WO2017119060A1 - Système de présentation d'informations - Google Patents

Système de présentation d'informations Download PDF

Info

Publication number
WO2017119060A1
WO2017119060A1 PCT/JP2016/050085 JP2016050085W WO2017119060A1 WO 2017119060 A1 WO2017119060 A1 WO 2017119060A1 JP 2016050085 W JP2016050085 W JP 2016050085W WO 2017119060 A1 WO2017119060 A1 WO 2017119060A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
argument
objection
processing unit
presentation system
Prior art date
Application number
PCT/JP2016/050085
Other languages
English (en)
Japanese (ja)
Inventor
利昇 三好
利彦 柳瀬
佐藤　美沙
孝介 柳井
芳樹 丹羽
Original Assignee
株式会社日立製作所
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 株式会社日立製作所 filed Critical 株式会社日立製作所
Priority to PCT/JP2016/050085 priority Critical patent/WO2017119060A1/fr
Publication of WO2017119060A1 publication Critical patent/WO2017119060A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/10—Text processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N5/00—Computing arrangements using knowledge-based models
    • G06N5/04—Inference or reasoning models

Definitions

  • the present invention relates to an information presentation system.
  • the construction processing unit 110 stores the document ID including this sentence and information on the position of the sentence in the document together with the syntax analysis result 501 of FIG. As a result, the document ID and the position information of the sentence in the document can be searched from the annotation knowledge database 206.
  • the construction processing unit 110 may store the position of the first character of the sentence (“E” in the above example) in the document in byte units as the sentence position information. Further, the construction processing unit 110 assigns each sentence in the document with an ID such as 1, 2, 3,... From the first sentence, and assigns the ID to the annotation knowledge database 206 as a sentence ID. It may be stored. Thereby, if the document ID and the sentence ID are designated, the sentence in the document can be uniquely identified, and the position of the sentence in the document can be specified by the position of the sentence in byte units.
  • the construction processing unit 110 may search for a predicate using the result of the syntax analysis 202 and extract the above argument using the dependency relationship between phrases.
  • software such as OpenIE may be used as the construction processing unit 110.
  • FIG. 7 shows an example in which the result of the relation information A 601 in FIG.
  • the table 701 includes id 702, tag 703, token 704, begin 705, and end 706 as configuration items.
  • id 702 is an id for identifying the token 704.
  • a tag 703 is a tag for the token 704. As an example, the tag 703 stores information about whether it is a predicate or a term. The tag 703 may store other information.
  • begin 705 is the start position of token 704.
  • end 706 is the end position of token 704.
  • the table 701 indicates that “a quick rest” and “productivity” are the first argument and the second argument as arguments of “increase”, respectively.
  • the construction processing unit 110 generates a node based on the argument obtained by the relationship information extraction 203. For each argument of the first argument or the second argument, each argument is processed as follows, for example.
  • the non-importance attribute indicates whether an expression that weakens the importance is included. For example, if the expression includes small, tiny, little, few, yet, lower, smaller, lowly, less, only, limited, low,, etc., the non-importance attribute is true, and these are not included In addition, the non-importance attribute is false. For example, in the example of 606 in FIG. 6, regarding the argument “only limited revenue”, the non-importance attribute is true.
  • edges that connect the same nodes in the same direction and that have the same core representation are considered the same.
  • Indirect objection is an objection from a different perspective than the opponent's claim.
  • Indirect objection is a method of objection from a different point of view, and in this example, the introduction of uniforms is equivalent to objection of suppressing individuality.
  • the objection from this other point of view is the objection of the other party because discipline and individuality are in a trade-off relationship. For example, if a uniform is adopted, a specific uniform will be protected and sold, so it is better to argue against individuality as in this example, rather than arguing against the freedom and fairness of the market. Is persuasive.
  • the argument generation processing unit 111 estimates a value that is in a trade-off relationship with the value given by the other party, and generates an objection based on the value.
  • trade-offs are welfare and tax burden, economic development and environmental destruction.
  • the argument generation processing unit 111 searches the knowledge that is the basis of the path and acquires the knowledge.
  • the edge has a reference attribute that specifies a predicate as a source stored in the annotation knowledge database 206.
  • the argument generation processing unit 111 can acquire a relationship, a sentence, an article, and the like including the predicate of the edge from the annotation knowledge database 206 using the reference attribute.
  • the node 1109 is connected to the node 1112.
  • the education at node 1112 has positive value as a value attribute.
  • the lottery of the node 1109 indicates that the education funding of the node 1112 is supported. Therefore, the path of node 1107-edge 1114-node 1109-edge 1117-node 1108 not only supports government revenue, but also reduces (has an adverse effect) the lottery revenue that supports the positive value of education. Therefore, the argument is stronger.
  • Example 2 In the above-described first embodiment, the method for generating the objection against the opponent argument has been described. The system described above can be used to measure the robustness of an argument when it is an argument.
  • the present invention is not limited to the above-described embodiments, and includes various modifications.
  • the above embodiments have been described in detail for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described.
  • a part of the configuration of one embodiment can be replaced with the configuration of another embodiment.
  • the structure of another Example can also be added to the structure of a certain Example.
  • another configuration can be added, deleted, or replaced.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Computational Linguistics (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Data Mining & Analysis (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 système de présentation d'informations comprenant : un processeur ; une unité de mémorisation pour la mémorisation de connaissances inférentielles comprenant des informations concernant des jetons d'une phrase et des polarités des jetons ; et une unité d'affichage. Le processeur émet un résultat d'analyse après l'analyse d'informations d'argument concernant un sujet donné, génère des informations de contre-argument pour le résultat d'analyse conformément aux connaissances inférentielles et affiche le contre-argument sur l'unité d'affichage.
PCT/JP2016/050085 2016-01-05 2016-01-05 Système de présentation d'informations WO2017119060A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/050085 WO2017119060A1 (fr) 2016-01-05 2016-01-05 Système de présentation d'informations

