TWI385540B - Article content value-added service system and method of the same - Google Patents

Description

Article content value-added service system and method thereof

The present invention relates to an article content value-added service, and in particular to an article content value-added service system and method thereof that can provide related wiki-like knowledge, related words, and advertisements.

For the average reader, when viewing an article, it is generally only possible to get a general idea of the main description of the article from the title of the article. It is not easy to grasp the key points of the article in advance and speed up the understanding of reading. Unless you can mark the keyword of the article before reading the entire article, the main content of the article can be grasped by the title of the article and the keyword words that are appropriately related to the title.

However, the keyword that simply indicates the content of the article is still of limited help to the reader, especially if the keyword in the article is a proper noun and the reader is not aware of the proper noun. This will cause the reader to use another tool to query the proper noun to understand. Therefore, there is still a need for a system or method that can solve the above problems to improve the speed and depth of reading articles.

The present invention provides an article content value-added service system and method thereof. In one aspect, the article content value-added service system of the present invention includes a database; a keyword capture indicator module is used to capture and mark keywords in the article content that are highly relevant to the article topic; wiki-based knowledge capture a module coupled to the keyword capture indicator module and the database for searching the database for wiki-like knowledge matching the keyword; and an advertisement providing module coupled to the keyword capture indicator module And a database for searching for advertisements matching the keywords in the database. In addition, the content value-added service system of the article further includes a related word feedback module coupled to the keyword capture indicator module and the database for searching the database for related words associated with the keyword, and multi-country A language translation module coupled to a wiki-based knowledge capture module for translating wiki-like knowledge. Furthermore, the keyword capture indicator module includes a content pre-processing module, a string comparison module, and a topic relevance scoring module.

In another aspect, the article content value-adding service method of the present invention includes: capturing and indicating keywords in the article content that are highly relevant to the article topic; searching for wiki-like knowledge matching the keywords, and searching for keywords related Associated words and search for ads that match the keywords; and display the above wiki-like knowledge, related words, and ads on the associated window of the keyword. In addition, the content value-added service method of this article also includes the instant translation of wiki-style knowledge into the language of user preference. Furthermore, the step of extracting and indicating the keyword with high relevance to the article topic in the article content includes pre-processing the article content, comparing the pre-processed article content with the corpus to search for the corresponding corpus. The word, and the relevance score for calculating the above words and the subject of the article.

One of the advantages of the present invention is that the content value-added service system and method thereof allow the user to obtain further information about the keywords in the article content when reading the article.

Another advantage of the present invention is that the content value-added service system and method thereof allow the user to quickly grasp the focus and meaning of the article by using the keyword indication when reading the article.

Yet another advantage of the present invention is that the content value-added service system and method thereof of the present invention allow a user to view an advertisement of interest to a user when reading an article.

The present invention will be described in detail with reference to the preferred embodiments and the accompanying claims Therefore, the invention may be embodied in other embodiments in addition to the preferred embodiments described herein.

The invention discloses an article content value-added service system and a method thereof. The article content value service system of the present invention can be installed in the operating system of the personal computer or the mobile device, so that the user of the personal computer or the mobile device can use the content value service system of the article to read or browse the value-added information. The content of the above article may be the content of a news article, technical file or any other type of digital file.

As shown in the first figure, in the preferred embodiment of the present invention, the article content value-added service system 10 includes a keyword capture indicator module 101, a wiki-based knowledge capture module 102, and a multi-language translation module 103. The advertisement providing module 104, the related word feedback module 105, and the database 106. In one embodiment, the article content value-added service system 10 can be coupled to a processor 107 of a mobile device or personal computer, as shown in the first figure. In the above mobile device or personal computer, the processor 107 is coupled to the memory 108, the display 109, the hard disk 110, the input/output interface 111, the audio output interface 112, and the pointing device 113. The video data of the value-added information processed by the article content value-added service system 10 can be displayed on the display 109, and the audio information of the value-added information processed by the article content value-added service system 10 can be transmitted through the audio output interface 112. The output and the article content value-added service system 10 can be temporarily stored in the memory 108 or stored in the hard disk 110 during operation. In addition, the user can use the pointing device 113 to select an object in the article content value-added service system 10 to manipulate it.

In one embodiment, the keyword capture indicator module 101 includes a content pre-processing module 101a, a string comparison module 101b, and a topic relevance scoring module 101c to determine and label keywords. The content pre-processing module 101a performs pre-processing on the content of the article read by the user, for example, full semi-converted conversion, English single-word initial conversion uppercase, English single-word disassembly, and bottom line segmentation. The string comparison module 101b compares the pre-processed article content with a pre-established corpus, such as a corpus downloaded from a Wikipedia website, using a string comparison algorithm to search for article content. Correspond to the words stored in the corpus. The above-described string comparison algorithm uses a conventional text string comparison algorithm. The topic relevance weight module 101c calculates the relevance weights of the compared words in the article content and the article theme, and sorts according to the weight of the article.

The consideration parameters for the calculation of the association weight may include, but are not limited to, the term frequency (TF), the inverse document frequency (IDF), the long word priority, the stop word, the frequency of occurrence of the text, the English abbreviation, and the wiki. Word relevance, etc. Among them, the wiki website is a public writing plan on the Internet. As long as it meets the needs and norms of the encyclopedia, anyone can write new terms on the wiki, or edit and modify existing ones. Each wiki word (Wiki Word) on the wiki has a page that will be introduced, and the editor will include other wikis in this word that are relevant to the word and stored on the wiki. Marked out, thus presenting that the wiki word has a connection with the remaining wiki words, and the link relationship is the above wiki term relevance. The above link is recorded in the public information of the wiki website with a page link data sheet. It is worth noting that the link between the introduction page and the rest of the wiki words is indicated by the editor and is therefore more relevant to the subject of the introduced wiki. In addition, the user can set a threshold in the keyword capture indicator module 101, so that the keyword capture indicator module 101 identifies the words whose weights calculated by the topic relevance weight module 101c are higher than the threshold score as a key. Words and delete words that do not reach the threshold. In an embodiment of the present invention, the calculation of the relevance weight is based on the public information of the Wikipedia website, and the word frequency, the reverse file frequency, and the long word priority are used as parameters for calculating the weight. In the following, an example will be given to illustrate the manner in which the correlation weights are calculated. However, those skilled in the art should understand that the following examples are intended to illustrate the invention and not to limit the invention. The formula is as follows: word frequency (TF) × reverse file frequency (IDF) × (1 + (wiki word relevance) + (long word score × third weight))

The wiki word relevance can be expressed as follows: (association number × first weight) + (contribution number × second weight)

In other words, the equation of the present invention is as follows: word frequency (TF) × reverse file frequency (IDF) × (1 + (association number × first weight) + (contribution number × second weight) + (long word score × third) Weights))

The values of the first weight, the second weight, and the third weight may be determined by the user according to the type of the article to be detected. Wherein, the word frequency (TF) refers to the number of times a given word appears in an article. The Reverse File Frequency (IDF) is a measure of the universal importance of a word, that is, the frequency that appears in all database files. For a wiki word, if it has a high word frequency in an article, but a low file frequency appears in the entire database file, that is, it only appears in part of the database file, then the wiki word can be generated. High-frequency word frequency (TF) × reverse file frequency (IDF). For example, if the total number of words in an article is 100 and the word "baseball" appears three times, the word "baseball" in the article has a frequency of 0.03 (= 3/100). A method of calculating the file frequency (DF) is to determine how many files have appeared in the word "baseball" and then divide by the total number of files contained in the file set. Therefore, if the term "baseball" appears in 1,000 documents and the total number of documents is 10,000,000, the file frequency is 0.0001 (=1000/10,000,000). Finally, the word frequency (TF) x reverse file frequency (IDF) score can be obtained by dividing the frequency of the word by the frequency of the file. In the above example, the word "baseball" has a word frequency (TF) x reverse file frequency (IDF) score of 300 (=0.03/0.0001) in the document set.

In addition, the Wikipedia publicly available data contains a web link data table that records the relevance of each wiki word. The number of associations refers to the number of related words in the linked data table of a wiki word. A contributing word means that a wiki word appears in the linked data table of other wiki terms. For example, "Zhan Yuran" appears in the linked data sheet of "Zhuang Jiarong" and "Zhuang Jiarong" appears in the linked data sheet of "Zhan Yuran". Therefore, if a contribution score of 0.5 is assumed, then "Zhan Yuran" and " Zhuang Jiarong's contribution scores are all 0.5. A long word score refers to a keyword whose length is greater than 3, and a specific score is given after judging whether it is an important keyword. For example, "Zhan Yuran" and "Zhuang Jiarong" are both three characters, so their long-term scores are all 1, but the "Yankee Stadium" is 4 words, so the long-term score is 1.5, and the "TV Sports" is 5 Word, so its long word score is 2. In other words, in the present embodiment, the long word score is increased by 0.5 for each additional word based on 3 words. Thereby, according to the weight value set by the user and each parameter value, the correlation weights of the words compared in the article can be calculated, and sorted according to the weight.

The keyword capture indicator module 101 is coupled to the wiki-based knowledge capture module 102, the advertisement providing module 104, and the related word feedback module 105 to respectively transmit the data of the keyword above the threshold score to the three. The wiki-based knowledge capture module 102, the advertisement providing module 104, and the related word feedback module 105 are each coupled to the database 106 to search for the required data in the database 106.

The wiki-based knowledge capture module 102 searches the database 106 for the wiki-like knowledge matching the keyword data by using the keyword data to display the wiki-style knowledge on the associated window of the keyword. The wiki-based knowledge data in the database 106 can be downloaded from the Wikipedia website via the Internet or a mobile phone network in advance, or periodically connected to the wiki via the Internet or a mobile phone network. The website (Wikipedia) is updated. The advertisement providing module 104 searches the database 106 for the advertisement matched with the keyword data by using the keyword data to display the advertisement below, above or to the side of the wiki knowledge in the associated window. The advertising material in the database 106 can be text, graphics, animation or a combination thereof, and can be provided by the advertiser to the advertising service provider beforehand, and then transmitted from the advertising service provider to the database through the internet or mobile phone network. 106, can also be regularly updated through the Internet or mobile phone network to the advertising service provider. Each advertisement material can be set by the advertiser to be compared with the above keyword data for the media to be suitable for the advertisement. Therefore, the above keyword data can be used to match advertisements that may attract users. Artists in this field should be able to understand that after the user clicks, reads or browses the advertising material, the user can receive feedback from the advertiser or system owner.

The related word feedback module 105 searches the database 106 for the related words associated with the keyword data by using the keyword data to display the related words below, above or to the side of the wiki knowledge in the associated window. . Similarly, the related word data in the database 106 can be stored in the database 106 in advance, or periodically updated via the Internet or a mobile phone network. In an embodiment, the related words include fuzzy words. In an embodiment, as shown in the first figure, the related word feedback module 105 can be coupled to the wiki-based knowledge capture module 102, whereby the related word feedback module 105 can also pass through the wiki-based knowledge capture module 102. Relevant words are retrieved from the wiki-like knowledge in the database 106 that matches the keyword data. Since the words in the wiki-style knowledge that match the keyword data are mostly similar to the keyword data, most of the related words are similar to the keyword data. In an embodiment, the wiki-based knowledge capture module 102 can use the related word data to search the database 106 for wiki-like knowledge matching the related word data, so as to display the wiki-like knowledge on the related words. On the associated window. The multi-language translation module 103 is coupled to the wiki-based knowledge capture module 102 to instantly translate the wiki-like knowledge acquired by the wiki-based knowledge capture module 102 with keywords or related words into a user-preferred language. For example, Traditional Chinese, Simplified Chinese, English, Japanese, German, or any other language. In an embodiment, the wiki-based knowledge capture module 102 can also include a multi-language selection mode for the user to select a preferred language, so that the wiki-based knowledge capture module 102 searches the database 106 for the preferred language version. Wikipedia knowledge.

Therefore, the article content value-added service system 10 provided by the present invention can extract and mark keywords in the article content that are highly correlated with the article topic, and provide wiki-like knowledge, related words, Relevant window of value-added information such as related advertisements, so that the user who reads the content of the article can browse or read the wiki-related knowledge related to the keyword and the wiki-related knowledge related to the keyword related words by using the mobile device or the personal computer. , or click on relevant ads to get rewards. Therefore, the content value-added service system 10 of the article allows the user to obtain further information about the keyword in the article content when reading the article, and quickly grasp the focus and meaning of the article by using the keyword indication, and can view Advertising that users are interested in.

As described above, in another embodiment of the present invention, the present invention discloses an electronic device (or computer) identifiable storage medium, and the electronic device (or computer) identifiable storage medium includes an article content value service program or software. Storing therein, the article content value-adding service program includes a keyword capture indicator module for capturing and marking keywords in the article content that are highly relevant to the article topic; a wiki-based knowledge capture module, It is coupled to the keyword capture indicator module and the database for searching the database for wiki-like knowledge matching the keyword; and an advertisement providing module coupled to the keyword capture indicator module and the database Used to search the database for ads that match the keywords. The electronic device identifiable storage medium further includes a related word feedback module coupled to the keyword capture indication module and the database for searching the database for related words associated with the keyword. The electronic device identifiable storage medium may also include a multi-language translation module coupled to the wiki-based knowledge capture module for translating wiki-like knowledge. The keyword capture indicator module includes a content pre-processing module or a string comparison module. In an embodiment, the keyword capture indicator module may also include a topic relevance weight module.

As shown in the second figure, in the article content value-adding service method of the present invention, first, in step 201, keywords with high relevance to the article topic in the article content are extracted and marked. Then, in step 202, the wiki-like knowledge matching the above keywords is searched, the related words associated with the above keywords are searched, and the advertisements matched with the above keywords are searched. Thereafter, in step 203, the wiki-like knowledge, related words, and advertisements are displayed on the associated window of the keyword. In addition, the content value service method of the article further includes the step 204 of instantly translating the wiki-like knowledge into a language preferred by the user. The step of extracting and indicating the keyword with high relevance to the article topic in the article content comprises the step 201a pre-processing the article content, and the step 201b comparing the pre-processed article content with the corpus to search for the corresponding existence. The words of the corpus, and step 201c calculate the relevance scores of the above words and the subject of the article. In addition, in an embodiment, the step of extracting and marking the keyword in the article content that is highly correlated with the article topic further includes step 201d sorting the word according to the relevance score; and step 201e setting the threshold and deleting The relevance score does not reach the above threshold. In an embodiment, if the content of the article is an English article content, the step of pre-processing the article content includes performing a full-half transformation; converting the English word prefix of the English article content to uppercase; and the content of the English article The English word is disassembled and separated by the bottom line.

The above description is a preferred embodiment of the invention. Those skilled in the art should be able to understand the invention and not to limit the scope of the patent claims claimed herein. The scope of patent protection is subject to the scope of the patent application and its equivalent fields. Any modification or refinement made by those skilled in the art without departing from the spirit or scope of the present invention is equivalent to the equivalent change or design made in the spirit of the present disclosure, and should be included in the following patent application scope. Inside.

10. . . Article content value-added service system

101. . . Keyword capture module

101a. . . Content pre-processing module

101b. . . String comparison module

101c. . . Topic relevance scoring module

102. . . Wiki-style knowledge capture module

103. . . Multi-language translation module

104. . . Advertising module

105. . . Related word feedback module

106. . . database

107. . . processor

108. . . Memory

109. . . monitor

110. . . Hard disk

111. . . Output interface

112. . . Audio output interface

113. . . Pointing device

201. . . step

201a. . . step

201b. . . step

201c. . . step

201d. . . step

201e. . . step

202. . . step

203. . . step

204. . . step

The invention may be understood by the description of the preferred embodiments and the detailed description and the accompanying drawings. However, those skilled in the art should understand that the preferred embodiments of the present invention are intended to be illustrative and not to limit the scope of the claims of the present invention, wherein the first figure is the value-added service of the article content according to the present invention. Block diagram of the system.

The second figure is a schematic diagram of an article content value service method according to the present invention.

10. . . Article content value-added service system

101. . . Keyword capture module

101a. . . Content pre-processing module

101b. . . String comparison module

101c. . . Topic relevance scoring module

102. . . Wiki-style knowledge capture module

103. . . Multi-language translation module

104. . . Advertising module

105. . . Related word feedback module

106. . . database

107. . . processor

108. . . Memory

109. . . monitor

110. . . Hard disk

111. . . Output interface

112. . . Audio output interface

113. . . Pointing device

Claims (23)

An article content value-added service system includes: a database; a keyword capture indicator module for capturing and marking keywords in the article content that are highly relevant to the article topic; a wiki-based knowledge capture module And coupled to the keyword capture indicator module and the database for searching the database for wiki-like knowledge matching the keyword; and an advertisement providing module coupled to the keyword 撷The tag module and the database are used to search for the ad that matches the keyword in the database. The article content value-added service system of claim 1, further comprising a related word feedback module coupled to the keyword capture indicator module and the database for searching and searching in the database The related words associated with the keyword. The article content value-added service system of claim 1, further comprising a multi-language translation module coupled to the wiki-based knowledge capture module for translating the wiki-style knowledge. The article content value-added service system of claim 1, wherein the keyword capture indicator module comprises a content pre-processing module. The article content value-added service system of claim 1, wherein the keyword capture indicator module comprises a string comparison module. The article content value-added service system of claim 1, wherein the keyword capture indicator module comprises a topic relevance weight module. An article content value-adding service method includes: capturing and marking keywords in an article content that are highly relevant to an article topic; searching for wiki-like knowledge matching the keyword, searching for related words associated with the keyword And searching for an advertisement that matches the keyword; and displaying the wiki-style knowledge, the related word, and the advertisement on the associated window of the keyword. The article content value-adding service method as claimed in claim 7 further includes promptly translating the wiki-style knowledge into a language preferred by the user. The article content value-adding service method according to claim 7, wherein the step of extracting and indicating a keyword having high relevance to the article topic in the article content comprises pre-processing the article content. The article content value-adding service method of claim 9, wherein the article content is an English article content. The article content value-adding service method as claimed in claim 10, wherein the step of pre-processing the content of the article comprises: performing a full-half-shaped conversion; converting an English single-word prefix of the English article content to uppercase; and the English article The English word of the content is disassembled and separated by the bottom line. The article content value-adding service method of claim 9, wherein the step of extracting and indicating the keyword of the article content that is highly correlated with the article topic further comprises comparing the pre-processed article content with the corpus to Search for the words that correspond to the corpus. The article content value-adding service method as claimed in claim 12, wherein the step of extracting and indicating the keyword in the article content that is highly correlated with the article topic further comprises calculating a relevance score of the word and the article topic. The article content value-adding service method of claim 13, wherein the step of extracting and indicating a keyword in the article content that is highly correlated with the article topic further comprises sorting the word according to the relevance score. The article content value-adding service method of claim 13, wherein the step of extracting and indicating the keyword with high relevance to the article topic in the article content further comprises setting a threshold and deleting the relevance score does not reach the threshold The word of the score. The article content value-adding service method as claimed in claim 13, wherein the parameter for calculating the relevance score of the word and the article topic comprises a word frequency, a reverse file frequency, a long word priority, a stop word, and a frequency of occurrence of the article. , with English abbreviations and wiki word relevance. The article content value-adding service method as claimed in claim 13, wherein the calculation formula for calculating the relevance score of the word and the article subject is the word frequency×reverse file frequency×(1+(wiki word relevance)+( Long word score × weight)). An electronic device identifiable storage medium includes an article content value-adding service program stored therein, comprising: a keyword capture indication module, configured to capture and mark keywords in the article content that are highly correlated with the article theme; a wiki-based knowledge capture module coupled to the keyword capture indicator module and a database for searching the database for wiki-style knowledge matching the keyword; and an advertisement providing module And coupled to the keyword capture indicator module and the database for searching for advertisements matching the keyword in the database. The electronic device identifiable storage medium of claim 18, further comprising a related word feedback module coupled to the keyword capture indication module and the database for searching and searching in the database Keyword phase Related words. The electronic device identifiable storage medium as claimed in claim 18, further comprising a multi-language translation module coupled to the wiki-based knowledge capture module for translating the wiki-like knowledge. The electronic device as claimed in claim 18 can identify the storage medium, wherein the keyword capture indicator module includes a content pre-processing module. The electronic device as claimed in claim 18 can identify the storage medium, wherein the keyword capture indicator module comprises a string comparison module. The electronic device as claimed in claim 18 can identify the storage medium, wherein the keyword capture indicator module includes a topic relevance weight module.