JP2012212333A - Hot-word association apparatus and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To associate hot-words with one another included in short blogs with limitation on the number of letters.SOLUTION: An hot-word association apparatus (1) includes hot-word extraction means (11) for extracting hot-words in a short blog service based on a change in a frequency of appearance of a word included in short blogs, short blog retrieval means (12) for retrieving short blogs which include a hot-word and a URL based on the extracted hot-word, and hot-word association means (13) for associating hot-words with one another included in multiple short blogs which include the same URL among the retrieved short blogs in order to associated hot-words with one another included in short blogs submitted in the short blog service.

Description

  The present invention relates to an apparatus and a method for associating rapidly increasing words that associate rapidly increasing words included in a short blog posted within a short blog service.

  In recent years, there has been a short blog service represented by TWITTER (registered trademark) and the like that facilitates communication between users by accepting postings of short sentences describing the current situation and miscellaneous notes from users and allowing other users to browse and reply. It is used frequently.

For example, TWITTER receives a tweeted word string (short blog) tweeted by a user via a communication network so that it can be viewed by the public. (Follow), another user quotes a short blog of one user and makes it a short blog of his own (retweet), adds his own comment to a given user's short blog (quotes) (Non-patent document 1).
Thus, in a short blog represented by TWITTER, it is possible to give a feeling as if a large number of conversations are made by developing responses in various forms.

Twitter full use technique, Japan, ASCII Media Works, February 26, 2010, 8-9 pages

Here, in communication on a network represented by a short blog, a soaring word that is temporarily used by many users, called a buzzword (buzzword), is generated. Such spikes may occur as a result of a temporary increase in usage frequency for events that occur in daily life, for example, when a company announces a new product and a large number of such companies or new products When a short blog is posted, the company name, new product name, etc. may become a sudden rise word.
Short blogs containing soaring words on the network are what many users want to see and browse, so there is a need for a method that can appropriately pick out short blogs containing such soaring words from among many short blogs. .

  By the way, since the short blog posted on the short blog service has a limit on the number of characters, even a short blog posted for the same event may use different words depending on the posted user. For example, while a user posts a short blog that clearly states the company name, such as “XX Company has announced a new product” and does not specify a new product name, another user says “New Product XX has been announced. In some cases, a short blog that specifies the name of the new product without posting the name of the company is posted.

  At this time, “XX Company” and “New Product XX” are inherently closely related words, so when taking out a short blog containing a rapidly rising word from many short blogs, It is required to take out short blogs in consideration of the relationship between the soaring words.

  The present invention has been made in view of such a demand, and an object of the present invention is to appropriately grasp the relevance between rapidly increasing words included in a short blog with a limited number of characters.

  The inventor of the present invention has completed the present invention by connecting the rapidly rising words based on a Web page (URL) quoted by a short blog including the rapidly rising word.

  (1) A rapidly increasing word associating device for associating rapidly increasing words included in a short blog posted within the short blog service, and included in the short blog by referring to a short blog database storing the posted short blogs. A spike word extraction means for extracting a spike word in the short blog service based on a change in the frequency of appearance of the word, and the short blog database based on the extracted spike word and including the spike word and URL Short blog search means for searching for the short blog, and among the searched short blogs, the spike word association associated with the spike words included in a plurality of short blogs including the same URL and stored in the association database Trending associated apparatus comprising a means.

According to the rapidly rising word associating device of (1), the rapidly rising word associating means associates the rapidly rising words included in a plurality of short blogs including the same URL.
Here, since a plurality of short blogs including the same URL (Web page) are posted for the same event, the rapidly increasing word used in each of the plurality of short blogs. Each other can be regarded as related. Therefore, according to the present invention, even if the sudden rise words used by the users are different, it is possible to appropriately associate the sudden rise words.

  (2) Search request accepting means for accepting a search request for a short blog designating the sudden rise word from the user terminal, and other related to the accepted sudden rise word by referring to the association database based on the accepted sudden rise word And a suggestion means for extracting the word and sending it for proposal to the user terminal.