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2016/050085 WO2017119060A1 (fr) 2016-01-05 2016-01-05 Système de présentation d'informations

Publications (1)

Publication Number Publication Date
WO2017119060A1 true WO2017119060A1 (fr) 2017-07-13

Family

ID=59273499

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2016/050085 WO2017119060A1 (fr) 2016-01-05 2016-01-05 Système de présentation d'informations

Country Status (1)

Country Link
WO (1) WO2017119060A1 (fr)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113158671A (zh) * 2021-03-25 2021-07-23 胡明昊 一种结合命名实体识别的开放域信息抽取方法
WO2022113314A1 (fr) * 2020-11-27 2022-06-02 日本電信電話株式会社 Procédé d'apprentissage, programme d'apprentissage et dispositif d'apprentissage

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015075854A (ja) * 2013-10-08 2015-04-20 独立行政法人情報通信研究機構 矛盾表現収集装置及びそのためのコンピュータプログラム
WO2015151268A1 (fr) * 2014-04-04 2015-10-08 株式会社日立製作所 Procédé, système de génération de contre-arguments

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2015075854A (ja) * 2013-10-08 2015-04-20 独立行政法人情報通信研究機構 矛盾表現収集装置及びそのためのコンピュータプログラム
WO2015151268A1 (fr) * 2014-04-04 2015-10-08 株式会社日立製作所 Procédé, système de génération de contre-arguments

Non-Patent Citations (4)

* Cited by examiner, † Cited by third party
Title
HIDEYUKI SHIBUKI ET AL.: "Joho Shinpyosei Handan Shien ni Okeru User ga Shoten o Meikakuka shita Jokoyoka deno Chotei Yoyaku Seisei Shuho", PROCEEDINGS OF THE 18TH ANNUAL MEETING OF THE ASSOCIATION FOR NATURAL LANGUAGE PROCESSING, 31 March 2012 (2012-03-31), pages 835 - 838 *
HIROSHI KANEYAMA ET AL.: "Universal Dependencies- Nippongo Universal Dependencies no Shian", PROCEEDINGS OF THE 21TH ANNUAL MEETING OF THE ASSOCIATION FOR NATURAL LANGUAGE PROCESSING, vol. 5, no. 21, 9 March 2015 (2015-03-09), pages 505 - 508, XP055599188, Retrieved from the Internet <URL:http://www.anlp.jp/proceedings/annual_meeting/2015/pdf_dir/E3-4.pdf> [retrieved on 20160318] *
KATSUMI NITTA ET AL.: "A Negotiation Support Tool using a computer Network", IPSJ SIG NOTES 98- I C S - 1 1 3, vol. 98, no. 65, 24 July 1998 (1998-07-24), pages 99 - 106, ISSN: 0919-6072 *
YUSUKE YAMAMOTO ET AL.: "WebAlert : Web Joho noInsho Shuyaku o Riyo shita Etsuran Page Naiyo no Taisuru Hantai Iken Teiji", JOURNALL O F THE DBSJ, vol. 7, no. 1, 27 June 2008 (2008-06-27), pages 251 - 256, ISSN: 1883-1060 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022113314A1 (fr) * 2020-11-27 2022-06-02 日本電信電話株式会社 Procédé d'apprentissage, programme d'apprentissage et dispositif d'apprentissage
CN113158671A (zh) * 2021-03-25 2021-07-23 胡明昊 一种结合命名实体识别的开放域信息抽取方法
CN113158671B (zh) * 2021-03-25 2023-08-11 胡明昊 一种结合命名实体识别的开放域信息抽取方法

Similar Documents

Publication Publication Date Title
Tabassum et al. A survey on text pre-processing & feature extraction techniques in natural language processing
CN109241538B (zh) 基于关键词和动词依存的中文实体关系抽取方法
Dunietz et al. A new entity salience task with millions of training examples
Johansson et al. Relational features in fine-grained opinion analysis
Johansson et al. Syntactic and semantic structure for opinion expression detection
KR101136007B1 (ko) 문서 감성 분석 시스템 및 그 방법
US20190392035A1 (en) Information object extraction using combination of classifiers analyzing local and non-local features
WO2016051551A1 (fr) Système de génération de texte
JP6663826B2 (ja) 計算機及び応答の生成方法
Zhang et al. Natural language processing: a machine learning perspective
Yin et al. An exploration of embeddings for generalized phrases
US20160350278A1 (en) Claim polarity identification
Goryachev et al. Implementation and evaluation of four different methods of negation detection
Bouarroudj et al. Named entity disambiguation in short texts over knowledge graphs
Tachicart et al. Moroccan Arabic vocabulary generation using a rule-based approach
Dickinson Error detection and correction in annotated corpora
Dejean Learning Rules and Their Exceptions.
US11669574B2 (en) Method, apparatus, and computer-readable medium for determining a data domain associated with data
WO2017119060A1 (fr) Système de présentation d&#39;informations
Selvaretnam et al. A linguistically driven framework for query expansion via grammatical constituent highlighting and role-based concept weighting
Chopra et al. The Natural Language Processing Workshop: Confidently design and build your own NLP projects with this easy-to-understand practical guide
Džeroski et al. Learning to lemmatise Slovene words
Zavrel et al. Feature-Rich Memory-Based Classification for Shallow NLP and Information Extraction.
Danenas et al. Enhancing the extraction of SBVR business vocabularies and business rules from UML use case diagrams with natural language processing
Nishy Reshmi et al. Textual entailment classification using syntactic structures and semantic relations

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: 16883565

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: 16883565

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP