JP2018190113A - Data processing apparatus, data processing system, data processing method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data processing apparatus, a data processing system, a data processing method and a program which locate the occurrence of an event from posted data collected over the Internet about cases, accidents, disasters and so on.SOLUTION: A data processing apparatus 1 for processing posted data collected over the Internet about an arbitrary event is configured to comprise a poster information analysis unit 11 which extracts geographical names from poster information of the posted data, a past posting analysis unit 12 which extracts geographical names from past posted data from posters of the posted data, and an occurrence locating unit 13 which locates the occurrence of the event in accordance with the posted data and data associated therewith. The occurrence locating unit 13 uses analysis results of the poster information analysis unit 11 and/or the past posting analysis unit 12 to locate the occurrence of the event.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a data processing apparatus, a data processing system, a data processing method, and a program for processing post data collected from the Internet. More specifically, the present invention relates to a technique for extracting specific information from post data and data accompanying it.

  Many sentences and images are posted on the Internet every day. In particular, the social networking service (SNS) can easily post comments and photos, so information about incidents, accidents, and disasters may be posted in real time. Has begun to be used for news and reporting.

  On the other hand, since postings to SNS are often short texts, it is difficult to identify where the event is occurring even if post data on incidents, accidents, and disasters is collected from the Internet. There's a problem. Therefore, conventionally, techniques have been proposed for specifying a posting position and a location related to the posting content for posting data collected from the Internet (see Patent Documents 1 to 4).

  For example, in Patent Document 1, a comment including place name information is extracted from post comments collected from SNS, and a place name information or a broadcast station name included in the news is further extracted from the comments. Thus, an information extraction device that improves the readability of regional information is disclosed except for comments derived from television broadcasts and newspaper articles. Patent Document 2 also collects post data related to cases originated from the reference area and its surrounding areas, and collects again the area with a large amount of collected post data as a new reference area. A search device is disclosed that can collect detailed information regarding the search.

  Patent Document 3 classifies post comments with position information for each predetermined area range, extracts the topic numbers with the largest number as area-specific topic numbers, and collects them based on the area-specific topic numbers in each area range. A method is disclosed in which all the posted comments are classified to extract only unexpected posted sentences that occurred in the area. Further, Patent Document 4 classifies comments with position information for each map route included in a predetermined range, extracts the largest topic number among them as a map route topic number, and sets the map route topic number of each map route as the map route topic number. A method of classifying each posted comment based on it is disclosed.

JP 2014-137632 A JP 2014-182525 A JP 2016-45938 A JP-A-2006-136358

  However, in the methods described in Patent Documents 1 to 4 described above, information necessary for determining the posting position or the like cannot be obtained sufficiently, and it may not be possible to specify a location related to the posted content. In particular, when posting data related to incidents, accidents, disasters, etc. is used in news and reports, it is important to specify the location of the event, and the location of the event can be identified in a short time from limited information. Possible technology is required.

  Therefore, the present invention provides a data processing device, a data processing system, a data processing method, and a program capable of specifying the occurrence position of the event from post data regarding incidents, accidents, and disasters collected via the Internet. Objective.