  According to the surge word associating device of (2), the surge word related to the spike word received from the user terminal, that is, the spike word used for the same event is proposed to the user terminal. Thereby, the user can acquire the short blog posted with respect to the event attracting attention by many users without omission.

  (3) a page similarity determination unit that determines a similarity between Web pages specified by each of the URLs included in the short blog searched by the short blog search unit, and the rapidly rising word association unit includes: The rapidly increasing word associating device according to (1) or (2), wherein the rapidly increasing words included in a short blog each including the URL specifying the Web page having a predetermined similarity or higher are associated and stored in the association database .

Since there are innumerable Web pages on the network, there may be a plurality of Web pages describing the same event. That is, there may be a Web page consisting of an article of newspaper company B for the same event in addition to a Web page consisting of an article of newspaper company A for an event. Such Web pages have different URLs but are related to the same event, so it is preferable to consider them when associating soaring words.
According to the rapidly increasing word associating device of (3), not only the same URL but also URLs with similar Web page contents are used to link the rapidly increasing words, so that the association range (coverage) is appropriately managed. And the association of the soaring word can be performed appropriately.

  (4) The short blog search means calculates a search score based on the sudden rise word when searching for the short blog including the sudden rise word and the URL, and the sudden rise word association means has a predetermined search score or more. The rapidly increasing word associating device according to any one of (1) to (3), wherein the rapidly increasing words included in the short blog each including the URL included in the short blog are associated and stored in the association database.

  According to the rapidly increasing word associating device of (4), a plurality of URLs (Web pages) cited by short blogs having a high search score based on the rapidly increasing word are regarded as being related to each other, and the short blog including the URL is regarded as related. Associate the included spikes. Thereby, the range (coverage) of association can be further expanded, and the association of the rapidly rising words can be performed appropriately.

  (5) A method of associating rapidly increasing words included in a short blog posted in a short blog service executed by a computer, referring to the short blog database storing the posted short blogs, and A step of extracting a sudden rise word in a short blog service based on a change in appearance frequency of a word included in the word, and referring to the short blog database based on the extracted sudden rise word, including the sudden rise word and URL A method comprising: searching for a short blog; and associating the rapidly increasing words included in a plurality of short blogs including the same URL among the searched short blogs and storing them in an association database.

  According to the method (5), the same effect as that of the surge word association device (1) can be obtained.

  According to the present invention, it is possible to associate rapidly increasing words included in a short blog with a limited number of characters.

It is a figure which shows the function structure of the sudden rise word correlation apparatus of 1st Embodiment. It is a figure which shows a short blog database. It is a figure which shows the sudden rise word extracted by the sudden rise word extraction means. It is a figure which shows an association database. It is a figure which shows operation | movement of the soaring word correlation apparatus of 1st Embodiment. It is a figure which shows operation | movement of the soaring word correlation apparatus of 1st Embodiment. It is a flowchart which shows the flow of a process of the rapid increase word correlation apparatus of 1st Embodiment. It is a flowchart which shows the flow of a process of the rapid increase word correlation apparatus of 1st Embodiment. It is a figure which shows the function structure of the sudden rise word correlation apparatus of 2nd Embodiment. It is a figure which shows operation | movement of the soaring word correlation apparatus of 2nd Embodiment. It is a figure which shows operation | movement of the soaring word correlation apparatus of 3rd Embodiment.

[First Embodiment]
First, with reference to FIG. 1 to FIG. 7, a rapidly increasing word associating apparatus 1 according to the first embodiment will be described.

[Configuration of the soaring word association device 1]
With reference to FIG. 1, the structure of the rapidly increasing word correlation apparatus 1 of this invention is demonstrated.
The soaring word associating apparatus 1 includes an associating unit 10 and a suggesting unit 20 in order to grasp the relevance of the soaring word in the short blog service and to allow the user to search the short blog without omission. .

[Association unit 10]
The associating unit 10 associates the rapidly rising words used in the short blog service with each other, the rapidly rising word extracting means 11, the short blog searching means 12, the rapidly rising word associating means 13, the short blog database 14, and the association database. 15 and so on.

The rapidly increasing word extracting means 11 accesses the short blog database 14 at a predetermined timing, and extracts a rapidly increasing word based on the change in the appearance frequency (number of appearances) of words included in the short blog posted in the short blog service. . Here, the “rapidly rising word” refers to a word whose appearance frequency has rapidly increased among words included in a short blog posted within the short blog service. It should be noted that the rate of increase that is assumed to rise rapidly can be set arbitrarily, and as an example, it can be assumed that the rate of increase has increased rapidly when the appearance frequency is 10 times or more that of the previous day.
In addition, a well-known method that has been known so far can be used to extract the rapidly increasing word. As an example, the rapidly rising word extracting unit 11 performs morphological analysis on the text information of the short blog, and extracts words from the short blog by connecting the analyzed morphemes based on a predetermined dictionary. And the sudden rise word extraction means 11 can extract the sudden rise word by counting the appearance frequency (number of appearances) of the extracted word.

  Here, as shown in FIG. 2, the short blog database 14 stores short blogs posted by users within the short blog service. The short blog includes text information with a character limit (for example, within 140 characters), and this text information may include a URL indicating the location of the Web page in addition to a comment created by the user. The URL included in the short blog is generally information related to a comment created by the user. For example, a short blog with a short blog ID “200” is a short blog in which a user creates a comment “Hanako is a shock” by quoting a Web page of “URL1”.

In the present embodiment, the sudden rise word extraction means 11 extracts the sudden rise word shown in FIG. 3 from the short blog stored in the short blog database 14 shown in FIG.
The sudden rise word extracting means 11 may extract the sudden rise word in consideration of not only the change in the appearance frequency but also the appearance frequency itself. That is, when the appearance frequency of the word itself is small, it may not be extracted as a sudden rise word even if the appearance frequency has risen sharply.

Returning to FIG. 1, the short blog search means 12 searches the short blog database 14 based on the sudden rise word extracted by the sudden rise word extraction means 11, and searches for a short blog including the sudden rise word and including the URL. .
Referring to FIG. 2 and FIG. 3, when the short blog database 14 is searched based on “XX Hanako” with the rapidly increasing word ID “300”, the short blogs with the short blog ID “200” and “203” are rapidly increasing. Since the word “Hanako” is included and “URL1” is included, the search is hit. On the other hand, the short blog with the short blog ID “204” includes the rapidly rising word “Hanako”, but does not include the URL, and therefore does not hit the search.

  The rapidly increasing word associating means 13 associates rapidly increasing words included in a plurality of short blogs including the same URL among the short blogs searched by the short blog searching means 12. That is, the rapidly increasing word associating unit 13 associates the rapidly increasing words included in each of the plurality of short blogs that cite the same Web page (URL). Referring to FIG. 2, the short blog with short blog ID “200” and the short blog with short blog ID “201” refer to the same Web page (URL1). Therefore, the rapidly rising word associating unit 13 associates the rapidly rising word “XX Hanako” included in the short blog ID “200” with the rapidly rising word “XX pitcher” included in the short blog ID “201”. In addition, when the number of short blogs quoting the same Web page (URL) among a plurality of short blogs including a predetermined rapid increase word exceeds a predetermined number, the rapid increase words included in the short blog The association may be performed.

  The association database 15 stores the sudden rise words associated by the sudden rise word association means 13. At this time, as shown in FIG. 4, the association database 15 stores the associated rapidly rising words in association with the URL (co-occurrence URL) used for the association. Referring to FIG. 4, the rapidly rising words “XX Hanako”, “ABC group”, and “XX pitcher” are associated with each other based on “URL1”.

[Suggest 20]
The suggestion unit 20 is communicably connected to the user terminal 100, and includes a search request accepting unit 21 and a suggestion unit 22 in order to propose a rapidly increasing word related to the search query received from the user terminal 100.

The search request receiving unit 21 receives a short blog search request specifying a search query from the user terminal 100.
The suggesting unit 22 determines whether or not the search query received by the search request receiving unit 21 is a sudden rise word. If the search query is a sudden rise word, the suggestion unit 22 refers to the association database 15 and relates to the other words related to the sudden rise word. In order to propose the other spike word to the user terminal 100, the other spike word is transmitted.

[Operation of the soaring word association device 1]
Next, with reference to FIG. 5 and FIG. 6, the operation of the rapidly increasing word association apparatus 1 will be described.

  With reference to FIG. 5 (1), the sudden rise word extracting means 11 extracts the sudden rise word from the short blog stored in the short blog database 14. In FIG. 5 (1), ascending words such as “XX Hanako”, “ABC group”, and “XX pitcher” are extracted.

  Referring to FIG. 5 (2), the short blog search means 12 searches the short blog database 14 based on the extracted sudden increase word, and searches for a short blog including the URL and including the URL. As a result, as shown in FIG. 5 (2), the URL hit in the search is acquired for each rapidly increasing word used in the search.

  With reference to FIG. 5 (3), the rapidly increasing word associating means 13 associates rapidly increasing words included in a plurality of short blogs including the same URL based on the search result of the short blog searching means 12. Specifically, the rapidly increasing word associating unit 13 associates the rapidly increasing words by collecting the rapidly increasing words used for the search for each URL acquired as the search result (sorting by URL). As a result, as shown in FIG. 5 (3), the rapidly increasing words such as “Hanako XX”, “ABC group”, and “XX pitcher” are associated based on “URL1”.

  The soaring words associated in this way are included in a plurality of short blogs that cite the same Web page (URL), and thus are associated with each other. As a result, in the short blog posted for the same event (Web page), even if the surge word used by the user is different due to the limitation on the number of characters, each of the surge words can be appropriately associated. .

  Subsequently, referring to FIGS. 6 (4) and 6 (5), when the search request receiving unit 21 receives a search request specifying a search query from the user terminal 100, the suggestion unit 22 indicates that the search query is a rapidly increasing word. To determine. If the search query is a sudden rise word, another sudden rise word related to the sudden rise word is proposed to the user terminal 100. 6 (4) and 5 (5), since the search query “XX Hanako” is received from the user terminal 100, and this “XX Hanako” is a rapidly increasing word, “ABC group”, “XX pitcher”, etc. Other soaring words related to “Hanako” have been proposed.

  Thereby, when a search request for designating a sudden rise word is received from the user terminal 100, the user can search for a short blog created using another sudden rise word for the same event (Web page). . As a result, the user who performs the search can obtain various short blogs posted for events that are attracting attention by many users without omission.

[Hardware Configuration of Rapidly Increasing Word Association Device 1]
The hardware of the soaring word association apparatus 1 described above can be configured by one or a plurality of general computers. For example, a general computer includes a central processing unit (CPU) as a control unit, and a memory (RAM, ROM), a hard disk (HDD), and an optical disk (CD, DVD, etc.) as a storage unit, and a network communication device. Various wired and wireless LAN devices as display devices, for example, various displays such as liquid crystal displays and plasma displays, and as input devices, for example, keyboards and pointing devices (mouse, tracking ball, etc.) Are connected by a bus line. In such a general computer, the CPU comprehensively controls the soaring word associating device 1 and reads and executes various programs as appropriate, thereby cooperating with the hardware described above and various functions according to the present invention. Is realized.

[Processing of the soaring word association device 1]
Next, with reference to FIG. 7 and FIG. 8, processing of the sudden rise word association device 1 of the present invention will be described.

[Association]
First, an association process for associating rapidly increasing words will be described with reference to FIG.

