TW201123060A - News rating system and method. - Google Patents

News rating system and method. Download PDF

Info

Publication number
TW201123060A
TW201123060A TW98145242A TW98145242A TW201123060A TW 201123060 A TW201123060 A TW 201123060A TW 98145242 A TW98145242 A TW 98145242A TW 98145242 A TW98145242 A TW 98145242A TW 201123060 A TW201123060 A TW 201123060A
Authority
TW
Taiwan
Prior art keywords
news
score
management module
scoring
stock
Prior art date
Application number
TW98145242A
Other languages
Chinese (zh)
Other versions
TWI409714B (en
Inventor
Zheng-Ren Zhang
Original Assignee
Hsin Tian Ti Information Co Ltd
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 Hsin Tian Ti Information Co Ltd filed Critical Hsin Tian Ti Information Co Ltd
Priority to TW98145242A priority Critical patent/TW201123060A/en
Publication of TW201123060A publication Critical patent/TW201123060A/en
Application granted granted Critical
Publication of TWI409714B publication Critical patent/TWI409714B/zh

Links

Abstract

The present invention relates to a news rating system and method, including a scoring thesaurus, a real-time news database, a score management module, a keyword weight management module and a classification management module; the score management module performs evaluation and scoring for a target object (such as stock, character, movie, etc.) in real-time news by using the keywords of the scoring thesaurus, and the classification management module summarizes the scored news respectively into a dedicated data table of the corresponding target object, further according to the content of the dedicated database of the target object and based on the news daily volume, the amount of positive news, the amount of negative news to draw line-charts for using as reference indicators of stock price, popularity, support rate. Also, a score value is assigned to a rating keyword in the scoring thesaurus first, and the score value will be adjusted by feedback comparing its ascending or descending range after a period of time since the news event, thereby improving the accuracy of the rating system.

Description

201123060 六、發明說明: 【發明所屬之技術領域】 物 受 本發明係關於一種新聞評分系統,尤指一種特定標的 新聞事件衫響其人氣、支持度之分析模型及相關技術 【先前技術】201123060 VI. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a news scoring system, and more particularly to an analysis model and related technology of a specific subject news event, which is popular and supportive. [Prior Art]

