WO2021136009A1 - Procédé et appareil de traitement d'informations de recherche, et dispositif électronique - Google Patents

Procédé et appareil de traitement d'informations de recherche, et dispositif électronique Download PDF

Info

Publication number
WO2021136009A1
WO2021136009A1 PCT/CN2020/138220 CN2020138220W WO2021136009A1 WO 2021136009 A1 WO2021136009 A1 WO 2021136009A1 CN 2020138220 W CN2020138220 W CN 2020138220W WO 2021136009 A1 WO2021136009 A1 WO 2021136009A1
Authority
WO
WIPO (PCT)
Prior art keywords
search information
search
keywords
seller
information
Prior art date
Application number
PCT/CN2020/138220
Other languages
English (en)
Chinese (zh)
Inventor
何佳贤
巩媛媛
曹洁
刘婕
聂一兵
夏明泽
周婷
林晓雯
董双
葛正晗
白鑫
徐磊
施杨斌
Original Assignee
阿里巴巴集团控股有限公司
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 阿里巴巴集团控股有限公司 filed Critical 阿里巴巴集团控股有限公司
Publication of WO2021136009A1 publication Critical patent/WO2021136009A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/3332—Query translation
    • G06F16/3337—Translation of the query language, e.g. Chinese to English
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/3331—Query processing
    • G06F16/334—Query execution
    • G06F16/3344—Query execution using natural language analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
    • G06F16/33—Querying
    • G06F16/338—Presentation of query results
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/279—Recognition of textual entities
    • G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
    • G06F40/295—Named entity recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0623—Item investigation
    • G06Q30/0625—Directed, with specific intent or strategy

Definitions

  • This application relates to the technical field of search information processing, and more specifically, to a search information processing method, search information processing device and electronic equipment.
  • Cross-border e-commerce refers to an international business activity in which transaction entities belonging to different customs borders complete transactions and settle payments through e-commerce platforms, and deliver goods and complete transactions through cross-border logistics.
  • cross-border sellers have major deficiencies in content translation, editing, and management tools, especially for product content in small languages, which basically rely on third-party translation companies or machine translation tools. Therefore, ignorance of local minor languages, cultural customs and habits, and ignorance of local people’s preferences have hindered the further development of these sellers, especially significantly affecting the search results of these sellers’ products.
  • the embodiments of the present application provide a processing method for searching information, a processing device for searching information, and electronic equipment, which obtain keywords for searching information and entities that obtain the keywords, so as to edit first search information based on the entities, thereby Optimized the search effect of search information.
  • a method for processing search information including: obtaining first search information in a target language; obtaining multiple keywords of the first search information; obtaining entities corresponding to the keywords; and, based on the entities
  • the first search information is edited to optimize the search effect to obtain the second search information, and the second search information is the first search information whose search effect has been optimized.
  • the editing for optimizing the search effect includes at least one of the following: adding keywords with excellent search effects; replacing the original keywords with keywords with better search effects ; And, delete keywords with bad search results.
  • obtaining search information in the target language includes: obtaining third search information in the target language, the third search information being search information that has not undergone word order adjustment; determining the category of the third search information; and, The word order of the third search information is adjusted based on the category to obtain the first search information.
  • obtaining the first search information in the target language includes: obtaining fourth search information in the source language, where the fourth search information is search information that has not undergone word order adjustment; and determining the category of the fourth search information; Adjust the word order of the fourth search information based on the category to obtain fifth search information in the source language, the fifth search information being the search information that has undergone word order adjustment; and, translating the fifth search information in the source language into the first search information in the target language Search Information.
  • obtaining the fourth search information in the source language includes: obtaining the sixth search information in the source language, the sixth search information is the search information originally input by the user; and the named entity recognition is performed on the sixth search information to Obtain the entity corresponding to the word in the sixth search information; and remove the word not oriented to the search user in the sixth text information based on the entity corresponding to the word to obtain the fourth search information in the source language.
  • translating the fifth search information in the source language into the first search information in the target language includes: translating the fifth search information in the source language into the target language using a thesaurus with a named entity identification tag First search for information.
  • editing the first search information based on the entity to optimize the search effect to obtain the second search information includes: determining whether the entity is a reference language; and in response to the entity being not a reference language, converting the entity to Reference language; and, editing the first search information in the target language for optimizing the search effect based on the entity of the reference language to obtain the second search information in the target language.
  • the editing for optimizing the search effect further includes at least one of the following: filtering keywords of the first search information based on at least one of time, region, and category; selecting a predetermined key Words to view the attribute data and associated data of keywords; based on at least one of time, region, and category, view one or more hot search words corresponding to the first search information in the target language, and combine one or more hot search At least one hot search word among the words is added to the first search information; and, the keywords are added, modified or deleted in other ways.
  • the first search information processing method after the first search information is edited based on the entity for optimizing the search effect to obtain the second search information, it further includes: determining the popularity of the keyword; and, combining one or more hot searches The popularity of the word is compared with the popularity of the keyword.
  • a processing method for online shopping product titles including: obtaining a first product title in a target language; obtaining multiple keywords of the first product title; obtaining an entity corresponding to the keyword ; And, based on the entity-based editing of the first product title for optimizing the search effect to obtain the second product title, the second product title is the first product title for which the search effect has been optimized.
  • the editing for optimizing search results includes at least one of the following: adding keywords with excellent search effects; replacing the original keywords with better searches Effective keywords; and, delete keywords with bad search effects.
  • the editing for optimizing search results further includes at least one of the following: filtering keywords of the first product title based on at least one of time, region, and category ; Select a predetermined keyword to view the attribute data and related data of the keyword; view one or more hot search terms corresponding to the first product title in the target language based on at least one of time, region, and category, and combine one or At least one of the hot search words is added to the first product title; and, keywords are added, modified or deleted in other ways.
  • a method for processing seller titles includes: obtaining the first seller title in the target language; obtaining multiple keywords of the first seller title; key to at least one of the multiple keywords Words perform named entity recognition to obtain the entity corresponding to at least one keyword; determine the seller's attention corresponding to at least one keyword based on the entity; and edit the first seller title to optimize the seller's attention based on the seller's attention
  • the second seller title is the first seller title that has optimized the seller's attention.
  • the multiple quality evaluation indicators include at least one of the following: whether the first predetermined number of characters in the search information include words in all categories; whether the keywords include hot search words; Whether the total number of words exceeds a predetermined threshold; and whether the keywords contain repeated words.
  • an interactive method for searching information including: receiving a user's input operation on first search information; displaying the first search information to the user; generating second search information with optimized search effect; And, the keywords and their entities of the second search information are displayed to the user.
  • displaying the first search information to the user includes: displaying at least one of a category, a source language, and a target language of the first search information.
  • displaying the keywords and entities of the second search information to the user includes: displaying other words with good search effects.
  • displaying other words with good search effect includes: displaying the popularity index of other words.
  • an apparatus for processing search information including: an information acquisition unit for acquiring first search information in a target language; a keyword acquisition unit for acquiring a plurality of first search information Keywords; entity obtaining unit, used to obtain the entity corresponding to the keyword; and, search editing unit, used to edit the first search information for optimizing the search effect based on the entity to obtain the second search information, the second search information It is the first search information that has optimized search performance.
  • the editing for optimizing the search effect includes at least one of the following: adding keywords with excellent search effects; replacing the original keywords with keywords with better search effects ; And, delete keywords with bad search results.
  • the information acquisition unit is used to: acquire fourth search information in the source language, the fourth search information being search information that has not undergone word order adjustment; determine the category of the fourth search information; adjust based on the category