The data processing apparatus according to the present invention is a data processing apparatus that processes post data relating to an arbitrary event collected via the Internet, and a poster information analysis unit that extracts a place name from poster information of the post data, A past post analysis unit that extracts a place name from past post data by a contributor of the post data, and an occurrence position specifying unit that specifies an occurrence position of the event from the post data and data associated therewith, The occurrence position specifying unit specifies an occurrence position of the event using an analysis result in the poster information analysis unit and / or the past posting analysis unit.
If the place name extracted by the poster information analysis unit and the place name extracted by the past post analysis unit are different, the occurrence position specifying unit preferentially uses the place name extracted by the poster information analysis unit. You may choose.
The data processing apparatus of the present invention may further include an image analysis unit that extracts character information from an image attached to the posted data. In this case, the generation position specifying unit is configured to analyze the poster information. When the event occurrence position cannot be identified from the analysis results of the image analysis unit and the past posting analysis unit, the event occurrence position is identified using the analysis result of the image analysis unit.
The data processing apparatus according to the present invention includes a follower information analysis unit that extracts a place name from profile information of a user who follows the poster of the posted data, and profile information of the user that the poster of the posted data follows. It may further include a follow user information analysis unit for extracting a place name, in which case, the occurrence position specifying unit is unable to identify the occurrence position of the event from the analysis result in the image analysis unit, The occurrence position of the event is specified using the analysis result in the follower information analysis unit and / or the follow user information analysis unit.
The data processing apparatus according to the present invention includes a follower post analysis unit that extracts a place name from post data of a user who is following the contributor of the post data, and post data of the user that the post person of the post data is following A follow user post analysis unit for extracting a place name, in which case, the occurrence position specifying unit is configured to analyze the event from the analysis results of the follower information analysis unit and the follow user information analysis unit. When the occurrence position cannot be specified, the occurrence position of the event is specified by using the analysis result in the follower post analysis unit and / or the follow user post analysis unit.
The data processing apparatus of the present invention includes a text analysis unit that extracts a place name from text included in the post data, and a position information analysis unit that extracts post position information from the post data. When the occurrence position of the event cannot be specified from the analysis result in the text analysis unit and the position information analysis unit, the analysis result in the poster information analysis unit and / or the past contribution analysis unit is used. Thus, the occurrence position of the event can be specified.

A data processing system according to the present invention includes the data processing device described above and a data collection device that collects post data relating to an arbitrary event via the Internet, and the post data or post data group collected by the data collection device. Are processed by the data processing device, and the occurrence position of the event is specified.
The data processing system may further include a distribution device that adds the occurrence position information specified by the data processing device and externally distributes post information related to the event.

A data processing method according to the present invention is a method of processing post data relating to an arbitrary event collected via the Internet using one or a plurality of data processing devices, wherein the post-data is processed by the data processing device. A poster information analysis step of extracting a place name from poster information, a past post analysis step of extracting a place name from past post data by a poster of the post data, the poster information analysis step and / or the past post analysis An occurrence position specifying step of specifying the occurrence position of the event using the analysis result obtained in the process is performed.
In the occurrence position specifying step, when the place name extracted in the poster information analysis step is different from the place name extracted in the past post analysis step, the place name extracted in the poster information analysis step is given priority. You may choose.
In the data processing method of the present invention, if the occurrence position of the event cannot be specified from the analysis results of the contributor information analysis step and the past post solution step, characters are further extracted from the image attached to the post data. It is also possible to perform an image analysis process for extracting information and use the analysis result of the image analysis process to specify the occurrence position of the event.
In the data processing method of the present invention, if the occurrence position of the event cannot be specified from the analysis result of the image analysis process, the place name is further extracted from the profile information of the user who follows the poster of the posted data. The follower information analysis step to extract, and the follow user information analysis step to extract the place name from the profile information of the user who the poster of the post data follows, the follower information analysis step and / or the follow user information analysis You may identify the generation | occurrence | production position of the said event using the analysis result of a process.
In the data processing method of the present invention, if the occurrence position of the event cannot be specified from the analysis results of the follower information analysis step and the follow user information analysis step, the user who follows the contributor of the post data A follower post analysis step of extracting a place name from the post data of the post, and a follow user post analysis step of extracting a place name from the post data of the user that the poster of the post data follows, and the follower post analysis step and / or Alternatively, the occurrence position of the event may be specified using the analysis result of the follow user posting analysis step.
In the data processing method of the present invention, first, a text analysis step of extracting a place name from text included in the post data, and a location information analysis step of extracting post location information from the post data, the text analysis step and When the occurrence position of the event cannot be specified from the analysis result of the position information analysis process, the poster information analysis process and the past contribution analysis process may be performed.