拜社會開放與科技之賜,各種資訊媒體如雨後春荀般 相繼成立,每天製造及提供大量的資訊,使現代人無疑身 處在-個資訊爆炸的時代裡。事實上,資訊的傳播通常會 對-些人、事、物造成非常顯著的影響,例如經濟相關新 聞常常造成股市振堡及個股澡跌,i負面#新聞盘曝光率 經常影響-些公幕人物(如藝人'政治人物)的人氣虚支持 度。其中,以股市(證券市場)為例,其無疑是最受消息面 影響的市場,任何與相關產業或個股有關的新聞或事件, 都可能引起股價的震盪與波動,當有利多消息出現時’依 實際經驗其股價多半有上揚的機會,一旦出現利空消息時 ’則股價大多會出現下跌的狀況,亦即利多新聞對個股股 價具有拉抬作用,利空新聞則將造成股價下挫。 另以藝人而言’透過新聞事件反應出來的曝光率亦通 常對一個藝人的人氣指數產生關鍵性的影響,純就新聞量 而言,係反應藝人的曝光率,而正面新聞與負面新聞的量 則將影響藝人的受歡迎程度。而在另外—個領域,例如政 治人物或政黨,受新聞事件的影響亦如藝人一般,同樣關 乎其支持度與社會形象。 201123060 由上述可知新聞事件與特定人事物的關聯性,對於關 切前述人事物受關注程度的人而t,如果能夠即時掌握與 特定人事物相關的新聞事件,將是非常重要且具有價值的 參考指標。 【發明内容】 因此本發明主要目的在提供一種新聞評分系統,其 透過即時搜尋與特定標的物_的新聞,並透過關鍵字分 析判斷其屬於正面或負® ’再經過歸納分析後提供參考之 用,由該系統的評分結果,遂對於該等評分結果有興趣的 人即具有一可信客觀之參考依據。 i^述目的&取的主要技術手段係令前述新 分糸統包括有: :即時新聞資料庫,係供搜尋與特定標的物 時新聞; 1 “ Γ刀用°司庫’係'由多數與特定標的物相關的關鍵字 組成,母一關鍵字均被賦予一評分值. 分用係針對即時新聞中的-標的物《評 〒的關鍵字進行正負面評分; 用以將r類管理模組’具有各種標的物之專屬資料表,而 應標的::::管理模组完成評分的新聞分別歸納至各對 <•專屬資料表中; 後,關鍵:::值官理模組’係根據新聞發生後-定期間 分值; ;標的物之影響程度修正評分用詞庫中的評 201123060 "刀&理槟組將進一步根據各標的物專 的内容,依每日之雜明真 ,^ Ba 貝抖庫 母曰之新聞量、正面新聞量、負面新聞量 線圖’以方便瞭特定標的物受新聞事件影響的程度。、 【實施方式】 圖 關於本發明一較佳實施例之系統架構,請參閱第一 所示,其包括: -即時新聞資料庫(10),係供搜尋與特定標的 原即時新聞事件’每一新聞事件包含時間及新聞標題、來 啊、, 秩的分用詞庫(20),係由多數分別代表不同標的物的 下:%與多數關鍵字組成,每一關鍵字均被賦予一評分 值;其中關鍵字係用以描述標的物的名詞,該關鍵字具: 面的涵意及負面的涵意,以個股為例,「營收」、「弃▲ 」寺關鍵字係正面形容詞 「網降 南 字調降」、「目標價」等關鍵 J為負面形容;又以人物為例,「 Λ ,「 1 Α益」為正面關鍵字 為負面關鍵字1等關鍵字均用來對新聞事件 2出現的特定標的物進行評分,·又如前述,每—關 句有-評分值,以反應其「正面」與「 評分锸葚、」的紅度,該 ^㈤° —主觀值’該值大小可依新聞源的重要性 、疋’以放大評分的效果; 襟的=分管理模组(30),係利用評分用詞庫(2〇)提供的 八稱及關鍵予,對新聞事件中的標的物進行正負面評 二其評分方式係可對各種新聞源分別賊予一基本分,者 —新聞事件中出現的關鍵字,則將關鍵字的權值乘上; 201123060 聞源的基本分,而給予評分;而當新聞事件中同時出現多 組關鍵字時’以最大絕對值為主; 一分類管理模組(40),具有各種標的物之專屬資料表 ’而用以將經過評分管理模組完成評分的新聞分別歸納至 各對應標的物之專屬資料表中; -關鍵字權值管理模組(5Q),係根據新聞發生後一定 期間後,關鍵字對於標的物之影響程度,以修正評分用詞 庫中各關鍵字的評分值;其中: 力 該評分管理模組(3〇)將進—步根據各標的物專屬資料 庫的内容’依每曰之新聞量、正面新聞量、負面新聞量給 成線圖’以方便瞭特定標的物受新聞事件影響的程度, s新聞發生時,作為評估其影響效應的參考依據。 芝 當應用在影視娛樂方面時,前述評分用詞庫(20)的内 容係如下列: v η 標的名稱:關穎、王心凌、隋棠 關鍵字:公益、吸毒、酒駕、慈善、代言 而由評分管理模組(30)自即時新聞資料。庫⑽ 絕對出評分值較高的新聞事件,就是對藝人身價影響較大 的新聞。又就此應用方式,分類管理模組(4〇 藝人專屬的資料表,依每曰之新 、 新聞量,正面的新 吳敦義 ::::::成,,藉此舆收視率、*"_行 係如下又;運用在政治人物時,前料分㈣庫㈣的内容 摞的名稱:馬英九、陳水扁、 201123060 表 評分詞:公益、貪污、骑選、施政滿意 出席代 在此運用領域中,仍由缔八其 d子刀g理振組(3〇)自即時新聞 資料庫(1〇)中篩選絕對出坪八佶于新聞 ,, 出。子刀值較两的新聞事件,就是對 物身價影響較大的新聞。又就此應用方式,分類管 理拉:⑽)係根據各個政治人物專屬的資料表,依每曰之 :聞里it面的新聞置、負面的新聞量繪成一線圖,藉此 與政治人物的支持度等進行比對分析。Thanks to the openness of society and the gift of technology, various information media have been established like the Spring and the Spring. They create and provide a large amount of information every day, making modern people undoubtedly in an era of information explosion. In fact, the dissemination of information usually has a very significant impact on some people, things, and things. For example, economic-related news often causes the stock market to rebound and individual stocks fall, i negative #新闻盘率率 often affects some public screen figures (such as the artist's politician)'s popularity support. Among them, taking the stock market (securities market) as an example, it is undoubtedly the market most affected by the news. Any news or events related to related industries or individual stocks may cause stock price fluctuations and fluctuations, when favorable news appears. According to actual experience, most of its stock price has an opportunity to rise. In the event of bad news, 'the stock price will mostly fall, that is, Lido News has a pulling effect on individual stock prices, and bad news will cause the stock price to fall. In addition, for artists, the exposure rate reflected by news events usually has a crucial impact on an artist's popularity index. In terms of news volume, it reflects the exposure of artists, while the amount of positive news and negative news. It will affect the popularity of the artist. In other areas, such as political figures or political parties, they are also influenced by news events, such as artists, as well as their support and social image. 201123060 From the above, we can know the relevance of news events to specific people's things. For those who are concerned about the degree of concern of people's things, it is a very important and valuable reference indicator if they can instantly grasp the news events related to specific people's things. . SUMMARY OF THE INVENTION Accordingly, it is a primary object of the present invention to provide a news scoring system that searches for news with a specific subject matter by an instant search and judges that it belongs to positive or negative by keyword analysis and then provides a reference for inductive analysis. By the scoring results of the system, the person who is interested in the results of the scoring has a credible and objective reference. The main technical means for the purpose of the i<>> are that the new sub-systems include: : an instant news database for news and search for specific subject matter; 1 For the keyword-related keyword composition, the parent-key keyword is given a rating value. The sub-use system is used for the positive-negative rating of the keyword in the instant news, and the r-type management module. 'The exclusive data sheet with various subject matter, and the news of the target:::: management module to complete the score is summarized into each pair of <• exclusive data table; after, the key::: value government module 'system According to the news after the occurrence of the period - the score of the period;; the degree of influence of the target object correction score in the vocabulary in the 201123060 " knife & Penang group will further according to the content of each subject matter, according to the daily miscellaneous , ^ Ba 抖 库 曰 曰 新闻 新闻 正面 正面 正面 正面 正面 新闻 新闻 新闻 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以 以system structure, Please refer to the first one, which includes: - an instant news database (10) for searching for original news events with specific targets. 'Every news event contains time and news headlines, come, and ranks. (20), which consists of the majority: % and the majority of the keywords, each of which is assigned a score; each keyword is assigned a score; the keyword is used to describe the noun of the subject, the keyword has: The meaning of the face and the negative meaning, taking individual stocks as an example, the key words of the "revenue" and "abandoned ▲" temple keywords are positive adjectives such as "net drop south word down" and "target price" are negative descriptions; Taking the character as an example, " Λ , " 1 Α benefits " is a positive keyword for negative keywords 1 and other keywords are used to score the specific subject matter of news event 2, and as described above, each sentence There is a -score value to reflect the redness of "positive" and "score 锸葚,", the ^(f) ° - subjective value 'this value can be based on the importance of the news source, 疋' to enlarge the effect of the score; 襟= sub-management module (30), using the wording of the score (2〇) The eight names and key points provided are positive and negative evaluations of the subject matter in the news event. The scoring method is to assign a basic score to the thieves of various news sources, and the keywords appearing in the news event. Then multiply the weight of the keyword; 201123060 Wenyuan's basic score, and give the score; and when there are multiple sets of keywords in the news event, 'the largest absolute value; a classification management module (40), The exclusive data sheet with various subject matter' is used to summarize the news scored by the score management module into the exclusive data table of each corresponding target; - Keyword weight management module (5Q), according to the news After a certain period of time after the occurrence of the keyword, the degree of influence of the keyword on the target object, to correct the score value of each keyword in the vocabulary of the score; wherein: the score management module (3〇) will advance according to each target The content of the database 'depends on the amount of news, positive news, and negative news to the line graph' to facilitate the extent to which specific subject matter is affected by news events, when the news occurs, as an assessment The reference basis for its effect. When the Zhishi application is used in film and television entertainment, the content of the above-mentioned scoring vocabulary (20) is as follows: v η The name of the target: Guan Ying, Wang Xinling, 隋棠 Keywords: public interest, drug addiction, drunk driving, charity, endorsement Management module (30) from instant news materials. Library (10) Absolutely a news event with a high score is a news that has a greater impact on the artist's worth. In this application mode, the classification management module (4 〇 artist's exclusive data sheet, according to the new, news, positive new Wu Dunyi:::::: Cheng,, by this 舆 ratings, *"_ The lineage is as follows; when used in politicians, the name of the former (4) library (4): Ma Ying-jeou, Chen Shui-bian, 201123060 Table scores: public welfare, corruption, riding, governance, satisfaction, attendance, in this field of application, still From the news of the instant news database (1〇), the screen is absolutely out of the news, and the news of the sub-knife value is the price of the object. The news that has a greater impact. In terms of this application method, the classification management pulls: (10)) according to the information table exclusive to each political figure, according to the news: the news of the news, the negative news volume is drawn into a line graph, borrowed This is compared with the support of political figures.

