WO2013161510A1 - 評価の極性に基づいた文章の分類方法、コンピュータ・プログラム、コンピュータ - Google Patents
評価の極性に基づいた文章の分類方法、コンピュータ・プログラム、コンピュータ Download PDFInfo
- Publication number
- WO2013161510A1 WO2013161510A1 PCT/JP2013/059472 JP2013059472W WO2013161510A1 WO 2013161510 A1 WO2013161510 A1 WO 2013161510A1 JP 2013059472 W JP2013059472 W JP 2013059472W WO 2013161510 A1 WO2013161510 A1 WO 2013161510A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- evaluation
- sentence
- degree
- expression
- computer
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Definitions
- the present invention relates to information processing technology, and more particularly to technology that supports reputation analysis using information processing.
- the present invention has been made in view of such problems, and one of its purposes is to analyze a large amount of review texts in order to efficiently analyze the review texts with limited time and resources.
- the object is to provide a technique for efficiently extracting a part of review texts to be referred to by a person (analyst).
- the present invention is a method for extracting a part of a sentence from a plurality of sentences by a computer, the step of primarily evaluating the degree of positive expression and the degree of negative expression of each sentence, Secondary evaluation of each sentence based on a plurality of evaluation functions using the degree of positive expression and the degree of negative expression as variables, and sentences having higher evaluation results based on the same evaluation function And extracting a sentence with priority over a sentence having a lower evaluation result.
- one of the partial evaluation functions may be a function that outputs a higher evaluation result with respect to sentences that include positive and negative expressions on average.
- the following formula can be adopted for this function. p + n C p ⁇ p (1 ⁇ ) n ⁇ (p + n): where ⁇ is a popular rate in all documents.
- the sum of the positive expression level and the negative expression level, or the difference between the positive expression level and the negative expression level can be selected.
- the degree of the positive expression and the degree of the negative expression are primarily evaluated based on the number of positive expressions and the number of negative expressions included in each sentence. You can also.
- sentences of each evaluation result based on different evaluation functions can be extracted based on a predetermined order.
- it may further include a step of outputting the extracted sentence to the user.
- a step of outputting the extracted sentence it is possible to output to the user whether the extracted sentence is extracted based on the evaluation result of which evaluation function.
- the positive expression and the negative expression in the extracted sentence can be output in different expression forms.
- FIG. 1 is a conceptual diagram illustrating a review site system.
- This system includes a review site server 2 and a user terminal, which are connected to each other via the Internet 4 so as to communicate with each other.
- the user terminal any form of computer having a communication function can be adopted.
- a personal data assistant PDA, personal digital assistant
- on-vehicle computer netbook, etc. (not shown) can be employed. .
- FIG. 2 explains the data structure of data stored in the hard disk devices 20 and 21 in the review site server 2.
- a transmission date / time create create (created_at) indicating the date / time when each review text was transmitted / posted
- a review ID (id) for identifying each review
- It includes a user ID (user_id) that identifies the user who sent the review, and text that is the content of the review.
- the user table (FIG. 2B) stored in the hard disk device 21 includes a user ID (user_id) for identifying the user, the gender, age, and residence (location) of the user. ) To identify each. In addition, you may add ID which specifies the product and service of review object to review ID.
- FIG. 3 is a block diagram illustrating the hardware configuration of the personal computer 1.
- the hardware configuration of the computer 1 includes a (low-speed and high-speed) bus 10, a CPU (arithmetic control device) 11 connected to the bus 10, a RAM (random access memory: storage device) 12, a ROM (read-only memory).
- a memory (storage device) 13, an HDD (hard disk drive: storage device) 14, a communication interface 15, and an input / output interface 16 are provided.
- a mouse 17 connected to the input / output interface 16, a flat panel display (display device) 18, a keyboard 19 and the like are provided.
- the computer 1 has been described as adopting a general personal computer architecture, for example, the CPU 11 and the HDD 14 can be multiplexed in order to obtain higher data processing capability and availability. In addition to the desktop type, various types of computer systems can be employed.
- the software configuration of the computer 1 includes an operating system (OS) that provides basic functions, application software that uses the functions of the OS, and driver software for input / output devices. These pieces of software are loaded onto the RAM 12 together with various data and executed by the CPU 11 or the like, and the computer 1 functions as a functional module shown in FIG. 4 as a whole and executes the processing shown in FIG.
- OS operating system
- driver software driver software for input / output devices.
- FIG. 4 is a functional block diagram illustrating functional modules of the computer 1 according to the embodiment.
- the computer 1 functions as a primary evaluation module 101, a secondary evaluation module 102, an extraction module 103, and an output module 104.
- FIG. 5 is a flowchart for explaining processing executed by the computer 1.
- FIG. 6 is a conceptual diagram for explaining steps S101 to S103 in the flowchart of FIG.
- the computer 1 obtains a plurality of review sentences D (1) to D (N) each including a specific proper noun (herein referred to as PPP) (S101).
- N is a sufficiently large value that it is difficult to display a list of review texts.
- the condition is transmitted from the computer 1 to the review site server 2.
- a condition for example, it is possible to attach a condition that the proper noun PPP is included in the review text and the transmitted period is within a specified period.
- the computer 1 receives from the review site server 2 data of a review text group that meets the above conditions.
- the review group data (see FIG. 2A) that meets the above conditions and the user profile associated with the review (see FIG. 2B) are received. These received data are stored in the HDD 14 of the computer 1. Further, the review may be sequentially sent from the review site server 2 to the computer 1 until the period includes the future and the end of the period comes. As a different aspect, a large amount of review sentence group data (see FIG. 2A) and a user profile (see FIG. 2B) associated with the review sentence are stored in advance in the HDD 14 of the computer 1. In some cases, a review group that meets the above conditions can be searched from these data.
- the primary evaluation module 101 performs primary evaluation on the degree of positive expression and the degree of negative expression of each review (step S101).
- a list of words indicating favorable (positive expression) and a list of words / expressions indicating unfavorable (negative expression) are stored in the HDD 14 in advance, and the number of popularity (p) in each review sentence Count the unpopular number (n).
- These evaluation words and expressions are preferably set in advance according to the product or service to be evaluated.
- a score according to popularity for each popular word / expression and a score according to unpopularity for each unfavorable word / expression are set in advance, and primary evaluation is performed based on these scores. Can also be done.
- each review text is secondarily evaluated by the secondary evaluation module 102 based on a plurality of evaluation functions (step S102).
- evaluation function m1 p Evaluation function m2: p + n Evaluation function m3: p / (p + n) Evaluation function m4: pn Evaluation function m5: p + n C p ⁇ p (1- ⁇ ) n ⁇ (p + n), ⁇ is the percentage of favorable reviews in all reviews
- each evaluation function is as follows. That is, the evaluation function m1 highly evaluates the review sentence D including many popular words.
- the rating function 5 gives a higher rating for reviews that on average contain positive and negative words.
- the density function of the P rate in k trials (k evaluation expressions) is obtained by squashing the density function of the binomial distribution B (k, ⁇ ) to 1 / k in the horizontal axis direction.
- the probability that the P rate P x / (P x + N x ) due to the unfavorable numbers P x and N x of the document x is observed is used as an index of the evaluation expression of the document x.
- the calculation of m C n takes time when m and n are large, and may be approximated by normal distribution or Poisson distribution as necessary.
- FIG. 7 is a graph illustrating the evaluation functions m1, m2, and m5.
- the number P of popular words is plotted on the vertical axis
- the number N of unpopular words is plotted on the horizontal axis
- each review sentence D is plotted on a mark x on the graph.
- the contour line of the evaluation function m1 is represented by a thin line
- the contour line of the evaluation function m2 is represented by a dotted line
- the contour line of the evaluation function m5 is represented by a thick line.
- the review text includes various reviews such as reviews with few evaluation expressions, reviews with dissatisfaction, etc., fan reviews with the evaluation function m1, reviews with many evaluation expressions with the evaluation function m2, and average opinions with the evaluation function m5 Shows that reviews with many evaluation expressions can be effectively separated.
- K review sentences are extracted by the extraction module 103 (step S103).
- Each review sentence D has five evaluation scores based on the evaluation functions m1 to m5.
- the extraction module 103 sorts the review sentences in order from the sentence with the highest evaluation score for each evaluation function.
- a review sentence having the highest evaluation score based on the evaluation function m5 is selected.
- the review sentence having the second highest evaluation score by the evaluation function m1 the review sentence having the second highest evaluation score by the evaluation function m2, the review sentence having the second highest evaluation score by the evaluation function m3, and the evaluation score by the evaluation function m4
- the review text that is the second highest, and the review text that is the second highest after the evaluation score by the evaluation function m5 are selected.
- sentences are selected until K cases are reached.
- FIG. 8 shows an example of the display screen. In this screen, some of the selected K review sentences are displayed. If you pay attention to the part surrounded by a thick square, the displayed review sentence will get a high evaluation score by which evaluation function. It is displayed so that you can see at a glance whether it has been selected.
- the review text displayed at the top of the screen has a check mark in the columns m1, m2, and m5, and is displayed based on the evaluation results by the evaluation functions m1, m2, and m5. I understand. It is also possible to delete an evaluation function that the user does not need and add a new evaluation function. Furthermore, it is possible to display only the evaluation result based on a specific evaluation function. Furthermore, the evaluation result can be indicated more simply by an icon. Note that the output module 104 can also display the graph shown in FIG. 7 on the display 18.
- the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment containing both hardware and software elements.
- the invention is implemented in software, including but not limited to firmware, resident software, microcode, parsing picocode, and the like.
- the present invention can also take the form of a computer program or computer-readable medium comprising program code for use by or in connection with a computer or any instruction execution system.
- a computer-readable medium is any apparatus that can contain, store, communicate, propagate, or transmit a program for use by or in connection with any instruction execution system, apparatus, or device. It can be.
- the syntax analysis control module described above constitutes an instruction execution system or a computer in this sense.
- the medium can be an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system (or apparatus or device) or a propagation medium.
- Examples of computer readable media include semiconductor or solid state memory, magnetic tape, removable computer diskette, random access memory (RAM), read-only memory (ROM), rigid magnetic disk. And optical discs. Current examples of optical disks include compact disk read only memory (CD-ROM: compact disk read only memory), compact disk read / write (CD-R / W) memory, DVD Is included.
- a data processing system suitable for storing and / or executing program code may include at least one processor coupled directly or indirectly to memory elements through a system bus. This memory element contains at least some of the local memory used in the actual execution of the program code, the bulk storage, and the number of times it must be read from the bulk storage during execution.
- the program code can include a cache memory that provides temporary storage.
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
以下、本発明を実施するための最良の形態を図面に基づいて詳細に説明するが、以下の実施形態は特許請求の範囲にかかる発明を限定するものではなく、また実施形態の中で説明されている特徴の組み合わせの全てが発明の解決手段に必須であるとは限らない。また、本発明は多くの異なる態様で実施することが可能であり、実施の形態の記載内容に限定して解釈されるべきものではない。また、実施の形態の中で説明されている特徴の組み合わせの全てが発明の解決手段に必須とは限らないことに留意されたい。実施の形態の説明の全体を通じて(特段の断りのない限り)同じ要素には同じ番号を付している。
レビュー文章1:好評5・不評0
(分量が多く)、(安い)ので(いいですね)!他のスキンケア商品との(相性もいい)ので(安心しています)
レビュー文章2:好評5・不評1
特に、(ニキビに悩んでいるのではないです)が、(価格も手ごろ)だったので購入。(刺激も少なく)、若いうちには(良いと思います)。現在は、乾燥肌なので、化粧水などつけたとしても<潤いが不足します>…よって、現在は使用していません。ニキビに悩んでいる方の使用も(良さそう)です。
レビュー文章3:好評4・不評1
肌荒れ、ニキビに(安心して使えます)。普段は拭き取り化粧水を使っています。軽く顔にすべらせたあとパッティングします。ニキビができているところに多めに使うと、たしかに(よくなります)。潤いに関しては乾燥肌の方には<物足りないかもしれません>が、脂性肌の私には(ちょうどいい)です。(便利な化粧水です)。
レビュー文章4:好評2・不評3
<あまり効果判りません>。ニキビや吹き出物も最近は出来ないので<必要性を感じません>。肌にヒリヒリを感じたことは無いので、拭き取り用として使ってはいます。リピーと購入は<微妙かな>?でもお値段が(安く)て(いい)??
