JP2006072579A - Name retrieval device, name retrieval method, and name retrieval program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a name retrieval device that allows outputting appropriate candidates for a name using certain related keywords even without knowledge of the name of a celebrity or a group. <P>SOLUTION: A list of names serving as candidates is stored in a name list 203 in advance. A URL search part 201 searches web pages on the Internet according to keywords and a text extracting part 202 obtains text information related to the keywords. A score counting part 204 counts from the text information the number of names that appear which are included in the list of names and an output part 205 outputs the web pages in decreasing order of the number of names that appear. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a person name search apparatus, a person name search method, and a person name search program for searching for names and group names of celebrities and celebrities.

  Currently, search sites for celebrities and celebrities are provided on the Internet, but most of them are classified according to the Japanese syllabary and genre (eg athletes, singers, announcers, politicians, etc.). The information about the person such as the profile is searched from the name.

  On the other hand, there are many cases where it is desirable to know the name of a person from information related to a celebrity. In fact, I want to show the name of a celebrity during a conversation with someone, but I can't remember the name. However, there is a need to find a person's name from these various types of information because people are coming to mind.

  However, it is inconvenient to search for a person whose name is unknown, using a conventional system based on the person's name. For example, if a site is classified by category, it was necessary to first narrow down by category and then check the target person name one by one.

  In addition, there is a method of searching information in the search site by keyword (attribute information such as profile for a person's name) and outputting a person's name that contains a lot of the keyword. There is a problem that the information in the site that is given does not easily hit the keyword specified by the user. Furthermore, in order to cope with the keywords that are constantly updated, there is a problem that the information on the site has to be constantly updated, which is troublesome.

  The present invention has been made in view of the above, and it is possible to output a suitable person name or group name candidate from any related keyword without knowing the names and group names of celebrities and celebrities. An object is to provide a device, a person name search method, and a person name search program.

  In order to solve the above-described problems and achieve the object, a personal name search device according to the first aspect of the present invention includes a personal name list storage unit that stores personal names to be output as candidates, and a keyword related to the name of a person who desires a search. In the content of the web page acquired by the search means, the search means for acquiring the content of the web page related to the keyword input by the input means, the personal name list storage means in the content of the web page acquired by the search means A counting means for counting the number of appearances of personal names stored for each personal name, and an output means for outputting personal names based on the number of appearances counted by the counting means.

  According to a second aspect of the present invention, there is provided a personal name search apparatus comprising: an input unit that receives an input of a keyword related to a name of a person who desires a search; and a web page content related to the keyword input by the input unit. From the contents of the web page acquired by the search means, an extraction means for extracting a person name, a counting means for counting the number of person names extracted by the extraction means for each person name, Output means for outputting a person's name based on the number counted by the counting means.

  According to a third aspect of the present invention, there is provided a personal name search method comprising: an input step for receiving an input of a keyword related to a name of a person who desires a search; and a content of a web page related to the keyword input by the input step on the Internet And a counting step of counting the number of appearances of personal names stored in the personal name list for storing personal names to be output as candidates in the contents of the web page acquired by the search step for each personal name, And an output step of outputting a person's name based on the number of appearances counted by the counting step.

  According to a fourth aspect of the present invention, there is provided a personal name search method comprising: an input step for receiving an input of a keyword related to a name of a person who desires a search; and a content of a web page related to the keyword input by the input step From the content of the web page acquired by the search step, an extraction step of extracting a person name, a counting step of counting the number of person names extracted by the extraction step for each person name, And an output step of outputting a person's name based on the number counted by the counting step.

  According to a fifth aspect of the present invention, a personal name search program causes a computer to execute any one of the personal name search methods according to the third or fourth aspect.

  According to the present invention, a list of candidate names is stored in advance, a web page on the Internet is searched based on a keyword, text information related to the keyword is acquired, and the list of person names is obtained from the text information. Since the number of occurrences of the names of people included is counted and output in descending order of the number of occurrences, if you do not know the name of a celebrity, just specify any keyword related to that person, There is an effect that it is possible to provide a personal name search device that estimates and outputs the personal name.