以下再針對本發明之新聞評分系統運用在股市的實施 狀況進一步說明如后,盆 其l括一即時新聞資料庫(1〇)、一 評分用詞庫(2〇)、—評分管理模組㈣、—分類管理模組 (4〇)及一關鍵字權值管理模組(5〇);其中: 該即時新聞資料庫⑽係利用新聞搜尋引擎收集與股 市相關的即時新聞。 該評分用詞庫(20)係收集個股名稱及股市新聞中常見 與利多或利空有關的關鍵字,並分別建立利多字資料庫與 利空字資料庫:而評分用詞庫(2〇)進一步提供一網頁介面 ,以顯示前述利多字資料庫及利空字資料庫,請參閱第二 圖所示,其中歸類於利多的關鍵字包括:「買超」、「入 主」、「上看」、「創新高」等;又歸類於利空的關鍵字 二包括.「短缺」、「大跌」、「創新低」等。前述關鍵 字為股市相關新聞中的經常性用語,且依經驗觀察,其分 】存在利多」或「利空」之涵意,因此,前述關鍵字將 破賦予一加權分數,用來對搜尋而來的新聞進行績效分析 再者該。平分用5司庫(10)在前述網頁介面所呈現的各個 201123060 利多字與利空字,分 建立一連結,當使用者按 鍵字時,將會連結至另—莰下任一關 面’而在該頁面顯示一特 間内,出現該關鍵字的所有 ,期 吓有新聞,例如當使用者按 主」關鍵字後,其連妹的百 入 ,、逑、·。的頁面係如第三圖所示,其 基本的統計訊息,如「這— 一有 ^個月内共有310筆,關於入 主的新聞。其中有19樓 、入 細奴不提到入主以下是斜 播提到的股票做出績效分新丨丨 屮〜 」,並進一步列出相關新聞In the following, the news scoring system of the present invention is further explained in the implementation status of the stock market, and the instant news database (1〇), a scoring vocabulary (2〇), and the scoring management module (4) are further described. - a classification management module (4〇) and a keyword weight management module (5〇); wherein: the instant news database (10) uses a news search engine to collect instant news related to the stock market. The scoring vocabulary (20) collects the names of individual stocks and keywords commonly associated with bullish or bearish stock market news, and establishes a Lido word database and a niche word database respectively: and the scoring vocabulary (2〇) provides further A web interface to display the aforementioned Lido database and the niche database. Please refer to the second figure. The keywords classified in Lido include: "Buy Super", "Enter", "Look", "Innovation is high", etc.; Keywords 2, which are classified as bad, include "shortfall", "big fall" and "low innovation". The aforementioned keywords are recurring terms in stock market related news, and they are observed by experience, and their sub-existing meanings are “profitable” or “bad”. Therefore, the above keywords will be given a weighted score for the search. The news is going to be a performance analysis. Separately use the 5 Treasurer (10) to create a link between each of the 201123060 Lido words and the negative words presented in the aforementioned web interface. When the user presses the word, it will be linked to the other - The page displays a special room, and all of the keywords appear, and there is news, such as when the user presses the main keyword, the sister's hundred, 逑, ·. The page is as shown in the third figure. Its basic statistical information, such as "This - there are a total of 310 pens in a month, about the news of the owner. Among them, there are 19th floor, the slaves are not mentioned below. Is the stock mentioned in the oblique broadcast to make a new performance ~", and further list relevant news