評価関数m2: p+n
評価関数m3: p/(p+n)
評価関数m4: p-n
評価関数m5: p+nCpαp(1-α)n× (p+n)、αは全レビュー中の好評の割合
11…CPU(演算制御装置)
12…RAM(ランダム・アクセス・メモリ:記憶装置)
13…ROM(リード・オンリ・メモリ:記憶装置)
14…HDD(ハード・ディスク・ドライブ:記憶装置)
15…通信インタフェース
16…入出力インタフェース
17…マウス
18…フラット・パネル・ディスプレイ(表示装置)
2…レビューサイト・サーバ
20、21…ハード・ディスク・ドライブ
31…スマートフォン
32…タブレット
33…(ノート型)パーソナル・コンピュータ
101…一次評価モジュール
102…二次評価モジュール
103…抽出モジュール
104…出力モジュール
Claims (12)
- コンピュータにより複数の文章から一部の文章を抽出する方法であり、
各文章の肯定的な表現の程度及び否定的な表現の程度を一次評価するステップと、
少なくとも一部は前記肯定的な表現の程度及び前記否定的な表現の程度を変数とする複数の評価関数に基づいて、各文章を二次評価するステップと、
同一の評価関数に基づいた各評価結果のより高い文章を前記評価結果のより低い文章よりも優先して文章を抽出するステップと
を含む方法。 - 前記一部の評価関数の一は、肯定的及び否定的な表現を平均的に含む文章に対してより高い評結果を出力する関数である請求項1に記載の方法。
- 前記一部の評価関数の一は、p+nCpαp(1-α)n
×(p+n):pは肯定的な表現数、nは否定的な表現数、αは全文書中の好評の割合、で与えられる請求項1に記載の方法。 - 前記一部の評価関数の一は、前記肯定的な表現の程度と前記否定的な表現の程度の和である請求項1に記載の方法。
- 前記一部の評価関数の一は、前記肯定的な表現の程度と前記否定的な表現の程度の差である請求項1に記載の方法。
- 前記一次評価するステップは、各文章に含まれる肯定的な表現の数及び否定的な表現の数に基づいて、前記肯定的な表現の程度及び否定的な表現の程度を一次評価する請求項1に記載の方法。
- 前記抽出するステップは、異なる評価関数に基づいた各評価結果の文章を予め定められた順序に基づいて抽出する請求項1に記載の方法。
- 前記抽出された文章をユーザに出力するステップを更に備える請求項1に記載の方法。
- 前記出力するステップでは、前記抽出された文章がいずれの評価関数の評価結果に基づいて抽出されたのかを併せてユーザに出力する請求項8に記載の方法。
- 前記出力するステップでは、前記抽出された文章中の前記肯定的な表現と前記否定的な表現とを相異なる表現形態で出力する請求項8に記載の方法。
- コンピュータに請求項1に記載の方法の各ステップを実行させるコンピュータ・プログラム。
- 複数の文章から一部の文章を抽出するコンピュータであり、
各文章の肯定的な表現の程度及び否定的な表現の程度を一次評価する手段と、
少なくとも一部は前記肯定的な表現の程度及び前記否定的な表現の程度を変数とする複数の評価関数に基づいて、各文章を二次評価する手段と、
同一の評価関数に基づいた各評価結果のより高い文章を前記評価結果のより低い文章よりも優先して文章を抽出する手段と
を含むコンピュータ。
Priority Applications (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201380021425.4A CN104272301B (zh) | 2012-04-25 | 2013-03-29 | 用于提取一部分文本的方法、计算机可读介质和计算机 |
JP2014512437A JP5607859B2 (ja) | 2012-04-25 | 2013-03-29 | 評価の極性に基づいた文章の分類方法、コンピュータ・プログラム、コンピュータ |
DE112013002187.0T DE112013002187T5 (de) | 2012-04-25 | 2013-03-29 | Verfahren zum Klassifizieren von Texteinheiten auf der Grundlage von Bewertungsgegensätzen, Computerprogrammprodukt und Computer dafür |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2012100288 | 2012-04-25 | ||
JP2012-100288 | 2012-04-25 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2013161510A1 true WO2013161510A1 (ja) | 2013-10-31 |
Family
ID=49478065
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2013/059472 WO2013161510A1 (ja) | 2012-04-25 | 2013-03-29 | 評価の極性に基づいた文章の分類方法、コンピュータ・プログラム、コンピュータ |
Country Status (5)
Country | Link |
---|---|
US (1) | US9740681B2 (ja) |
JP (1) | JP5607859B2 (ja) |
CN (1) | CN104272301B (ja) |
DE (1) | DE112013002187T5 (ja) |
WO (1) | WO2013161510A1 (ja) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019095919A (ja) * | 2017-11-20 | 2019-06-20 | ヤフー株式会社 | 情報処理装置、情報処理方法および情報処理プログラム |
JP2020091730A (ja) * | 2018-12-06 | 2020-06-11 | ヤフー株式会社 | 情報処理装置、情報処理方法、およびプログラム |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2008234090A (ja) * | 2007-03-19 | 2008-10-02 | Fujitsu Ltd | 最新評判情報通知プログラム、記録媒体、装置及び方法 |
JP2009510637A (ja) * | 2005-09-30 | 2009-03-12 | グーグル インコーポレイテッド | 表示のための高品質レビューの選択 |
