WO2019103380A1 - Appareil destiné à générer une histoire basée sur un article d'actualité, procédé associé, et support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre ce procédé - Google Patents

Appareil destiné à générer une histoire basée sur un article d'actualité, procédé associé, et support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre ce procédé Download PDF

Info

Publication number
WO2019103380A1
WO2019103380A1 PCT/KR2018/013748 KR2018013748W WO2019103380A1 WO 2019103380 A1 WO2019103380 A1 WO 2019103380A1 KR 2018013748 W KR2018013748 W KR 2018013748W WO 2019103380 A1 WO2019103380 A1 WO 2019103380A1
Authority
WO
WIPO (PCT)
Prior art keywords
texts
text
main
story
representative value
Prior art date
Application number
PCT/KR2018/013748
Other languages
English (en)
Korean (ko)
Inventor
장달원
이종설
Original Assignee
전자부품연구원
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 전자부품연구원 filed Critical 전자부품연구원
Publication of WO2019103380A1 publication Critical patent/WO2019103380A1/fr

Links

Images

Classifications

    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0241—Advertisements
    • G06Q30/0251—Targeted advertisements
    • G06Q30/0255—Targeted advertisements based on user history
    • 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
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services

Definitions

  • the present invention relates to a device for collecting news articles and generating a story based on the collected news articles, a method therefor, and a computer readable recording medium storing a program for performing the method
  • the present invention relates to a recording medium.
  • An object of the present invention is to provide a device for collecting a news article, analyzing the collected news article and generating a story from the analyzed news article, a method therefor, and a computer-readable recording medium on which a program for performing the method is recorded .
  • an apparatus for generating a story comprising: a plurality of news articles; a list of tags attached to the text; A text processing module for loading a plurality of texts linked with the selected tag when at least one tag among the plurality of tags in the tag list is selected; A text classification module for classifying a plurality of texts belonging to a cluster including a number of texts into a plurality of main texts and dividing a plurality of texts belonging to remaining clusters into a plurality of additional texts; A main text processing module for extracting a representative value of each main text, A supplementary text processing module for extracting a representative value of any one additional text among the supplementary texts, a representative value of each of the plurality of extracted main texts and a representative value of the extracted supplementary text, And a story generation module for generating story information.
  • the main text processing module detects duplicate texts in the plurality of main texts and performs deduplication for eliminating duplicated texts and generates a plurality of clusters by clustering the plurality of main texts according to contents of text, And the center of the cluster is selected as a representative value from the generated plurality of clusters.
  • the additional text processing module detects duplicate texts in the plurality of additional texts to perform duplication elimination for eliminating duplicated texts, and generates the plurality of clusters by clustering the plurality of additional texts according to the contents of the text, A cluster including the largest number of texts among the plurality of clusters is selected, and the center of the cluster is selected as a representative value from the plurality of generated clusters.
  • the apparatus comprising: a communication unit for communicating with a web server that serves a web page including a news article; and a web page downloading unit that accesses the web server through the communication unit to download a web page including the news article, And a text collection module for generating a list of texts of a plurality of news articles, a list of tags assigned to the text, and a list of texts associated with the tags, and storing the list in the storage unit.
  • a method of generating a story comprising: storing a text of a plurality of news articles, a list of tags assigned to the text, Loading a plurality of texts associated with the selected tag when at least one tag among the plurality of tags in the tag list is selected; clustering the loaded plurality of texts according to contents; Classifying a plurality of texts belonging to a cluster including text into a plurality of main texts and dividing a plurality of texts belonging to remaining clusters into a plurality of additional texts; A step of extracting a representative value, A step of extracting the representative value of the host, the extracted representative value of the plurality of the main text of each representative value, and the extracted one additional text, arranged according to the chronological order, and a step of generating a story.
  • the step of extracting the representative value of each of the plurality of main texts comprises the steps of: detecting redundant texts in the plurality of main texts to perform deduplication for eliminating duplicated texts; Generating a plurality of clusters by clustering; and selecting a center of the cluster from among the generated plurality of clusters as a representative value.
  • the step of extracting a representative value of the one additional text may further include the steps of performing duplicate removal to delete duplicate text by detecting duplicate text in the plurality of additional texts, Generating a plurality of clusters, selecting a cluster including the largest number of texts among the plurality of clusters, and selecting a center of the cluster from among the generated plurality of clusters as a representative value.
  • the storing step comprises the steps of: downloading a web page including the news article by accessing a web server that serves a web page including a news article; extracting text from the web page, And generating and storing a list of tags attached to the text and a list of text associated with the tags.
  • various news articles appear on the basis of Internet news, and the main article and the non-main article are divided and processed.
  • you create a story by finding out non-general texts and supplementary texts and managing them separately, you insert a part of supplementary text so that a rich story can come out.
  • a story based on news can be recommended to a user rather than recommending only one news.
  • FIG. 1 is a block diagram illustrating a configuration of an apparatus for generating a story according to an embodiment of the present invention.
  • FIG. 2 is a block diagram illustrating a detailed configuration of a controller of a device for generating a story according to an embodiment of the present invention.
  • FIG. 3 is a flowchart illustrating a method of collecting basic data for generating a story according to an embodiment of the present invention.
  • FIG. 4 is a flowchart illustrating a method for generating a story based on a news article according to an embodiment of the present invention.
  • an apparatus 100 for generating a story according to an embodiment of the present invention is basically a device for performing computing operations, And the like.
  • the story device 100 may be a personal computer, a notebook, a workstation, a smart phone, a tablet PC, or the like.
  • the story generation apparatus 100 includes a communication unit 110, an input unit 120, a display unit 130, a storage unit 140, and a control unit 150.
  • the communication unit 110 is a means for accessing a web server that provides a news service via a network in a wired or wireless manner.
  • the communication unit 110 can access the web server and download news from the web server.
  • the communication unit 110 includes an RF (Radio Frequency) transmitter (Tx) for up-converting and amplifying the frequency of a transmitted signal, a RF receiver (Rx) for low-noise amplifying the received signal and down- And a modem for modulating the transmitted signal and demodulating the received signal.
  • the communication unit 110 may include a modem for modulating a signal transmitted according to a protocol for wired communication and demodulating a received signal for wired communication.
  • the communication unit 110 can access a web server providing a news service under the control of the control unit 150 and download a web page including news. Accordingly, the control unit 150 can crawl or scrape news from the web page.
  • the input unit 120 receives a user's key operation for controlling the story device 100, generates an input signal, and transmits the generated input signal to the control unit 150.
  • the input unit 120 may include various kinds of keys for controlling the story device 100.
  • the display unit 130 is a touch screen, the functions of the various keys can be performed in the display unit 130. In the case where all functions can be performed only by the touch screen, the input unit 120 is omitted It is possible.
  • the display unit 130 visually provides a menu of the story device 100, input data, function setting information, and various other information to the user.
  • the display unit 130 displays a screen such as a boot screen, a standby screen, a menu screen, and the like of the story device 100.
  • the display unit 130 outputs a meter reading image according to an embodiment of the present invention on a screen.
  • the display unit 130 may include a liquid crystal display (LCD), an organic light emitting diode (OLED), and an active matrix organic light emitting diode (AMOLED).
  • the display unit 130 may be implemented as a touch screen.
  • the display unit 130 may include a touch sensor. The touch sensor senses the user's touch input.
  • the touch sensor may be constituted by a touch sensing sensor such as a capacitive overlay, a pressure type, a resistive overlay, or an infrared beam, or may be constituted by a pressure sensor .
  • a touch sensing sensor such as a capacitive overlay, a pressure type, a resistive overlay, or an infrared beam
  • a pressure sensor a sensor that can be used as the touch sensor of the present invention.
  • the touch sensor senses the touch input of the user, generates a sensing signal, and transmits the sensing signal to the controller 150.
  • the display unit 130 is a touch screen, some or all of the functions of the input unit 120 may be performed through the display unit 130.
  • the storage unit 140 plays a role of storing programs and data necessary for the operation of the story apparatus 100.
  • the storage unit 140 is an area in which user data generated according to use of the story device 100, for example, usage amount of gas, water, electricity, and images photographed for meter reading are stored.
  • Each kind of data stored in the storage unit 140 can be deleted, changed or added according to a user's operation.
  • the control unit 150 may control the overall operation of the story device 100 and the signal flow between the internal blocks of the story device 100 and may perform a data processing function for processing the data. In addition, the controller 150 basically controls various functions of the story device 100.
  • the control unit 150 may be a central processing unit (CPU), a digital signal processor (DSP), or the like.