出現入主子眼的個股及其績效分析,供使用者參考。 該-f刀管理核組(3〇)係根據評分用詞庫㈣提供 空關鍵字進行績效分析,並給 、夕 σ 丁。干刀,例如一則股市相 新聞中提到了個股名稱及一個 W Μ上的利多或利空關鍵字, 例如在一則新聞中出現了 「營 &收新岗」,則表示該則 聞標題中同時含有「營收,月「# 一 丁 3 , S收」及「新咼」兩個利多關鍵字, 又如新聞中出現「調降財測」則是指該則新聞同時含有 「調降」及「財測」兩個利空關鍵字;前述則每一個利多 或利空關鍵字本身已被賦予—加權分數,&可用以該則新 聞進行評分。 另評分管理模組(30)對於該則新聞進行評分的同時, 亦進一步與新聞公布當時相關個股的股價建立關聯,因此 個股在新聞公布之後股價異動狀況,將由關鍵字權值管理 模組(5 0)來對關鍵字的加權分數進行修正調整。 再者,該評分管理模組(30)將進一步針對每一個股分 別建立一個股線圖’請參閱第四圆所示,該評分管理模組 (3〇)係提供一頁面’以顯示該個股線圖,於本實施例中, 該個股線圖係重疊在個股的K線圖上,其計有三條線,八 201123060 別代表當日新聞量(A)、利多 由於-碰ώ姑/ 斤聞置(Β)及利空新聞量(c), 由於二條曲線係隨Κ線 ]重(’ 透過該個股線㈣出在_ fa1軸變動,因此使用者可 及利多、利空新聞與股價的;:(可選擇曰或周),新聞量 線圖更進一步提對關係及影響程度。該個股 線圖’可隨游標選擇領*茔罝於個股 定時Η虛路…内的任一時間點’以觀察指 ::間處發生的新聞量與股價變化…一頁面則可列出 的詳細内容。 麵—步透過連結職該新聞 該刀類&理私組(40)係與評分管理模組(30)連結,其 將-平为f理換組(3〇)完成評分的新聞事件分別歸納到各個 專屬的資料表中’並提供-頁面顯示經過評分管理模垣 (3〇)完成評分的精選新聞,冑參閱第五圖所示該精選新 聞頁面在每一則新聞之後分別顯示該則新聞在經過績效分 析後所得的評分’除此以外,另提供一 結,該連結 内谷係整理該則新聞中所提個股在一定期間内所出現的所 有新聞及%聞出S時的股價5見值;換言之,前述精選新聞 頁面將即時展示經過績效分析的新聞,而每一則新聞至少 提及一標的個股,當使用者按下「分析、結,該分類管理模 、、且(4 0)即連結另一頁面,如第六圖所示,在該頁面顯示一 定期間(例如一個月)内出現前述個股的所有新聞,並在每 則新聞中同時顯示新聞發生當時的股價現值。藉此,使用 者不僅可透過前述介面瞭解個股在目前或—定期間内發生 了那些利多或利空新聞,且在利多或利空新聞發生時,當 時的股價現值變化,而作為投資人選股換股的參考依據。 201123060 —月υ边實施例中,系統係透過網頁方式顯示精選新聞内 谷,除此以外,系統亦可進-步建置一通信模組,而透過 通㈣組將前述精選新聞傳送給使用纟,例如利用簡訊傳 ^㈣㈣手機上或㈣其他即時通信方式(例如瞻) 傳送給使用者,至於精選新聞的内容,可由使用者透過設 定關鍵字以決定其想收到的新聞内容。 、根據前述的系統制可知,本發明亦提供—新聞分析 方法其如第七圖所示,係包括以下步驟:Individual stocks appearing in the main eye and their performance analysis for users' reference. The -f knife management core group (3〇) provides performance analysis based on the score vocabulary (4) to provide an empty keyword, and gives it to the evening. A dry knife, such as a stock market news, mentions the name of a stock and a bullish or bearish keyword on the W. For example, if a "camp & new post" appears in a news item, it means that the title also contains "Revenue, monthly "# Ding 3, S" and "New" are two keywords. In the news, "downgrade" means that the news also contains "down" and " "Financial" two bad keywords; in the foregoing, each of the bullish or bearish keywords themselves has been given a weighted score, & can be scored with this news. In addition, the rating management module (30) scored the news and further related to the stock price of the relevant stocks at the time of the news announcement. Therefore, the stock price change status after the news announcement will be determined by the keyword weight management module (5). 0) To correct the weighted score of the keyword. Furthermore, the rating management module (30) will further establish a stock chart for each stock separately. Please refer to the fourth circle. The rating management module (3〇) provides a page to display the stock. Line diagram, in this embodiment, the stock pattern is superimposed on the K-line chart of individual stocks, which has three lines, and eight 201123060 does not represent the daily news volume (A), Lido due to - touch the ώ / / 斤 闻(Β) and bad news (c), because the two curves are followed by the line] (the movement through the stock line (four) is on the _fa1 axis, so users can access the bullish, bad news and stock prices; Select 曰 or 周), the news volume chart further mentions the relationship and the degree of influence. The stock chart 'can be selected with the cursor 茔罝 茔罝 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个 个: News volume and stock price changes happening... The details can be listed on one page. Face-to-step link to the news The knife & privacy group (40) is linked to the score management module (30) , which will summarize the news events of the scores of the flats (3〇) Go to the 'Specialized Data Sheet' and select the page to display the selected news that has been scored by the rating management module (3〇). Please refer to the selected news page shown in the fifth picture to show the news after each news. After the performance analysis, the score obtained 'except for this, another one is provided. The link in the link is to sort out all the news that the stocks mentioned in the news appear in a certain period of time and the stock price when the S smells out. In other words, the selected news page will immediately display the performance analysis news, and each news will mention at least one target stock. When the user presses "analysis, knot, the classification management module, and (4 0) is the link. Another page, as shown in the sixth figure, displays all the news of the aforementioned stocks in a certain period (for example, one month) on the page, and simultaneously displays the current value of the stock price at the time of the news in each news. Not only can the above-mentioned interface be used to understand the bullish or bad news that happened to individual stocks during the current or fixed period, and when Lido or bad news happened, then The current value of the stock price changes, and as a reference for the investor's stock selection and conversion. 201123060 - In the monthly example, the system displays the selected news valley through the webpage. In addition, the system can also build a new one. The communication module transmits the selected news to the user through the group (4), for example, by using the message transmission (4) (4) on the mobile phone or (4) other instant communication methods (for example, viewing) to transmit to the user, and the content of the selected news can be used by the user. By setting a keyword to determine the news content that it wants to receive. According to the foregoing system, the present invention also provides a news analysis method, as shown in the seventh figure, which includes the following steps:

即時搜哥與指定標的物相關的新聞(7〇 1); 利用正面及負面關鍵字對於前述新聞進行評分(702广 I 一㈣標的物之新聞量建立—線圖,並與該標的物的至 少一指標建立關聯,以顯示新聞與該指標的互動關係(703) 提供經過評分的新聞,並針對各則新聞提供評分及標 的物的分析内容(704);其中: $ 前述標的物之指標,係指個股之股價、政治人物支持 度、藝人身價及人氣等。由前述的系統說明可知,前述步 驟(702)中的評分’係對各個正面或負面關鍵字賦予加權分 數,再以加權分數對新聞進行評分;另一方面,對新聞進 订績效分析時同時引進入當時指標的變動狀況,因此新聞 中所出現特定關鍵字對於股價的影響程度與速度,將用以 回饋對關鍵字進行加權分數調整,藉以提高評分的準確性 與參考價值。 至於步驟(704)中提供精選新聞給使用者的方式,除以 網頁方式供使用者登入後瀏覽外,亦可透過手機簡訊或其 201123060 他即時通信方式(例如MSN)傳送給使用者 【圖式簡單說明】 第-圖.係本發明一較佳實施例之系統 第二圖:係本發明一較佳實施例之關鍵字;^圖。 。 頁面示意圖 第二圖’係本發明一較佳實施例之關 意圖》 予新聞頁面示 第四圖:係本發明一較佳實施例之個股線 圖顯 示意圖》 ,_示介面 第五圖.係本發明一較佳實施例之 圖 精選新聞頁面示意 圖 第六圖.係本發明一較佳實施例之個股分析頁面示意 第七圖:係本發明一較佳實施例之流程圖。 【主要元件符號說明】 (10)即時新聞資料庫(20)評分用詞庫 (30)評分管理模組 (40)分類管理模組 (50)關鍵字權值管理模組Instantly search for news related to the specified subject matter (7〇1); use the positive and negative keywords to score the above news (the 702 wide I (four) target news volume build-line map, and at least with the subject matter An indicator is linked to show the interaction of the news with the indicator (703) to provide scored news, and to provide a rating and analysis of the subject matter for each news item (704); wherein: $ the indicator of the aforementioned subject matter Refers to the stock price of individual stocks, politician support, artist's worth and popularity, etc. As can be seen from the foregoing system description, the score in the above step (702) is to give weighted scores to each positive or negative keyword, and then to weight the score to the news. On the other hand, the performance analysis of the news is simultaneously introduced into the change of the indicator at that time, so the degree and speed of the impact of the specific keyword on the stock price in the news will be used to feedback the weighted score adjustment of the keyword. In order to improve the accuracy and reference value of the score. As for the way to provide the selected news to the user in step (704), The webpage can be used for users to log in and browse, and can also be transmitted to the user via a mobile phone newsletter or its 201123060 instant messaging method (for example, MSN). [Simplified Schematic] - Figure 1. A preferred embodiment of the present invention The second diagram of the system is a keyword according to a preferred embodiment of the present invention; the second diagram of the page is 'the intention of a preferred embodiment of the present invention>> The fourth page of the news page is the present invention A stock diagram of a preferred embodiment shows an intent, a fifth diagram of a preferred embodiment of the present invention. FIG. 6 is a schematic diagram of a selected news page of a preferred embodiment of the present invention. 7 is a flow chart of a preferred embodiment of the present invention. [Explanation of main component symbols] (10) Instant news database (20) Grading vocabulary (30) Scoring management module (40) Classification management mode Group (50) keyword weight management module