  In addition, according to the present invention, since the text information is subjected to morphological analysis to extract a person name, a person name is estimated from an arbitrary keyword and output without preparing a list of candidate person names in advance. There is an effect that a personal name search device can be provided.

  Exemplary embodiments of a person name search device according to the present invention will be explained below in detail with reference to the accompanying drawings.

(Embodiment 1)
FIG. 1 is a configuration diagram showing the entire network system including a person name search device according to Embodiment 1 of the present invention. In FIG. 1, 101 is a mobile phone capable of inputting a keyword and showing a search result, 102 is a system of a mobile phone operator such as a base station linking the mobile phone and the Internet, and 103 is connectable to the Internet A personal computer or a PDA, 104 is the Internet including a huge amount of information such as a web page, and 105 is a server provided with the personal name search device of the present invention.

  In this network system, a user who wants to know a person's name first operates the cellular phone 101 and inputs a matter related to the person as a keyword. For example, what kind of television program or commercial appears. Next, the keyword is transmitted to the server 105. The keyword is sent to the server 105 via the mobile phone operator system 102 and the Internet 104. The personal name search device in the server 105 that has received the keyword searches information on the Internet 104 with the keyword, obtains a personal name candidate based on the result, and sends the result to the mobile phone 101. The candidate names are displayed on the mobile phone 101 via the Internet 104 and the mobile phone operator system 102.

  Further, instead of the mobile phone 101, a personal computer or PDA 103 that can be connected to the Internet can be used. Similarly, a keyword related to a person's name is input from a personal computer or PDA 103, transmitted to the server 105 including the person name search device via the Internet, and candidate names are received from the server 105 via the Internet.

  As another network system configuration, the server 105 may be included in the system 102 of the mobile phone operator, and the personal name search device or a part thereof may be provided in the mobile phone 101. Good.

  FIG. 2 is a functional block diagram functionally showing the configuration of the personal name search apparatus according to the first embodiment of the present invention. The personal name search apparatus according to the first embodiment includes a URL search unit 201, a text extraction unit 202, a personal name list 203, a score counting unit 204, and an output unit 205.

  The URL search unit 201 performs a search from a vast amount of information obtained from a web page on the Internet based on the received keyword, acquires a plurality of URLs of the web page including the keyword, and stores the URL as a URL list. This is output to the extraction unit 202. The URL search unit 201 can be realized by using the technology of an existing Internet search engine (Google (registered trademark), Yahoo! (registered trademark), etc.). The number of URLs to be obtained may be set in advance, or may be specified separately by the user.

  The text extraction unit 202 sequentially accesses the URL addresses according to the URL list, and acquires (extracts) the text information of the page. Here, as the text information, if the web page is in the HTML format, the information obtained by removing the tag information from the web page can be used. Note that only the text information around the keyword may be extracted without acquiring all the text information. As the periphery, a predetermined number of characters and words before and after the keyword can be used. Then, the text extraction unit 202 outputs the acquired text information to the score counting unit 204.

  Note that the contents of the URL can be stored in the cache in advance without accessing the URL address each time. In this case, the text may be acquired from the cache instead of extracting the text via the Internet.

  The personal name list 203 stores personal names that can be candidates in a list form. You can simply store a large number of person names side by side. In addition, as a personal name to be recorded in the personal name list, not only a personal name but also a group name can be designated. Personal names may include stage names, stage name abbreviations or nicknames, or role names such as movies and dramas.

  The score counting unit 204 collates the text information with the person name list 203 and counts the number of occurrences of the person name in the text information for each person name. This number is output as a score for that person's name.

  The output unit 205 outputs a specified number of personal names in descending order of score. Further, not only the person name but also information related to the person name (for example, the link destination of the official site, the profile, the image of the person, etc.) may be output together.

  The URL search unit 201, text extraction unit 202, score counting unit 204, and output unit 205 function as the CPU executes a program stored in a ROM, RAM, hard disk, or removable storage medium (not shown). To realize.

  Next, FIG. 3 is a flowchart showing a processing procedure of the personal name search apparatus according to the first embodiment of the present invention.

  In the flowchart of FIG. 3, first, a keyword is input (step S301). Here, in the case of a system configuration of a terminal such as a mobile phone and a server, a keyword is received. A plurality of keywords can be specified.

  Thereafter, a search is performed from information on a large number of web pages on the Internet based on the input keyword, and a plurality of URLs of web pages including the keyword are acquired (step S302).

  Subsequently, the URL obtained in step S302 is accessed to obtain text information of the web page (step S303). Here, the entire text may be acquired, or only the text in the vicinity where the keyword appears may be extracted. In the case where the URL search unit 201 is provided with a cache function for storing the text information of the web page, the cache data may be read from the storage unit without accessing the Internet.

  Thereafter, it is determined whether or not the text information of all URLs in the URL list obtained in step S302 has been acquired (step S304). If it is determined that the text information has not been acquired to the end (step S304: No), step S303 is performed. Return to. On the other hand, when it is determined that the URL list has been acquired to the end (step S304: Yes), the process proceeds to step S305.

  Subsequently, one person name is read from the person name list 203 one by one, the number of appearances of how many person names are included in the text information acquired in step S303 is counted, and the number is used as a score for the person name (step S305). ).

  Thereafter, it is determined whether or not all personal names have been read from the personal name list 203 (step S306). If it is determined that all names have not been acquired (step S306: No), the process returns to step S305. On the other hand, if it is determined that all the names have been read from the name list (step S306: Yes), the process proceeds to step S307.

  Note that it is also possible to speed up the processing by performing the processing of step S305 and step S306 in parallel using a character string search technique such as using a hash table without counting personal names one by one.

  Subsequently, personal names are output in descending order of score (step S307), and a series of processing ends. Here, in the case of the system configuration as shown in FIG. 1, the personal names are transmitted to the terminal side such as the mobile phone 101.

  In addition to outputting a person name, in the case of outputting additional information related to the person name, the corresponding additional information may be read from the means for storing additional information for each person name in the person name list 204 and output together with the person name. .

  Next, an example of an input / output screen of the personal name search device according to the first embodiment of the present invention will be described. FIG. 4 is an example of an input screen. Here, the input box 401 is a part for inputting a keyword. When this is selected, the screen moves to the keyword input screen shown in FIG. 5, and the keyword is actually input here. A button 402 is a button for instructing a search. When the search button is pressed after inputting a keyword, the keyword is sent to the server and the search process is started. After the search process, the screen moves to the output screen of FIG.

  FIG. 5 is an example of a keyword input screen. Here, in the character input screen 501, various keywords related to the name of a person who wants to search are input. When entering multiple keywords, separate them with spaces or line breaks. A button 502 is a confirmation button. When this button is selected, the screen returns to the input screen of FIG.

  FIG. 6 is an example of an output screen. Here, candidate names are displayed in the output 601 in descending order of score. The order may be displayed as shown in FIG. When a candidate for a person name is selected, information (related links, images, etc.) related to the person name may be displayed. When the button 602 is selected, candidates for the next rank are displayed.

  According to the first embodiment described above, since a list of candidate names is prepared in advance, a large amount of web page information scattered on the Internet and automatically updated is used as a database. This makes it possible to output candidate names from the keywords.

  In addition, it is different from the conventional method that keyword matching is not performed within the site, but personal name matching is performed from information searched by the keyword, and the human name can be searched from various keywords with high accuracy.

  In addition, since Internet information is used in the database instead of in-site information, it is always possible to search from new information.

  Note that after the text information is acquired by the text extraction unit 202 in FIG. 2, morphological analysis of the text information may be performed, and only the part-of-speech information related to the person name may be extracted in advance. Thereby, the processing in the score counting unit 204 can be speeded up.