  • the control unit 150 includes a text acquisition module 210, a text processing module 220, a text classification module 230, a main text processing module 240, an additional text processing module 250, Module 260, as shown in FIG.
  • the text collection module 210 is for collecting basic data for generating a story according to an embodiment of the present invention. More specifically, the text acquisition module 210 accesses a web server that serves a web page including a news article through the communication unit 110, and downloads a web page including a news article. The text collection module 210 then extracts the text from the web page containing the news article, analyzes the content of the extracted text, and tags each text. For example, a tag can be a subject, a material, an issue, a person, a keyword, and the like. The text collection module 210 creates a list of tags and a text list associated with each tag.
  • the text collection module 210 accumulates the tag list, the text list and the text in the storage unit 140.
  • the text processing module 220 selects at least one tag among the plurality of tags in the tag list. At this time, the text processing module 220 can sequentially select tags one by one from the tags associated with the largest number of texts in the tag list. When the tag is selected, the text processing module 220 loads all the plurality of texts associated with the selected tag from the storage unit 140.
  • the text classification module 230 divides a plurality of texts into main texts and supplementary texts. At this time, the text classification module 230 clusters a plurality of texts according to contents, classifies a plurality of texts belonging to a cluster including the largest number of texts into a plurality of main texts, It is classified into a plurality of supplementary texts.
  • the main text processing module 240 is for deriving a representative value of each of the plurality of main texts from a plurality of main texts. To this end, the main text processing module 240 detects redundant texts in a plurality of main texts, eliminates redundant texts to perform deduplication, clusters a plurality of main texts according to contents of texts, And the center of the cluster from the generated plurality of clusters can be derived as a representative value.
  • the additional text processing module 250 is for deriving a representative value of one additional text among a plurality of additional texts.
  • the supplementary text processing module 250 detects redundant texts in a plurality of supplemental texts, eliminates redundant texts to perform deduplication, clusters a plurality of supplemental texts according to contents of text to generate a plurality of clusters, A cluster including the largest number of texts among a plurality of clusters may be selected and the center of the cluster may be derived as a representative value from the generated plurality of clusters.
  • the story generation module 260 is for generating a story using the above-mentioned representative value.
  • the story generation module 260 can generate a story by arranging representative values of each of the plurality of extracted main texts and representative values of one extracted supplementary text according to time order.
  • the time may be the article creation time on which the text is based.
  • FIG. 3 is a flowchart illustrating a method of collecting basic data for generating a story according to an embodiment of the present invention.
  • the text acquisition module 210 of the control unit 150 accesses a web server serving a web page including a news article through the communication unit 110 in step S110, Download the web page.
  • the text acquisition module 210 extracts the text from the web page including the news article in step S130. Then, the text acquisition module 210 analyzes the contents of the extracted text in step S140 and assigns tags to the respective texts. For example, the text collection module 210 analyzes the text of a news article and generates and assigns tags such as a subject, a material, an issue, a person, and a keyword. Analysis of the text to generate the appropriate tag may be made by one of ordinary skill in the art.
  • the text collection module 210 generates a list of tags in step S150. Then, in step S150, the text collection module 210 creates a text list associated with each tag. For example, when any one of the tags is a character 'Trump' (# Trump), a list of texts to which the Trump is assigned is generated among the plurality of texts.
  • a character 'Trump' # Trump
  • the text collection module 210 accumulates the tag list, the text list, and the text in step S170, and stores the tag list, the text list, and the text in the storage unit 140, .
  • FIG. 4 is a flowchart illustrating a method for generating a story based on a news article according to an embodiment of the present invention.
  • the text processing module 220 of the controller 150 selects at least one tag among a plurality of tags in the tag list in step S210.
  • the text processing module 220 can sequentially select tags sequentially from the tag associated with the largest number of texts in the tag list.
  • the selected tag may be one or two or more tags.
  • the text processing module 220 loads all the plurality of texts connected to the selected tag from the storage unit 140 in step S220.
  • the text classification module 230 divides the plurality of texts loaded in step S220 into a main text and an outlier text.
  • the main text means a text in which general contents are written among a plurality of texts linked with the tag
  • the supplementary text means text in which non-general contents are written. Accordingly, the text classification module 153 clusters a plurality of texts according to contents. Then, a plurality of texts belonging to the cluster including the largest number of texts are classified into main texts, and a plurality of texts belonging to remaining clusters are classified as additional texts.