r Ir I

1111

Claims (1)

201123060 七、申請專利範圍: 1·—種新聞評分系統,包括有: 一即時新聞資料*,係供搜尋 時新聞; 寻與特疋標的物相關的即 -乎刀用詞庫,係由多數特定標的物名 的物相關的關鐽宝知士、 冉及與特定標 =組成’每一關鍵字均被賦予 ^官理模組’係針對即時新聞中 分用詞庫中的關鍵字進行正負面評分; ^的物以評 刀類S理模組’具有各種標的物之專 用以將經過評分普理措4 i 貝科表,而 0砰刀官理模組完成評分的新聞分別歸 應標的物之專屬資料表中; ,各對 一關鍵字權值管理模組,係根據新聞發生後_ 後’關鍵字對於標的物之影響程度修正評分 的二 分值;其中: τ的汁 前述評分管理模組將進一步根據各標的物專屬資料庫 的内容,依每日之新聞量、正面新聞量、負面新聞 線圖。 2.如申請專利範圍第彳項所述之新聞評分系統,係 運用個股之股價變化觀察,其中: ' 該即時新聞資料庫係供搜尋與股市相關的即時新聞; 該評分用詞庫係收集個股及股市新聞中常見與個股利 多或多空有關的關鍵字,並分別建立利多字資料庫與利空 字資料庫; 該評分管理模組係根據評分用詞庫提供的多空關鍵字 進行評分’另進一步與新聞公布當時相關個股的股價建立 12 201123060 關聯;該評分管理模組進一 股線圖,以顯示新聞發佈時 程度; 步針對每一個股分別建立一個 間與數量對於個股股價的影響 以建立一精 刀系統,該 利多字資料 該分類管理模組係與評分管理模組連結 選新聞群組,而提供予投資人。 3.如申請專利範圍第2項所述之新聞評 評分用詞庫進一步提供一網頁介面,以顯示一 庫及一利空字資料庫;201123060 VII. Patent application scope: 1. A kind of news scoring system, including: an instant news material*, which is used for searching news; The subject matter of the subject matter is related to the key words of the key words in the instant news. ^ The object of the evaluation tool is a special type of object that will be used to score the Pu Licuo 4 i Beco table, and the news that the 0 rake official module completes the score is respectively assigned to the subject matter. In the data table, each pair of a keyword weight management module corrects the score of the score according to the degree of influence of the keyword on the target after the news occurs; wherein: the juice rating management module of τ Further based on the contents of the exclusive database of each subject matter, based on daily news volume, positive news volume, and negative news line graph. 2. The news scoring system as described in the scope of the patent application section is based on the stock price change observation of individual stocks, where: ' The instant news database is used to search for real-time news related to the stock market; the score is used to collect stocks. And the stock market news often has keywords related to multiple stocks or long and short positions, and respectively establish a Lido word database and a bad word database; the score management module is based on the long and short keywords provided by the score vocabulary to score 'another Further related to the stock price of the relevant stocks at the time of the establishment of 12 201123060; the score management module into a line chart to show the degree of news release; step for each stock to establish a separate and quantity impact on the stock price to establish a The fine knife system, the multi-word data, the classification management module is linked to the rating management module to select a news group, and is provided to the investor. 3. The news evaluation vocabulary as described in item 2 of the patent application scope further provides a web page interface for displaying a library and a profitable word database; 該評分用詞庫在前述網頁介面所呈現的各個利多字 利空字’分別建立一連結,以顯示出現該關鍵字鉍 聞及其績效分析。 ’新 4.如申請專利範圍第3 鍵字分別被賦予一加權分數 評分。 項所述之新聞評分系統,各關 ’用來對搜尋而來的新聞進行 5.如申請專利範圍第4項所述之新聞評分系統,= 關鍵字權值管理模組係利用個股在新聞公布之後股價異该 _ 狀況對評分用詞中的關鍵字加權分數作調整。 6·如申請專利範圍第2至5項中任一項所述之新聞 評分系統’該個股線圖係重疊在個股的K線圖上。 7.如申請專利範圍第6項所述之新聞評分系統,: 個股線圖包括三條曲線,分別代表當曰新聞量、 Λ 曰 夕新聞 篁及利空新聞量,而隨Κ線圖上的時間軸變動。 8·如申請專利範圍第7項所述之新聞評分系統,^ 評分管理模組提供一頁面,顯示前述個股線圖,同—頁Λ 並列出各條新聞的標題及其評分’並提供連結供 — % 一步潜J 13 201123060 覽該新聞的詳細内容。 ^ 9_如申請專利範圍帛2 1 5項中任一項所述之新聞 各平分丰έ在 二、'·、,該分類管理模組提供一頁面顯示經過評分管理 | ’且疋成坪分的新聞,該新聞頁面在每—則新聞之後分別 ^不=則新聞在經過績效分析後所得的評分及連結,該連 '° ''顯示各則新聞中所提個股在一定期間内所出現的 所有新聞及新聞出現時的股價現值。 —,1:·如中請專利範圍帛9項所述之新聞評分系統,進 步包括-通信模組,用以將前述新聞傳送給使用者。 .如申印專利範圍第9項所述之新聞評分系統,該 通信模組係利用簡訊傳送至使用者的手機上。 、二2_如申明專利範圍第9項所述之新聞評分系統,該 通信模組係利用即時通信方式傳送給使用者。 .μ請專㈣圍第j項所述之新聞評分系統,係 運用在政治人物之支持度變化觀察。 14. 如申請專利範圍第,項所述之新聞評分系統,係 運用在藝人之人氣變化觀察。 15. —種新聞評分方法,包括: 即時搜尋指定標的物相關的新聞; 利用正面及負面關鍵字對於前述新聞進行評分; 針對標的物之新聞量建立一绫周,# & , 银圖’並與標的物之指標 建立關聯,以顯示新聞與該指標的互動關係,· 提供經過評分的精選新聞,並針對 f對各則新聞提供評分 及標的分析内容。 16.如申請專利範圍第15項所 所述之新聞評分方法, 14 201123060 該β平刀係對各個利空或利多關鍵字 楢分數對新聞進行評分。 予加權分數,再以加 1 7’如申請專利範圍第1 5 n %义 野新聞進杆 斤述之新聞評分方法, 狀況,利用新進入払的物之指標在當時的變化 程度與迷户,科予對於扣標變化的影響 疋度對關鍵字進行加權分數調整。 1 8.如申請專利範 % 聞評分大、土 ⑽第15至17項中任-項所述之新 去,該標的物係指個股,其指標係指股價。 聞評1 八\如中請專利範圍第15至17項中任-項所述之新 ”法’該標的物係人,其指標係指人氣或支持度。 八、 式··(如次頁)The scoring uses a vocabulary to create a link for each of the various tilable words presented in the aforementioned web interface to display the keyword smear and its performance analysis. 'New 4. If the third key word of the patent application scope is given a weighted score. The news scoring system described in the article is used to carry out the news from the search. 5. The news scoring system described in item 4 of the patent application scope, = the keyword weight management module uses the stocks in the press release. After that, the stock price is different. The status is adjusted for the keyword weighted score in the score word. 6. The news scoring system as described in any one of claims 2 to 5, wherein the stock pattern is superimposed on a K-line chart of individual stocks. 7. For the news scoring system described in item 6 of the patent application, the stock chart includes three curves representing the news volume, the news, the news, and the bad news, and the timeline on the chart. change. 8. If the news scoring system described in item 7 of the patent application scope is provided, the scoring management module provides a page displaying the aforementioned stock chart, the same page, and lists the titles of the news and their ratings' and provides links for — % One step dive J 13 201123060 See the details of the news. ^ 9_If the application of the patent scope 帛 2 1 5, the news is equally divided in the second, '·,, the classification management module provides a page display after the score management | 'and Cheng Chengping points The news, the news page after each news, respectively ^ not = the scores and links obtained after the performance analysis of the news, the even '° '' shows that the stocks mentioned in each news appear in a certain period of time The current value of the stock price when all news and news appear. -, 1: The news scoring system described in the scope of Patent Application 帛 9 further includes a communication module for transmitting the aforementioned news to the user. According to the news scoring system described in claim 9 of the patent application, the communication module is transmitted to the user's mobile phone by using a short message. 2, 2, for example, the news scoring system described in claim 9 of the patent scope, the communication module is transmitted to the user by means of instant communication. .μ Please use the news scoring system described in item j of the fourth (4), which is used in the observation of changes in the support of politicians. 14. The news scoring system described in the scope of application for patents is used in the observation of the popularity of artists. 15. A news scoring method, including: Instantly searching for news related to the specified subject matter; Using the positive and negative keywords to score the aforementioned news; Establishing a week for the news of the subject matter, # & , 银图' and Linked to the target's metrics to show the interaction of the news with the metric, provide selected news that has been scored, and provide ratings and analytics for each news item. 16. The news scoring method as described in claim 15 of the patent application, 14 201123060 The β flat knife scores the news for each bearish or bullish keyword 楢 score. To give a weighted score, and then add 1 7', as in the scope of the patent application, the 15th n % of the wilderness news, the news score method, the situation, the use of new indicators of the entry into the shackles at the time of change and the fans, Keyu's effect on the change of the deduction changes the weighted score of the keyword. 1 8. If the patent application is a new one, the new item mentioned in the item (10), items 15 to 17 refers to individual stocks, and the index refers to the stock price. Comment 1 8 If you refer to the new "Law" in the scope of Articles 15 to 17 of the patent scope, the indicator refers to the popularity or support. Eight, the formula · (such as the next page ) 1515
TW98145242A 2009-12-28 2009-12-28 News rating system and method. TW201123060A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
TW98145242A TW201123060A (en) 2009-12-28 2009-12-28 News rating system and method.