  The program according to the present invention is a program for causing a computer to process post data relating to an arbitrary event collected via the Internet, and for extracting the place name from the poster information of the post data, Identify the location of the event using the past post analysis function that extracts the place name from the past post data by the contributor of the post data and the analysis result by the contributor information analysis function and / or the past post analysis function The generation position specifying function is executed.

  The “event” in the present invention includes various events such as incidents, accidents, disasters, etc. that are taken up in the press, and their occurrence locations are not limited to domestic but also include overseas.

  According to the present invention, it is possible to specify the occurrence position of an event such as an incident / accident / disaster in a short time even for post data for which the position cannot be specified by the prior art.

It is a conceptual diagram which shows the structural example of the data processor of the 1st Embodiment of this invention. It is a flowchart which shows the method of processing posting data using the data processor 1 shown in FIG. It is a conceptual diagram which shows the structural example of the data processor of the 1st modification of the 1st Embodiment of this invention. It is a flowchart which shows the method of processing posting data using the data processor 2 shown in FIG. It is a conceptual diagram which shows the structural example of the data processor of the 2nd modification of the 1st Embodiment of this invention. It is a flowchart which shows the method of processing post data using the data processing device 3 shown in FIG. It is a conceptual diagram which shows the structural example of the data processing system which concerns on the 2nd Embodiment of this invention.

  DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the accompanying drawings. Note that the present invention is not limited to the embodiments described below.

(First embodiment)
First, a data processing apparatus according to the first embodiment of the present invention will be described. FIG. 1 is a conceptual diagram illustrating a configuration example of a data processing apparatus according to the present embodiment. As shown in FIG. 1, the data processing apparatus 1 of the present embodiment includes at least a poster information analysis unit 11, a past posting analysis unit 12, and an occurrence position specifying unit 13, and from post data regarding an arbitrary event, This is to specify the location of the event. The data processing apparatus 1 can be provided with an image analysis unit 14 as necessary.

[Posted data]
The post data processed by the data processing apparatus 1 of the present embodiment relates to an arbitrary event such as an incident, an accident, or a disaster collected via the Internet. In particular, “post position information (GPS data) is included. The location where the event occurred cannot be specified by the conventional method for reasons such as “not yet” or “the place name is not included in the text”.

  The post data used in the data processing device 1 includes text data such as post texts and comments, image data attached thereto, and information related to the contributor such as a profile. In addition, text data and image data include not only posts related to events that have been collected, but also past posts by the same poster.

[Contributor Information Analysis Unit 11]
The poster information analysis unit 11 extracts a place name from the poster information. This is because the poster may have encountered the specific target event in his / her own living area, and in that case, the event occurrence position is the residence, work place, hometown or the surrounding area.

[Past Post Analysis Unit 12]
The past posting analysis unit 12 extracts a place name from a posted sentence or a comment posted by the same poster before a posting related to a specific target event. This is because the past posts may include information on the specific target event and information on the poster's life sphere.

[Generation position specifying unit 13]
The occurrence position specifying unit specifies an occurrence position of an event from post data and data accompanying it. For example, when the same place name is extracted by the poster information analysis unit 11 and the past post analysis unit 12, or when the place name is extracted by only one of the poster information analysis unit 11 and the past post analysis unit 12. The generation position specifying unit 13 specifies the generation position from the extracted place name.

  When the place name extracted by the poster information analysis unit 11 and the place name extracted by the past post analysis unit 12 are different, the occurrence position specifying unit 13 preferentially uses the place name extracted by the poster information analysis unit 11. Select If neither the poster information analysis unit 11 nor the past posting analysis unit 12 can extract the place name, the generation position may be specified from the place name extracted by the image analysis unit 14 described later.

[Image Analysis Unit 14]
The image analysis unit 14 extracts character information related to the place name from the image attached to the post data. Specifically, character information is extracted from image data by an optical character recognition technique such as OCR. Here, examples of the character information extracted from the image include a car number, an address display board, a sign, and a place name described on a store signboard.