S1: The rapidly increasing word extracting means 11 accesses the short blog database 14 at a predetermined timing, and determines a rapidly increasing word based on the change in the appearance frequency (number of appearances) of words included in the short blog posted within the short blog service. Extract.
S2: Subsequently, the short blog search means 12 searches the short blog database 14 based on the rapidly increasing word extracted in S1, and searches for the short blog including the URL and including the URL.
S3: Subsequently, the rapidly increasing word associating unit 13 associates the rapidly increasing words included in each of the plurality of short blogs that cite the same Web page (URL). That is, the rapidly increasing word associating unit 13 associates the rapidly increasing words having the same URL (co-occurrence URL) acquired as the search result of S2. Thereafter, the sudden rise word association means 13 stores the associated sudden rise word in the association database 15 and ends the association process.

[Suggest processing]
Next, with reference to FIG. 8, a suggestion process for proposing a rapidly rising word related to a search query received from the user terminal 100 will be described.

S11: The search request receiving unit 21 receives a search query from the user terminal 100.
S12: Subsequently, the suggestion unit 22 determines whether or not the search query received in S11 is a rapidly increasing word. At this time, if the search query accepted in S11 is not a rapidly increasing word, the process is terminated.
S13: On the other hand, when the search query received in S11 is a sudden rise word, the suggestion unit 22 refers to the association database 15 and sends another sudden rise word associated with the sudden rise word to the user terminal 100. suggest.

[Effect of soaring word association device 1]
According to the soaring word associating device 1 as described above, since soaring words are associated with each other based on the Web page (URL) quoted by the short blog, it is possible to appropriately associate the soaring words used for the same event. .
When a search request for specifying a sudden rise word is received from the user terminal 100, another sudden rise word is proposed based on this relevance. Therefore, what is a sudden rise word in a search request for the same event (Web page)? Even a short blog that uses different soaring words can be made to search by the user. As a result, the user who performs the search can obtain various short blogs posted for events that are attracting attention by many users without omission.

[Modified Embodiment]
In the above, the sudden rise word correlation apparatus 1 of 1st Embodiment was demonstrated. In the rapidly increasing word associating apparatus 1 of the first embodiment, the rapidly increasing words are associated with each other depending on whether or not they are the same URL. However, the range for associating the rapidly increasing words is not limited to the same URL, and the following second The range can be as in the embodiment and the third embodiment. That is, in the second embodiment and the third embodiment described below, the range for associating rapidly rising words is different from that in the first embodiment.
In the following description, the same components as those in the first embodiment are denoted by the same reference numerals, and description thereof is omitted or simplified.

[Second Embodiment]
With reference to FIG.9 and FIG.10, 1 A of sudden rise word correlation apparatuses of 2nd Embodiment are demonstrated. In the rapidly rising word associating apparatus 1A of the second embodiment, the range for associating rapidly rising words to similar Web pages (URLs) is expanded.
Here, since there are innumerable Web pages on the network, there may be a plurality of Web pages describing the same event. That is, there may be a Web page consisting of an article of newspaper company B for the same event in addition to a Web page consisting of an article of newspaper company A for an event. Since such Web pages have different URLs but are for the same event, the rapidly increasing word associating apparatus 1A of the second embodiment associates the rapidly increasing words in consideration of the similarity of the Web pages. .

[Configuration of Rapidly Increasing Word Association Device 1A]
First, with reference to FIG. 9, the configuration of the rapidly increasing word associating apparatus 1A of the second embodiment will be described. In order to expand the range for associating the rapidly rising word to a similar Web page (URL), the rapidly increasing word associating apparatus 1A of the second embodiment further includes the page similarity in addition to the configuration of the rapidly increasing word associating apparatus 1 of the first embodiment. A determination unit 16A is provided.

  The page similarity determination unit 16A determines the similarity between Web pages specified by each URL included in the short blog searched by the short blog search unit 12. The determination of the similarity between Web pages can be performed by any method. For example, data on similarity between Web pages is prepared in advance, and the similarity is determined by referring to the data. Also, as is well known to those skilled in the art, the similarity may be determined based on the type and frequency of words used in the Web page.

  When the page similarity determination unit 16A determines the similarity between Web pages, the sudden increase word association unit 13 causes the rapid increase included in each short blog including URLs corresponding to Web pages whose determined similarity is equal to or higher than the predetermined similarity. Associate words. That is, the rapidly increasing word associating unit 13 associates rapidly increasing words included in a short blog each including two or more URLs corresponding to two or more Web pages having a relationship of a predetermined similarity or higher. At this time, when the associating unit 13 associates the rapidly increasing words with each other, the associating unit 13 stores the associated rapidly increasing words in the association database 15. The predetermined similarity can be arbitrarily set.

[Operation of Rapidly Increasing Word Association Device 1A]
Next, with reference to FIG. 10, the operation of the rapidly increasing word association apparatus 1A will be described. In FIG. 10, only the part different from the operation of the sudden rise word association apparatus 1 of the first embodiment is shown. That is, in the operation of the rapidly rising word associating apparatus 1A, the rapidly rising word extracting unit 11 extracts the rapidly rising word from the short blog stored in the short blog database 14 (FIG. 5 (1)), and the short blog searching unit 12 extracts it. The operation of searching the short blog database 14 based on the sudden rise word and searching for the short blog including the sudden rise word and including the URL (FIG. 5 (2)) is the sudden rise word associating device 1 of the first embodiment. Since the operation is the same as that in FIG.

  Referring to FIG. 10A, the page similarity determination unit 16A determines the similarity of the Web page specified by the co-occurrence URL acquired as the search result of the short blog search unit 12. Here, “URL1” is a Web page indicating an article of a sports newspaper, and “URL2” is a Web page indicating an article of a weekly magazine. In each Web page, a secret meeting of “XX Hanako and XX pitcher” It deals with common events such as “date”. Therefore, in FIG. 10A, the page similarity determination unit 16A is similar to the Web page specified by “URL1” and the Web page specified by “URL2” (with a relationship of a predetermined similarity or higher). Yes).

  Subsequently, referring to FIG. 10 (2), the rapidly increasing word associating unit 13 determines that “URL1” and “URL2” have a relationship of a predetermined similarity or higher based on the determination result of the page similarity determination unit 16A. Are associated with each other and stored in the association database 15. As a result, the rapidly increasing words “ABC group” and “Restaurant LMN”, which were not related when they are associated with the same URL, are associated with each other.

  As described above, according to the rapidly increasing word associating apparatus 1A of the second embodiment, not only the same URL but also URLs with similar contents on the Web page are used to associate the rapidly increasing words with each other. It is possible to manage appropriately, and the association of soaring words can be performed appropriately.

[Third Embodiment]
Next, a rapidly increasing word associating apparatus 1B according to the third embodiment will be described with reference to FIG. In the rapidly increasing word associating apparatus 1B of the third embodiment, a range in which the rapidly increasing words are associated with each other based on the matching rate (search score) between the rapidly increasing word used for the short blog search by the short blog search means 12 and the text information of the short blog. decide. That is, in the rapidly increasing word associating device 1A of the second embodiment, the range of association is determined in consideration of the contents of the Web page specified by the URL, whereas in the rapidly increasing word associating device 1B of the third embodiment, The range to be associated is determined in consideration of the content of the text information of the short blog without considering the content of the Web page.
In addition, since the structure of the rapid increase word correlation apparatus 1B of 3rd Embodiment is the same as that of the rapid increase word correlation apparatus 1 of 1st Embodiment, illustration is abbreviate | omitted.

[Operation of Rapidly Increasing Word Association Device 1B]
With reference to FIG. 11, the operation of the rapidly increasing word association apparatus 1B will be described. In FIG. 11, only the parts different from the operation of the sudden rise word association apparatus 1 of the first embodiment are shown. That is, in the operation of the rapidly increasing word associating apparatus 1B, the operation of the rapidly increasing word extracting unit 11 extracting the rapidly increasing word from the short blog stored in the short blog database 14 (FIG. 5 (1)) is the rapid increase of the first embodiment. Since the operation is the same as that of the word association apparatus 1, the illustration is omitted.