  • the word order of the fourth search information is to obtain the fifth search information in the source language, the fifth search information is the search information that has undergone word order adjustment; and the fifth search information in the source language is translated into the first search information in the target language.
  • the information acquisition unit translating the fifth search information in the source language into the first search information in the target language includes: translating the fifth search information in the source language into the word database with a named entity identification tag. The first search message in the target language.
  • the search editing unit is used to: determine whether the entity is a reference language; in response to the entity is not a reference language, convert the entity into a reference language; and, based on the entity's first search of the target language in the reference language The information is edited to optimize the search effect to obtain the second search information in the target language.
  • the editing for optimizing the search effect further includes at least one of the following: filtering keywords of the first search information based on at least one of time, region, and category; selecting a predetermined key Words to view the attribute data and associated data of keywords; based on at least one of time, region, and category, view one or more hot search words corresponding to the first search information in the target language, and combine one or more hot search At least one hot search word among the words is added to the first search information; and, the keywords are added, modified or deleted in other ways.
  • search information processing device it further includes: a popularity comparison unit for determining the popularity of the keyword; and comparing the popularity of one or more popular search words with the popularity of the keyword.
  • search information processing device it further includes: a quality determination unit configured to determine the quality of the second search information based on the search effect of the second search information.
  • the editing for optimizing search results includes at least one of the following: adding keywords with excellent search effects; replacing the original keywords with better searches Effective keywords; and, delete keywords with bad search effects.
  • the editing for optimizing search effects further includes at least one of the following: filtering keywords of the first product title based on at least one of time, region, and category ; Select a predetermined keyword to view the attribute data and associated data of the keyword; view one or more hot search terms corresponding to the first product title in the target language based on at least one of time, region, and category, and combine one or At least one of the hot search words is added to the first product title; and, keywords are added, modified or deleted in other ways.
  • a device for processing seller titles including: a seller title obtaining unit for obtaining the first seller title in the target language; and a seller keyword obtaining unit for obtaining the first seller title.
  • Multiple keywords used to perform named entity recognition on at least one of the multiple keywords to obtain the entity corresponding to at least one keyword; attention degree determining unit, used to determine at least one key based on the entity The seller attention degree corresponding to the word; and, the seller title editing unit, used to edit the first seller title based on the seller attention degree for optimizing the seller attention degree to obtain the second seller title, the second seller title is the optimized seller attention The first seller title of the degree.
  • the editing for optimizing seller attention further includes at least one of the following: filtering keywords of the first seller title based on at least one of time, region, and category; selecting a reservation Keywords to view the attribute data and associated data of keywords; based on at least one of time, region, and category to view one or more words with high seller attention corresponding to the first seller title in the target language, and At least one of one or more words with high seller attention is added to the first seller title; and, keywords are added, modified or deleted in other ways.
  • an apparatus for determining the quality of search information including: a search information acquisition unit for acquiring search information with optimized search results; a word and entity acquisition unit for acquiring search information Keywords and their entities; quality index determination unit, used to determine whether to meet multiple quality evaluation indicators of the search information based on search information, keywords and their entities; and a determination result display unit, used to display search information, keywords Its entity and whether it meets the determination results of multiple quality evaluation indicators.
  • the multiple quality evaluation indicators include at least one of the following: whether the first predetermined number of characters of the search information include words in all categories; whether the keywords include hot search words; search information Whether the total number of words in is more than a predetermined threshold; and whether the keywords contain repeated words.
  • the first display unit is configured to display at least one of the category, source language, and target language of the first search information.
  • the second display unit displaying other words with good search effects includes: displaying the popularity index of the other words.
  • a third input unit configured to receive an operation of confirming the third search information by the user after the third search information is displayed.
  • an electronic device including: a processor; and, a memory, in which computer program instructions are stored, and the computer program instructions cause the processor to execute the above-mentioned Search information processing method, online shopping product title processing method, seller title processing method, search information quality determination method, and search information interaction method.
  • a computer-readable medium having computer program instructions stored thereon.
  • the processor executes the above-mentioned search information processing method.
  • Fig. 2 illustrates a flowchart of a method for processing search information according to an embodiment of the present application
  • FIG. 3 illustrates a flowchart of an example of a preprocessing process of the same language in the method for processing search information according to an embodiment of the present application
  • FIG. 4 illustrates a flowchart of an example of a cross-language preprocessing process in a method for processing search information according to an embodiment of the present application
  • Figure 5 illustrates a flowchart of a method for processing online shopping product titles according to an embodiment of the present application
  • FIG. 8 illustrates a schematic diagram of a method for determining the quality of search information according to an embodiment of the present application
  • FIG. 11 illustrates a schematic diagram of multi-terminal interaction of search information according to an embodiment of the present application
  • Fig. 12 illustrates a block diagram of a processing device for searching information according to an embodiment of the present application
  • FIG. 13 illustrates a block diagram of a processing device for online shopping product titles according to an embodiment of the present application
  • FIG. 15 illustrates a block diagram of an apparatus for determining the quality of search information according to an embodiment of the present application
  • Fig. 16 illustrates a block diagram of an interactive device for searching information according to an embodiment of the present application
  • Fig. 17 illustrates a block diagram of an electronic device according to an embodiment of the present application.
  • the hot search terms of the existing titles are also crawled from the data in the keyword analysis tool of the seller's backend on the e-commerce platform.
  • the basic idea of this application is to first obtain the keywords of the search information, and then obtain the entities of the keywords, and then edit the search information based on the entities to optimize the search effect.
  • the search information processing method, search information processing device, and electronic equipment provided in this application first obtain the first search information in the target language, then obtain multiple keywords of the first search information, and then obtain the entity corresponding to the keyword Finally, based on the entity, edit the first search information for optimizing the search effect to obtain the second search information, and the second search information is the first search information whose search effect has been optimized.
  • the search information can be the title of the product sold on the e-commerce platform, or it can be what the user wants to improve the search effect.
  • the commodities mentioned here include not only physical commodities but also service commodities.
  • Fig. 1 illustrates a schematic diagram of an application scenario of a method for processing search information according to an embodiment of the present application.
  • the method for processing search information according to the embodiment of the present application can be applied to a cloud server C, such as a cloud server of a network service provider of a shopping website.
  • the cloud server C receives the unoptimized first search information I1, I2, etc. input by the user U1, the user U2, etc. through their terminal devices, such as a smart phone M1, or a computer M2.
  • the cloud server C processes the unoptimized first search information I1, I2, etc. based on the search information processing method according to the embodiment of the present application, and generates optimized second search information I3, I4, etc.
  • the optimized second search information I3, I4, etc. can be stored in the database DB corresponding to the cloud server C to be used by the users U1, U2, etc. on the shopping website, or can be returned to the users U1, U2 for use in Use on terminal equipment M1, M2, or other terminal equipment.
  • Fig. 2 illustrates a flowchart of a method for processing search information according to an embodiment of the present application.
  • the method for processing search information according to the embodiments of the present application can be applied to various terminal devices of end users or website service providers, such as end users’ smart phones, laptops, desktop computers, etc., and can also be network services.
  • the method for processing search information includes the following steps.
  • Step S110 Acquire first search information in the target language.
  • the first search information in the target language is the search information that the user wants to optimize the search effect. It can be the source search information entered by the user, it can also be the source search information for further processing or translation processing. Search afterwards.
  • the first search information in the target language is a translated text, it will be described in further detail below.
  • the target language is a language used for searching the first search information. For example, if it is applied to an English search engine and the first search information is searched in English, the target language is English.
  • Step S120 Obtain multiple keywords of the first search information.
  • the keyword is the keyword used for searching in the first search information.
  • the keywords are “2019 new style”, “leather leather jacket”, “Women's coat”, “short style”.
  • the keywords of the first search information can be obtained by word segmentation, and the word segmentation is based on the target language of the first search information. For example, if the target language of the first search information is English, English-based word segmentation is performed, and if the target language of the first search information is Chinese, then Chinese-based word segmentation is performed.
  • the multiple keywords of the first search information obtained include not only words with entity meaning, but also more important non-entity meanings. Words, such as particles, prepositions, modal particles, etc. Therefore, in the embodiments of the present application, keywords refer to all words that have an effect on searching for search information, and not only words that have an entity meaning.
  • the identified entities are different.
  • the identified entities may include core product words, category words, attribute words, material words, modifier words, crowd words, marketing words, etc.
  • named entity recognition methods can be used, such as a traditional rule-based method, a statistical method based on a large-scale corpus, or an artificial intelligence learning method based on a deep neural network.
  • named entity recognition does not limit the type of recognized entity, and can include any entity that is helpful for the understanding of search information, that is, used as named entity recognition and recognition in the open domain. classification.
  • each keyword multiple entities corresponding to a certain keyword may also be obtained through named entity recognition.
  • the entity of a certain keyword may represent both the category of the product and the material of the product. Therefore, in the embodiments of the present application, it is not limited that each keyword can only obtain a corresponding entity.
  • Step S140 editing the first search information for optimizing the search effect based on the entity to obtain the second search information, the second search information being the first search information whose search effect has been optimized.
  • the keywords of the first search information and their corresponding entities are obtained.
  • the keywords are obtained as “2019 new style”, “leather leather jacket”, “female jacket”, “short style”, and identify each key
  • the entities corresponding to the words are respectively "promotional words” (ie, 2019 new models), “category words” (ie, leather jackets), “category words” (ie, women's coats), and “attribute words” (ie, short styles).
  • “leather leather” can further identify the entity as a “material word”.
  • the user can conveniently edit the first search information, for example, add, delete or modify its keywords, so as to obtain a second search with better search results.
  • the user can conveniently edit the first search information, for example, add, delete or modify its keywords, so as to obtain a second search with better search results.
  • the user thinks that "2019 new style” as a "promotional word” may limit the search time range, you can modify "2019 new style” to "latest style”.
  • “leather leather clothing” as the “category word” may limit the scope of the searched category, you can modify “leather leather clothing” to “leather” to only indicate the material of the product instead of the category of the product.
  • the product category is indicated by the keyword "women's outerwear”.
  • the method for processing search information obtains the keywords of the search information and obtains the entities corresponding to the keywords, and the first search information can be edited based on the entities, so that the keywords of the first search information It can better match the search attributes and improve the search effect of search information.
  • FIG. 3 illustrates a flowchart of an example of the same language preprocessing process in the search information processing method for optimizing search effects according to an embodiment of the present application.
  • step S110 further includes the following steps.
  • Step S1111 Obtain third search information in the target language, where the third search information is search information that has not undergone word order adjustment.
  • the third search information may be "Ladies short leather jacket 2019 new style”.
  • the search information processing of the search information in the same language is performed, so the third search information before preprocessing is also in the target language.
  • Step S1113 Adjust the word order of the third search information based on the category to obtain the first search information. That is, in this example, only the word order of the third search information is adjusted, so as to further facilitate the subsequent word segmentation and named entity recognition processing for the first search information. For example, for the above-mentioned product title “Ladies Short Leather Leather Jacket 2019 New”, according to the category “Clothing”, specifically “Ladies Jacket”, and further subdivided, "Short Samuel Jacket” , Adjust its word order to "2019 new short female jacket leather leather jacket”.
  • Fig. 4 illustrates a flowchart of an example of a cross-language preprocessing process in a method for processing search information according to an embodiment of the present application.
  • Step S1121 Obtain fourth search information in the source language, where the fourth search information is search information that has not undergone word order adjustment. That is to say, in this example, search information processing for searching information across languages is performed. Therefore, the fourth search information before preprocessing is in a source language different from the target language.
  • the target language can be English
  • the source language can be Chinese.
  • step S1122 the category of the fourth search information is confirmed.
  • the category to which the product belongs is determined to facilitate editing of the fourth search information, which is similar to the step S1112 mentioned above.
  • Step S1123 Adjust the word order of the fourth search information based on the category to obtain the fifth search information in the source language. That is, in this example, by adjusting the word order of the fourth search information, subsequent word segmentation and named entity recognition processing can be facilitated.
  • Step S1124 Translate the fifth search information in the source language into the first search information in the target language.
  • named entity recognition may also be performed on the fourth search information in the source language to optimize the fourth search information.
  • the sixth search information is the search information in the source language before optimization, that is, the search information originally input by the user, such as the Chinese " ⁇ " in Chinese as described above. 2019 new style leather jackets". Then, perform named entity recognition on the sixth search information to obtain the entity corresponding to the word in the sixth search information.
  • the named entity recognition performed on the sixth search information is similar to the named entity recognition described previously, and will not be repeated here. Finally, based on the entity corresponding to the word, the meaningless words in the sixth search information are removed to obtain the fourth search information in the source language.
  • nonsense words refer to words in the sixth search information that are not oriented to search users, such as marketing words that are seen by wholesalers, such as "one-click generation" and so on.
  • the named entity identification tag refers to a tag of a word corresponding to an entity with a specific meaning, such as a tag of core product words, category words, attribute words, and material words.
  • the named entity identification tag may be "women's jacket”, “leather leather jacket”, “short style” and so on.
  • translating the fifth search information in the source language into the first search information in the target language includes: translating the first search information in the source language with a thesaurus with a named entity identification tag. 5.
  • the search information is translated into the first search information in the target language.
  • the entity corresponding to the keyword of the first search information can be translated into any language to facilitate the user to edit. Specifically, it can be determined whether the entity is a reference language, such as the native language of the user, so as to translate the entity into the reference language.
  • editing the first search information based on the entity for optimizing the search effect to obtain the second search information includes: determining whether the entity is a reference language; and responding to the fact that the entity is not The reference language converts the entity into the reference language; and the entity based on the reference language edits the first search information in the target language for optimizing the search effect to obtain the second search information in the target language.
  • the first search information can be edited to optimize the search effect.
  • the editing of the first search information may include the following operations.
  • the keywords of the first search information can be filtered based on time, region, category, etc. For example, if the product expects to emphasize the new style of the current year, you can filter the keywords with the year of the current year. Or, if the product is sold in a specific area, you can filter keywords based on the area.
  • the search information processing method for optimizing search results according to the embodiments of the present application can be based on massive data, such as e-commerce field data, so as to obtain attribute data and associated data of each keyword, so that users can better grasp the key.
  • massive data such as e-commerce field data
  • the attribute and related data of the keyword "short style” indicates that it belongs to the style of clothing, and the related data Shows the specific popular data of this style at present. If based on attribute data and associated data, the user determines that the style may limit the search effect, then the keyword "short style" can be removed.
  • one or more hot search words corresponding to the first search information in the target language can be viewed based on at least one of time, region, and category. That is, due to the local language, cultural customs, preferences of local people, etc., for the first search information in the target language, there may be hot search terms that the user does not understand. Therefore, in the embodiment of the present application, the user is provided with the function of viewing the hot search terms corresponding to the first search information in the target language.
  • the popularity value of the hot search words can be displayed at the same time, such as search popularity, search index, click-through rate, payment conversion rate, competition index, etc.
  • the above-mentioned indexes of the keywords in the first search information such as search popularity, search index, click-through rate, payment conversion rate, competition index, etc., can also be obtained, and compared with the above-mentioned indexes of hot search words.
  • the translation of one or more hot search words and the corresponding entities obtained through named entity recognition can be displayed at the same time, so as to facilitate Users understand the hot search terms better.
  • the user can further add one or some hot search terms to the first search information, so as to improve the search effect of the search information through the hot search terms with traffic attributes.
  • hot search term “same style fur collar”
  • users can add the hot search term "same style fur collar” to the search information "2019 new leather leather jacket short style” ", for example, edit it as "2019 new leather and leather women's jacket short with the same fur collar” to improve the search effect.
  • the editing for optimizing the search effect further includes at least one of the following: a key to the first search information based on at least one of time, region, and category Word filtering; select a predetermined keyword to view the attribute data and related data of the keyword; view one or more hot search words corresponding to the first search information in the target language based on at least one of time, region, and category, and Add at least one of the one or more hot search words to the first search information; and, add, modify, or delete keywords in other ways.
  • the processing method after the first search information is edited based on the entity for optimizing the search effect to obtain the second search information, it further includes: determining the popularity of the keyword; and comparing the popularity of one or more hot search terms with Compare the popularity of keywords.
  • the quality of the second search information can be further determined, which will be described in further detail below.
  • the method for processing the product title of online shopping includes: S210, acquiring the first product title in the target language; S220, acquiring multiple keywords of the first product title; S230, acquiring The entity corresponding to the keyword; and, S240, edit the first product title to optimize the search effect based on the entity to obtain the second product title, which is the first product title whose search effect has been optimized.
  • the search information in the second exemplary method is the product title of online shopping.
  • other details in the second exemplary method are different from those in the second exemplary method.
  • the corresponding details in the first exemplary method are completely the same, and will not be repeated in order to avoid redundancy.
  • the method for processing the seller title includes: S310, acquiring the first seller title in the target language; S320, acquiring multiple keywords of the first seller title; S330, applying to multiple keywords Perform named entity recognition on at least one keyword in at least one keyword to obtain the entity corresponding to at least one keyword; S340, determine the seller's attention corresponding to at least one keyword based on the entity; and, S350, use the first seller title based on the seller's attention
  • the editor for optimizing the seller's attention to obtain the second seller title, which is the first seller's title that has optimized the seller's attention.
  • FIG. 8 illustrates a schematic diagram of a method for determining the quality of search information according to an embodiment of the present application.
  • search information that has optimized search results, such as product titles.
  • obtain the keywords and their entities in the product title such as the keywords “manuel”, “polipasto”, and “cadena” shown in Figure 8, and their entities “style”, “others”, and “category”.
  • the translations of the keywords and their entities can be displayed, so that users can clearly understand the meaning of the keywords and their entities, which can be called "Title Perspective".
  • search information based on the search information, keywords and their entities, it can be determined whether the search information meets multiple quality evaluation indicators.
  • a series of evaluation indicators can be used to determine its quality, for example, whether the first 50 characters contain all category subjects (that is, the category to which the product belongs, such as the image 8 "chain", "3m winch"), whether it contains hot search words, whether the number of characters exceeds the limit, whether it contains repeated words, etc.
  • the multiple quality evaluation indicators include at least one of the following: whether the first predetermined number of characters of the search information contains words of all categories; the key Whether the word contains hot search words; whether the total number of words in the search information exceeds a predetermined threshold; and whether the keywords contain repeated words.
  • the quality of the second search information that has been optimized for the search effect as described above can be determined.
  • the quality refers to the second search information relative to the search effect.
  • Text quality for example, the first predetermined number of characters in the search information contains words of all categories, the keywords include hot search words, the total number of search information does not exceed a predetermined threshold, and the keywords do not contain repeated words are all helpful to improve the search
  • the search effect of information thereby indicating that the quality of the search information is high.
  • Fig. 9 illustrates a flowchart of a method for searching for information according to an embodiment of the present application.
  • the search information interaction method includes: S510, receiving a user's input operation on the first search information; S520, displaying the first search information to the user; S530, generating an optimized search effect The second search information; and, S540, display the keywords and their entities of the second search information to the user.