  The image has useful information regarding the occurrence position of the event, but character recognition processing such as OCR requires more time than the place name extraction processing described above. Therefore, in the data processing apparatus 1 according to the present embodiment, the analysis result in the image analysis unit 14 is used only for a post in which the event occurrence position cannot be specified from the poster information or the past post. The processing in the image analysis unit 14 may be performed after the analysis in the poster information analysis unit 11 and the past posting analysis unit 12, but the processing may be started at the same time.

[Operation]
Next, an operation of the data processing apparatus 1 according to the present embodiment, that is, a method for processing the posted data using the data processing apparatus 1 and specifying the occurrence position of the event will be described. FIG. 2 is a flowchart showing the data processing method of this embodiment. The data processing method of the present embodiment is a method of processing post data relating to an arbitrary event collected via the Internet using one or a plurality of data processing devices 1, including a poster information analysis step S 1 a and past posts An analysis step S1b and a generation position specifying step (I) S2a are performed.

<Contributor Information Analysis Step S1a>
In the poster information analysis step S1a, information on the poster such as a profile is analyzed to extract a place name. At that time, when a plurality of place names are included in the poster information, the latest information is preferentially selected, and when the priority cannot be determined from the temporal transition, a larger round is adopted.

  For example, when the latest residence of the poster is “A city” and the birth place is “B city”, “A city” is selected. Further, when priority cannot be determined and both “A city” and “B city” are extracted, it is only necessary to determine that these cities have “X prefecture”. Furthermore, if the poster information includes “X prefecture A city” and “Y prefecture C city” and one of them cannot be selected, it may be determined as “Z district” including X prefecture and Y prefecture, Both may be selected and determined as “A city or C city” or “X prefecture or Y prefecture”. When the place name is not included in the poster information, it is determined that “no place name”.

<Past post analysis step S1b>
In the past posting analysis step S1b, a place name is extracted by analyzing past posting data such as a posted sentence and a comment posted in the past by a poster who has made a posting related to the specific target event. If the place name is not included in the past post data, it is judged as “no place name”, and if there are multiple place names, the one that is close in time to the post related to the specific event is preferentially selected. .

<Generating position specifying step (I) S2a>
In the occurrence position specifying step (I) S2a, the occurrence position of the event is specified using the analysis result obtained in the above-described poster information analysis step S1a and / or the past contribution analysis step S1b. Specifically, when one of the poster information analysis process S1a and the past contribution analysis process S1b is “no place name”, the analysis result in the process of extracting the place name is used to determine the event occurrence position. Identify.

  When the same place name is extracted in the poster information analysis step S1a and the past post analysis step S1b, the occurrence position of the event is specified based on the place name. On the other hand, when different place names are extracted in the poster information analysis step S1a and the past post analysis step S1b, the analysis result in the poster information analysis step S1a having a high probability of being the occurrence position of the specific target event is preferentially used. To identify the position of occurrence. For example, when “place name A” is extracted in the poster information analysis step S1a and “place name B” is extracted in the past post analysis step S1b, “place name A” is set as the event occurrence position.

  If the place name cannot be extracted in either the poster information analysis step S1a or the past post analysis step S1b, it is determined as “unidentifiable” in the generation position specifying step (I) S2a. In that case, an image analysis step S1c and a generation position specifying step (II) S2b described later are performed.

<Image analysis step S1c>
Image analysis process S1c extracts the character information regarding a place name using the optical character recognition techniques, such as OCR, from the image accompanying the posting regarding a specific target event. Specifically, a place name is extracted from a car number, an address display board, a sign, a store signboard, and the like included in the image.

  This image analysis step S1c may be performed after the occurrence position specifying step (I) S2a, or may be performed simultaneously with the contributor information analysis step S1a and the past post analysis step S1b. When performing in parallel with the contributor information analysis step S1a and the past post analysis step S1b, in the generation position specifying step (I) S2a, together with the analysis results of the contributor information analysis step S1a and the past post analysis step S1b, an image The analysis result in the analysis step S1c may be used to specify the event occurrence position.