  Also, prepare a name list for each field (for example, singer, talent, athlete, etc.) so that the field can be specified along with the keyword, and the score is obtained only for the name list in the specified field. It can also be.

(Embodiment 2)
In the first embodiment described above, the personal name list 203 is prepared. However, in the second embodiment described below, the personal name list 203 may not be used by using the morphological analysis technique.

  FIG. 7 is a functional block diagram functionally showing the configuration of the personal name search device according to the second embodiment of the present invention.

  The personal name search apparatus according to the second embodiment includes a URL search unit 701, a text extraction unit 702, a morpheme analysis unit 703, a personal name extraction unit 704, a duplication count unit 705, and an output unit 706.

  Since the URL search unit 701 and the text extraction unit 702 are the same as the URL search unit 201 and the text extraction unit 202 of the first embodiment, respectively, their description is omitted here.

  The morpheme analysis unit 703 performs morphological analysis of the text information, divides the text information into words, and outputs a word and the part of speech of the word. An existing morpheme analysis technique can be used for this morpheme analysis unit. Thus, for example, if “Taro Yamada” is included in the text information, the divided words and parts of speech such as [Yamada: proper noun / person name / surname] [Taro: proper noun / person name / first name] Is output.

  The personal name extraction unit 704 extracts only words related to personal names from the output of the morphological analysis unit 703 and outputs them as a candidate list. Therefore, this output is a list with only the names of people extracted from the text information. Note that the term “person name” here includes not only the name of a proper noun but also a group name or an organization name.

  If the morphological analysis unit 703 is specially constructed for personal name analysis, the morpheme analysis unit 703 and the personal name extraction unit 704 can be combined into one.

  Duplicate counting section 705 counts the number of duplicate names for each person name from the candidate list that is output from person name extracting section 704, and outputs it as a score for that person name.

  The output unit 706 has a function of outputting a specified number of person names in descending order of score. Further, not only a person name but also information related to the person name (for example, a link destination of an official site, a profile, an image of the person, etc.) may be added and output.

  Next, FIG. 8 is a flowchart showing a processing procedure of the personal name retrieval apparatus according to the second embodiment of the present invention.

  In the flowchart of FIG. 8, the process from the keyword input in step S801 to the final determination in the URL list in step S804 is the same as that in steps S301 to S304 in the first embodiment, and thus the description thereof is omitted.

  After step S803 is repeatedly performed to acquire text information, the text information is subjected to morphological analysis (step S805). Text information is divided into units of words by morphological analysis, and each part of speech is given. Subsequently, from the result of morphological analysis, the part of speech extracts words related to the names of persons and groups, and lists them (step S806).

  Thereafter, when there are a plurality of the same person names (or group names) in the listed words, the number is counted and the number is set as the score of the person name (step S807). Subsequently, the names are output in descending order of score, and the series of processing ends (step S808). Here, in the case of the system configuration as shown in FIG. 1, the personal names are transmitted to the terminal side such as the mobile phone 101.

  Since the input / output example of the personal name retrieval apparatus according to the second embodiment of the present invention is the same as that of the first embodiment shown in FIGS. 4, 5, and 6, description thereof will be omitted.

  According to the second embodiment described above, by extracting a person name by morphological analysis without preparing a list of candidate person names, it is possible to output arbitrary candidates without limiting the person name candidates. it can.

  In addition, in the score counting unit 204 in FIG. 2 or the duplication counting unit 705 in FIG. 7, the score may be increased not only by the complete match of the first and last names but also by the match of either the last name and the first name. In this case, the score increment may be smaller than the complete match, and the weight of the score when the first name matches the last name may be increased. As a result, an appropriate score can be given even to a person name that is often expressed only by a surname or a first name.

  Further, in the score counting unit 204 in FIG. 2 or the duplication counting unit 705 in FIG. 7, the weight of the score when the matched person name is the title in the web page may be increased. This makes it possible to estimate a person name with higher accuracy.

  In some cases, the keyword includes a person's name. For example, when searching for a group name, the name of another member may be specified, or the name of a co-star may be specified. In this case, since it is not necessary to display the person name specified by the keyword as the person name to be output, the output unit 205 in FIG. 2 or the output unit 706 in FIG. 7 is configured not to output the person name included in the keyword. The score counting unit 204 of 2 or the duplication counting unit 705 of FIG. 7 may not count personal names included in the keywords.

  The person name search method described in this embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program is recorded on a computer-readable recording medium such as a hard disk, a flexible disk, a CD-ROM, an MO, and a DVD, and is executed by being read from the recording medium by the computer. Further, this program may be a transmission medium that can be distributed via a network such as the Internet.

  As described above, the person name search device, person name search method, and person name search program according to the present invention estimate the person name from various keywords related to the person name in a situation where the name of the celebrity or the like or the group name is unknown. Is suitable.

BRIEF DESCRIPTION OF THE DRAWINGS It is a block diagram which shows the whole network system concerning Embodiment 1 of this invention. It is a functional block diagram which shows functionally the structure of the personal name search device concerning Embodiment 1 of this invention. It is a flowchart which shows the procedure of the process of the personal name search device concerning Embodiment 1 of this invention. It is explanatory drawing which shows the example of the input screen concerning Embodiment 1 of this invention. It is explanatory drawing which shows the example of the keyword input screen concerning Embodiment 1 of this invention. It is explanatory drawing which shows the example of the output screen concerning Embodiment 1 of this invention. It is a functional block diagram which shows functionally the structure of the personal name search device concerning Embodiment 2 of this invention. It is a flowchart which shows the procedure of the process of the personal name search device concerning Embodiment 2 of this invention.

Explanation of symbols

101 Mobile phone 102 Mobile phone operator system 103 Personal computer or PDA
104 Internet 105 Server 201 Including Name Search Device of the Present Invention URL Search Unit 202 Text Extraction Unit 203 Person Name List 204 Score Counting Unit 205 Output Unit 401 Input Box 402 for Entering Keywords Input Button 501 for Instructing Search Input Screen 502 Confirm button 601 Output result display 602 Next selection button 701 URL search unit 702 Text extraction unit 703 Morphological analysis unit 704 Person name extraction unit 705 Duplicate count unit 706 Output unit

Claims (5)

Personal name list storage means for storing personal names to be output as candidates;
An input means for receiving an input of a keyword related to a name of a person who desires a search;
Search means for acquiring the contents of a web page related to the keyword input by the input means from the Internet;
In the contents of the web page acquired by the search means, a counting means for counting the number of appearances of personal names stored in the personal name list storage means for each personal name,
Output means for outputting a person's name based on the number of appearances counted by the counting means;
A personal name search device characterized by comprising: An input means for receiving an input of a keyword related to a name of a person who desires a search;
Search means for acquiring the contents of a web page related to the keyword input by the input means from the Internet;
Extraction means for extracting a person's name from the content of the web page acquired by the search means;
Counting means for counting the number of person names extracted by the extracting means for each person name;
Output means for outputting a person's name based on the number counted by the counting means;
A personal name search device characterized by comprising: An input process for receiving an input of a keyword related to a name of a person who desires a search;
A search step of acquiring the content of a web page related to the keyword input by the input step from the Internet;
In the content of the web page acquired by the search step, a counting step of counting the number of appearances of personal names stored in the personal name list that stores personal names to be output as candidates,
An output step of outputting a person name based on the number of appearances counted by the counting step;
Name search method characterized by including. An input process for receiving an input of a keyword related to a name of a person who desires a search;
A search step of acquiring the content of a web page related to the keyword input by the input step from the Internet;
From the content of the web page acquired by the search step, an extraction step of extracting a person name;
A counting step of counting the number of person names extracted by the extraction step for each person name;
An output step of outputting a person's name based on the number counted by the counting step;
Name search method characterized by including. A personal name search program that causes a computer to execute any one of the personal name search methods according to claim 3.

Images