  • the following process is performed for each of the main text and the supplementary text.
  • the main text processing module 240 detects duplicate texts in a plurality of main texts in step S310, and performs deduplication to delete duplicated texts.
  • the main text processing module 240 may select one of a plurality of main texts, that is, the same text, leaving only one of the texts and erasing all the others.
  • it is possible to perform deduplication by considering texts that are not identical but whose similarity is greater than or equal to a predetermined value as duplicates.
  • the main text processing module 240 may calculate the similarity between a plurality of main texts, and may delete all the remaining texts, leaving only one of a plurality of main texts whose similarities are equal to or greater than a predetermined value. Because the Internet articles have many texts with redundant contents, duplicates are detected and removed in advance for the processing speed and accuracy of the subsequent clustering process.
  • step S320 the main text processing module 240 creates a plurality of clusters by clustering the plurality of main texts according to the contents of the text. Accordingly, each of the plurality of clusters includes a plurality of main texts whose contents are equal to or greater than a predetermined value.
  • the main text processing module 240 selects a representative value of each cluster in step S330. At this time, the main text processing module 240 selects the center of a plurality of main texts belonging to each cluster as a representative value of each cluster.
  • the supplementary text processing module 250 performs deduplication to detect redundant texts in a plurality of supplemental texts and delete the redundant texts in step S410.
  • the supplementary text processing module 250 may select one of a plurality of supplementary texts, that is, the same text, leaving only one of the texts and erase the rest.
  • it is possible to perform deduplication by considering texts that are not identical but whose similarity is greater than or equal to a predetermined value as duplicates.
  • the supplementary text processing module 250 may calculate the similarity between a plurality of supplementary texts, and may erase all the remaining texts, leaving only one of a plurality of supplementary texts with a degree of similarity higher than a predetermined value. Because the Internet articles have many texts with redundant contents, duplicates are detected and removed in advance for the processing speed and accuracy of the subsequent clustering process.
  • step S420 the supplementary text processing module 250 generates a plurality of clusters by clustering a plurality of additional texts according to the contents of the text. Accordingly, each of the plurality of clusters includes a plurality of supplementary texts whose contents are equal to or greater than a predetermined value.
  • the supplementary text processing module 250 selects a cluster including the largest number of texts among the plurality of clusters at step S430.
  • the supplementary text processing module 250 selects a representative value of the cluster selected in operation S440.
  • the additional text processing module 250 selects the center of the cluster among the plurality of additional texts belonging to the cluster as a representative value.
  • the story generation module 260 in step S510, Generate story by connecting representative values in chronological order. Accordingly, a story reflecting the main text and additional text is generated.
  • the story generation module 260 selects a keyword that can represent the generated story. For example, the story generation module 260 can select a keyword that is the most used noun in the story together with the tag that was initially used.
  • a story is produced based on an Internet news article
  • various news articles can be provided as the basis of the story. It divides the main text that contains common contents among various news articles and the supplementary text which is not so different, and inserts some of the non-general texts, that is, the supplementary text, in the main text so that a rich story can come out. Furthermore, it is possible to produce various news stories about a single topic, issue, and material, and to provide news from various viewpoints. In addition, according to the present invention, not only one news but also a story based on news can be provided to the user.
  • the various methods according to the embodiments of the present invention described above can be implemented in a form of a program readable by various computer means and recorded on a computer-readable recording medium.
  • the recording medium may include program commands, data files, data structures, and the like, alone or in combination.
  • Program instructions to be recorded on a recording medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software.
  • the recording medium may be a magnetic medium such as a hard disk, a floppy disk and a magnetic tape, an optical medium such as a CD-ROM or a DVD, a magneto-optical medium such as a floppy disk magneto-optical media, and hardware devices that are specially configured to store and execute program instructions such as ROM, RAM, flash memory, and the like.
  • Examples of program instructions may include machine language wires such as those produced by a compiler, as well as high-level language wires that may be executed by a computer using an interpreter or the like.
  • Such a hardware device may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.
  • Text collection module 220 Text processing module
  • Text classification module 240 Main text processing module