  • Fig. 10 illustrates a schematic diagram of a search information interaction method according to an embodiment of the present application.
  • the search information interaction method is used to optimize the search effect of the title of the commodity sold on the cross-border e-commerce platform.
  • the user enters the first search information, which is the original title of the product.
  • the category, source language, and target language of the product can be determined.
  • the category, source language, and target language of the product can also be input by the user.
  • the title of the product is "Free shipping and return 2018 autumn and winter new leather leather jacket women's short slim high waist sheep leather motorcycle small wallet plus cashmere”
  • the category is "Fashion>Women's Outift)” and the language direction is from Chinese to English.
  • displaying the first search information to the user includes: displaying at least one of a category, a source language, and a target language of the first search information.
  • an optimized target title can be generated, "2018 New Leather Coats Women's Jackets Short Slims High Waist Sheep Skin Locomotive Small Wallet Slim High Waist Sheep Skin Locomotive Wallet Kiga Keke Cashmere”.
  • each keyword, corresponding Chinese translation and entity can be displayed through the "title perspective" as described above.
  • “2018New” is translated as “2018 new”, and its entity is a marketing word
  • “Leather Coats” is translated as “leather leather”, and its entity is a category word
  • “Short Slims” is translated as “ ⁇ ”, Its entity is an attribute word
  • “Sheep Skin” is translated as “Sheep Skin” and its entity is a material word
  • “Locomotive” is translated as “Locomotive” and its entity is a modifier, etc.
  • the search information processing method can search for local hot search words and add them to the title, and can further classify the hot search words, for example, as shown in Figure 10, the classification is " "Hot search words”, “sales words”, “long-tail words”, “attribute words”, “promotional words”, “brand words”, etc.
  • the classification is " "Hot search words”, “sales words”, “long-tail words”, “attribute words”, “promotional words”, "brand words”, etc.
  • its translation, search popularity, competition index and other hot indicators can be further displayed, so that users can determine whether to add it to the title.
  • displaying the keywords and entities of the second search information to the user includes: displaying other words with good search effects.
  • displaying other words with good search effect includes displaying the popularity index of other words.
  • the user can choose to add one or more words to the product title. That is, in the search information interaction method according to the embodiment of the present application, after displaying the keyword and the entity of the second search information to the user, it further includes: receiving an operation of adding at least one of the other words by the user; A word is added to the second search information to obtain the third search information; and, the third search information is displayed.
  • the search information interaction method after displaying the third search information, it further includes: receiving an operation of the user confirming the third search information.
  • the above-mentioned search information interaction method after receiving the user's confirmation of the third search information operation, it further includes: receiving the user's input operation of the fourth search information.
  • the rewriting and translation of the product title and the optimization process for the search effect can be unified, which is convenient for the user's operation.
  • it can help sellers operate product titles in various languages for products of different stages and attributes.
  • Fig. 11 illustrates a schematic diagram of multi-terminal interaction of search information according to an embodiment of the present application.
  • the user's terminal equipment UE receives first search information from the user, for example, the title of the product to be optimized by the user. Then, in S602, the first search information input by the user is displayed to the user on the terminal device UE.
  • the processing procedure on the server S is the same as the first exemplary method described previously.
  • the first search information may be translated and sorted in S604 first.
  • S605 perform word segmentation on the first search information to obtain keywords.
  • S606 the entity corresponding to the keyword is obtained through named entity recognition.
  • the keywords and their entities can be sent from the server S to the terminal device UE, and in S608, the keywords and corresponding keywords of the first search information can be displayed on the terminal device UE. Entity.
  • the terminal device UE receives the user's editing operations, such as adding, deleting and modifying keywords, and in S610, sends the user's editing results to the server S.
  • the server S edits the first search information based on the user's edit result to generate second search information. Then, in S612, the second search information is sent from the server S to the terminal device UE. And in S613, the second search information that optimizes the search effect is displayed on the terminal device UE.
  • Fig. 12 illustrates a block diagram of a processing device for searching information according to an embodiment of the present application.
  • the search information processing device 700 includes: an information acquisition unit 710, configured to acquire first search information in a target language; a keyword acquisition unit 720, configured to acquire first search information A plurality of keywords; the entity obtaining unit 730 is used to obtain the entity corresponding to the keyword; and the search editing unit 740 is used to edit the first search information for optimizing the search effect based on the entity to obtain the second search information,
  • the second search information is the first search information whose search effect has been optimized.
  • the editing for optimizing the search effect includes at least one of the following: adding keywords with excellent search effects; replacing the original keywords with better ones. Keywords with search effects; and, delete keywords with bad search effects.
  • the information obtaining unit 710 is configured to: obtain the third search information of the target language, the third search information is search information that has not undergone word order adjustment; and the third search information is determined Category; and adjusting the word order of the third search information based on the category to obtain the first search information.
  • the information acquisition unit 710 is configured to: acquire fourth search information in the source language, the fourth search information being search information that has not undergone word order adjustment; and determining the fourth search information Category; adjust the word order of the fourth search information based on the category to obtain the fifth search information in the source language, the fifth search information is the search information that has been adjusted in the word order; and, the fifth search information in the source language is translated into the target language The first search information.
  • the information obtaining unit 710 obtaining the fourth search information in the source language includes: obtaining sixth search information in the source language, where the sixth search information is the search information originally input by the user; Perform named entity recognition on the sixth search information to obtain the entity corresponding to the word in the sixth search information; and remove the non-searching user-oriented words in the sixth text information based on the entity corresponding to the word to obtain a fourth search in the source language information.
  • the information acquisition unit 710 translates the fifth search information in the source language into the first search information in the target language, including: translating the source language into a thesaurus with a named entity identification tag.
  • the fifth search information is translated into the first search information in the target language.
  • the search editing unit 740 is configured to: determine whether the entity is a reference language; in response to the entity is not a reference language, convert the entity into a reference language; and, based on the reference language entity The first search information in the target language is edited to optimize the search effect to obtain the second search information in the target language.
  • the editing for optimizing the search effect further includes at least one of the following: a keyword of the first search information based on at least one of time, region, and category Filter; select a predetermined keyword to view the attribute data and related data of the keyword; view one or more hot search words corresponding to the first search information in the target language based on at least one of time, region, and category, and At least one of the one or more hot search words is added to the first search information; and, keywords are added, modified or deleted in other ways.
  • the above-mentioned search information processing device 700 further includes: a popularity comparison unit for determining the popularity of a keyword; and comparing the popularity of one or more popular search words with the popularity of the keyword.
  • Fig. 13 illustrates a block diagram of a processing device for online shopping product titles according to an embodiment of the present application.
