time period in all the brand names in the sixth brand name set is collected, wherein the target information of each brand name comprises at least one of purchase quantity, transaction quantity, good-scoring condition and repurchase condition of each brand name in a certain time period (such as 3 months) and the sixth brand name set is a fourth brand name set or a fifth brand name set, and the ending time of the certain time period is the time when the number of all the sub-brand names in the fourth brand name set is judged to be larger than 1;
and determining the welcome level of each brand name according to the target information corresponding to each brand name in the sixth brand name set, and determining the brand name corresponding to the target welcome level with the highest level from all the brand names in the sixth brand name set as the brand result of the object to be identified according to all the welcome levels.
As can be seen, in this alternative embodiment, when it is determined that the brand name of the vending platform is not included in the brand set or the number of the brand names in the brand set is still greater than 1 after the brand name of the vending platform is removed, the brand result of the article to be identified can be determined by determining the brand name with the highest popularity level in the brand set as the brand result of the article to be identified, and the determination probability of the brand result of the article to be identified can be improved.
In this optional embodiment, optionally, each content included in the target information corresponding to each brand name in the sixth brand name set has a corresponding weight value, and the sum of the weight values of all the contents of each brand name is equal to 1. Optionally, determining the popularity rating of each brand name according to the target information corresponding to each brand name in the sixth brand name set, including:
determining a numerical value corresponding to each content, and calculating a product value of the numerical value corresponding to each content and the weight value of each content;
and calculating the product value of all the contents as the total value of each brand name, and determining the welcome level of each brand name according to the total value of each brand name.
Therefore, the optional embodiment determines the total value of each brand name according to the weight value of each content and the numerical value of each content included in the target information of each brand name, so that the determination accuracy of the popularity level of each brand name is favorably improved, and the identification accuracy of the brand result of the object to be identified is favorably improved.
In yet another alternative embodiment, the method may further include the operations of:
when the brand name matched with the common word is judged not to be matched with the brand information of the object to be identified, deleting the brand name matched with the common word from the first brand name set to obtain a seventh brand name set;
determining a brand result of the object to be identified according to the number of all brand names in the first brand name set, wherein the determining comprises the following steps:
and determining the brand result of the object to be identified according to the number of the brand names in the seventh brand name set.
For example, the text information of the object to be identified is: the mask is repaired after the facial mask is sunned in the European Li ya, two brand names of 'European Li ya' and 'after' are obtained by recognition, and the information of the object to be recognized input by a user is as follows: and Oulia, deleting the name of the 'back' brand, wherein the brand result of the object to be identified is as follows: is European and Li ya.
Therefore, in the optional embodiment, when it is judged that the brand name matched with the common word is not matched with the brand information of the article to be identified, the brand name matched with the common word is directly deleted, an accurate brand name set can be obtained, the brand result of the article to be identified is determined according to the number of the brand names in the brand name set, and the accuracy of determining the brand result of the article to be identified according to the number of the brand names can be further improved.
EXAMPLE III
Referring to fig. 3, fig. 3 is a schematic structural diagram of a brand identification device for an article according to an embodiment of the present invention. As shown in fig. 3, the brand identification device of the article may include:
the matching module 301 is configured to, when the category of the to-be-identified item is determined, perform multi-mode matching analysis on the text information of the to-be-identified item and the identification information of at least one brand in the industry brand library corresponding to the category of the to-be-identified item to obtain a first brand name set matched with the text information of the to-be-identified item, where the first brand name set includes at least one brand name.
The judging module 302 is configured to judge whether a brand name matching the acquired brand information of the to-be-identified item input by the user exists in all the brand names of the first brand name set;
a determining module 303, configured to determine, when the judging module 302 judges that there is a brand name matching the brand information of the article to be identified, that the brand name matching the brand information of the article to be identified is a brand result of the article to be identified.
It can be seen that, after the brand recognition device described in fig. 3 is implemented to recognize the category of the object to be recognized, the multi-mode matching analysis is performed on the text information of the object to be recognized and the identification information of the brand in the industry brand library corresponding to the text information of the object to be recognized, so that the accuracy and efficiency of extracting the brand name from the text information of the object to be recognized can be improved, and when the brand name matched with the brand information of the object to be recognized, which is input by the user, exists in the set of brand names obtained through the multi-mode matching analysis, the matched brand name is determined to be the brand result of the object to be recognized, so that the accuracy and efficiency of recognizing the brand result of the object to be recognized can be improved, and whether the brand name included in the text information of the object to be recognized is the correct brand of the object to be recognized can be accurately determined.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to, before determining whether a brand name matching the acquired brand information of the to-be-identified item input by the user exists in all the brand names of the first set of brand names, determine whether a brand name of the determined brand name type exists in all the brand names of the first set of brand names according to the text information of the to-be-identified item, and when it is determined that the brand name does not exist, perform the above operation of determining whether a brand name matching the acquired brand information of the to-be-identified item input by the user exists in all the brand names of the first set of brand names.
And, as shown in fig. 4, the apparatus further comprises:
the first deleting module 304 is configured to, when the judging module 302 judges that a brand name of the brand name type exists, delete the brand name of the brand name type from all the brand names in the first brand name set to obtain a second brand name set, and trigger the judging module 302 to perform the operation of judging whether a brand name matching the obtained brand information of the to-be-identified item input by the user exists in all the brand names in the first brand name set, where the second brand name set is the first brand name set.
It can be seen that, by implementing the brand identification apparatus described in fig. 4, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the object to be identified input by the user, it can be determined whether an unmatched brand name exists in the brand name set obtained by the multi-mode matching analysis, if not, the subsequent brand analysis step is directly executed, and if so, the unqualified brand name is deleted from the brand name set obtained by the multi-mode matching analysis, so that the occurrence of the brand result identification error of the object to be identified and the reduction of the identification efficiency of the brand result due to the fact that the unqualified brand name exists in the brand name set obtained by the multi-mode matching analysis but the brand result analysis is still continued can be reduced, which is beneficial to further improving the identification accuracy and efficiency of the brand result of the object to be identified.
In yet another alternative embodiment, as shown in fig. 4, the manner that the determining module 302 determines whether all the brand names in the first set of brand names have the determined brand name type according to the text information of the article to be identified is specifically as follows:
judging whether a certain word in the determined word set exists in the text information of the object to be identified, and when the certain word exists, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, wherein the brand names in front of the certain word are the brand names of the determined brand name types; or,
judging whether all the brand names in the first brand name set have contained brand names according to text information of the object to be identified, and when the contained brand names are judged to exist, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, and the contained brand names are the brand names of the determined brand name types.
It can be seen that, by implementing the brand recognition device described in fig. 4, when words such as a substitution word and a money word exist in the text information of the object to be recognized and/or when included brands exist, the determination that the set of brand names obtained by the multi-mode matching analysis has the unsatisfactory brand names can be realized, and the determination manner that the set of brand names obtained by the multi-mode matching analysis has the unsatisfactory brand names can be enriched, so that the determination probability that the set of brand names obtained by the multi-mode matching analysis has the unsatisfactory brand names is improved.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to, after the matching module 301 performs multi-mode matching analysis on the text information of the object to be recognized and the identification information of at least one brand in the industry brand library corresponding to the category of the object to be recognized, obtain a first set of brand names that matches the text information of the object to be recognized, determine whether the number of all the brand names in the first set of brand names is greater than the determined threshold value of the number of brand names, and when it is determined that the number of all the brand names in the first set of brand names is less than or equal to the threshold value of the number of brand names, perform the above operation of determining whether there is a brand name that matches the acquired brand information of the object to be recognized, which is input by the user.
As shown in fig. 4, the apparatus further includes:
the first collecting module 305 is configured to collect appearance information of the to-be-identified item when the judging module 302 judges that the number of the brand names is greater than the threshold value.
And the screening module 306 is used for screening the brand names matched with the appearance information of the to-be-identified article from all the brand names in the first brand name set.
And the updating module 307 is configured to update all the brand names of the first brand name set based on the screened brand names to obtain an updated first brand name set, and trigger the judging module 302 to execute the operation of judging whether a brand name matching the obtained brand information of the to-be-identified item input by the user exists in all the brand names of the first brand name set.
It can be seen that, after the brand recognition device described in fig. 4 is implemented to obtain the brand name set through multi-mode matching analysis, whether the number of the brand names in the brand name set is large is further judged, if not, the subsequent operation of determining the brand result of the article to be recognized is directly executed, if yes, the brand name matched with the article to be recognized is screened according to the appearance information of the article to be recognized, the accurate brand name can be extracted, the data volume of the brand name can be reduced, then the brand result of the article to be recognized is determined, and the efficiency of recognizing the brand result of the article to be recognized can be improved.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to determine whether an industry to which the to-be-identified item belongs is a certain industry in the determined industry set before determining whether there is a brand name matching the acquired brand information of the to-be-identified item input by the user in all the brand names in the first brand name set.
The determining module 302 is further configured to determine whether the text information of the to-be-identified item includes the determined content matching with a certain industry when it is determined that the industry to which the to-be-identified item belongs is a certain industry, and execute the above operation of determining whether the brand name matching the obtained brand information of the to-be-identified item input by the user exists in all the brand names of the first brand name set when it is determined that the text information of the to-be-identified item does not include the content matching with a certain industry.
It can be seen that, with the brand recognition apparatus described in fig. 4, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the object to be recognized input by the user, it can be determined whether the industries to which the object to be recognized belongs are some special industries, and if yes, further automatically search the text information of the object to be recognized for the contents of the special industries, such as: if not, the subsequent brand analysis operation is continuously executed, so that the occurrence situation that the brand result of the object to be identified is mistakenly identified because the text information of the object to be identified is used for indicating that the object to be identified is a peripheral product of a special industry or only represents the model, the color number and the like of the object can be reduced, and the identification accuracy of the brand result of the object to be identified can be further improved.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to determine whether all target names are primary names before determining whether there is a name matching the acquired brand information of the to-be-identified item input by the user among all names in the first set of names, where all target names include all names in the first set of names and the acquired brand information of the to-be-identified item input by the user, and when the determination result is yes, perform the above operation of determining whether there is a name matching the acquired brand information of the to-be-identified item input by the user among all names in the first set of names.
And, as shown in fig. 4, the apparatus further comprises:
a conversion module 308, configured to, when the determination result of the determination module 302 is negative, perform a primary name conversion on a brand name that is not the primary brand name among all the target brand names, to obtain a third brand name set of the to-be-identified item and/or target brand information of the to-be-identified item, where the third brand name set includes at least one primary brand name.
The method for judging whether the brand name matched with the acquired brand information of the to-be-identified item input by the user exists in all the brand names of the first brand name set by the judging module 302 specifically includes:
and judging whether the brand name matched with the target brand information of the object to be identified exists in all the main brand names in the third brand name set.
It can be seen that, with the brand identification device described in fig. 4, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the object to be identified input by the user, it can be determined whether the brand name set obtained by the multi-mode matching analysis and the brand information input by the user are both the primary brand names, if yes, the subsequent brand analysis operation is continuously performed, and if not, the primary name conversion is performed first, so that the brand names in the brand name set obtained by the multi-mode matching analysis can be deduplicated, the calculation amount is reduced, and the brand result identification accuracy and efficiency of the object to be identified can be further improved.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to, before determining whether there is a brand name that matches the acquired brand information of the to-be-identified item input by the user in all the brand names of the first set of brand names, determine whether the acquired brand information of the to-be-identified item input by the user matches identification information of a certain brand in an industry brand library corresponding to the category of the to-be-identified item, and when determining that there is a match, perform the above operation of determining whether there is a brand name that matches the acquired brand information of the to-be-identified item input by the user in all the brand names of the first set of brand names.
It can be seen that, by implementing the brand recognition apparatus described in fig. 4, before analyzing the brand name set obtained by the multi-mode matching analysis and the brand information of the object to be recognized input by the user, it can be automatically determined whether the brand information of the object to be recognized matches with the identification information of a certain brand in the industry brand library corresponding to the category of the object to be recognized, and if so, the user continues to perform subsequent brand analysis operations, which is beneficial to further improving the recognition accuracy and efficiency of the brand result of the object to be recognized.
In yet another alternative embodiment, as shown in fig. 4, the apparatus further comprises:
an adding module 309, configured to add the brand information of the to-be-identified item to the first set of brand names when the determining module 302 determines that there is no brand name that matches the brand information of the to-be-identified item.
The determining module 302 is further configured to determine whether the number of all the brand names in the first set of brand names is equal to 1.
The determining module 303 is further configured to determine that the brand name of the first set of brand names is a brand result of the to-be-identified item when the determining module 302 determines that the number of all the brand names in the first set of brand names is equal to 1.
It can be seen that, after the brand identification device described in fig. 4 is implemented to judge that no brand name matching the brand information of the to-be-identified item input by the user exists in the set of brand names obtained by the multi-mode matching analysis, the brand information input by the user is further added to the set of brand names obtained by the multi-mode matching analysis, and when it is judged that only one brand number exists in the set of brand names of the to-be-identified item, the unique brand name is directly determined as the brand result of the to-be-identified item, on the basis of obtaining an accurate brand result of the to-be-identified item, the determination efficiency of the brand result of the to-be-identified item can be improved, the determination mode of the brand result of the to-be-identified item is enriched, the determination possibility of the brand result of the to-be-identified item is improved, and the intelligent function of the brand identification device is further enriched.
In yet another alternative embodiment, as shown in fig. 4, the determining module 302 is further configured to determine whether there are any brand names in the first set of brand names that match the determined common word after the adding module 309 adds the brand information of the to-be-identified item to the first set of brand names and before the determining module 302 determines whether the number of all the brand names in the first set of brand names is equal to 1.
The judging module 302 is further configured to, when it is determined that there is a brand name matching the common word, judge whether the brand name matching the common word matches the brand information of the article to be recognized, and when it is determined that the brand name matching the common word matches the brand information of the article to be recognized, trigger the judging module 302 to perform the above-described operation of judging whether the number of all the brand names in the first set of brand names is equal to 1.
It can be seen that, when the brand recognition apparatus described in fig. 4 is implemented to first determine that the brand name matched with the common word exists in the brand name set of the object to be recognized and the brand name is matched with the brand information input by the user, the user determines the brand result of the object to be recognized according to the number of the brand names in the brand name set of the object to be recognized, and the accuracy of determining the brand result of the object to be recognized according to the number of the brand names can be improved.
In yet another alternative embodiment, as shown in fig. 4, the apparatus further comprises:
the converting module 308 is configured to, when the determining module 302 determines that the number of all the brand names in the first set of brand names is greater than 1, convert all the brand names of the first set of brand names, which are the parent brand names, into corresponding child brand names.
The second deleting module 310 is configured to delete all parent brand names from the first set of brand names, so as to obtain a fourth set of brand names.
And a calculating module 311, configured to calculate the number of all the sub-brand names in the fourth set of brand names.
The determining module 303 is further configured to determine a brand result of the object to be identified according to the number of all sub-brand names in the fourth set of brand names.
It can be seen that, when the brand recognition device described in fig. 4 is implemented to judge the number 1 of brand names in the brand name set obtained by the multi-mode matching analysis, the determination method of the accurate brand result of the article to be recognized can be enriched and the probability of the accurate brand result can be improved by converting the parent brand name in the brand name set into the child brand name, deleting the parent brand name and determining the brand result of the article to be recognized according to the number of the brand names in the set.
In yet another alternative embodiment, as shown in fig. 4, the determining module 303 determines the brand result of the to-be-identified item according to the number of all the sub-brand names in the fourth set of brand names in a specific manner:
judging whether the number of all the sub-brand names in the fourth brand name set is equal to 1, and determining the sub-brand names of the fourth brand name set as the brand result of the article to be identified when the number of all the sub-brand names in the fourth brand name set is equal to 1;
when the number of all the sub-brand names in the fourth brand name set is judged to be larger than 1, whether the brand names belonging to the determined selling platform exist in all the sub-brand names in the fourth brand name set is judged, and when the brand names of the selling platform exist, the brand names belonging to the selling platform are deleted from all the sub-brand names in the fourth brand name set, so that a fifth brand name set is obtained;
and calculating the number of all the sub-brand names in the fifth brand name set, judging whether the number of all the sub-brand names in the fifth brand name set is equal to 1, and determining the sub-brand names in the fifth brand name set as the brand result of the object to be identified when judging that the number of all the sub-brand names in the fifth brand name set is equal to 1.
It can be seen that, after the brand recognition device described in fig. 4 is implemented to delete the parent brand names in the brand name set, if the number of the brand names in the brand name set is equal to 1, it is directly determined that the brand names in the brand name set are the brand result of the to-be-recognized article, and if the number of the brand names in the brand name set is still greater than 1, the brand names belonging to the selling platform in the brand name set are deleted, so that the accuracy of extracting the brand names in the text information can be further improved, and then the brand result of the to-be-recognized article is determined according to the number of the brand names in the set, so that the determination method for identifying the accurate brand result of the to-be-recognized article can be enriched, and the probability and the accuracy for accurately identifying the brand result are improved.
In yet another alternative embodiment, as shown in FIG. 4, the brand identification device further includes:
a second collecting module 312, configured to collect target information of each brand name in all the brand names of a sixth brand name set within a certain time period when the determining module 302 determines that there is no brand name of the vending platform, or when the determining module 302 determines that the number of all the sub-brand names in the sixth brand name set is greater than 1, where the sixth brand name set is a fifth brand name set or a fourth brand name set, and an end time of the certain time period is a time when the number of all the sub-brand names in the fifth brand name set is greater than 1;
the determining module 303 is further configured to determine a popularity rating of each brand name according to the target information corresponding to each brand name in the sixth brand name set, and determine, according to all popularity ratings, a brand name corresponding to a target popularity rating with a highest rating from all brand names in the sixth brand name set as a brand result of the article to be identified.
It can be seen that, with the brand identification apparatus described in fig. 4, when it is determined that the number of brand names in the brand set is still greater than 1 after the brand name of the vending platform is not included in the brand set or removed, the brand result of the article to be identified can be determined by determining the brand name with the highest popularity level in the brand set as the brand result of the article to be identified, so that the determination of the brand result of the article to be identified can be achieved, and the determination probability of the brand result of the article to be identified can be improved.
Example four
Referring to fig. 5, fig. 5 is a schematic structural diagram of a brand identification device of another article according to an embodiment of the present disclosure. As shown in fig. 5, the apparatus may include:
a memory 501 in which executable program code is stored;
a processor 502 coupled to a memory 501;
further, an input interface 503 and an output interface 504 coupled to the processor 502 may be included;
the processor 502 calls the executable program code stored in the memory 501 to execute the steps of the brand identification method for the article disclosed in any one of the first embodiment and the second embodiment of the present invention.
EXAMPLE five
The embodiment of the invention discloses a computer storage medium, which stores computer instructions, and the computer instructions are used for executing the steps of the brand identification method of the article disclosed by any one of the first embodiment and the second embodiment of the invention when being called.
The above-described embodiments of the apparatus are merely illustrative, and the modules described as separate components may or may not be physically separate, and the components shown as modules may or may not be physical modules, may be located in one place, or may be distributed on a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above detailed description of the embodiments, those skilled in the art will clearly understand that the embodiments may be implemented by software plus a necessary general hardware platform, and may also be implemented by hardware. Based on such understanding, the above technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, where the storage medium includes a Read-Only Memory (ROM), a Random Access Memory (RAM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), a One-time Programmable Read-Only Memory (OTPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a Compact Disc-Read-Only Memory (CD-ROM), or other disk memories, CD-ROMs, or other magnetic disks, A tape memory, or any other medium readable by a computer that can be used to carry or store data.
Finally, it should be noted that: the method and apparatus for identifying the brand of an article disclosed in the embodiments of the present invention are only preferred embodiments of the present invention, and are only used for illustrating the technical solutions of the present invention, not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art; the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (15)