  Referring to FIG. 11 (1), the short blog search means 12 searches the short blog database 14 based on the sudden rise word extracted by the sudden rise word extraction means 11, includes the sudden rise word, and also includes the short URL. Search for blogs. At this time, the short blog search means 12 calculates a search score based on the matching rate between the rapidly increasing word used for the search and the text information of the short blog.

Subsequently, referring to FIG. 11 (2), the rapidly increasing word associating means 13 associates the rapidly increasing words included in the short blogs each including the URL included in the short blog having a predetermined search score or higher. That is, the rapidly increasing word associating unit 13 associates the rapidly increasing words with each other based on two or more URLs included in two or more short blogs whose matching rate (search score) with the rapidly increasing word used for the search is a predetermined value or more. And stored in the association database 15. The predetermined search score can be arbitrarily set.
In FIG. 11A, in the search by the rapidly increasing word “XX pitcher”, the short blog ID “201” is the search score “80”, and the short blog ID “206” is the search score “75”. At this time, if the predetermined search score is set to “75” or more, the rapidly increasing word association unit 13 includes “URL1” included in the short blog ID “201” and “short blog ID“ 206 ”. Based on “URL3”, the rapidly increasing words are associated with each other. As a result, the rapidly increasing words “XX Hanako”, “ABC group”, and “XYZ team”, which were not related when they were associated with the same URL, are associated.

  Thus, according to the rapidly increasing word associating apparatus 1B of the third embodiment, not only the same URL but also the URL included in the short blog in which the search score by the rapidly increasing word used for the search is equal to or greater than a predetermined value is rapidly increased. Since the words are associated with each other, the association range (coverage) can be appropriately managed, and the association of the rapidly increasing words can be appropriately performed.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

1, 1A, 1B Rapidly rising word association device 10 Association unit 11 Rapidly rising word extraction means 12 Short blog search means 13 Rapidly rising word association means 14 Short blog database 15 Association database 16A Page similarity determination means 20 Suggestion part 21 Search request acceptance means 22 Suggestion Means 100 User terminal

Claims (5)

A soaring word association device that associates soaring words included in a short blog posted within a short blog service,
With reference to a short blog database that stores posted short blogs, a rapidly rising word extracting means for extracting a rapidly rising word in the short blog service based on a change in the appearance frequency of words included in the short blog,
Short blog search means for searching the short blog including the spike word and URL by referring to the short blog database based on the extracted spike word;
Among the searched short blogs, a sudden word association means for associating the sudden words included in a plurality of short blogs including the same URL and storing them in an association database;
A soaring word association device comprising: Search request accepting means for accepting a search request for a short blog designating the rapidly increasing word from the user terminal;
A suggestion means for referring to the association database based on the received spike word, extracting other spike words related to the received spike word and transmitting the proposal to the user terminal; and
The soaring word association apparatus according to claim 1. Page similarity determination means for determining the similarity between Web pages specified by each of the URLs included in the short blog searched by the short blog search means,
The spike word association means associates the spike words included in the short blog including each of the URLs that specify the Web pages having a predetermined similarity or higher and stores them in the association database.
The sudden rise word association apparatus according to claim 1 or 2. The short blog search means calculates a search score based on the sudden rise word when searching the short blog including the sudden rise word and the URL,
The spike word association means associates the spike words included in the short blogs each including the URL included in the short blog having a predetermined search score or higher and stores them in the association database.
The rapidly increasing word associating device according to any one of claims 1 to 3. A method of associating soaring words included in a short blog posted in a short blog service executed by a computer,
Referring to a short blog database storing posted short blogs, extracting a rapidly rising word in the short blog service based on a change in the appearance frequency of words included in the short blog;
Searching the short blog including the spike word and URL by referring to the short blog database based on the extracted spike word;
A step of associating the rapidly increasing words included in a plurality of short blogs including the same URL among the searched short blogs and storing them in an association database;
Including methods.

Images