  • Fig. 14 illustrates a block diagram of an apparatus for processing seller titles according to an embodiment of the present application.
  • the apparatus 900 for processing a seller title includes: a seller title obtaining unit 910 for obtaining the first seller title in the target language; and a seller keyword obtaining unit 920 for obtaining the first seller Multiple keywords of the title; named entity recognition unit 930, used to perform named entity recognition on at least one of the multiple keywords to obtain the entity corresponding to the at least one keyword; attention degree determining unit 940, used to determine entity-based Determine the seller attention degree corresponding to at least one keyword; and, the seller title editing unit 950, configured to edit the first seller title based on the seller attention degree for optimizing the seller attention degree to obtain the second seller title, the second seller title It is the first seller title that has optimized seller attention.
  • the editing for optimizing seller attention includes at least one of the following: adding keywords with higher seller attention; replacing original keywords with Keywords with higher seller attention; and, delete keywords with lower seller attention.
  • the editing for optimizing seller attention further includes at least one of the following: the key to the first seller title based on at least one of time, region, and category Filter by word; select a predetermined keyword to view the attribute data and related data of the keyword; view one or more sellers with high seller attention corresponding to the first seller title in the target language based on at least one of time, region, and category And add at least one of the one or more words with higher seller attention to the first seller title; and, add, modify, or delete keywords in other ways.
  • FIG. 15 illustrates a block diagram of an apparatus for determining the quality of search information according to an embodiment of the present application.
  • the device 1000 for determining the quality of search information includes: a search information obtaining unit 1010, configured to obtain search information that has optimized search results; and a word and entity obtaining unit 1020, configured to obtain search information. Keywords and their entities in the information; a quality index determination unit 1030 for determining whether they meet multiple quality evaluation indicators of the search information based on the search information, keywords and their entities; and a determination result display unit 1040 for displaying the search Information, keywords and their entities, and the results of determining whether they meet multiple quality evaluation indicators.
  • the multiple quality evaluation indicators include at least one of the following: whether the first predetermined number of characters of the search information include words in all categories; whether keywords Contains popular search words; whether the total number of words in the search information exceeds a predetermined threshold; and whether the keywords contain repeated words.
  • the interactive device 1100 for searching information includes: a first input unit 1110, configured to receive a user's input operation on the first search information; and a first display unit 1120, configured to display to the user The first search information; the first generating unit 1130 is used to generate the second search information with optimized search effect; and the second display unit 1140 is used to display the keywords and their entities of the second search information to the user.
  • the second display unit 1140 is used to display other words with good search effects.
  • the second display unit 1140 displaying other words with good search effects includes: displaying the popularity index of the other words.
  • the above-mentioned interactive device for searching information 1100 further includes: a second input unit, configured to receive at least one of other words added by the user after displaying the keywords and their entities of the second search information to the user Operation of words; a word adding unit for adding at least one word to the second search information to obtain the third search information; and a third display unit for displaying the third search information.
  • the above-mentioned interactive device for searching information 1100 further includes: a third input unit, configured to receive an operation of confirming the third search information by the user after the third search information is displayed.
  • the above-mentioned interactive device 1100 for searching information further includes: a fourth input unit, configured to receive an input operation of the fourth search information by the user after the user confirms the operation of the third search information.
  • each of the exemplary apparatus 1 to the exemplary apparatus 5 can be implemented in various terminal devices, such as a server used to run an e-commerce platform.
  • each device according to the embodiment of the present application can be integrated into the terminal device as a software module and/or hardware module.
  • each device may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, each device may also be one of many hardware modules of the terminal device.
  • Fig. 17 illustrates a block diagram of an electronic device according to an embodiment of the present application.
  • the electronic device 10 includes one or more processors 11 and memory 12.
  • the processor 11 may be a central processing unit (CPU) or other form of processing unit with data processing capability and/or instruction execution capability, and may control other components in the electronic device 10 to perform desired functions.
  • CPU central processing unit
  • the processor 11 may control other components in the electronic device 10 to perform desired functions.
  • the memory 12 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and/or non-volatile memory.
  • Volatile memory may include random access memory (RAM) and/or cache memory (cache), for example.
  • the non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, and the like.
  • One or more computer program instructions can be stored on a computer-readable storage medium, and the processor 11 can run the program instructions to implement the search information processing methods and online shopping product titles of the various embodiments of the application described above.
  • search information, keywords, entities, etc. can also be stored.
  • the input device 13 may include, for example, a keyboard, a mouse, and so on.
  • the output device 14 can output various information to the outside, including the second search information that optimizes the search effect and so on.
  • the output device 14 may include, for example, a display, a speaker, a printer, a communication network and a remote output device connected to it, and so on.
  • the electronic device 10 may also include any other appropriate components.
  • the embodiments of the present application may also be computer program products, which include computer program instructions.
  • the processor executes what is described in the “exemplary method” section of this specification. Steps in the search information processing method, online shopping product title processing method, seller title processing method, search information quality determination method, and search information interaction method according to various embodiments of the present application.
  • the computer program product can be written in any combination of one or more programming languages to write program codes for performing the operations of the embodiments of the present application.
  • the programming languages include object-oriented programming languages, such as Java, C++, etc., as well as conventional
  • the program code can be executed entirely on the first user computing device, partly executed on the first user device, executed as an independent software package, partly executed on the first user computing device and partly executed on the remote computing device, or entirely Execute on a remote computing device or server.
  • embodiments of the present application may also be computer-readable storage media on which computer program instructions are stored.
  • the processor executes the basis described in the “exemplary method” section of this specification. The steps in the search information processing method, online shopping product title processing method, seller title processing method, search information quality determination method, and search information interaction method in various embodiments of the present application.
  • the computer-readable storage medium may adopt any combination of one or more readable media.
  • the readable medium may be a readable signal medium or a readable storage medium.
  • the readable storage medium may include, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or a combination of any of the above, for example. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable Type programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above.
  • each component or each step can be decomposed and/or recombined.
  • decompositions and/or recombinations shall be regarded as equivalent solutions of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Computational Linguistics (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Databases & Information Systems (AREA)
  • Accounting & Taxation (AREA)
  • Finance (AREA)
  • Artificial Intelligence (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • General Health & Medical Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