Abstract

La présente invention concerne un appareil destiné à générer une histoire basée sur un article d'actualité, un procédé associé, et un support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre le procédé. La présente invention est fondée sur des actualités Internet, et traite ainsi divers articles d'actualité, la présente invention traitant séparément des articles principaux et les autres articles. Lorsqu'une histoire est fabriquée par recherche et gestion séparée de morceaux de texte non général et de texte supplémentaire, une partie du texte supplémentaire peut être insérée dans l'histoire de façon à permettre la création d'une histoire riche. La présente invention peut produire divers reportages pour un même sujet, problème et matériel, et peut ainsi fournir des actualités de divers points de vue. En outre, la présente invention peut non seulement recommander une actualité mais peut également recommander, à un utilisateur, une histoire générée sur la base d'actualités.
PCT/KR2018/013748 2017-11-22 2018-11-13 Appareil destiné à générer une histoire basée sur un article d'actualité, procédé associé, et support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre ce procédé WO2019103380A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
KR1020170156240A KR102319849B1 (ko) 2017-11-22 2017-11-22 뉴스 기사 기반 스토리를 생성하기 위한 장치, 이를 위한 방법 및 이 방법을 수행하는 프로그램이 기록된 컴퓨터 판독 가능한 기록매체
KR10-2017-0156240 2017-11-22

Publications (1)

Publication Number Publication Date
WO2019103380A1 true WO2019103380A1 (fr) 2019-05-31

Family

ID=66632032

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/KR2018/013748 WO2019103380A1 (fr) 2017-11-22 2018-11-13 Appareil destiné à générer une histoire basée sur un article d'actualité, procédé associé, et support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre ce procédé

Country Status (2)

Country Link
KR (1) KR102319849B1 (fr)
WO (1) WO2019103380A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111414736B (zh) * 2020-03-23 2022-05-20 腾讯科技（深圳）有限公司 故事生成模型训练方法、装置、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008021256A (ja) * 2006-07-14 2008-01-31 Toshiba Corp 情報共有システム及びプログラム
KR20110009912A (ko) * 2009-07-23 2011-01-31 한국과학기술원 매체의 편향 현상을 완화하기 위한 단면수준 뉴스브라우징 서비스 시스템 및 서비스 제공방법
KR20150041267A (ko) * 2013-10-07 2015-04-16 주식회사 솔트룩스 주제 기반 기사 연대기 생성 시스템
KR20160120529A (ko) * 2015-04-08 2016-10-18 조남희 분야별 분류 기반의 인터넷 기사 제공 시스템 및 방법
KR101682659B1 (ko) * 2015-07-15 2016-12-05 주식회사 에이제이케이 키워드 기반 맞춤 뉴스 알림 방법 및 그를 위한 뉴스 검색 관리 서버

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150015423A (ko) 2014-12-08 2015-02-10 한양대학교 에리카산학협력단 뉴스 추천 시스템 및 뉴스 추천 방법

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008021256A (ja) * 2006-07-14 2008-01-31 Toshiba Corp 情報共有システム及びプログラム
KR20110009912A (ko) * 2009-07-23 2011-01-31 한국과학기술원 매체의 편향 현상을 완화하기 위한 단면수준 뉴스브라우징 서비스 시스템 및 서비스 제공방법
KR20150041267A (ko) * 2013-10-07 2015-04-16 주식회사 솔트룩스 주제 기반 기사 연대기 생성 시스템
KR20160120529A (ko) * 2015-04-08 2016-10-18 조남희 분야별 분류 기반의 인터넷 기사 제공 시스템 및 방법
KR101682659B1 (ko) * 2015-07-15 2016-12-05 주식회사 에이제이케이 키워드 기반 맞춤 뉴스 알림 방법 및 그를 위한 뉴스 검색 관리 서버

Also Published As

Publication number Publication date
KR20190058870A (ko) 2019-05-30
KR102319849B1 (ko) 2021-11-02

Similar Documents

Publication Publication Date Title
CN107209905A (zh) 针对个性化和任务完成服务而对应用去主题归类
CN104462547B (zh) 一种可配置的网页数据采集的方法及系统
CN103842993A (zh) 用于上下文个性化搜索的系统和方法
WO2020166883A1 (fr) Procédé et système d'édition de vidéo sur la base d'un contexte obtenu à l'aide d'une intelligence artificielle
WO2016186362A1 (fr) Dispositif et procédé permettant de fournir des informations sur l'état psychologique d'un utilisateur reflété par du texte
KR20150086441A (ko) 콘텐츠 및 상관 거리에 기반하는 사람 연결
US20180285362A1 (en) Scoring mechanism for discovery of extremist content
CN107077345A (zh) 用于在当前应用中插入内容的个性化上下文菜单
CN109492118A (zh) 一种数据检测方法及检测装置
WO2020222386A1 (fr) Procédé et appareil de mise à jour d'un modèle de probabilité de grappe
CN109983459A (zh) 用于标识语料库中出现的n-gram的计数的方法和设备
CN111026967A (zh) 一种获得用户兴趣标签的方法、装置、设备和介质
CN107111649A (zh) 将用户和系统数据从源位置上传到目的地位置
WO2019103380A1 (fr) Appareil destiné à générer une histoire basée sur un article d'actualité, procédé associé, et support d'enregistrement lisible par ordinateur sur lequel est enregistré un programme pour mettre en œuvre ce procédé
US9898467B1 (en) System for data normalization
CN114428705A (zh) 一种网络数据的监测方法、装置、设备以及存储介质
CN111949849B (zh) 鱼类信息的获取方法、装置、电子设备及可读存储介质
JP7033115B2 (ja) クリップボードデータに基づく検索処理方法および装置
KR102337536B1 (ko) 장기간 관련 있는 이슈 단위의 클러스터를 이용한 문서 타임라인을 제공하는 방법 및 시스템
CN105550183A (zh) 一种网页中识别信息的标识方法及电子设备
CN105745684A (zh) 针对电子书的基于实体的摘要
KR20200072724A (ko) 스팸 문구가 포함된 스팸뉴스 탐지를 위한 장치, 이를 위한 방법 및 이 방법을 수행하는 프로그램이 기록된 컴퓨터 판독 가능한 기록매체
CN114943978A (zh) 一种表格重建的方法及电子设备
CN114745558B (zh) 直播监控方法、装置、系统、设备及介质
CN109446356A (zh) 一种多媒体文件检索方法及装置

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: 18880544

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: 18880544

Country of ref document: EP

Kind code of ref document: A1