1. A method of brand identification of an item, the method comprising:
when the category of an article to be identified is determined, conducting multi-mode matching analysis on text information of the article to be identified and identification information of at least one brand in an industry brand library corresponding to the category of the article to be identified to obtain a first brand name set matched with the text information of the article to be identified, wherein the first brand name set comprises at least one brand name;
judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set;
and when judging that the brand name matched with the brand information of the article to be identified exists, determining the brand name matched with the brand information of the article to be identified as the brand result of the article to be identified.
2. The method for brand identification of an article according to claim 1, wherein before determining whether there is a brand name matching the acquired brand information of the article to be identified, which is input by the user, among all the brand names in the first set of brand names, the method further comprises:
judging whether the brand name of the determined brand name type exists in all the brand names of the first brand name set or not according to the text information of the article to be identified;
when the first brand name set does not exist, the operation of judging whether the brand name matched with the acquired brand information of the article to be identified input by the user exists in all the brand names of the first brand name set is executed;
and when the brand name exists, deleting the brand name with the type of the brand name from all the brand names in the first brand name set to obtain a second brand name set, and executing the operation of judging whether the brand name matched with the acquired brand information of the article to be identified input by the user exists in all the brand names in the first brand name set, wherein the second brand name set is the first brand name set.
3. The method for brand identification of an article according to claim 2, wherein the determining whether all the brand names in the first set of brand names have brand names of the determined brand name type according to the text information of the article to be identified comprises:
judging whether a certain word in the determined word set exists in the text information of the object to be identified, and when the certain word exists, determining that all the brand names in the first brand name set have brand names of the determined brand name type, wherein the brand name in front of the certain word is the brand name of the determined brand name type; or,
and judging whether all the brand names in the first brand name set have contained brand names according to the text information of the article to be identified, and when the contained brand names are judged to exist, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, wherein the contained brand names are the brand names of the determined brand name types.
4. The method for brand identification of an article according to any one of claims 1 to 3, wherein after conducting multi-mode matching analysis on the text information of the article to be identified and the identification information of at least one brand in an industry brand library corresponding to the category of the article to be identified, and obtaining a first set of brand names matching the text information of the article to be identified, the method further comprises:
judging whether the number of all the brand names in the first brand name set is larger than a determined brand name number threshold value or not, and executing the operation of judging whether the brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names in the first brand name set or not when the number of all the brand names in the first brand name set is judged to be smaller than or equal to the brand name number threshold value;
the method further comprises the following steps:
when the number of the brand names is judged to be larger than the threshold value, collecting appearance information of the to-be-identified article, and screening brand names matched with the appearance information of the to-be-identified article from all the brand names in the first brand name set;
updating all the brand names in the first brand name set based on the screened brand names to obtain the updated first brand name set, and executing the operation of judging whether the brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names in the first brand name set.
5. The method for brand identification of an item according to any one of claims 1 to 3, wherein before the determining whether there is a brand name matching the acquired brand information of the item to be identified, which is input by the user, among all the brand names in the first set of brand names, the method further comprises:
judging whether all target brand names are main brand names or not, wherein all the target brand names comprise all the brand names of the first brand name set and the acquired brand information of the to-be-identified object input by the user;
when the judgment result is yes, executing the operation of judging whether the brand name matched with the acquired brand information of the article to be identified input by the user exists in all the brand names in the first brand name set;
when the judgment result is negative, performing primary name conversion on the brand names which are not the primary brand name in all the target brand names to obtain a third brand name set of the object to be identified and/or target brand information of the object to be identified, wherein the third brand name set comprises at least one primary brand name;
the judging whether the brand name matched with the brand information of the to-be-identified object input by the user exists in all the brand names of the first brand name set or not comprises the following steps:
and judging whether the brand name matched with the target brand information of the object to be identified exists in all the main brand names in the third brand name set.
6. The method of brand identification of an item according to claim 5, further comprising:
adding the brand information of the article to be identified to the first brand name set when judging that no brand name matched with the brand information of the article to be identified exists;
judging whether the number of all the brand names in the first brand name set is equal to 1, and determining the brand names of the first brand name set as the brand results of the to-be-identified articles when the number of all the brand names in the first brand name set is judged to be equal to 1.
7. The method for brand identification of an item according to claim 6, wherein after the brand information of the item to be identified is added to the first set of brand names and before the determining whether the number of all the brand names in the first set of brand names is equal to 1, the method further comprises:
judging whether all the brand names in the first brand name set have brand names matched with the determined common words or not, and judging whether the brand names matched with the common words are matched with the brand information of the to-be-identified object or not when the brand names matched with the common words are judged to exist;
and when judging that the brand name matched with the common word is matched with the brand information of the article to be identified, executing the operation of judging whether the number of all the brand names in the first brand name set is equal to 1.
8. The method of brand identification of an item according to claim 6 or 7, characterized in that it further comprises:
when the number of all the brand names in the first brand name set is judged to be larger than 1, all the brand names with the brand names being father brand names in the first brand name set are converted into corresponding son brand names, all the father brand names are deleted from the first brand name set, and a fourth brand name set is obtained;
and calculating the number of all the sub-brand names in the fourth brand name set, and determining the brand result of the object to be identified according to the number of all the sub-brand names in the fourth brand name set.
9. The method for brand identification of an article according to claim 8, wherein the determining the brand result of the article to be identified according to the number of all the sub-brand names in the fourth set of brand names comprises:
judging whether the number of all the sub-brand names in the fourth brand name set is equal to 1, and determining the sub-brand names in the fourth brand name set as the brand results of the to-be-identified articles when the number of all the sub-brand names in the fourth brand name set is judged to be equal to 1;
the method further comprises the following steps:
when the number of all the sub-brand names in the fourth brand name set is larger than 1, judging whether the sub-brand names in the fourth brand name set belong to a determined selling platform or not, and when the sub-brand names in the fourth brand name set are judged to exist, deleting the brand names belonging to the selling platform from all the sub-brand names in the fourth brand name set to obtain a fifth brand name set;
and calculating the number of all the sub-brand names in the fifth brand name set, judging whether the number of all the sub-brand names in the fifth brand name set is equal to 1, and determining the sub-brand names in the fifth brand name set as the brand result of the article to be identified when the number of all the sub-brand names in the fifth brand name set is equal to 1.
10. The method of brand identification of an item according to claim 9, further comprising:
when it is judged that no brand name of the selling platform exists, or when it is judged that the number of all sub-brand names in the fifth brand name set is greater than 1, acquiring target information of each brand name in all the brand names in a sixth brand name set within a certain time period, wherein the sixth brand name set is the fourth brand name set or the fifth brand name set, and the ending time of the certain time period is the time when the number of all sub-brand names in the fourth brand name set is greater than 1;
determining the popularity rating of each brand name according to the target information corresponding to each brand name in the sixth brand name set, and determining the brand name corresponding to the target popularity rating with the highest rating from all the brand names in the sixth brand name set as the brand result of the article to be identified according to all the popularity ratings.
11. An apparatus for brand identification of an article, the apparatus comprising:
the matching module is used for carrying out multi-mode matching analysis on the text information of the object to be identified and the identification information of at least one brand in an industry brand library corresponding to the category of the object to be identified when the category of the object to be identified is determined, so as to obtain a first brand name set matched with the text information of the object to be identified, wherein the first brand name set comprises at least one brand name;
the judging module is used for judging whether brand names matched with the acquired brand information of the to-be-identified object input by the user exist in all the brand names of the first brand name set;
and the determining module is used for determining the brand name matched with the brand information of the article to be identified as the brand result of the article to be identified when the judging module judges that the brand name matched with the brand information of the article to be identified exists.
12. The brand recognition device of an article according to claim 11, wherein the determining module is further configured to determine whether there is a brand name of a determined brand name type in all the brand names of the first set of brand names according to text information of the article to be recognized before determining whether there is a brand name matching the obtained brand information of the article to be recognized input by the user in all the brand names of the first set of brand names, and when it is determined that there is no brand name, perform the operation of determining whether there is a brand name matching the obtained brand information of the article to be recognized input by the user in all the brand names of the first set of brand names;
and, the apparatus further comprises:
the first deleting module is used for deleting the brand names with the brand name types from all the brand names in the first brand name set to obtain a second brand name set when the judging module judges that the brand names with the brand name types exist, and triggering the judging module to execute the operation of judging whether the brand names matched with the acquired brand information of the object to be identified, which is input by the user, exist in all the brand names in the first brand name set, wherein the second brand name set is the first brand name set.
13. The brand identification device of an article according to claim 12, wherein the manner in which the determining module determines whether all the brand names in the first set of brand names have the brand name of the determined brand name type according to the text information of the article to be identified is specifically:
judging whether a certain word in the determined word set exists in the text information of the object to be identified, and when the certain word exists, determining that all the brand names in the first brand name set have brand names of the determined brand name type, wherein the brand name in front of the certain word is the brand name of the determined brand name type; or,
and judging whether all the brand names in the first brand name set have contained brand names according to the text information of the article to be identified, and when the contained brand names are judged to exist, determining that all the brand names in the first brand name set have the brand names of the determined brand name types, wherein the contained brand names are the brand names of the determined brand name types.
14. The brand recognition device of an article according to any one of claims 11 to 13, wherein the determining module is further configured to, after the matching module performs multi-mode matching analysis on the text information of the article to be recognized and the identification information of at least one brand in an industry brand library corresponding to the category of the article to be recognized to obtain a first set of brand names matching the text information of the article to be recognized, determine whether the number of all the brand names in the first set of brand names is greater than a determined threshold value of the number of brand names, and when the number of brand names is determined to be less than or equal to the threshold value of the number of brand names, perform the operation of determining whether there is a brand name matching the acquired brand information of the article to be recognized, which is input by the user, in all the brand names in the first set of brand names;
the device further comprises:
the first acquisition module is used for acquiring the appearance information of the article to be identified when the judgment module judges that the number of the brand names is larger than the threshold value;
the screening module is used for screening the brand names matched with the appearance information of the to-be-identified article from all the brand names in the first brand name set;
and the updating module is used for updating all the brand names in the first brand name set based on the screened brand names to obtain the updated first brand name set, and triggering the judging module to execute the operation of judging whether the brand names matched with the acquired brand information of the article to be identified input by the user exist in all the brand names in the first brand name set.
15. The method according to any one of claims 11 to 13, wherein the determining module is further configured to determine whether all target brand names are primary brand names before determining whether there is a brand name matching the obtained brand information of the to-be-identified item input by the user in all the brand names in the first set of brand names, where all the target brand names include all the brand names in the first set of brand names and the obtained brand information of the to-be-identified item input by the user, and when the determining result is yes, perform the operation of determining whether there is a brand name matching the obtained brand information of the to-be-identified item input by the user in all the brand names in the first set of brand names;
and, the apparatus further comprises:
the conversion module is used for carrying out primary name conversion on the brand names which are not the primary brand names in all the target brand names to obtain a third brand name set of the object to be identified and/or target brand information of the object to be identified when the judgment result of the judgment module is negative, wherein the third brand name set comprises at least one primary brand name;
the method for judging whether the brand name matched with the acquired brand information of the article to be identified input by the user exists in all the brand names of the first brand name set by the judging module specifically comprises the following steps:
and judging whether the brand name matched with the target brand information of the object to be identified exists in all the main brand names in the third brand name set.
CN202110643892.7A 2021-06-09 2021-06-09 Method and device for identifying brand of article Active CN113297493B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110643892.7A CN113297493B (en) 2021-06-09 2021-06-09 Method and device for identifying brand of article

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110643892.7A CN113297493B (en) 2021-06-09 2021-06-09 Method and device for identifying brand of article

Publications (2)

Publication Number Publication Date
CN113297493A true CN113297493A (en) 2021-08-24
CN113297493B CN113297493B (en) 2023-01-10

Family

ID=77327853

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110643892.7A Active CN113297493B (en) 2021-06-09 2021-06-09 Method and device for identifying brand of article

Country Status (1)

Country Link
CN (1) CN113297493B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001167128A (en) * 1999-12-07 2001-06-22 Kiyomi Ogushi Identification code of apparel industry and data base
US20170103412A1 (en) * 2015-10-09 2017-04-13 Madison & Sunset (DBA HollyFy) Online entertainment and brand matching
CN109409964A (en) * 2018-11-27 2019-03-01 口碑(上海)信息技术有限公司 The recognition methods of Premium Brands and device
CN109766550A (en) * 2019-01-07 2019-05-17 有米科技股份有限公司 A kind of text brand identification method, identification device and storage medium
CN110750985A (en) * 2018-07-04 2020-02-04 阿里巴巴集团控股有限公司 Brand word recognition method, device, equipment and storage medium
CN111782760A (en) * 2019-05-09 2020-10-16 北京沃东天骏信息技术有限公司 Core product word recognition method, device and equipment

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2001167128A (en) * 1999-12-07 2001-06-22 Kiyomi Ogushi Identification code of apparel industry and data base
US20170103412A1 (en) * 2015-10-09 2017-04-13 Madison & Sunset (DBA HollyFy) Online entertainment and brand matching
CN110750985A (en) * 2018-07-04 2020-02-04 阿里巴巴集团控股有限公司 Brand word recognition method, device, equipment and storage medium
CN109409964A (en) * 2018-11-27 2019-03-01 口碑(上海)信息技术有限公司 The recognition methods of Premium Brands and device
CN109766550A (en) * 2019-01-07 2019-05-17 有米科技股份有限公司 A kind of text brand identification method, identification device and storage medium
CN111782760A (en) * 2019-05-09 2020-10-16 北京沃东天骏信息技术有限公司 Core product word recognition method, device and equipment

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
贾娜: "基于本体的食品投诉文档文本分类研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 *

Also Published As

Publication number Publication date
CN113297493B (en) 2023-01-10

Similar Documents

Publication Publication Date Title
US11341170B2 (en) Automated extraction, inference and normalization of structured attributes for product data
CN112102037B (en) Live E-commerce platform commodity content intelligent pushing management system based on big data
CN108346075B (en) Information recommendation method and device
CN113570413B (en) Advertisement keyword generation method and device, storage medium and electronic equipment
CN104978356B (en) A kind of recognition methods of synonym and device
CN103678335A (en) Method and device for identifying commodity with labels and method for commodity navigation
CN112015721A (en) E-commerce platform storage database optimization method based on big data
CN106649276A (en) Identification method and device for core product word in title
CN105931082B (en) Commodity category keyword extraction method and device
KR20190140701A (en) Hierarchical Category Cluster Based Shopping Basket Associated Recommendation Method
CN113762309A (en) Object matching method, device and equipment
CN111626813A (en) Product recommendation method and system
CN114529359A (en) Shoe money type recommendation system and method based on user characteristics
CN106997350A (en) A kind of method and device of data processing
CN116739626A (en) Commodity data mining processing method and device, electronic equipment and readable medium
CN117151826B (en) Multi-mode electronic commerce commodity alignment method and device, electronic equipment and storage medium
CN113297493B (en) Method and device for identifying brand of article
CN117273865A (en) Commodity recommendation method and device, electronic equipment and storage medium
CN112492342A (en) E-commerce video live broadcast platform based on cloud computing data analysis
CN110968670B (en) Method, device, equipment and storage medium for acquiring attributes of popular commodities
CN111429200B (en) Content association method and device, storage medium and computer equipment
CN113313187B (en) Brand identification method and device based on text information
CN113495987A (en) Data searching method, device, equipment and storage medium
CN113420146A (en) Material brand identification method and device
CN111694928A (en) Data index recommendation method and device, computer equipment and readable storage medium

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