L'invention concerne un procédé et un appareil de traitement d'informations de recherche, et un dispositif électronique. Le procédé de traitement d'informations de recherche consiste à : acquérir des premières informations de recherche dans une langue cible (S110) ; obtenir une pluralité de mots-clés liés aux premières informations de recherche (S120) ; obtenir une entité correspondant aux mots-clés (S130) ; et éditer, sur la base de l'entité, les premières informations de recherche pour optimiser un effet de recherche afin d'obtenir des secondes informations de recherche, les secondes informations de recherche étant les premières informations de recherche présentant l'effet de recherche optimisé (S140). De cette manière, l'effet de recherche d'informations de recherche est optimisé.
PCT/CN2020/138220 2019-12-31 2020-12-22 Procédé et appareil de traitement d'informations de recherche, et dispositif électronique WO2021136009A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911411521.5 2019-12-31
CN201911411521.5A CN113127597B (zh) 2019-12-31 2019-12-31 搜索信息的处理方法、装置及电子设备

Publications (1)

Publication Number Publication Date
WO2021136009A1 true WO2021136009A1 (fr) 2021-07-08

Family

ID=76687368

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/138220 WO2021136009A1 (fr) 2019-12-31 2020-12-22 Procédé et appareil de traitement d'informations de recherche, et dispositif électronique

Country Status (2)

Country Link
CN (1) CN113127597B (fr)
WO (1) WO2021136009A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113657099A (zh) * 2021-08-13 2021-11-16 华世界数字科技(深圳)有限公司 一种商品标题生成方法、装置、智能设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060206479A1 (en) * 2005-03-10 2006-09-14 Efficient Frontier Keyword effectiveness prediction method and apparatus
CN103593410A (zh) * 2013-10-22 2014-02-19 上海交通大学 通过替换概念性词语进行搜索推荐系统
CN104615755A (zh) * 2015-02-12 2015-05-13 北京航空航天大学 一种基于众包的新型问答系统
CN106294358A (zh) * 2015-05-14 2017-01-04 北京大学 一种信息的检索方法及系统
CN106557480A (zh) * 2015-09-25 2017-04-05 阿里巴巴集团控股有限公司 查询改写的实现方法及装置
CN109033161A (zh) * 2018-06-19 2018-12-18 深圳市元征科技股份有限公司 一种数据处理方法、服务器及计算机可读介质

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103425664A (zh) * 2012-05-16 2013-12-04 阿里巴巴集团控股有限公司 一种对实体数据单元进行搜索、显示的方法和设备
US10402407B2 (en) * 2013-06-17 2019-09-03 Lenovo (Singapore) Pte. Ltd. Contextual smart tags for content retrieval
CN105677725A (zh) * 2015-12-30 2016-06-15 南京途牛科技有限公司 一种用于旅游垂直搜索引擎的前置解析方法
US10671653B2 (en) * 2016-02-18 2020-06-02 Adobe Inc. Analyzing search queries to provide potential search query modifications via interactive user-interfaces
CN109213920A (zh) * 2017-06-29 2019-01-15 阿里巴巴集团控股有限公司 搜索方法、客户端、服务器和存储介质
CN110399548A (zh) * 2018-04-20 2019-11-01 北京搜狗科技发展有限公司 一种搜索处理方法、装置、电子设备以及存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060206479A1 (en) * 2005-03-10 2006-09-14 Efficient Frontier Keyword effectiveness prediction method and apparatus
CN103593410A (zh) * 2013-10-22 2014-02-19 上海交通大学 通过替换概念性词语进行搜索推荐系统
CN104615755A (zh) * 2015-02-12 2015-05-13 北京航空航天大学 一种基于众包的新型问答系统
CN106294358A (zh) * 2015-05-14 2017-01-04 北京大学 一种信息的检索方法及系统
CN106557480A (zh) * 2015-09-25 2017-04-05 阿里巴巴集团控股有限公司 查询改写的实现方法及装置
CN109033161A (zh) * 2018-06-19 2018-12-18 深圳市元征科技股份有限公司 一种数据处理方法、服务器及计算机可读介质

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113657099A (zh) * 2021-08-13 2021-11-16 华世界数字科技(深圳)有限公司 一种商品标题生成方法、装置、智能设备及存储介质

Also Published As

Publication number Publication date
CN113127597A (zh) 2021-07-16
CN113127597B (zh) 2024-07-05

Similar Documents

Publication Publication Date Title
US11341170B2 (en) Automated extraction, inference and normalization of structured attributes for product data
CN111177569B (zh) 基于人工智能的推荐处理方法、装置及设备
Wijeratne et al. Emojinet: An open service and api for emoji sense discovery
US11200259B2 (en) System and method for processing contract documents
US9898773B2 (en) Multilingual content based recommendation system
US20240028651A1 (en) System and method for processing documents
US20190220490A1 (en) Combining website characteristics in an automatically generated website
US20180260698A1 (en) Recurrent neural network architectures which provide text describing images
CN109446341A (zh) 知识图谱的构建方法及装置
US11023503B2 (en) Suggesting text in an electronic document
US11182540B2 (en) Passively suggesting text in an electronic document
Yang et al. A decision method for online purchases considering dynamic information preference based on sentiment orientation classification and discrete DIFWA operators
US11481722B2 (en) Automated extraction, inference and normalization of structured attributes for product data
CN113919336A (zh) 基于深度学习的文章生成方法、装置及相关设备
Wong et al. Learning to extract and summarize hot item features from multiple auction web sites
CN115309885A (zh) 一种用于科技服务的知识图谱构建、检索和可视化方法及系统
CN108470289B (zh) 基于电商购物平台的虚拟物品发放方法及设备
WO2021136009A1 (fr) Procédé et appareil de traitement d'informations de recherche, et dispositif électronique
US20210271637A1 (en) Creating descriptors for business analytics applications
CN117420998A (zh) 一种客户端ui交互组件生成方法、装置、终端及介质
CN111858930A (zh) 一种社交电商用户画像的建立方法
US20210263915A1 (en) Search Text Generation System and Search Text Generation Method
CN110413795A (zh) 一种数据驱动的专业知识图谱构建方法
Rudniy et al. Trend tracking tools for the fashion industry: the impact of social media
JP7117168B2 (ja) 情報処理装置および情報処理方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20911148

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20911148

Country of ref document: EP

Kind code of ref document: A1