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
TW98145242A TW201123060A (en) 2009-12-28 2009-12-28 News rating system and method.

Publications (2)

Publication Number Publication Date
TW201123060A true TW201123060A (en) 2011-07-01
TWI409714B TWI409714B (en) 2013-09-21

Family

ID=45046518

Family Applications (1)

Application Number Title Priority Date Filing Date
TW98145242A TW201123060A (en) 2009-12-28 2009-12-28 News rating system and method.

Country Status (1)

Country Link
TW (1) TW201123060A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509010A (en) * 2017-09-15 2019-03-22 腾讯科技(北京)有限公司 A kind of method for processing multimedia information, terminal and storage medium

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI758676B (en) * 2020-01-03 2022-03-21 華南商業銀行股份有限公司 Financial transaction volume warning system
TWI765847B (en) * 2020-01-03 2022-05-21 華南商業銀行股份有限公司 Financial transaction volume warning system with promotion function
TWI765645B (en) * 2021-04-07 2022-05-21 元智大學 Investment scoring method of financial text

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6460036B1 (en) * 1994-11-29 2002-10-01 Pinpoint Incorporated System and method for providing customized electronic newspapers and target advertisements
TWI264656B (en) * 2004-01-16 2006-10-21 Ching-Shyang Hwang System and method for providing news service
US20060048053A1 (en) * 2004-08-30 2006-03-02 Xerox Corporation Individually personalized customized report document system
US20060048042A1 (en) * 2004-08-30 2006-03-02 Xerox Corporation Individually personalized customized report document system with user feedback
US20080201078A1 (en) * 2006-09-28 2008-08-21 Dudley Fitzpatrick Apparatuses, Methods and Systems for Determining and Announcing Proximity Between Trajectories
TW200832162A (en) * 2007-01-24 2008-08-01 Telepaq Technology Inc System and method for sorting and searching stock news
US7860872B2 (en) * 2007-01-29 2010-12-28 Nikip Technology Ltd. Automated media analysis and document management system
US9342551B2 (en) * 2007-08-14 2016-05-17 John Nicholas and Kristin Gross Trust User based document verifier and method

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109509010A (en) * 2017-09-15 2019-03-22 腾讯科技(北京)有限公司 A kind of method for processing multimedia information, terminal and storage medium
CN109509010B (en) * 2017-09-15 2023-04-18 腾讯科技(北京)有限公司 Multimedia information processing method, terminal and storage medium

Also Published As

Publication number Publication date
TWI409714B (en) 2013-09-21

Similar Documents

Publication Publication Date Title
US9201880B2 (en) Processing a content item with regard to an event and a location
US11176142B2 (en) Method of data query based on evaluation and device
JP5632574B2 (en) System and method for improving ranking of news articles
Jahanbakhsh et al. The predictive power of social media: On the predictability of us presidential elections using twitter
TWI351619B (en) Search engine that applies feedback from users to
US20160162582A1 (en) Method and system for conducting an opinion search engine and a display thereof
US20120042020A1 (en) Micro-blog message filtering
CN103729360A (en) Interest label recommendation method and system
WO2009023070A1 (en) Systems and methods for keyword selection in a web-based social network
Nicolini et al. Framing the Women’s March on Washington: Media coverage and organizational messaging alignment
WO2012075302A2 (en) Re-publishing content in an activity stream
JP2009520264A5 (en)
CN102929966B (en) A kind of for providing the method and system of personalized search list
JP6563428B2 (en) Screen display of live events in search results
JP2009099088A (en) Sns user profile extraction device, extraction method and extraction program, and device using user profile
TW201123060A (en) News rating system and method.
US10127322B2 (en) Efficient retrieval of fresh internet content
Wu et al. Keyword extraction for contextual advertisement
Qiu et al. Incorporate the syntactic knowledge in opinion mining in user-generated content
CA2862763A1 (en) Computerized internet search system and method
CN102298621B (en) System for obtaining page user focus degree PageFocus by method for aggregating and displaying same source information search engine based on focus degree
Yang et al. A study of query term deletion using large-scale e-commerce search logs
CN106708939A (en) Target person scoring and pushing methods, apparatuses and systems
CN116089723A (en) Recommendation system recommendation method and device
Matsumoto et al. Supporting human recollection of the impressive events using the number of photos

Legal Events

Date Code Title Description
MM4A Annulment or lapse of patent due to non-payment of fees