WO2019090737A1 - 基于文字内容推荐场景信息的方法与装置 - Google Patents

基于文字内容推荐场景信息的方法与装置 Download PDF

Info

Publication number
WO2019090737A1
WO2019090737A1 PCT/CN2017/110567 CN2017110567W WO2019090737A1 WO 2019090737 A1 WO2019090737 A1 WO 2019090737A1 CN 2017110567 W CN2017110567 W CN 2017110567W WO 2019090737 A1 WO2019090737 A1 WO 2019090737A1
Authority
WO
WIPO (PCT)
Prior art keywords
scene
text content
information
recommending
phrase
Prior art date
Application number
PCT/CN2017/110567
Other languages
English (en)
French (fr)
Inventor
张研
Original Assignee
深圳市华阅文化传媒有限公司
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 深圳市华阅文化传媒有限公司 filed Critical 深圳市华阅文化传媒有限公司
Priority to PCT/CN2017/110567 priority Critical patent/WO2019090737A1/zh
Publication of WO2019090737A1 publication Critical patent/WO2019090737A1/zh

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/06Buying, selling or leasing transactions
    • G06Q30/0601Electronic shopping [e-shopping]
    • G06Q30/0631Item recommendations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/10Text processing
    • G06F40/12Use of codes for handling textual entities
    • G06F40/134Hyperlinking
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/10Text processing
    • G06F40/166Editing, e.g. inserting or deleting
    • G06F40/169Annotation, e.g. comment data or footnotes
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06QINFORMATION 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
    • G06Q30/00Commerce
    • G06Q30/02Marketing; Price estimation or determination; Fundraising
    • G06Q30/0282Rating or review of business operators or products
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies

Definitions

  • the present invention relates to the field of electronic reading technology, and in particular, to a method and apparatus for recommending scene information based on text content.
  • the related content can be pushed based on the type of the article that the user likes to read, and the type of the book; and the specific content in the reading content is not carefully considered, for example, in a document content about a war story, It may involve various types of scenes such as battle screens, urban bar leisure, hotels, restaurants, etc.
  • the specific content in the reading content is not carefully considered, for example, in a document content about a war story, It may involve various types of scenes such as battle screens, urban bar leisure, hotels, restaurants, etc.
  • the main object of the present invention is to provide a method for recommending scene information based on text content, which aims to solve the technical problem that the related electronic scene reading technology does not recommend related scene information according to the specific scene corresponding to the text content in the text.
  • the present invention provides a method for recommending scene information based on text content, including:
  • the method before the step of receiving an instruction of the display scene generated by the data link of the scene corresponding to the specified text content on the electronic reading interface, the method includes:
  • the step of establishing the data link of the specified text content and the corresponding scene comprises: [0011] analyzing the phrase information in the specified text content; [0012] matching, by the phrase information, a scene type of the corresponding scene;
  • the method before the step of filtering the scene samples in the database according to the scenario type, the method includes:
  • the method includes:
  • the present invention also provides an apparatus for recommending scene information based on text content, including:
  • a receiving module configured to receive an instruction generated by a user clicking a data link of a scene corresponding to the specified text content on the electronic reading interface
  • a display module configured to display the scene information in a specified display area according to the instruction.
  • the device for recommending scene information based on text content includes:
  • a establishing module configured to establish a data link between the specified text content and the corresponding scene.
  • the establishing module includes:
  • an analyzing unit configured to analyze phrase information in the specified text content
  • a first matching unit configured to match, by using the phrase information, a scene type of the corresponding scene
  • a filtering unit configured to filter a scene sample in the database according to the scene type
  • a first linking unit configured to link a storage path of the scene sample to the phrase information.
  • the establishing module includes:
  • a distinguishing unit configured to distinguish, according to the degree of association of each scenario, a scenario type corresponding to each scene sample in the database
  • the identifier unit is configured to identify a type label corresponding to each of the scene types.
  • the device for recommending scene information based on text content includes:
  • a discriminating module configured to distinguish phrase information in new text content
  • a second matching module configured to automatically match the scenario associated with the phrase information according to the phrase information
  • a second linking module configured to automatically link a storage path of the scene sample corresponding to the scene to the phrase information.
  • the present invention establishes a scene database by analyzing different scenes of the article content, and associates scene information such as an application, an activity post, a topic, or a character associated with each scene with a specified text content through a data link, when the reader reads
  • the specified text content can be recommended to the reader according to the content of the text in the text, so that the reader can more accurately accept the required scene information.
  • the present invention can analyze a typical scene through self-learning, and then analyze and learn the text content in other new books in the network, and automatically match the scene information of the associated application activity post, topic or character with the scene type label, and increase Scene recognition and matching accuracy enable full automatic recognition of text content in new online books and increase the commercial value of new online books.
  • FIG. 1 is a schematic flow chart of a method for recommending scene information based on text content according to an embodiment of the present invention
  • FIG. 2 is a schematic diagram of an optimization process of a method for recommending scene information based on text content according to an embodiment of the present invention
  • step S10 is a schematic flow chart of step S10 in an embodiment of the present invention.
  • step S10 is a schematic flowchart of step S10 in another embodiment of the present invention.
  • FIG. 5 is a schematic flowchart of a method for recommending scene information based on text content in still another embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of an apparatus for recommending scene information based on text content according to an embodiment of the present invention
  • FIG. 7 is a schematic diagram of an apparatus optimization structure for recommending scene information based on text content according to an embodiment of the present invention.
  • FIG. 8 is a schematic structural diagram of a module built in an embodiment of the present invention.
  • FIG. 9 is a schematic structural diagram of establishing a module in another embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of an apparatus for recommending scene information based on text content in still another embodiment of the present invention.
  • a method for recommending scene information based on text content including: [0053] S1: receiving an instruction generated by a user clicking a data link of a scene corresponding to a specified text content on an electronic reading interface .
  • scene information such as an application, an active post, a topic, or a character associated with each scene is associated with the specified text content through a data link, and the user clicks on the related data link to start each scene information.
  • S2 displaying the scene information in a specified display area according to the instruction.
  • the specified display area display includes: displaying the corresponding scene information on the electronic reading interface or entering the link interface to display corresponding scene information,
  • the reader is more eager to receive the required scene information. For example, if the user reads the bar scene, it can provide related bar party activities or dating software, and enter the relevant bar party event or the display interface of the dating software through the corresponding link; when the bar has some specific wine cellars, the product link or prompt can be directly given. For example, link the product image, and display the corresponding image information on the e-reading interface after the snoring link.
  • the scene information of the non-reading content related to the scene can be recommended according to the scene, and the scene information is commercialized.
  • step S1 of this embodiment the method includes:
  • S10 Establish a data link between the specified text content and the corresponding scene.
  • a scene sample database is manually established, and a data link is established between the specified text content and the corresponding scene by analyzing characteristics of a typical scene, including scene information content, scene association information, and the like.
  • step S10 of the embodiment further includes:
  • This step refers to refining the scene by analyzing a large number of phrases in the text content.
  • the phrase information includes the mutual distances of the phrases appearing in the specified paragraphs of text and the frequency of occurrence of similar similar phrases. It refers to the analysis of scenes through a large amount of text analysis, as well as the judgment of relevant activity posts, topics and software relevance. Broken. All kinds of phrases, not the content of the article is similar, but the scenes in the article are similar, for example, the same label used in the contest.
  • S101 Match the scene type of the corresponding scene by the phrase information.
  • S102 Filter the scene samples in the database according to the scene type.
  • a war scene sample associated with a war scene is selected in a database, such as a recently released war film, an international military news, and the like.
  • S103 Link the storage path of the scene sample to the phrase information.
  • the storage path refers to the storage address of the scene sample in the database.
  • step S102 a method for recommending scene information based on text content is proposed in another embodiment of the present invention.
  • the method includes:
  • S104 Differentiate the scene type corresponding to each scene sample in the database according to the degree of association of each scene.