JP2011085986A (ja) * | 2009-10-13 | 2011-04-28 | Nippon Telegr & Teleph Corp <Ntt> | テキスト要約方法、その装置およびプログラム |
Family Cites Families (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1027181A (ja) | 1996-07-11 | 1998-01-27 | Fuji Xerox Co Ltd | 文書評価装置 |
JP2002140465A (ja) | 2000-08-21 | 2002-05-17 | Fujitsu Ltd | 自然文処理装置及び自然文処理用プログラム |
US6622140B1 (en) | 2000-11-15 | 2003-09-16 | Justsystem Corporation | Method and apparatus for analyzing affect and emotion in text |
US7346492B2 (en) * | 2001-01-24 | 2008-03-18 | Shaw Stroz Llc | System and method for computerized psychological content analysis of computer and media generated communications to produce communications management support, indications, and warnings of dangerous behavior, assessment of media images, and personnel selection support |
US7058566B2 (en) * | 2001-01-24 | 2006-06-06 | Consulting & Clinical Psychology, Ltd. | System and method for computer analysis of computer generated communications to produce indications and warning of dangerous behavior |
US7149804B2 (en) * | 2001-04-30 | 2006-12-12 | Sony Computer Entertainment America Inc. | Method and system for providing evaluation of text-based products |
JP3962382B2 (ja) * | 2004-02-20 | 2007-08-22 | インターナショナル・ビジネス・マシーンズ・コーポレーション | 表現抽出装置、表現抽出方法、プログラム及び記録媒体 |
JP4148522B2 (ja) * | 2004-11-19 | 2008-09-10 | インターナショナル・ビジネス・マシーンズ・コーポレーション | 表現検出システム、表現検出方法、及びプログラム |
JP2007299071A (ja) | 2006-04-27 | 2007-11-15 | Fuji Xerox Co Ltd | 評判情報処理システム、評判情報処理方法及び評判情報処理プログラム |
US20100017391A1 (en) | 2006-12-18 | 2010-01-21 | Nec Corporation | Polarity estimation system, information delivery system, polarity estimation method, polarity estimation program and evaluation polarity estimatiom program |
US7996210B2 (en) * | 2007-04-24 | 2011-08-09 | The Research Foundation Of The State University Of New York | Large-scale sentiment analysis |
US20080288481A1 (en) * | 2007-05-15 | 2008-11-20 | Microsoft Corporation | Ranking online advertisement using product and seller reputation |
US8280885B2 (en) * | 2007-10-29 | 2012-10-02 | Cornell University | System and method for automatically summarizing fine-grained opinions in digital text |
JP5229504B2 (ja) | 2007-11-05 | 2013-07-03 | 日本電気株式会社 | 広告提示方法、広告提示システム及びプログラム |
US20100318526A1 (en) * | 2008-01-30 | 2010-12-16 | Satoshi Nakazawa | Information analysis device, search system, information analysis method, and information analysis program |
US8117207B2 (en) * | 2008-04-18 | 2012-02-14 | Biz360 Inc. | System and methods for evaluating feature opinions for products, services, and entities |
WO2010036012A2 (ko) * | 2008-09-23 | 2010-04-01 | 주식회사 버즈니 | 인터넷을 이용한 의견 검색 시스템, 의견 검색 및 광고 서비스 시스템과 그 방법 |
JP4683394B2 (ja) * | 2008-09-26 | 2011-05-18 | Necビッグローブ株式会社 | 情報処理装置、情報処理方法およびプログラム |
JP5359399B2 (ja) * | 2009-03-11 | 2013-12-04 | ソニー株式会社 | テキスト分析装置および方法、並びにプログラム |