<Generation position specifying step (II) S2b>
In the occurrence position specifying step (II) S2b, the occurrence position of the event is specified using the place name ("place name C") extracted in the image analysis step S1c described above. At that time, if multiple place names are included in the image, the one with higher certainty such as an address display board or a sign is preferentially adopted, and if there is no reliable information, plural information Is judged comprehensively and a larger round is used.

  For example, if the place name read from the address display board is “A city” and the place name read from the car number is “B city”, “A city” with higher certainty is selected. In addition, when there is no highly reliable character information such as an address display board or a sign and “A city” and “B city” are extracted from the car number, it is determined that these cities have “X prefecture”. be able to. Furthermore, when “X prefecture A city” and “Y prefecture C city” are included in the image and either one cannot be selected, it may be determined as “Z district” including X prefecture and Y prefecture. Both of them may be selected and determined as “A city or C city” or “X prefecture or Y prefecture”. If the place name is not included in the image, it is determined that there is no place name.

[program]
Each process described above can be implemented by creating a computer program for realizing each function of the data processing apparatus and mounting it on one or more computers. In other words, the data processing method of the present embodiment includes a poster information analysis function that extracts a place name from poster information of posted data, and a past post analysis that extracts a place name from past posted data by a poster of posted data. This can be implemented by executing the function and the occurrence position specifying function for specifying the occurrence position of the event using the analysis result by the poster information analysis function and / or the past contribution analysis function.

  In this computer program, if the place name extracted by the poster information analysis function is different from the place name extracted by the past contribution analysis function, the place name extracted by the poster information analysis function will be given priority in the occurrence location specification function. You may make it choose. In addition to the above functions, the computer program according to the present embodiment causes the computer to execute an image analysis function that extracts character information from an image attached to the post data, thereby providing a poster information analysis function and a past post solution. When the event occurrence position cannot be identified from the analysis result by the function, the occurrence position identification function may identify the event occurrence position using the analysis result by the image analysis function.

  Each function described above need not be installed in one program, and may be executed by creating a program for each function and linking them. In that case, each program can be divided and installed in two or more computers or apparatuses to be operated.

  As described above in detail, the data processing apparatus of the present embodiment extracts the place name from the information related to the poster and past posts of the same poster, and the accompanying images as necessary. It is possible to specify the location of the event even if the post cannot be specified.

(First modification of the first embodiment)
Next, a data processing apparatus according to a first modification of the first embodiment of the present invention will be described. FIG. 3 is a conceptual diagram showing a configuration example of the data processing apparatus according to this modification. In FIG. 3, the same components as those of the data processing apparatus 1 shown in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted. As shown in FIG. 3, the data processing device 2 of the present modification includes a text analysis unit 25 and a position information analysis unit 26 in addition to the units of the data processing device 1 of the first embodiment described above.

[Text analysis unit 25]
A place name is extracted from text included in a post about a specific target event, such as a posted sentence or a comment. The place name included in the posting of the specific target event is highly likely to indicate the occurrence position of the event, and is information that is particularly effective for specifying the occurrence position.

[Position information analysis unit 26]
Post position information (GPS data) is extracted from post data. The poster is highly likely to post from the event occurrence position, and the GPS data attached to the posting of the specific target event is the most effective information for specifying the occurrence position.

  In the data processing device 2 of the present modification, first, the occurrence position of the event is specified from the analysis results in the text analysis unit 25 and the position information analysis unit 26, and when the generation position cannot be specified by these, The occurrence position specifying unit 13 specifies the occurrence position of the event using the analysis result in the poster information analysis unit 11 and / or the past post analysis unit 12.

[Operation]
FIG. 4 is a flowchart showing a method of processing post data using the data processing apparatus 2 shown in FIG. As shown in FIG. 4, the data processing device 2 of the present modification includes a text analysis step S10a that extracts a place name from text included in post data, and a position information analysis step S10b that extracts post position information from post data. And an occurrence position specifying step S20a for specifying the occurrence position of the event from these analysis results. Then, when the occurrence position of the event cannot be specified in the generation position specifying step S20a, the above-described poster information analysis step S1a and the past post analysis step S1b are performed.

[program]
Each process described above can be implemented by creating a computer program for realizing each function of the data processing apparatus and mounting it on one or more computers. That is, the data processing method of the present modification includes a text analysis function for extracting a place name from text included in post data and a post from post data in addition to the functions of the program of the first embodiment described above. Executes the location information analysis function that extracts location information, and executes the poster information analysis function and the past contribution analysis function when the event occurrence location cannot be identified from the analysis results by the text analysis function and the location information analysis function. Can be implemented.

  In the data processing apparatus 2 of the present modification, for post data having information that can easily specify an event occurrence position such as text or GPS data, the event occurrence position is specified based on such information. In addition, since the poster information and past posts are analyzed only for post data for which the occurrence position cannot be specified, the occurrence position of the event can be specified efficiently and accurately. The configuration and effects other than those described above in the present modification are the same as those in the first embodiment described above.

(Second modification of the first embodiment)
Next, a data processing apparatus according to a second modification of the first embodiment of the present invention will be described. FIG. 5 is a conceptual diagram showing a configuration example of a data processing apparatus according to this modification. In FIG. 5, the same components as those of the data processing apparatus 1 shown in FIG. 1 are denoted by the same reference numerals, and detailed description thereof is omitted. As shown in FIG. 5, the data processing device 3 of the present modification includes a follower analysis unit 37 and a follow user analysis unit 38 in addition to the units of the data processing device 1 of the first embodiment described above.

[Follower Analysis Unit 37]
The follower analysis unit 37 is a follower information analysis unit that extracts a place name from profile information of a user (follower) who is following a contributor of a post regarding a specific target event, and a follower post analysis unit that extracts a place name from post data of the follower With. The follower may have the same sphere of life as the contributor, and the place where the event occurred is included in the place name extracted from the information such as the place of birth and residence of the follower, the text of the posted text and comments, etc. Because there is a possibility that.

[Follow user analysis unit 38]
The follow user analysis unit 38 is a follower information analysis unit that extracts a place name from the profile information of a user who is following a post about a specific target event, and a follow user post analysis unit that extracts a place name from the post data of the follow user With. The contributor may be following a user who has a common earthquake and living area, and the place name extracted from the text such as the post user's place of birth, residence, etc. This is because there is a possibility that the occurrence position is included.

  In the data processing device 3 according to the present modification, the occurrence position specifying unit 13 cannot determine the occurrence position of the event from the analysis results of the poster information analysis unit 11 and the past posting analysis unit 12, and the image analysis unit 14. The location of the event is specified from the analysis result of. Then, when the occurrence position cannot be specified even by the analysis result in the image analysis unit 14, the occurrence position of the event is specified by using the analysis result in the follower analysis unit 37 or the follow user analysis unit 38.

[Operation]
FIG. 6 is a flowchart showing a method of processing post data using the data processing device 3 shown in FIG. As shown in FIG. 6, the data processing device 3 of the present modified example further includes a follower when the occurrence position of the event cannot be specified from the analysis result of the image analysis process S1c in the generation position specifying step (II) S2b. An analysis step S3a and a follow user analysis step S3b are performed, and an occurrence position specifying step (III) S4 for specifying an occurrence position of an event from these analysis results is performed.

  Specifically, the place name is extracted from the profile information of the follower, the place name is extracted from the profile information of the follow user, and the occurrence position of the event is specified from the extracted place name. If the place name is not extracted from the profile information of the follower and the follow user, the place name is extracted from the post data of the follower and the post data of the follow user, and the occurrence position of the event is specified using the result.

[program]
Each process described above can be implemented by creating a computer program for realizing each function of the data processing apparatus and mounting it on one or more computers. In other words, the program of this modification example, when the event occurrence position cannot be identified from the analysis result of the image analysis function, the follower information analysis function for extracting the place name from the profile information of the follower and the profile of the follow user Follow user information analysis function to extract place name from information. Then, the occurrence position specifying function may specify the occurrence position of the event using the analysis result of the follower information analysis function or the follow user information analysis function.

  Furthermore, the program of this modification extracts the place name from the post data of the follower to the computer when the event occurrence position cannot be identified from the analysis result of the follower information analysis function and the follow user information analysis function. The function and the follow user post analysis function for extracting the place name from the post data of the follow user are executed. Then, the occurrence position specifying function may specify the occurrence position of the event using the analysis result of the follower post analysis function or the follow user post analysis function.

  Since the data processing apparatus 3 of this modification analyzes the information and posts of followers and follow users, it is possible to specify the occurrence position even for the event where the occurrence position cannot be specified from the poster information or the post data. It becomes. Thereby, the identification accuracy of the event occurrence position can be increased.

  The configuration and effects other than those described above in the present modification are the same as those in the first embodiment described above. In addition, the data processing device of this modification can be combined with the configuration of the first modification described above. In that case, a text analysis unit and a position information analysis unit are further provided, and only a post whose generation position cannot be specified by these analysis results may be analyzed by the poster information analysis unit and the past post analysis unit.

(Second Embodiment)
Next, a data processing system according to the second embodiment of the present invention will be described. FIG. 7 is a conceptual diagram showing a configuration example of the data processing system of this embodiment. As shown in FIG. 7, the data processing system 40 of the present embodiment collects post data 5a and 5b related to an arbitrary event via the data processing apparatus 1 of the first embodiment and the Internet 4 described above. A device 41 is provided.

  The data processing system 40 processes post data 5a, 5b collected by the data collection device 41 or a post data group into which these are classified by the data processing device 1, and identifies the occurrence position of the event. Further, the data processing system 40 of the present embodiment may be provided with a distribution device 42 that adds the occurrence position information specified by the data processing device 1 and externally distributes post information related to the event.

  The data processing system of the present embodiment is a data for extracting a place name related to a contributor by a contributor information analysis unit, a past post analysis unit, and an image analysis unit as necessary, and using the result to specify an occurrence position of an event Since the processing apparatus is provided, it is possible to specify the event occurrence position even in a post in which the occurrence position cannot be specified by the conventional method. The configuration and effects other than those described above in the present embodiment are the same as those in the first embodiment described above.

  FIG. 7 shows a configuration using the data processing device 1 of the first embodiment shown in FIG. 1, but the present invention is not limited to this. Specifically, the data processing device 2 shown in FIG. 3 includes a text analysis unit and a positional information analysis unit, and the data processing device 3 shown in FIG. 5 includes a follower analysis unit and a follow user analysis unit. In this case, the same effect can be obtained.

1-3 Data processing device 4 Internet 5a, 5b Post data 11 Contributor information analysis unit 12 Past post analysis unit 13 Generation location specifying unit 14 Image analysis unit 25 Text analysis unit 26 Location information analysis unit 37 Followers information analysis unit 38 Follow user Information analysis unit 40 Data processing system 41 Data collection device 42 Distribution device

Claims (15)

A data processing device for processing post data relating to an arbitrary event collected via the Internet,
A contributor information analysis unit for extracting a place name from contributor information of the post data;
A past post analysis unit for extracting a place name from past post data by a contributor of the post data;
An occurrence position identifying unit for identifying the occurrence position of the event from the posted data and data accompanying it,
The data generation device, wherein the occurrence position specifying unit specifies an occurrence position of the event using an analysis result in the poster information analysis unit and / or the past contribution analysis unit.   If the place name extracted by the poster information analysis unit and the place name extracted by the past post analysis unit are different, the occurrence position specifying unit preferentially uses the place name extracted by the poster information analysis unit. The data processing apparatus according to claim 1 to be selected. Furthermore, it has an image analysis unit for extracting character information from an image attached to the post data,
If the occurrence position of the event cannot be specified from the analysis results in the poster information analysis unit and the past contribution analysis unit, the generation position specifying unit uses the analysis result in the image analysis unit. The data processing apparatus according to claim 1, wherein the occurrence position of the event is specified. A follower information analysis unit that extracts a place name from profile information of a user who follows the poster of the post data;
A follow user information analysis unit that extracts a place name from the profile information of the user that the poster of the post data is following,
The occurrence position specifying unit uses the analysis result in the follower information analysis unit and / or the follow user information analysis unit when the occurrence position of the event cannot be specified from the analysis result in the image analysis unit. The data processing apparatus according to claim 3, wherein the occurrence position of the event is specified. A follower post analysis unit that extracts a place name from post data of a user who follows the poster of the post data;
A follow user post analysis unit that extracts a place name from post data of a user that the contributor of the post data follows;
When the occurrence position of the event cannot be specified from the analysis results of the follower information analysis unit and the follow user information analysis unit, the follower post analysis unit and / or the follow user post The data processing apparatus according to claim 4, wherein an occurrence position of the event is specified using an analysis result in an analysis unit. A text analysis unit that extracts a place name from the text included in the post data;
A position information analysis unit for extracting post position information from the post data;
If the occurrence position of the event cannot be specified from the analysis results in the text analysis unit and the position information analysis unit, the poster information analysis unit and / or the past contribution analysis unit The data processing apparatus according to any one of claims 1 to 5, wherein an occurrence position of the event is specified by using an analysis result in (1). The data processing device according to any one of claims 1 to 6,
A data collection device that collects post data on any event via the Internet,
A data processing system for processing post data or a post data group collected by the data collection device with the data processing device to identify the occurrence position of the event.   The data processing system according to claim 7, further comprising a distribution device that externally distributes post information related to the event by adding the occurrence position information specified by the data processing device. A method of processing post data relating to an arbitrary event collected via the Internet using one or more data processing devices,
By the data processing device,
A poster information analysis step of extracting a place name from the poster information of the posted data;
A past posting analysis step of extracting a place name from past posting data by a contributor of the posting data;
A data processing method for performing an occurrence position specifying step of specifying an occurrence position of the event using an analysis result obtained in the poster information analysis step and / or the past contribution analysis step.   In the occurrence position specifying step, when the place name extracted in the poster information analysis step is different from the place name extracted in the past post analysis step, the place name extracted in the poster information analysis step is given priority. The data processing method according to claim 9 to be selected. If the event occurrence position could not be identified from the analysis result of the poster information analysis step and the past posting solution step,
Furthermore, an image analysis step of extracting character information from an image attached to the post data is performed,
The data processing method according to claim 9, wherein an occurrence position of the event is specified using an analysis result of the image analysis process. If the event occurrence position could not be identified from the analysis result of the image analysis process,
Furthermore, a follower information analysis step for extracting a place name from the profile information of a user who follows the contributor of the post data;
The follow user information analysis step of extracting the place name from the profile information of the user that the contributor of the post data is following,
The data processing method according to claim 11, wherein an occurrence position of the event is specified using an analysis result of the follower information analysis step and / or the follow user information analysis step. If the occurrence position of the event could not be identified from the analysis results of the follower information analysis step and the follow user information analysis step,
A follower post analysis step of extracting a place name from post data of a user who follows the contributor of the post data;
A follow user post analysis step of extracting a place name from post data of a user that the poster of the post data follows,
The data processing method according to claim 12, wherein an occurrence position of the event is specified using an analysis result of the follower post analysis step and / or the follow user post analysis step. A text analysis step of extracting a place name from the text included in the post data;
A position information analysis step of extracting posted position information from the posted data,
The poster information analysis step and the past post analysis step are performed when the occurrence position of the event cannot be specified from the analysis results of the text analysis step and the position information analysis step. The data processing method according to item 1. A program that causes a computer to process post data relating to an arbitrary event collected via the Internet,
A contributor information analysis function for extracting a place name from contributor information of the post data;
A past post analysis function for extracting a place name from past post data by a contributor of the post data;
A program for executing an occurrence position specifying function for specifying an occurrence position of the event using an analysis result by the poster information analysis function and / or the past contribution analysis function.

Images