  • the degree of association includes similar, interrelated.
  • the bar scene type can include related bar parties, dating software associated with the bar environment, merchandise links for bar-specific wines, and bar culture book recommendations.
  • S105 Identify a type tag corresponding to each of the scene types.
  • the scenes with the high degree of association with the bar are identified as the bar scene type label to facilitate quick query of the relevant scene in the database.
  • step S a method for recommending scene information based on text content is proposed, step S
  • S12 automatically matching the scene associated with the phrase information according to the phrase information.
  • This step refers to automatically matching the scene associated with the phrase information according to the existing data link established by the scene and the phrase information in the database after reading the new website.
  • the bar scene type may include related bar party activities, and Matching dating software, bar-specific wine links, bar culture book recommendations, etc., match the 'bar' phrase in the new online book.
  • the existing data link can be automatically applied to the new network book.
  • a standard sample is manually established to analyze a typical scene; through self-learning, the newly-created phrase information in the new network book can be analyzed and learned to increase the matching degree of the scene, the phrase recognition degree and the scene and the phrase, so as to achieve Fully automatic recognition of phrases and scenes in new books, for example, by expanding the coverage of synonyms of related phrases during use, by using user-authenticated comments or systematic self-correcting phrases to match scenes, to improve the The matching accuracy of related words and scenes.
  • an apparatus for recommending scene information based on text content includes:
  • the receiving module 1 is configured to receive an instruction generated by a user clicking a data link of a scene corresponding to the specified text content on the electronic reading interface.
  • scene information such as an application, an active post, a topic, or a character associated with each scene is associated with the specified text content through a data link, and the receiving module 1 receives the user clicks on the related data link to start each scene information.
  • the display module 2 is configured to display the scene information in a specified display area according to the instruction.
  • the display module 2 recommends, according to the content of the text in the text, the scene information corresponding to the specific scene, and the display area display includes: displaying the corresponding scene information on the electronic reading interface or displaying the corresponding interface on the electronic reading interface.
  • the scene information makes the reader more comfortable to receive the required scene information. For example, if the user reads the bar scene, it can provide related bar party activities or dating software, and enter the relevant bar party event or the display interface of the dating software through the corresponding link; when the bar has some specific wine cellars, the product link or prompt can be directly given. For example, link the product image, and display the corresponding image information on the e-reading interface after the snoring link.
  • the scene information of the non-reading content related to the scene may be recommended according to the scene, and the scene information is commercialized.
  • the apparatus for recommending scene information based on the text content in the embodiment includes: [0087]
  • the establishing module 10 is configured to establish a data link between the specified text content and the corresponding scene.
  • the scene sample database is established by the establishing module 10, and the characteristics of the typical scene are analyzed.
  • the scene information content, the scene association information, and the like are included, and a data link is established between the specified text content and the corresponding scene.
  • the establishing module 10 includes:
  • the analyzing unit 100 is configured to analyze the phrase information in the specified text content.
  • the analysis unit 100 analyzes a large number of phrases in the text content to extract the scene.
  • the phrase information includes the mutual distance of the phrase in the specified paragraph of text and the frequency of appearance of similar phrases. It refers to the analysis of scenes through a large amount of text analysis, as well as the judgment of relevant activity posts, topics and software relevance. All kinds of phrases, not the content of the article is similar, but the scenes in the article are similar, for example, the same label used in the contest.
  • the first matching unit 101 is configured to match, by using the phrase information, a scene type of the corresponding scene.
  • the filtering unit 102 is configured to filter the scene samples in the database according to the scene type.
  • the screening unit 102 filters the scene samples of the war related to the war scene in the database according to the war scene, such as the recently released war film, international military news, and the like.
  • the first linking unit 103 is configured to link the storage path of the scene sample to the phrase information.
  • the storage path refers to the storage address of the scene sample in the database.
  • the establishing module 10 of another embodiment of the present invention includes:
  • the distinguishing unit 104 is configured to distinguish the scene type corresponding to each scene sample in the database according to the degree of association of each scene.
  • the degree of association includes similar, interrelated.
  • the bar scene type can include related bar parties, dating software associated with the bar environment, merchandise links for bar-specific wines, and bar culture book recommendations.
  • the identifier unit 105 is configured to identify a type label corresponding to each of the scene types.
  • the identification unit 105 identifies the above-mentioned scenes having a high degree of association with the bar as the bar scene type label, so as to conveniently query the related scenes in the database.
  • an apparatus for recommending scene information based on text content includes: [0104]
  • the discriminating module 11 is configured to distinguish the phrase information in the new text content.
  • the discriminating module 11 discriminates whether the new book in the network contains a phrase f ⁇ information that establishes a corresponding data link with the scene in the database.
  • the second matching module 12 is configured to automatically match the scenario associated with the phrase information according to the phrase information.
  • the second matching module 12 can automatically match the scene associated with the phrase information according to the existing data link established by the scene and the phrase information in the database.
  • the bar scene type may include a related bar party event, a dating software associated with the bar environment, a bar-specific wine product link, a bar culture book recommendation, and the like. The 'bar' phrase in the new online book is matched.
  • the second link module 13 is configured to automatically link a storage path of the scene sample corresponding to the scene to the phrase information.
  • the second link module 13 can automatically apply the existing data link to the new network book.
  • a standard sample is manually established to analyze a typical scene; through self-learning, the newly-created phrase information in the new network book can be analyzed and learned to increase the matching degree of the scene, the phrase recognition degree and the scene and the phrase, so as to achieve Fully automatic recognition of phrases and scenes in new books, for example, by expanding the coverage of synonyms of related phrases during use, by using user-authenticated comments or systematic self-correcting phrases to match scenes, to improve the The matching accuracy of related words and scenes.
  • the present invention establishes a scene database by analyzing different scenes of the article content, and associates scene information such as an application, an activity post, a topic, or a character associated with each scene with a specified text content through a data link, when the reader reads
  • the specified text content can be recommended to the reader according to the content of the text in the text, so that the reader can more accurately accept the required scene information.
  • the present invention can analyze a typical scene through self-learning, and then analyze and learn the text content in other new books in the network, and automatically match the scene information of the associated application activity post, topic or character with the scene type label, and increase Scene recognition and matching accuracy enable full automatic recognition of text content in new online books and increase the commercial value of new online books.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Development Economics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Artificial Intelligence (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Game Theory and Decision Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

本发明揭示了基于文字内容推荐场景信息的方法,包括:接收用户点击电子阅读界面上指定文字内容所对应场景的数据链接所产生的指令;根据指令在指定显示区域展示场景信息。本发明通过分析文章内容的不同场景,建立场景数据库,并将各场景相关联的应用程序、活动帖子、话题或人物等场景信息通过数据链接与指定文字内容相联系。

Description

基于文字内容推荐场景信息的方法与装置 技术领域
[0001] 本发明涉及到电子阅读技术领域, 特别是涉及到基于文字内容推荐场景信息的 方法与装置。
背景技术
[0002] 现有电子阅读技术中, 可基于用户喜欢读的文章类型、 书的类型推送相关内容 ; 而未细致考虑阅读内容中的某些具体场景, 例如一篇关于战争故事的文件内 容中, 可能涉及战斗画面、 都市酒吧休闲、 旅店、 餐饮等多种类型的场景, 目 前没有根据文中文字内容所对应的具体场景进行相关场景信息推荐的功能。
[0003] 因此, 现有技术还有待改进。
技术问题
[0004] 本发明的主要目的为提供基于文字内容推荐场景信息的方法, 旨在解决现有电 子阅读技术中没有根据文中文字内容所对应的具体场景进行相关场景信息推荐 的技术问题。
问题的解决方案
技术解决方案
[0005] 本发明提出基于文字内容推荐场景信息的方法, 包括:
[0006] 接收用户点击电子阅读界面上指定文字内容所对应场景的数据链接所产生的指 令;
[0007] 根据所述指令在指定显示区域展示所述场景信息。
[0008] 优选地, 所述接收用户点击电子阅读界面上指定文字内容所对应场景的数据链 接所产生的展示场景的指令的步骤之前, 包括:
[0009] 建立所述指定文字内容与所述对应场景的数据链接。
[0010] 优选地, 所述建立所述指定文字内容与所述对应场景的数据链接的步骤, 包括 [0011] 分析所述指定文字内容中的词组信息; [0012] 通过所述词组信息匹配所对应场景的场景类型;
[0013] 根据所述场景类型筛选数据库中的场景样本;
[0014] 将所述场景样本的存放路径链接到所述词组信息上。
[0015] 优选地, 所述根据所述场景类型筛选数据库中的场景样本的步骤之前, 包括:
[0016] 根据各场景的关联度区分数据库中各场景样本所对应场景类型;
[0017] 标识各所述场景类型所对应的类型标签。
[0018] 优选地, 所述接收用户点击电子阅读界面上指定文字内容所对应场景的数据链 接所产生的展示场景的指令的步骤之前, 包括:
[0019] 辨别新文本内容中的词组信息;
[0020] 根据所述词组信息自动匹配与所述词组信息相关联的所述场景;
[0021] 将所述场景对应的场景样本的存储路径自动链接到所述词组信息上。
[0022] 本发明还提供了一种基于文字内容推荐场景信息的装置, 包括:
[0023] 接收模块, 用于接收用户点击电子阅读界面上指定文字内容所对应场景的数据 链接所产生的指令;
[0024] 展示模块, 用于根据所述指令在指定显示区域展示所述场景信息。
[0025] 优选地, 所述基于文字内容推荐场景信息的装置, 包括:
[0026] 建立模块, 用于建立所述指定文字内容与所述对应场景的数据链接。
[0027] 优选地, 所述建立模块, 包括:
[0028] 分析单元, 用于分析所述指定文字内容中的词组信息;
[0029] 第一匹配单元, 用于通过所述词组信息匹配所对应场景的场景类型;
[0030] 筛选单元, 用于根据所述场景类型筛选数据库中的场景样本;
[0031] 第一链接单元, 用于将所述场景样本的存放路径链接到所述词组信息上。
[0032] 优选地, 所述建立模块, 包括:
[0033] 区分单元, 用于根据各场景的关联度区分数据库中各场景样本所对应场景类型
[0034] 标识单元, 用于标识各所述场景类型所对应的类型标签。
[0035] 优选地, 所述基于文字内容推荐场景信息的装置, 包括:
[0036] 辨别模块, 用于辨别新文本内容中的词组信息, [0037] 第二匹配模块, 用于根据所述词组信息自动匹配与所述词组信息相关联的所述 场景;
[0038] 第二链接模块, 用于将所述场景对应的场景样本的存储路径自动链接到所述词 组信息上。
发明的有益效果
有益效果
[0039] 本发明通过分析文章内容的不同场景, 建立场景数据库, 并将各场景相关联的 应用程序、 活动帖子、 话题或人物等场景信息通过数据链接与指定文字内容相 联系, 当读者阅读到上述指定文字内容吋, 可根据文中文字内容向读者推荐所 对应的具体场景的场景信息, 使读者更及吋地接受到所需要的场景信息。 同吋 , 本发明可通过自学习, 分析出典型场景, 再对其他网络新书中的文字内容进 行分析学习, 结合场景类型标签自动匹配相关联的应用程序活动帖子、 话题或 人物等场景信息, 增加场景识别度和匹配准确度, 可实现完全自动识别网络新 书中的文字内容, 并提高网络新书的商业价值。
对附图的简要说明
附图说明
[0040] 图 1本发明一实施例中基于文字内容推荐场景信息的方法的流程示意图;
[0041] 图 2本发明一实施例中基于文字内容推荐场景信息的方法的优化流程示意图;
[0042] 图 3本发明一实施例中步骤 S10的流程示意图;
[0043] 图 4本发明另一实施例中步骤 S10的流程示意图;
[0044] 图 5本发明再一实施例中基于文字内容推荐场景信息的方法的流程示意图;
[0045] 图 6本发明一实施例中基于文字内容推荐场景信息的装置结构示意图;
[0046] 图 7本发明一实施例中基于文字内容推荐场景信息的装置优化结构示意图;
[0047] 图 8本发明一实施例中建立模块的结构示意图;
[0048] 图 9本发明另一实施例中建立模块的结构示意图;
[0049] 图 10本发明再一实施例中基于文字内容推荐场景信息的装置结构示意图。
[0050] 本发明目的的实现、 功能特点及优点将结合实施例, 参照附图做进一步说明。 实施该发明的最佳实施例
本发明的最佳实施方式
[0051] 应当理解, 此处所描述的具体实施例仅仅用以解释本发明, 并不用于限定本发 明。
[0052] 参照图 1, 本发明一实施例中提出基于文字内容推荐场景信息的方法, 包括: [0053] S1 : 接收用户点击电子阅读界面上指定文字内容所对应场景的数据链接所产生 的指令。
[0054] 本实施例中将各场景相关联的应用程序、 活动帖子、 话题或人物等场景信息通 过数据链接与指定文字内容相联系, 用户点击相关数据链接启动各场景信息。
[0055] S2: 根据所述指令在指定显示区域展示所述场景信息。
[0056] 本步骤中通过根据文中文字内容向读者推荐所对应的具体场景的场景信息, 指 定显示区域展示包括, 在所述电子阅读界面上展示相应场景信息或进入链接界 面展示相应的场景信息, 使读者更及吋地接受到所需要的场景信息。 例如用户 读取到酒吧场景, 可提供相关酒吧聚会活动或者交友软件, 并通过相应链接进 入相关酒吧聚会活动或者交友软件的显示界面; 当酒吧出现一些特定酒吋, 可 以直接给予商品链接或提示等, 比如链接商品图片, 打幵链接后在电子阅读界 面上展示相应的图片信息。 可根据场景推荐与场景相关的非阅读内容的场景信 息, 实现场景信息商业化。
[0057] 参照图 2, 进一步地, 本实施例的步骤 S1之前, 包括:
[0058] S10: 建立所述指定文字内容与所述对应场景的数据链接。
[0059] 本实施例中通过人工建立场景样本数据库, 通过分析典型场景的特征, 包括场 景信息内容、 场景关联信息等, 将指定文字内容与所述对应场景之间建立数据 链接。
[0060] 参照图 3, 进一步地, 本实施例的步骤 S10, 包括:
[0061] S100: 分析所述指定文字内容中的词组信息。
[0062] 本步骤指通过分析文本内容中的大量词组, 提炼出场景。 词组信息包括词组在 指定文字段落中出现的相互距离以及同类相近词组的出现频度等。 是指通过大 量的文本分析, 提炼出场景判断, 以及对相关活动帖子、 话题和软件相关度判 断。 各类词组, 不是文章内容相近, 而是文章中的场景相近, 例如比武场就打 的相同标签。
[0063] S101 : 通过所述词组信息匹配所对应场景的场景类型。
[0064] 比如, 频繁出现'战争'、 '战斗'、 '炮弹 '等相近词组, 并将上述场景匹配归类为 战争类场景。 再比如词组信息出现'都市暧昧', 且判定该场景吋间属于晚上, 则 向读者推荐'交友软件'等都市暧昧场景。
[0065] S102: 根据所述场景类型筛选数据库中的场景样本。
[0066] 比如根据战争类场景在数据库中筛选跟战争类场景相关联的战争方面的场景样 本, 比如最近上映的战争影片、 国际军事要闻等。
[0067] S103: 将所述场景样本的存放路径链接到所述词组信息上。
[0068] 存放路径指场景样本在数据库中的存放地址。
[0069] 参照图 4, 本发明另一实施例中提出基于文字内容推荐场景信息的方法, 步骤 S 102之前, 包括:
[0070] S104: 根据各场景的关联度区分数据库中各场景样本所对应场景类型。
[0071] 关联度包括相似的、 有相互联系的。 比如, 酒吧场景类型中可包括相关的酒吧 聚会活动、 与酒吧环境相联系的交友软件、 酒吧特定酒的商品链接、 酒吧文化 书籍推荐等。
[0072] S105: 标识各所述场景类型所对应的类型标签。
[0073] 将上述与酒吧有较高关联度的场景同一标识为酒吧场景类型标签, 以方便在数 据库中快速査询相关场景。
[0074] 参照图 5, 本发明再一实施例中提出基于文字内容推荐场景信息的方法, 步骤 S
1之前, 包括:
[0075] S11 : 辨别新文本内容中的词组信息。
[0076] 指辨别网络新书中是否含有与数据库中场景建立相应数据链接的词组信息。
[0077] S12: 根据所述词组信息自动匹配与所述词组信息相关联的所述场景。
[0078] 本步骤是指在阅读网络新书吋, 可根据数据库中的场景与词组信息建立的已存 在的数据链接, 自动匹配与词组信息相关联的场景。 以酒吧场景举例, 当网络 新书中出现酒吧等词组吋, 可将酒吧场景类型中包括相关的酒吧聚会活动、 与 酒吧环境相联系的交友软件、 酒吧特定酒的商品链接、 酒吧文化书籍推荐等与 网络新书中的 '酒吧 '词组进行匹配。
[0079] S13: 将所述场景对应的场景样本的存储路径自动链接到所述词组信息上。
[0080] 当出现已与数据库中相应场景建立链接的词组, 或与上述词组信息相近的词组 吋, 可将已存在的数据链接自动应用到网络新书中。 本发明初期通过人工建立 标准样本, 分析出典型场景; 通过自学习, 可对网络新书中新出现的词组信息 进行分析学习, 以增加场景、 词组识别度以及场景与词组的匹配准确度, 以实 现对新书中的词组和场景完全自动识别, 比如, 在使用过程中通过不断扩大关 联词组的近义词等覆盖范围, 通过使用中用户验证评论或系统自矫正词组与场 景的匹配, 以提高对新书中的相关词汇与场景的匹配准确度。
[0081] 参照图 6, 本发明一实施例中基于文字内容推荐场景信息的装置, 包括:
[0082] 接收模块 1, 用于接收用户点击电子阅读界面上指定文字内容所对应场景的数 据链接所产生的指令。
[0083] 本实施例中将各场景相关联的应用程序、 活动帖子、 话题或人物等场景信息通 过数据链接与指定文字内容相联系, 接收模块 1接收用户点击相关数据链接启动 各场景信息。
[0084] 展示模块 2, 用于根据所述指令在指定显示区域展示所述场景信息。
[0085] 本实施例中通过展示模块 2根据文中文字内容向读者推荐所对应的具体场景的 场景信息, 指定显示区域展示包括, 在所述电子阅读界面上展示相应场景信息 或进入链接界面展示相应的场景信息, 使读者更及吋地接受到所需要的场景信 息。 例如用户读取到酒吧场景, 可提供相关酒吧聚会活动或者交友软件, 并通 过相应链接进入相关酒吧聚会活动或者交友软件的显示界面; 当酒吧出现一些 特定酒吋, 可以直接给予商品链接或提示等, 比如链接商品图片, 打幵链接后 在电子阅读界面上展示相应的图片信息。 可根据场景推荐与场景相关的非阅读 内容的场景信息, 实现场景信息商业化。
[0086] 参照图 7, 进一步地, 本实施例的基于文字内容推荐场景信息的装置, 包括: [0087] 建立模块 10, 用于建立所述指定文字内容与所述对应场景的数据链接。
[0088] 本实施例中通过建立模块 10建立场景样本数据库, 通过分析典型场景的特征, 包括场景信息内容、 场景关联信息等, 将指定文字内容与所述对应场景之间建 立数据链接。
[0089] 参照图 8, 所述建立模块 10, 包括:
[0090] 分析单元 100, 用于分析所述指定文字内容中的词组信息。
[0091] 本实施例通过分析单元 100分析文本内容中的大量词组, 提炼出场景。 词组信 息包括词组在指定文字段落中出现的相互距离以及同类相近词组的出现频度等 。 是指通过大量的文本分析, 提炼出场景判断, 以及对相关活动帖子、 话题和 软件相关度判断。 各类词组, 不是文章内容相近, 而是文章中的场景相近, 例 如比武场就打的相同标签。
[0092] 第一匹配单元 101, 用于通过所述词组信息匹配所对应场景的场景类型。
[0093] 比如, 频繁出现'战争'、 '战斗'、 '炮弹 '等相近词组, 并将上述场景通过第一匹 配单元 101匹配归类为战争类场景。 再比如词组信息出现'都市暧昧', 且判定该 场景吋间属于晚上, 则向读者推荐 '交友软件'等都市暧昧场景。
[0094] 筛选单元 102, 用于根据所述场景类型筛选数据库中的场景样本。
[0095] 比如筛选单元 102根据战争类场景在数据库中筛选跟战争类场景相关联的战争 方面的场景样本, 比如最近上映的战争影片、 国际军事要闻等。
[0096] 第一链接单元 103, 用于将所述场景样本的存放路径链接到所述词组信息上。
[0097] 存放路径指场景样本在数据库中的存放地址。
[0098] 参照图 9, 本发明另一实施例的所述建立模块 10, 包括:
[0099] 区分单元 104, 用于根据各场景的关联度区分数据库中各场景样本所对应场景 类型。
[0100] 关联度包括相似的、 有相互联系的。 比如, 酒吧场景类型中可包括相关的酒吧 聚会活动、 与酒吧环境相联系的交友软件、 酒吧特定酒的商品链接、 酒吧文化 书籍推荐等。
[0101] 标识单元 105, 用于标识各所述场景类型所对应的类型标签。
[0102] 比如, 标识单元 105将上述与酒吧有较高关联度的场景同一标识为酒吧场景类 型标签, 以方便在数据库中快速査询相关场景。
[0103] 参照图 10, 本发明再一实施例的基于文字内容推荐场景信息的装置, 包括: [0104] 辨别模块 11, 用于辨别新文本内容中的词组信息。
[0105] 辨别模块 11辨别网络新书中是否含有与数据库中场景建立相应数据链接的词组 f π息。
[0106] 第二匹配模块 12, 用于根据所述词组信息自动匹配与所述词组信息相关联的所 述场景。
[0107] 本实施例中, 在阅读网络新书吋, 第二匹配模块 12可根据数据库中的场景与词 组信息建立的已存在的数据链接, 自动匹配与词组信息相关联的场景。 以酒吧 场景举例, 当网络新书中出现酒吧等词组吋, 可将酒吧场景类型中包括相关的 酒吧聚会活动、 与酒吧环境相联系的交友软件、 酒吧特定酒的商品链接、 酒吧 文化书籍推荐等与网络新书中的'酒吧'词组进行匹配。
[0108] 第二链接模块 13, 用于将所述场景对应的场景样本的存储路径自动链接到所述 词组信息上。
[0109] 当出现已与数据库中相应场景建立链接的词组, 或与上述词组信息相近的词组 吋, 第二链接模块 13可将已存在的数据链接自动应用到网络新书中。 本发明初 期通过人工建立标准样本, 分析出典型场景; 通过自学习, 可对网络新书中新 出现的词组信息进行分析学习, 以增加场景、 词组识别度以及场景与词组的匹 配准确度, 以实现对新书中的词组和场景完全自动识别, 比如, 在使用过程中 通过不断扩大关联词组的近义词等覆盖范围, 通过使用中用户验证评论或系统 自矫正词组与场景的匹配, 以提高对新书中的相关词汇与场景的匹配准确度。
[0110] 本发明通过分析文章内容的不同场景, 建立场景数据库, 并将各场景相关联的 应用程序、 活动帖子、 话题或人物等场景信息通过数据链接与指定文字内容相 联系, 当读者阅读到上述指定文字内容吋, 可根据文中文字内容向读者推荐所 对应的具体场景的场景信息, 使读者更及吋地接受到所需要的场景信息。 同吋 , 本发明可通过自学习, 分析出典型场景, 再对其他网络新书中的文字内容进 行分析学习, 结合场景类型标签自动匹配相关联的应用程序活动帖子、 话题或 人物等场景信息, 增加场景识别度和匹配准确度, 可实现完全自动识别网络新 书中的文字内容, 并提高网络新书的商业价值。
[0111] 以上所述仅为本发明的优选实施例, 并非因此限制本发明的专利范围, 凡是利 用本发明说明书及附图内容所作的等效结构或等效流程变换, 或直接或间接运 用在其他相关的技术领域, 均同理包括在本发明的专利保护范围内。

Claims

权利要求书
一种基于文字内容推荐场景信息的方法, 其特征在于, 包括: 接收用户点击电子阅读界面上指定文字内容所对应场景的数据链接所 产生的指令;
根据所述指令在指定显示区域展示所述场景信息。
根据权利要求 1所述的基于文字内容推荐场景信息的方法, 其特征在 于, 所述接收用户点击电子阅读界面上指定文字内容所对应场景的数 据链接所产生的展示场景的指令的步骤之前, 包括:
建立所述指定文字内容与所述对应场景的数据链接。
根据权利要求 2所述的基于文字内容推荐场景信息的方法, 其特征在 于, 所述建立所述指定文字内容与所述对应场景的数据链接的步骤, 包括:
分析所述指定文字内容中的词组信息;
通过所述词组信息匹配所对应场景的场景类型;
根据所述场景类型筛选数据库中的场景样本;
将所述场景样本的存放路径链接到所述词组信息上。
根据权利要求 3所述的基于文字内容推荐场景的方法, 其特征在于, 所述根据所述场景类型筛选数据库中的场景样本的步骤之前, 包括: 根据各场景的关联度区分数据库中各场景样本所对应场景类型; 标识各所述场景类型所对应的类型标签。
根据权利要求 1所述的基于文字内容推荐场景信息的方法, 其特征在 于, 所述接收用户点击电子阅读界面上指定文字内容所对应场景的数 据链接所产生的展示场景的指令的步骤之前, 包括:
辨别新文本内容中的词组信息;
根据所述词组信息自动匹配与所述词组信息相关联的所述场景; 将所述场景对应的场景样本的存储路径自动链接到所述词组信息上。 一种基于文字内容推荐场景信息的装置, 其特征在于, 包括: 接收模块, 用于接收用户点击电子阅读界面上指定文字内容所对应场 景的数据链接所产生的指令;
展示模块, 用于根据所述指令在指定显示区域展示所述场景信息。
[权利要求 7] 根据权利要求 6所述的基于文字内容推荐场景信息的装置, 其特征在 于, 包括:
建立模块, 用于建立所述指定文字内容与所述对应场景的数据链接。
[权利要求 8] 根据权利要求 7所述的基于文字内容推荐场景信息的装置, 其特征在 于, 所述建立模块, 包括:
分析单元, 用于分析所述指定文字内容中的词组信息;
第一匹配单元, 用于通过所述词组信息匹配所对应场景的场景类型; 筛选单元, 用于根据所述场景类型筛选数据库中的场景样本; 第一链接单元, 用于将所述场景样本的存放路径链接到所述词组信息 上。
[权利要求 9] 根据权利要求 8所述的基于文字内容推荐场景的装置, 其特征在于, 所述建立模块, 包括:
区分单元, 用于根据各场景的关联度区分数据库中各场景样本所对应 场景类型;
标识单元, 用于标识各所述场景类型所对应的类型标签。
[权利要求 10] 根据权利要求 6所述的基于文字内容推荐场景信息的装置, 其特征在 于, 包括:
辨别模块, 用于辨别新文本内容中的词组信息, 第二匹配模块, 用于根据所述词组信息自动匹配与所述词组信息相关 联的所述场景;
第二链接模块, 用于将所述场景对应的场景样本的存储路径自动链接 到所述词组信息上。
PCT/CN2017/110567 2017-11-10 2017-11-10 基于文字内容推荐场景信息的方法与装置 WO2019090737A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/110567 WO2019090737A1 (zh) 2017-11-10 2017-11-10 基于文字内容推荐场景信息的方法与装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2017/110567 WO2019090737A1 (zh) 2017-11-10 2017-11-10 基于文字内容推荐场景信息的方法与装置

Publications (1)

Publication Number Publication Date
WO2019090737A1 true WO2019090737A1 (zh) 2019-05-16

Family

ID=66439054

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/110567 WO2019090737A1 (zh) 2017-11-10 2017-11-10 基于文字内容推荐场景信息的方法与装置

Country Status (1)

Country Link
WO (1) WO2019090737A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801142A (zh) * 2004-12-31 2006-07-12 北京联动在线通讯科技有限公司 对屏幕显示的特定匹配词语实现快速链接的方法
CN102136199A (zh) * 2011-03-10 2011-07-27 刘超 线上电子书阅读器和线上电子书编辑器
US20110209037A1 (en) * 2010-02-19 2011-08-25 Samsung Electronics Co., Ltd. Method for providing link and electronic apparatus thereof
CN103853544A (zh) * 2012-12-07 2014-06-11 华为技术有限公司 生成超链接的方法、装置和终端
CN104951439A (zh) * 2014-03-24 2015-09-30 明博教育科技有限公司 一种电子书籍及其相关电子资源的整合获取方法及系统
CN105335421A (zh) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 制作和展示电子书扩展内容的方法和装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1801142A (zh) * 2004-12-31 2006-07-12 北京联动在线通讯科技有限公司 对屏幕显示的特定匹配词语实现快速链接的方法
US20110209037A1 (en) * 2010-02-19 2011-08-25 Samsung Electronics Co., Ltd. Method for providing link and electronic apparatus thereof
CN102136199A (zh) * 2011-03-10 2011-07-27 刘超 线上电子书阅读器和线上电子书编辑器
CN103853544A (zh) * 2012-12-07 2014-06-11 华为技术有限公司 生成超链接的方法、装置和终端
CN104951439A (zh) * 2014-03-24 2015-09-30 明博教育科技有限公司 一种电子书籍及其相关电子资源的整合获取方法及系统
CN105335421A (zh) * 2014-08-06 2016-02-17 阿里巴巴集团控股有限公司 制作和展示电子书扩展内容的方法和装置

Similar Documents

Publication Publication Date Title
Rodgers et al. Incidental vocabulary learning through viewing television
Crow et al. Media in the policy process: Using framing and narratives to understand policy influences
CN106156354B (zh) 一种学习资源推荐系统
US20170004128A1 (en) Device and method for analyzing reputation for objects by data mining
Chen et al. Mining students' learning patterns and performance in Web-based instruction: a cognitive style approach
KR20150096295A (ko) 문답 데이터베이스 구축 시스템 및 방법, 그리고 이를 이용한 검색 시스템 및 방법
Vakulenko et al. Enriching iTunes App Store Categories via Topic Modeling.
CN110390025A (zh) 封面配图确定方法、装置、设备及计算机可读存储介质
CN112052414A (zh) 一种数据处理方法、装置以及可读存储介质
Sallabank The role of social networks in endangered language maintenance and revitalization: The case of Guernesiais in the Channel Islands
CN106096609A (zh) 一种基于ocr的商品查询关键字自动生成方法
CN112001739A (zh) 一种生成用户学习画像的方法和系统
CN111882445A (zh) 跨系统保险用户信息管理方法、装置、设备及可读介质
CN106776760A (zh) 应用于智能终端的搜题方法及装置
CN108121715A (zh) 一种文字标签方法及文字标签装置
CN109992752A (zh) 合同文件的标签标记方法、装置、计算机装置及存储介质
CN103699370B (zh) 一种基于xml的问卷脚本语言设计和开发方法
US9165053B2 (en) Multi-source contextual information item grouping for document analysis
JP4953428B2 (ja) コミュニティへの関連情報提供システム
WO2019090737A1 (zh) 基于文字内容推荐场景信息的方法与装置
CN111507062A (zh) 文本的展示方法及装置、系统、存储介质、电子装置
WO2020175662A1 (ja) 辞書作成装置、辞書作成方法及び辞書作成プログラム
CN109241438B (zh) 基于要素的跨通道热点事件发现方法、装置及存储介质
CN109598658B (zh) 一种新生入学服务系统和方法
CN108021617A (zh) 基于文字内容推荐场景信息的方法与装置

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (=EPO FORM 1205A DATED 22.10.2020)

122 Ep: pct application non-entry in european phase

Ref document number: 17931119

Country of ref document: EP

Kind code of ref document: A1