US20110082687A1 (en) * | 2009-10-05 | 2011-04-07 | Marcelo Pham | Method and system for taking actions based on analysis of enterprise communication messages |
US8990124B2 (en) * | 2010-01-14 | 2015-03-24 | Microsoft Technology Licensing, Llc | Assessing quality of user reviews |
CN102163189B (zh) * | 2010-02-24 | 2014-07-23 | 富士通株式会社 | 从评论性文本中提取评价性信息的方法和装置 |
US8402035B2 (en) * | 2010-03-12 | 2013-03-19 | General Sentiment, Inc. | Methods and systems for determing media value |
US20110320542A1 (en) * | 2010-06-28 | 2011-12-29 | Bank Of America Corporation | Analyzing Social Networking Information |
CN101894102A (zh) * | 2010-07-16 | 2010-11-24 | 浙江工商大学 | 一种主观性文本情感倾向性分析方法和装置 |
US8949211B2 (en) * | 2011-01-31 | 2015-02-03 | Hewlett-Packard Development Company, L.P. | Objective-function based sentiment |
US8650023B2 (en) * | 2011-03-21 | 2014-02-11 | Xerox Corporation | Customer review authoring assistant |
US20120246054A1 (en) * | 2011-03-22 | 2012-09-27 | Gautham Sastri | Reaction indicator for sentiment of social media messages |
WO2012143069A1 (en) * | 2011-04-21 | 2012-10-26 | Sony Corporation | A method for determining a sentiment from a text |
US20120290606A1 (en) * | 2011-05-11 | 2012-11-15 | Searchreviews LLC | Providing sentiment-related content using sentiment and factor-based analysis of contextually-relevant user-generated data |
US20120290910A1 (en) * | 2011-05-11 | 2012-11-15 | Searchreviews LLC | Ranking sentiment-related content using sentiment and factor-based analysis of contextually-relevant user-generated data |
US8600796B1 (en) * | 2012-01-30 | 2013-12-03 | Bazaarvoice, Inc. | System, method and computer program product for identifying products associated with polarized sentiments |
-
2013
- 2013-03-29 CN CN201380021425.4A patent/CN104272301B/zh not_active Expired - Fee Related
- 2013-03-29 WO PCT/JP2013/059472 patent/WO2013161510A1/ja active Application Filing
- 2013-03-29 DE DE112013002187.0T patent/DE112013002187T5/de not_active Ceased
- 2013-03-29 JP JP2014512437A patent/JP5607859B2/ja not_active Expired - Fee Related
- 2013-04-12 US US13/861,430 patent/US9740681B2/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2009510637A (ja) * | 2005-09-30 | 2009-03-12 | グーグル インコーポレイテッド | 表示のための高品質レビューの選択 |
JP2008234090A (ja) * | 2007-03-19 | 2008-10-02 | Fujitsu Ltd | 最新評判情報通知プログラム、記録媒体、装置及び方法 |
JP2011085986A (ja) * | 2009-10-13 | 2011-04-28 | Nippon Telegr & Teleph Corp <Ntt> | テキスト要約方法、その装置およびプログラム |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019095919A (ja) * | 2017-11-20 | 2019-06-20 | ヤフー株式会社 | 情報処理装置、情報処理方法および情報処理プログラム |
JP2020091730A (ja) * | 2018-12-06 | 2020-06-11 | ヤフー株式会社 | 情報処理装置、情報処理方法、およびプログラム |
JP6993955B2 (ja) | 2018-12-06 | 2022-01-14 | ヤフー株式会社 | 情報処理装置、情報処理方法、およびプログラム |
Also Published As
Publication number | Publication date |
---|---|
JP5607859B2 (ja) | 2014-10-15 |
US9740681B2 (en) | 2017-08-22 |
CN104272301A (zh) | 2015-01-07 |
CN104272301B (zh) | 2018-01-23 |
DE112013002187T5 (de) | 2015-01-08 |
US20130289978A1 (en) | 2013-10-31 |
JPWO2013161510A1 (ja) | 2015-12-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Nam et al. | Harvesting brand information from social tags | |
Cacciatore et al. | Coverage of emerging technologies: A comparison between print and online media | |
Tirunillai et al. | Mining marketing meaning from online chatter: Strategic brand analysis of big data using latent dirichlet allocation | |
Jerath et al. | Consumer click behavior at a search engine: The role of keyword popularity | |
Khang et al. | Social media research in advertising, communication, marketing, and public relations, 1997–2010 | |
Faulconbridge | Global architects: learning and innovation through communities and constellations of practice | |
Burda et al. | Sustaining accessibility of information through digital preservation: A literature review | |
Wang et al. | Attribute embedding: Learning hierarchical representations of product attributes from consumer reviews | |
Harvey et al. | Fear and derision: a quantitative content analysis of provaccine and antivaccine internet memes | |
Dahlstrom et al. | Third-person perception of science narratives: The case of climate change denial | |
Keshavarz et al. | Credibility evaluation of scientific information on websites: Designing and evaluating an exploratory model | |
Paul et al. | TexTonic: Interactive visualization for exploration and discovery of very large text collections | |
Tong et al. | A data-driven approach for integrating hedonic quality and pragmatic quality in user experience modeling | |
Karimi et al. | Online news media website ranking using user-generated content | |
JP5607859B2 (ja) | 評価の極性に基づいた文章の分類方法、コンピュータ・プログラム、コンピュータ | |
Downer et al. | All work and no play: A text analysis | |
Wedel et al. | A bilingual comparison of sentiment and topics for a product event on Twitter | |
Zhang et al. | Providing consumers with a representative subset from online reviews | |
Flaxton | HD aesthetics | |
CN112802454B (zh) | 一种唤醒词的推荐方法、装置、终端设备及存储介质 | |
Liang et al. | Exploring online reviews for user experience modeling | |
Tonkin | A day at work (with text): A brief introduction | |
Fu | The cultural influences of narrative content on consumers’ perceptions of helpfulness | |
Read et al. | Labeling emotions in suicide notes: Cost-sensitive learning with heterogeneous features | |
Singh et al. | A clustering and opinion mining approach to socio-political analysis of the blogosphere |
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: 13780607 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2014512437 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 112013002187 Country of ref document: DE Ref document number: 1120130021870 Country of ref document: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 13780607 Country of ref document: EP Kind code of ref document: A1 |