JP5739843B2 - Information providing apparatus, information providing method, and information providing program - Google Patents

Description

  The present invention relates to an information providing apparatus, an information providing method, and an information providing program.

  There is an information providing apparatus that provides information in real time by keyword search from updated or added information as needed.

  For example, in the technique described in Patent Literature 1, topicality that changes from moment to moment is determined from real-time information, and information extraction that considers the change in topicality is performed. For example, there is an information providing apparatus that extracts information including a keyword related to an event and provides the information in real time when a predetermined event occurs. Such an information providing apparatus repeats the process of extracting information including the first keyword, acquiring the second keyword common to the extracted information, and using it for further information extraction, and related to the event. Some provide information to do.

Japanese Patent No. 2729356

  However, in the information providing apparatus that is used for information extraction by adding keywords at any time as described above, it is assumed that the added keywords gradually decrease in relevance with events. For this reason, there is a problem that information related to an event cannot be obtained gradually in the information extraction using the added keyword.

  The present invention has been made in view of such problems, and an object thereof is to provide information related to an event within an appropriate range.

In order to achieve the above object, an information providing apparatus according to a first aspect of the present invention is an information providing apparatus that provides information that occurs in association with a predetermined event in which the situation changes. A first storage means for storing a main related keyword and a sub related keyword that are considered to be highly related to each other, and information stored in the first storage means among information generated within a predetermined period at every predetermined timing. Information including at least one of the main related keyword and the sub related keyword is extracted, and among the extracted information, keywords included in the information of a first predetermined number or more are used as the sub related keywords. extracting means for repeating the extraction and storage processing in the storage means, of the information generated within the predetermined period, small of the main relevant keywords and the sub related keywords Extraction Kutomo extracting the number of information including any one when the condition is met as a second less than the predetermined number, it performs process of excluding the sub related keywords from application of the extraction of information by the extracting means And an end control means.

According to such a configuration, the information providing apparatus extracts information including at least one of the main related keyword and the sub related keyword, and among the extracted information, the keyword included in the information of the first predetermined number or more. The extraction and storage process of storing the above as a sub-related keyword in the first storage means is repeated. However, in view of the fact that the relevance of the secondary related keyword is expected to gradually decrease, the information providing apparatus , at least the primary related keyword and the secondary related keyword among the information generated within the predetermined period. When the condition that the number of extracted information including any one is less than the second predetermined number is satisfied, a process of excluding the sub-related keyword from the application target at the time of extracting the information is performed. For this reason, it is possible to prevent the information related to the event from being obtained by using the sub-related keyword having a low relationship with the event, and to provide the information related to the event in an appropriate range.

Since sub-related keywords are added as needed, it is assumed that the number of information items including at least one of the main related keywords and the sub-related keywords will gradually increase. On the other hand, in spite of such assumptions, when the number of information including at least one of the main related keyword and the sub related keyword among the information generated within the predetermined period becomes less than the second predetermined number , It can be considered that the relevance between the secondary related keyword and the event is low. In this case, the information related to the event can be provided in an appropriate range by excluding the sub-related keyword from the application target when the information is extracted by the extraction unit .

The extraction end control means, when the condition that the number of extracted information including the sub-related keyword is less than a third predetermined number among the information generated within the predetermined period is satisfied, satisfies the condition. You may make it perform the process which excludes a related keyword from the application object at the time of the extraction of the information by the said extraction means .

When the number of pieces of information extracted using the sub-related keyword is less than the third predetermined number , it can be considered that the relationship between the sub-related keyword and the event is low. In this case, the information related to the event can be provided in an appropriate range by excluding the sub-related keyword from the application target at the time of information extraction .

The extraction end control unit, when extracting the number of information including at least one of the main relevant keywords and the sub related keywords are extracted for each of the predetermined timing represents a predetermined downward trend, the secondary related keywords You may make it exclude from the application object at the time of extraction of information.

Since sub-related keywords are added as needed, it is assumed that the number of information items including at least one of the main related keywords and the sub-related keywords will gradually increase. On the other hand, in spite of this assumption, when the number of information including at least one of the main related keyword and the sub related keyword shows a predetermined decreasing tendency, the relevance between the sub related keyword and the event becomes low. Can be considered. In this case, information related to the event can be provided in an appropriate range by excluding the sub-related keyword from the application target at the time of information extraction .

The extraction end control means, when the number of extracted information including the sub-related keyword extracted at each predetermined timing shows a predetermined decreasing tendency, the sub-related keyword is applied from the application target at the time of information extraction You may make it exclude.

When the number of pieces of information extracted using the secondary related keyword shows a predetermined decreasing tendency, it can be considered that the relationship between the secondary related keyword and the event is low. In this case, the information related to the event can be provided in an appropriate range by excluding the sub-related keyword from the application target at the time of information extraction .

Second storage means for storing an unrelated keyword that is considered to be less relevant to the predetermined event is provided, and the extraction means includes the main related keyword and the sub-relation stored in the first storage means. The information including at least one of keywords and not including any of the unrelated keywords stored in the second storage unit may be extracted.

Information that includes non-related keywords that are less relevant to the event is excluded from the extraction target even if it contains the main related keywords and the sub-related keywords , so that the extracted information is more relevant to the event. It can be limited to information that is considered high.

The predetermined event is an event in which a region where the event occurs is specified, and the information to be extracted includes position information indicating a position related to the information, and the extraction unit is stored in the first storage unit The information including at least one of the main related keyword and the sub related keyword , and the position indicated by the position information included in an area where the predetermined event occurs may be extracted.

  Considering the area where the event occurs and the position related to the information, the information indicated by the position information is extracted from the area where the event occurs, so that the extracted information It can be limited to information that is considered highly relevant to the event.

  Notification receiving means for receiving event occurrence information indicating that the predetermined event has occurred, and the extraction means starts the extraction and storage processing when the event occurrence information is received by the notification reception means You may make it do.

  Extraction and storage processing is performed only when it is certain that an event has occurred by starting extraction and storage processing only when the occurrence of a predetermined event is confirmed, and the event is related to the user. Highly contributed post information can be provided.

Third storage means for storing an event occurrence estimation keyword that is considered highly relevant to the predetermined event is provided, and the extraction means is stored in the third storage means among the information to be extracted. Alternatively , the information including the event occurrence estimation keyword may be extracted, and the extraction and storage process may be started when the number of pieces of the extracted information is equal to or greater than a fourth predetermined number .

When the number of extracted information including an event occurrence estimation keyword that is preliminarily stored and considered to be highly relevant to a predetermined event is greater than or equal to a fourth predetermined number , the occurrence of the predetermined event is estimated. The Therefore, although the occurrence of an event is not certain, extraction and storage processing is started when the occurrence is estimated, so that it is possible to quickly provide information highly related to the event to the user. Become.

In order to achieve the above object, an information providing method according to a second aspect of the present invention is an information providing method in an information providing apparatus that provides information that occurs in association with a predetermined event whose situation changes, a first storing step of storing the main relevant keywords and sub related keywords that are considered highly relevant to the predetermined event in the storage means, at every predetermined timing, among the information generated within a predetermined time period, wherein Information including at least one of the main related keyword and the sub related keyword stored in the storage means is extracted, and among the extracted information, keywords included in the first predetermined number of information or more are used as sub related keywords. an extraction step of repeating extraction and storage processing is stored in the storage means, of the information generated within the predetermined period, the main related keywords及When the number of extracted information including at least one of the sub-related keywords satisfies a condition that is less than a second predetermined number, the sub-related keywords are excluded from the application target at the time of extracting information in the extraction step. And an extraction end control step for performing processing .

In order to achieve the above object, an information providing program according to a third aspect of the present invention stores a main related keyword and a sub related keyword that are considered to be highly related to a predetermined event whose situation changes. Storage means for extracting information including at least one of the main related keyword and the sub related keyword stored in the storage means from information generated within a predetermined period for each predetermined timing, and extracting the information Of the information, an extraction unit that repeats extraction and storage processing for storing in the storage unit keywords included in information of a first predetermined number or more as sub-related keywords, and among the information generated within the predetermined period, the main association The condition that the number of extracted information including at least one of the keyword and the sub-related keyword is less than the second predetermined number is satisfied. The case, characterized in that to function the sub related keyword as an extracting end control means, for performing process of excluding from application of the extraction of information by the extracting means.

  According to the present invention, information related to an event can be provided in an appropriate range.

It is a figure which shows the structure of an information provision system. It is a figure which shows the structure of an information provision server. It is a figure which shows an example of event information. It is a figure which shows an example of crawler information. It is a figure which shows an example of main relevant keyword information. It is a figure which shows an example of sub related keyword information. It is a figure which shows an example of NG keyword information. It is a figure which shows an example of contribution information. It is a figure which shows an example of extraction result information. It is a figure which shows an example of candidate related keyword information. It is a 1st flowchart which shows operation | movement of an information provision server. It is a 2nd flowchart which shows operation | movement of an information provision server. It is a 3rd flowchart which shows operation | movement of an information provision server. It is a figure which shows an example of a contribution information screen. It is a 4th flowchart which shows operation | movement of an information provision server. It is a 5th flowchart which shows operation | movement of an information provision server.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In addition, this invention is not limited by the following embodiment and drawing. It goes without saying that the following embodiments and drawings can be modified without changing the gist of the present invention. Moreover, the same code | symbol is attached | subjected to the same or an equivalent part in a figure.

(1) First Embodiment (1-1) Configuration of Information Providing System FIG. 1 is a diagram showing a configuration of an information providing system 1 according to the first embodiment of the present invention. The information providing system 1 includes an information providing server 10, terminal devices 100-1, 100-2, and 100-3 (hereinafter, the terminal devices 100-1, 100-2, and 100-3 are collectively referred to as “terminal device 100” as appropriate. Radio base stations 200-1 and 200-2 (hereinafter, the radio base stations 200-1 and 200-2 are collectively referred to as “radio base stations 200” as appropriate), the communication network 250, and the event notification server 300. The posting server 400 is provided. The information providing server 10 corresponds to an information providing apparatus.

  The terminal devices 100-1 to 100-3 are a mobile phone such as a so-called smartphone, a personal computer having a communication function, or the like. The communication network 250 is, for example, the Internet. The information providing server 10 can communicate with the terminal device 100 via the communication network 250 and the radio base station 200. Further, the information providing server 10 can communicate with the event notification server 300 and the posting server 400 via the communication network 250. The terminal device 100 can communicate with the posting server 400 via the radio base station 200 and the communication network 250.

  In the information providing system 1, the terminal device 100 transmits post information, which is information generated by an operation by the user of the terminal device 100, to the posting server 400. Post server 400 receives and stores post information from terminal device 100. Also, the posting server 400 transmits the stored posting information to the terminal device 100 in response to access from the terminal device 100.

  The event notification server 300 transmits information indicating the occurrence of the disaster (event occurrence information) to the information providing server 10 when a disaster such as an earthquake occurs as an event. At this time, the event notification server 300 may perform unicast communication in which the transmission destination is designated as the information providing server 10 or may perform multicast communication to be transmitted simultaneously to a plurality of devices.

  Upon receiving event occurrence information indicating the occurrence of a disaster from the event notification server 300, the information providing server 10 is transmitted from the terminal devices 100-1 to 100-3 to the posting server 400 and stored in the posting server 400. The post information related to the disaster is extracted and transmitted to the terminal devices 100-1 to 100-3.

(1-2) Configuration of Information Providing Server FIG. 2 is a diagram showing a configuration of the information providing server 10 in the first embodiment. The information providing server 10 includes a control unit 20, a storage unit 40, a communication unit 60, and an operation unit 70. The control unit 20 is configured by, for example, a CPU (Central Processing Unit). The control unit 20 reads and executes the program stored in the storage unit 40 and controls the entire information providing server 10 by processing various data stored in the storage unit 40.

  The storage unit 40 is, for example, a RAM (Random Access Memory). The storage unit 40 stores programs and various data necessary for the operation of the information providing server 10. The storage unit 40 includes an event information database (DB) 42, a crawler information DB 44, an extraction result DB 46, a main related keyword DB 48, a secondary related keyword DB 50, an NG keyword DB 52, and a candidate related keyword DB 54. The main related keyword DB 48 and the sub related keyword DB 50 correspond to the first storage unit, and the NG keyword DB 54 corresponds to the second storage unit.

  The communication unit 60 is, for example, a LAN (Local Area Network) card. The communication unit 60 controls communication among the terminal device 100, the event notification server 300, and the posting server 400. For communication between the communication unit 60 and the terminal device 100, the event notification server 300, and the posting server 400, packet communication using TCP / IP (Transmission Control Protocol / Internet Protocol) or the like is employed. In addition, a secure communication path such as SSL (Secure Socket Layer) is used for communication between the communication unit 60 and the terminal device 100, the event notification server 300, and the posting server 400 as necessary. The operation unit 70 is, for example, a keyboard, and outputs an operation signal corresponding to the operation of the operator to the control unit 20.

  The control unit 20 implements the functions of an event registration unit 22, a crawler information registration unit 24, a secondary related keyword generation unit 26, a post information extraction unit 28, a post information provision unit 30, an extraction end control unit 32, and a keyword editing unit 34. . The sub-related keyword generation unit 26 and the posted information extraction unit 28 correspond to an extraction unit, and the extraction end control unit 32 corresponds to an extraction end control unit.

  The communication unit 60 receives event occurrence information from the event notification server 300 and outputs it to the control unit 20. The event occurrence information is, for example, an event date and time indicating the date and time of occurrence of a disaster as an event, a set of longitude and latitude of the outer edge of the disaster occurrence area, identification information (for example, JIS code) of the disaster occurrence area, and the like It includes event information character strings that are character strings indicating disaster contents such as location information, disaster types such as “earthquake”, “tsunami”, “seismic intensity”, and names of disaster-occurring areas. Note that the event position information may include not only information on the area where the disaster occurred, but also information on areas around the area where the disaster occurred and areas related to the area where the disaster occurred.

  When event occurrence information is input, the event registration unit 22 in the control unit 20 sets a crawler ID that is event identification information. For example, the event registration unit 22 acquires the current date and time from a clock (not shown) built in the information providing server 10 and generates a crawler ID including the acquired date and time. Next, the event registration unit 22 generates event information by adding the set crawler ID to the input event occurrence information. The event registration unit 22 registers the generated event information in the event information DB 42 in the storage unit 40. FIG. 3 is a diagram illustrating an example of event information. The event information shown in FIG. 3 is configured by associating a crawler ID with an event date / time, event position information, and an event information character string in the event occurrence information.

  The crawler information registration unit 24 in the control unit 20 uses the event information to generate a crawler information that corresponds to the event disaster in a one-to-one manner, in other words, the event information and the one-to-one information. To do. The crawler information registration unit 24 registers the generated crawler information in the crawler information DB 44 in the storage unit 40.

  FIG. 4 is a diagram illustrating an example of the crawler information. The crawler information shown in FIG. 4 includes a crawler ID, generation date / time, event type, event position information, event information character string, execution flag, execution flag update date / time, end flag, end flag update date / time, mismatch flag, and mismatch flag update date / time. Consists of including.

  As the crawler ID, a crawler ID in the event information is set. The generation date / time is the date / time when the crawler information is generated. For example, the crawler information registration unit 24 acquires the current date and time from a clock built in the information providing server 10 and sets the date and time of generation. The event type is information indicating the type of disaster that is an event, and is different information for each type of disaster such as an earthquake, tsunami, or typhoon. For example, the crawler information registration unit 24 determines a disaster type based on an event information character string or the like in the event information, and sets information corresponding to the type as an event type. The event position information and event information character string are event position information and event information character string in the event information.

  The execution flag is information indicating whether or not the crawler information has been used when the post information extraction unit 28, which will be described later, extracts post information, and “0” indicating that it is not used in the initial state is used. After that, it becomes “1”. The execution flag update date and time is the date and time when the execution flag is updated. The end flag is information indicating whether or not the post information extraction using the crawler information by the post information extraction unit 28 is completed. In the initial state, the end flag is “0” indicating that it has not ended. “1”. The end flag update date and time is the date and time when the end flag is updated. The discrepancy flag indicates whether the number of items to be extracted is zero when the post information extraction unit 28 extracts post information using the crawler information, or whether the latest post information has been extracted before a predetermined period in the past. This is information indicating whether or not the number of extracted items is 0, or “1” if the latest extracted post information occurs before a predetermined period in the past, and “0” otherwise. . The mismatch flag update date and time is the date and time when the mismatch flag is updated.

  Main related keyword information is registered in the main related keyword DB 48 in the storage unit 40 in advance. FIG. 5 is a diagram illustrating an example of main related keyword information. The main related keyword information shown in FIG. 5 includes an event type, position information, and main related keywords. The event type is information indicating the type of disaster that is an event for which the main related keyword should be used when extracting posted information. The location information is information indicating a disaster occurrence area that is an event for which the main related keyword should be used when extracting posted information, and is a set of longitude and latitude of the outer edge of the disaster occurrence area, and identification information of the disaster occurrence area (For example, JIS code). The main related keyword is a character string that is considered to be highly related to the event corresponding to the event type, and corresponds to the first related keyword. In the post information extraction, post information including a main related keyword is extracted.

  The sub-related keyword generation unit 26 in the control unit 20 generates sub-related keyword information based on the event information prior to the extraction of the posted information, and registers it in the sub-related keyword DB 50 in the storage unit 40.

  FIG. 6 is a diagram illustrating an example of the sub-related keyword information. The sub-related keyword information shown in FIG. 6 includes a crawler ID and a sub-related keyword. As the crawler ID, a crawler ID in the event information is set. The sub-related keyword is a character string that is considered highly related to the event corresponding to the crawler ID, and corresponds to the first related keyword. In the post information extraction, post information including sub-related keywords is extracted. In the initial state, the number of extracted cases is zero.

  In the NG keyword DB 52 in the storage unit 40, NG keyword information is registered in advance. FIG. 7 is a diagram illustrating an example of the NG keyword information. The NG keyword information shown in FIG. 7 includes an event type and an NG keyword. The event type is information indicating the type of disaster that is an event for which the NG keyword should be used when extracting posted information. The NG keyword is a character string that is considered to have a low association with the event corresponding to the event type, and corresponds to an unrelated keyword. In the extraction of the posted information, the posted information including the NG keyword is not extracted even if the main related keyword and the sub related keyword are included.

  The post information extracting unit 28 in the control unit 20 extracts post information for each crawler information after the event information, the crawler information, and the sub-related keyword information are generated and registered. Specifically, the posting information extraction unit 28 includes at least one of the main related keyword and the sub related keyword among the posting information transmitted by the terminal device 100 and stored in the posting server 400, and the NG keyword. Extract post information not included.

  FIG. 8 is a diagram illustrating an example of post information. The posting information shown in FIG. 8 includes posting date / time, posting position information, and a posting information character string. The posting date and time is the date and time when the terminal device 100 transmits the posting information or the date and time when the posting server 400 receives the posting information. The posting position information is information indicating the position of the terminal device 100 that has transmitted the posting information, and includes longitude and latitude. The post information character string is a character string input to the terminal device 100 by an operation of the user of the terminal device 100. The posted information may include content data other than the posted information character string such as the account name of the poster and the data of the posted image.

  Further, the posting information extraction unit 28 counts the number of posting information including the main related keyword and the sub related keyword for each main related keyword and the sub related keyword, and uses the count value as the main related keyword and the sub related keyword. The data are stored in the storage unit 40 in association with each other. Further, the posting information extraction unit 28 generates extraction result information including event information and posting information extracted corresponding to the crawler information including the crawler ID in the event information, and the extraction result DB 46 in the storage unit 40. Register with. FIG. 9 is a diagram illustrating an example of the extraction result information. The posting information providing unit 30 in the control unit 20 transmits the extraction result information in the extraction result DB 46 to the terminal device 100 via the communication unit 60, the communication network 250, and the radio base station 200.

  Further, the sub-related keyword generation unit 26 in the control unit 20 analyzes the extracted post information (for example, morpheme analysis), and specifies words included in a predetermined number or more of post information as new sub-related keywords. . Further, the secondary related keyword generation unit 26 generates new secondary related keyword information including the identified new secondary related keyword and the crawler ID in the corresponding crawler information, and the secondary related keyword DB 60 in the storage unit 40. Register with.

  Thereafter, the post information extraction unit 28 described above extracts the post information, the post information provision unit 30 transmits the post information, the sub related keyword generation unit 26 generates and registers new sub related keyword information (hereinafter referred to as “post” as appropriate). This is referred to as “information extraction / keyword storage processing”).

  When the post information extraction / keyword storage process is repeated, the extraction end control unit 32 in the control unit 20 performs a case where the post information extraction result satisfies a predetermined condition for finishing the post information extraction. The post information extraction unit 28 performs control to end the extraction of the post information. Details of post information extraction / keyword storage processing and post information extraction end control will be described later.

  The keyword editing unit 34 in the control unit 20 analyzes the extracted post information (for example, morphological analysis), and specifies words included in a predetermined number or more of post information as candidate related keywords. Further, the keyword editing unit 34 generates candidate related keyword information including the identified candidate related keyword and registers it in the candidate related keyword DB 54 in the storage unit 40.

  FIG. 10 is a diagram illustrating an example of candidate-related keyword information. The candidate related keyword information shown in FIG. 10 includes an event type, position information, and candidate related keywords. As the event type and position information, the event type and event position information in the crawler information are set. The candidate related keyword is a character string.

  The operator can operate the operation unit 70 to select a candidate related keyword in the candidate related keyword information registered in the candidate related keyword DB 54 and upgrade it to a new main related keyword. In this case, the keyword editing unit 34 mainly selects candidate related keyword information including the selected candidate related keyword from the candidate related keyword information registered in the candidate related keyword DB 54 in accordance with the operation of the operation unit 70 by the operator. It is registered in the main related keyword DB 48 as related keyword information.

    Further, the keyword editing unit 34 adds a new NG keyword to the NG keyword registered in the NG keyword DB 52 and deletes the NG keyword registered in the NG keyword DB 52 according to the operation of the operation unit 70 by the operator. The editing of the NG keyword such as correction of the NG keyword registered in the NG keyword DB 52 is performed.

(1-3) Operation of Information Providing Server Next, the operation of the information providing server 10 in the first embodiment will be described with reference to a flowchart. FIG. 11 to FIG. 13 are flowcharts showing the operation at the time of post information extraction / keyword storage processing by the information providing server 10 in the first embodiment. The operation shown in FIGS. 11 to 13 is repeatedly executed at a predetermined interval such as a 10-minute cycle.

  As illustrated in FIG. 11, the posting information extraction unit 28 in the control unit 20 acquires the start date and time (current date and time) of processing (step S201). For example, the posting information extraction unit 28 can acquire the start date and time from a clock built in the information providing server 10.

  Next, the posting information extraction unit 28 uses the crawler information registered in the crawler information DB 44 with the execution flag “0” or the end flag “0”, in other words, the crawler information. The crawler information for which the extraction of the posted information has not started, or the crawler information for which the extraction of the posted information using the crawler information has not been completed is acquired. Further, the posting information extraction unit 28 generates a list including the acquired crawler information (step S202). The list includes, for example, the crawler ID of the acquired crawler information. The list is stored in the storage unit 40.

  Next, the posting information extraction unit 28 determines whether or not the number of crawler information acquired in step S202 is 0 (step S203). When the number of acquired crawler information is 0 (step S203: YES), a series of operations ends. On the other hand, when the number of acquired crawler information is not 0 (step S203: NO), the posted information extracting unit 28 selects one unselected crawler information from the list (step S204).

  Next, the posted information extraction unit 28 determines whether the mismatch flag in the crawler information selected in step S204 is “1” and the mismatch flag update date is a predetermined time or more before (for example, 6 hours or more before). It is determined whether or not (step S205). When the inconsistency flag in the crawler information is “1” and the inconsistency flag update date and time is more than a predetermined time (step S205: YES), the posted information extraction unit 28 includes the crawler information in the crawler information selected in step S204. The end flag is set to “1”, and the end flag update date / time is set to the current date / time (step S206). Next, the posting information extraction unit 28 deletes the crawler information selected in step S204 from the list (step S207).

  After deletion of the crawler information in step S207, or when it is determined in step S205 that the mismatch flag in the crawler information is not “1” and that the mismatch flag update date / time is not more than a predetermined time ago. (Step S205: NO), the posting information extraction unit 28 determines whether there is unselected crawler information in the list (Step S208). If there is unselected crawler information in the list (step S208: YES), the operations after selection of one unselected crawler information in step S204 are repeated.

  On the other hand, when there is no unselected crawler information in step S204 (step S208: NO), the operation in FIG. 12 is performed. The posted information extraction unit 28 selects crawler information with a predetermined priority from the list (step S251). For example, the posting information extraction unit 28 compares the generation date and time in each crawler information included in the list, and selects the crawler information one by one in the order of generation date and time. Further, the posted information extraction unit 28 may compare the execution flag update date and time in each crawler information included in the list and select the crawler information one by one in the order of the execution flag update date and time. Further, the posted information extraction unit 28 may preferentially select the crawler information included in the list with a predetermined event type, for example, “tsunami” or “evacuation recommendation”.

  Hereinafter, the operation from step S252 to step S307 in FIG. 13 is performed for each crawler information selected in step S251. The posted information extraction unit 28 sets the execution flag in the crawler information selected in step S251 to “1”, and sets the execution flag update date and time to the current date and time (step S252).

  Next, the posting information extraction unit 28 includes the event type in the crawler information selected in step S251 out of the main related keyword information registered in the main related keyword DB 48, and the region indicated by the position information is selected in step S251. The main related keyword information included in the area indicated by the event position information in the crawler information is specified. For example, when the position information in the main related keyword information and the event position information in the crawler information are a set of the longitude and latitude of the outer edge of the disaster occurrence area, the posted information extraction unit 28 is registered in the main related keyword DB 48. Among the main related keyword information, the area including the event type in the crawler information selected in step S251, and the region composed of the outer edge longitude and latitude as the position information is the outer edge longitude and latitude as the event position information in the crawler information. The main related keyword information included in or coincident with the region consisting of Further, when the position information in the main related keyword information and the event position information in the crawler information are identification information (JIS code etc.) of the disaster occurrence area, the posted information extraction unit 28 is registered in the main related keyword DB 48. Among the main related keyword information, the main related keyword information that includes the event type in the crawler information selected in step S251 and whose identification information that is position information matches the identification information that is event position information in the crawler information is specified. Furthermore, the posting information extraction unit 28 acquires a main related keyword in the specified main related keyword information (step S253). One or more main related keywords may be acquired.

  Next, the posting information extraction unit 28 specifies sub-related keyword information including the crawler ID in the crawler information selected in step S251 from the sub-related keyword information registered in the sub-related keyword DB 50, and the specified sub-keyword information. A secondary related keyword in the related keyword information is acquired (step S254). One or more sub-related keywords may be acquired.

  Next, the posting information extraction unit 28 specifies NG keyword information including the event type in the crawler information selected in step S251 from the NG keyword information registered in the NG keyword DB 52, and includes the specified NG keyword information. NG keyword is acquired (step S255). One or a plurality of NG keywords may be acquired.

  Next, the posting information extraction unit 28 accesses the posting server 400 via the communication unit 60 and the communication network 250. The posted information extracting unit 28 occurs in the past first predetermined period (for example, within 24 hours from the current date and time) among the posted information stored in the posted server 400, and the position indicated by the posted position information is It is included in the area indicated by the event position information in the crawler information selected in step S251, includes at least one of the main related keyword acquired in step S253 and the sub related keyword acquired in step S254, and acquired in step S255. Posting information not including the NG keyword is extracted (step S256).

  In the post information extraction described above, for example, when the event position information in the crawler information is a set of the longitude and latitude of the outer edge of the disaster occurrence area, the post information extraction unit 28 indicates that the position indicated by the post position information is the crawler information. Posting information included in an area consisting of a set of longitude and latitude of the outer edge, which is the event position information, is extracted. Further, when the event position information in the crawler information is identification information (JIS code or the like) of the disaster occurrence area, the posted information extraction unit 28 determines the longitude and latitude of the outer edge of the area corresponding to the identification information and the identification information. With reference to a table (not shown) in which the set is associated, the posting information included in the area composed of the set of the longitude and latitude of the outer edge in which the position indicated by the posting position information is associated with the identification information is extracted. To do. Here, the posting information extraction unit 28 may acquire all the posting information stored in the posting server 400 and extract the posting information that matches the above-described extraction condition from the acquired posting information. Alternatively, the posting information extraction unit 28 transmits information on the first predetermined period in the past, information on the area indicated by the event position information, main related keywords, sub related keywords, and NG keywords to the posting server 400 as extraction condition information. May be. In this case, the posting server 400 extracts the posting information that matches the extraction condition from the stored posting information and transmits it to the information providing server 10, and the posting information extraction unit 28 transmits the posted information. Receive information.

  Next, the posting information extraction unit 28 analyzes the posting information character string in the posting information extracted in step S256 (for example, morpheme analysis), and the main related keyword acquired in step S253 and the sub related keyword acquired in step S254. Every time, the number of post information (the number of extractions) including the main related keyword and the sub related keyword in the post information character string is counted. Further, the posting information extraction unit 28 determines the number of posting information including the main related keyword in the posting information character string, the main related keyword information including the main related keyword, and the crawler ID in the crawler information selected in step S251. The data are stored in the storage unit 40 in association with each other. Further, the posted information extraction unit 28 sets the number of posted information including the secondary related keyword in the post information character string, the secondary related keyword information including the secondary related keyword, and the crawler ID in the crawler information selected in step S251. The data are stored in the storage unit 40 in association with each other (step S257).

  Next, the posted information extraction unit 28 has the number of posted information extracted in step S256 as 0, or the latest posted information extracted from the posted information extracted in step S256 before the second predetermined period in the past (for example, It is determined whether or not the alarm occurred before 6 hours (step S258).

  The number of post information extractions in step S256 satisfies 0 or at least one of the post information extracted in step S256 is the latest post information that occurred before the second predetermined period in the past. In the case (step S258: YES), the posted information extraction unit 28 sets the mismatch flag in the crawler information selected in step S251 to “1”, and sets the mismatch flag update date and time to the current date and time (step S259). Thereafter, a series of operations ends.

  On the other hand, when the number of pieces of post information extracted in step S256 is not 0, and the post information extracted in step S256 is not the latest post information generated before the second predetermined period in the past (step S258). : NO), the posting information extraction unit 28 extracts event information including the crawler ID in the crawler information selected in step S251 from the event information registered in the event information DB 42. Further, the posting information extraction unit 28 generates extraction result information by associating the extracted event information with the posting information extracted in step S256 (step S260).

  After step S260, the operation of FIG. 13 is performed. The posted information providing unit 30 transmits the extraction result information to the terminal device 100 via the communication unit 60, the communication network 250, and the radio base station 200 (step S301). For example, the user of the terminal device 100 can limit post information to be acquired from a regional viewpoint. In this case, the storage unit 40 includes, as user information, information on a region desired by the user of the terminal device 100 together with a user ID and the like. When the posting information providing unit 30 transmits the extraction result information based on the information on the area, the user of the terminal device 100 desires the area indicated by the event position information and the posting position information included in the extraction result information. Determine if it is a region. Furthermore, when the area indicated by the event position information or the posting position information included in the extraction result information is an area desired by the user of the terminal device 100, the posting information providing unit 30 extracts the extraction result information toward the terminal device 100. Send.

  Upon receiving the extraction result information, the control unit (not shown) in the terminal device 100 generates an extraction result screen based on the extraction result information and displays it on a display (not shown). FIG. 14 is a diagram illustrating an example of the extraction result screen. FIG. 14 includes images 500, 510, and 520 corresponding to three pieces of post information included in the extraction result information. For example, an image 500 corresponding to one piece of post information includes an account name image 501 formed based on the post information, a post information character string image 502, a post position information image 503, and an image 504 corresponding to content data. , An image 505 of the posting date is included.

  Returning to FIG. 13, the description will be continued. The extraction end control unit 32 in the control unit 20 calculates the total number of post information including the main related keyword and the sub related keyword obtained for each of the main related keyword and the sub related keyword in step S257 of FIG. (Step S302). Next, the extraction end control unit 32 determines whether or not the total number calculated in step S302 is less than the first predetermined number (step S303).

  When the total number calculated in step S302 is not less than the first predetermined number (step S303: NO), the sub-related keyword generation unit 26 in the control unit 20 analyzes the post information extracted in step S256 of FIG. For example, morphological analysis is performed, and words included in a predetermined number or more of post information are specified as new sub-related keywords. Further, the sub-related keyword generation unit 26 generates new sub-related keyword information including the specified new sub-related keyword, and registers it in the sub-related keyword DB 60 in the storage unit 40 (step S304). Thereafter, a series of operations ends.

  On the other hand, when the total number calculated in step S302 is less than the first predetermined number (step S303: YES), the extraction end control unit 32 includes the crawler ID in the crawler information selected in step S251 of FIG. The sub related keyword information is deleted (step S305). Next, the extraction end control unit 32 in the control unit 20 sets the end flag in the crawler information selected in step S251 to “1”, and sets the end flag update date and time to the current date and time (step S306). . Thereafter, a series of operations ends. When the end flag in the crawler information is set to “1”, the execution flag is already “1”, and therefore the crawler in step S202 of FIG. 11 in the subsequent post information extraction / keyword storage processing operation. Information is not set in the list. Accordingly, the post information extraction and keyword storage processing corresponding to the crawler information is not executed.

(1-4) Actions / Effects As described above, in the first embodiment, the information providing server 10 in the information providing system 1 has a main related keyword that is considered highly relevant to an event for each crawler information, and A process of extracting post information that includes at least one of the sub-related keywords and does not include an NG keyword that is considered to be less relevant to the event, and further extracts a new sub-related keyword included in the extracted post information. repeat.

  As described above, since sub-related keywords are added as needed, it is assumed that post information including at least one of the main related keywords and the sub-related keywords and not including the NG keyword gradually increases. On the other hand, in spite of such an assumption, when the total number of extracted items for each of the main related keywords and the sub related keywords is less than the first predetermined number, the event status changes, for example, the event ends. Therefore, it can be considered that the relationship between the main related keyword and the sub related keyword and the event is low. In this case, the information providing server 10 can provide information related to the event within an appropriate range by terminating the posting information extraction and the keyword storage processing. In addition, secondary related keywords are added at any time and used to extract posted information.For example, secondary information such as train delay information caused by earthquakes, river flooding caused by guerrilla heavy rain, etc. Post information related to disasters can be obtained.

  Further, the information providing server 10 excludes the posting information including the NG keyword from the extraction target even if it includes the main related keyword and the sub related keyword. For this reason, the extracted posting information can be limited to posting information that is considered to be more highly related to the event.

  Further, the information providing server 10 extracts post information in which the position indicated by the post position information is included in the area indicated by the event position information based on the event position information in the crawler information and the post position information in the post information. . For this reason, the extracted posting information can be limited to posting information that is considered more highly relevant to the event from the viewpoint of position.

  Further, when the event providing server 10 receives the event occurrence information from the event notification server 300, the information providing server 10 starts the posting information extraction and the keyword storing process. Thereby, only when it is certain that an event has occurred, the extraction of post information and the keyword storage process are performed, and post information highly related to the event can be provided to the user of the terminal device 100.

(2) Second Embodiment In the first embodiment, the extraction end control unit 32 in the control unit 20 is such that the total number of extracted post information including the main related keyword and the sub related keyword is less than the first predetermined number. In some cases, by setting the end flag in the crawler information to “1”, the post information extraction and keyword storage processing corresponding to the crawler information is not executed thereafter.

  On the other hand, in the second embodiment, the extraction end control unit 32 acquires the number of extracted post information including the sub-related keyword for each sub-related keyword corresponding to the crawler information, and the number of extracted cases is the second. If the number is less than the predetermined number, the sub-related keyword information including the sub-related keyword is deleted, and the post information including the sub-related keyword is not extracted thereafter.

  The operation of the information providing server 10 in the second embodiment will be described with reference to the flowchart. In the second embodiment, similarly to the first embodiment, the operation shown in FIG. 11 is performed, and thereafter, the operations shown in FIGS. 15 and 16 are performed. The operations in steps S351 through S356 in FIG. 15 are the same as the operations in steps S251 through S256 in FIG.

  The posting information extraction unit 28 in the control unit 20 analyzes the posting information character string in the posting information extracted in step S356, and the number of pieces of posting information including the sub related keyword for each sub related keyword acquired in step S354. Count. Furthermore, the posting information extraction unit 28 sets the number of pieces of posting information (extraction number) including the sub-related keyword in the post information character string, the sub-related keyword information including the sub-related keyword, and the crawler information selected in step S351. Are associated with the crawler ID and stored in the storage unit 40 (step S357).

  The operation from step S358 to step S401 in FIG. 16 is the same as the operation from step S258 in FIG. 12 to step S301 in FIG.

The extraction end control unit 32 in the control unit 20 selects one unselected sub-related keyword information from the sub-related keyword information acquired in step S354 of FIG. 16 (step S402). Next, the extraction end control unit 32 determines whether or not the number of extraction cases associated in step S357 with the sub-related keyword information selected in step S402 is less than a second predetermined number (step S403).

  When the number of extracted items associated with the sub-related keyword information selected in step S402 is less than the second predetermined number (step S403: YES), the extraction end control unit 32 selects the sub-related item selected in step S402. The keyword information is deleted from the subsidiary related keyword DB 50 (step S404). After the sub-related keyword information is deleted, the sub-related keyword deleted in the post information extraction and keyword storage processing corresponding to the crawler information having the crawler ID included in the deleted sub-related keyword information. Secondary related keywords in the information are not used.

  After sub-related keyword information is deleted in step S404, or after it is determined that the number of extractions corresponding to the sub-related keyword information selected in step S402 is not less than the second predetermined number (step S403: NO), extraction is performed The termination control unit 32 determines whether there is sub-related keyword information that has not been selected in step S402 (step S405). If there is unselected sub-related keyword information in step S402 (step S405: YES), the operation after the selection of one unselected sub-related keyword information in step S402 is repeated.

  On the other hand, if there is no unselected sub-related keyword information in step S402 (step S405: NO), the sub-related keyword generation unit 26 in the control unit 20 analyzes the post information extracted in step S356 of FIG. (For example, morphological analysis) is performed, and words included in the posting information of the first predetermined number or more are specified as new sub-related keywords. Further, the sub-related keyword generation unit 26 generates new sub-related keyword information including the specified new sub-related keyword, and registers it in the sub-related keyword DB 60 in the storage unit 40 (step S406). Thereafter, a series of operations ends.

  When the number of post information extracted using the added sub-related keyword is less than the second predetermined number, the relevance between the sub-related keyword and the event is changed due to a change in the event status, etc. Can be considered low. In this case, in the second embodiment, the information providing server 10 in the information providing system 1 deletes the sub related keyword information including the sub related keyword from the sub related keyword DB 50, and thereafter includes the sub related keyword. By preventing the posting information from being extracted, information related to the event can be provided in an appropriate range.

(3) Third Embodiment In the first embodiment, the extraction end control unit 32 in the control unit 20 has the first total number of extracted post information including the main related keyword and the sub related keyword for each crawler information. When the number is less than the predetermined number, the end flag in the crawler information is set to “1” so that the post information extraction and keyword storage processing corresponding to the crawler information is not executed thereafter.

  On the other hand, in the third embodiment, each time the extraction end control unit 32 extracts post information for each crawler information, the extraction end control unit 32 determines the number of post information extracted including the main related keyword and the sub related keyword corresponding to the crawler information. The total number is stored in the storage unit 40. Further, the extraction end control unit 32 sets the end flag in the crawler information to “1” when the transition of the total number shows a predetermined decreasing tendency. Here, the predetermined decreasing trend is a case where the total number gradually decreases without increasing or decreasing, or a case where the total number continues to be equal to or less than the predetermined number or more.

  As in the first embodiment, since sub-related keywords are added at any time, it is assumed that post information including at least one of the main related keywords and the sub-related keywords and not including the NG keyword gradually increases. Despite this assumption, if the total number of extracted items for each of the main related keywords and sub related keywords shows a certain downward trend, the main related keywords and sub related It can be considered that the relevance between the keyword and the event is low. In this case, in the third embodiment, the information providing server 10 sets the end flag in the crawler information to “1”, and thereafter, the post information extraction and the keyword storage processing corresponding to the crawler information are not executed. Thus, information related to the event can be provided in an appropriate range.

(4) Fourth Embodiment In the second embodiment, the extraction end control unit 32 in the control unit 20 determines that the sub-relevance when the number of posts of post information including the sub-related keyword is less than the second predetermined number. The sub-related keyword information including the keyword is deleted from the sub-related keyword DB 50 so that the post information including the sub-related keyword is not extracted thereafter.

  On the other hand, in the fourth embodiment, the extraction end control unit 32 extracts post information including the sub-related keyword for each sub-related keyword corresponding to the crawler information every time the post information is extracted for each crawler information. The total number of extracted cases is stored in the storage unit 40. Further, the extraction end control unit 32 deletes the secondary related keyword from the secondary related keyword DB 50 when the transition of the total number shows a predetermined decreasing tendency. Here, as in the third embodiment, the predetermined decreasing trend is when the total number gradually decreases without increasing or decreasing, or when the total number continues to be equal to or less than the predetermined number.

  When the number of post information extracted using the added sub-related keyword shows a predetermined decreasing trend, the relevance between the sub-related keyword and the event becomes low due to a change in the event status, etc. Can be considered. In this case, in the fourth embodiment, the information providing server 10 in the information providing system 1 deletes the sub related keyword information including the sub related keyword from the sub related keyword DB 50 and then includes the sub related keyword. By preventing the posting information from being extracted, information related to the event can be provided in an appropriate range.

(5) Fifth Embodiment In the first embodiment, when the information providing server 10 receives event occurrence information from the event notification server 300, the information providing server 10 starts extracting post information and keyword storage processing. On the other hand, in the fifth embodiment, the information providing server 10 starts the extraction of posted information and the keyword storage process when the occurrence of a disaster as an event is estimated even if the event occurrence information is not received. To do.

  Specifically, a keyword for event occurrence estimation is stored in the storage unit 40 in advance. The event occurrence estimation keyword is a character string that is considered highly related to the event, and corresponds to the second related keyword. An event occurrence estimation keyword is prepared for each event type, and an event type is added.

  The event registration unit 22 in the control unit 20 selects keywords for event occurrence estimation one by one. Next, the event registration unit 22 extracts post information including the selected event occurrence estimation keyword in the post information character string from the post information transmitted by the terminal device 100 and stored in the post server 400. Next, the event registration unit 22 counts the number of extracted pieces of posted information, and if the number is equal to or more than a third predetermined number, the event type added to the selected event occurrence estimation keyword is Presume that an event of the indicated type has occurred.

  After the event occurrence is estimated, the event registration unit 22 generates a crawler ID and generates event information. However, unlike the first embodiment, the event registration unit 22 analyzes the extracted post information, and displays the appropriate information obtained from the analysis result as the event date and time, event position information, and event information character string in the event information. Set to. For example, the event registration unit 22 sets the current date and time as the event date and time. Moreover, the event registration part 22 sets the information of the area | region corresponding to the place name contained in the predetermined number or more of posting information among the extracted posting information to event position information. In addition, the event registration unit 22 sets a character string of a keyword for estimating the occurrence of an event, a character string of words included in a post information character string in a predetermined number or more of post information, and the like in the event information character string.

  Thereafter, as in the first embodiment, the crawler information registration unit 24 generates crawler information, the sub-related keyword information generation unit 26 generates and registers sub-related keyword information, and further, as shown in FIGS. Post information extraction / keyword storage processing is performed by the sub-related keyword generation unit 26, the post information extraction unit 28, the post information provision unit 30, and the extraction end control unit 32.

  Thus, in the fifth embodiment, when the number of extracted post information including a keyword for event occurrence estimation that is considered highly relevant to the event is a predetermined number or more, the occurrence of the event is estimated, Posting information extraction / keyword storage processing is performed. Therefore, the event occurrence information is not received because the event occurrence information is not received, but when the occurrence is estimated, the post information extraction / keyword storage processing is performed, and the event is related to the user. It is possible to provide highly reliable information quickly.

(6) Other Embodiments Although the embodiment of the present invention has been described above, the present invention is not limited to the embodiment. The present invention includes a combination of the embodiments and the following modifications as appropriate, and includes equivalents thereof.

  For example, in the above-described embodiment, the post information has been described as including the post position information indicating the position of the terminal device 100 that transmitted the post information. In step S256, the posting information extraction unit 28 performs analysis such as morphological analysis on the posting information character string in the posting information stored in the posting server 400 to identify a region name, and the like. The identification information (JIS code or the like) of the corresponding area may be set in the posting information as posting position information.

  Further, for example, in the above-described embodiment, the sub-related keyword is prepared in advance, and the sub-related keyword is used from the first posting information extraction. However, in the initial state, there is no sub-related keyword, and the sub-related keyword is used in the first time. Posting information may be extracted without using related keywords. Further, the sub-related keyword generation unit 26 may generate a sub-related keyword from the post information obtained by the first post information extraction, and may be used for the second and subsequent post information extraction. .

  Further, for example, in the above-described embodiment, the posting information extraction and the keyword storage processing are terminated or the posting information extraction and the keyword storage using a predetermined sub-related keyword are performed according to the number of pieces of the posting information to be extracted and the decreasing tendency. Finished processing. However, in addition to the number of posted information items to be extracted and a decreasing trend, when the post information extraction result satisfies a predetermined condition for terminating the post information extraction and keyword storage processing, the post information extraction and keyword storage processing is performed. You may make it complete | finish.

  For example, in the above-described embodiment, the case where the predetermined event is a disaster has been described. However, the present invention can be similarly applied to the case where post information related to an event other than a disaster is provided.

  Further, the configuration of the information providing system 1 is not limited to that shown in FIG. For example, the information providing server 10 and the posting server 400 may be configured by a single server. Further, for example, a program for realizing the function of the information providing server 10 may be stored in a storage medium such as a CD-ROM or downloaded via the communication network 250.

DESCRIPTION OF SYMBOLS 1 Information provision system 10 Information provision server 20 Control part 22 Event registration part 24 Crawler information registration part 26 Secondary related keyword production | generation part 28 Posting information extraction part 30 Posting information provision part 32 Extraction completion control part 34 Keyword editing part 40 Storage part 60 Communication Unit 70 Operation unit 100-1, 100-2, 100-3 Terminal device 200-1, 200-2 Wireless base station 250 Communication network 300 Event server 400 Posting server

Claims (10)

An information providing apparatus that provides information that occurs in association with a predetermined event in which the situation changes,
First storage means for storing a main related keyword and a sub related keyword that are considered to be highly related to the predetermined event;
Information that includes at least one of the main related keyword and the sub related keyword stored in the first storage means is extracted from information generated within a predetermined period at each predetermined timing, and the extracted information An extraction unit that repeats the extraction and storage process of storing the keyword included in the information of the first predetermined number or more as a sub-related keyword in the first storage unit,
Of the information generated within the predetermined period, the sub-related keyword is satisfied when the number of extracted information including at least one of the main related keyword and the sub-related keyword is less than a second predetermined number. Extraction termination control means for performing processing to exclude the information from the application target when extracting information by the extraction means,
An information providing apparatus comprising: The extraction end control means, when the condition that the number of extracted information including the sub-related keyword is less than a third predetermined number among the information generated within the predetermined period is satisfied, satisfies the condition. The information providing apparatus according to claim 1, wherein a process of excluding a related keyword from an application target when information is extracted by the extraction unit is performed . The extraction end control unit, when extracting the number of information including at least one of the main relevant keywords and the sub related keywords are extracted for each of the predetermined timing represents a predetermined downward trend, the secondary related keywords The information providing apparatus according to claim 1, wherein the information providing apparatus is excluded from an application target when extracting information. The extraction end control means, when the number of extracted information including the sub-related keyword extracted at each predetermined timing shows a predetermined decreasing tendency, the sub-related keyword is applied from the application target at the time of information extraction The information providing apparatus according to claim 1, wherein the information providing apparatus is excluded. Second storage means for storing unrelated keywords that are considered to be less relevant to the predetermined event;
The extraction means includes at least one of the main related keyword and the sub related keyword stored in the first storage means, and any of the unrelated keywords stored in the second storage means 5. The information providing apparatus according to claim 1, wherein the information not included is extracted. The predetermined event is an event in which a region where the event occurs is specified,
The information to be extracted includes position information indicating a position related to the information,
The extraction unit includes at least one of the main related keyword and the sub related keyword stored in the first storage unit, and the position indicated by the position information is in an area where the predetermined event occurs. The information providing apparatus according to claim 1, wherein the information included is extracted. Notification receiving means for receiving event occurrence information indicating that the predetermined event has occurred,
The information providing apparatus according to claim 1 , wherein the extraction unit starts the extraction and storage process when the event occurrence information is received by the notification reception unit. Third storage means for storing an event occurrence estimation keyword that is considered to be highly related to the predetermined event;
The extraction means extracts information including the event occurrence estimation keyword stored in the third storage means from the information to be extracted, and the number of pieces of the extracted information is a fourth predetermined number or more. The information providing apparatus according to claim 1 , wherein the extraction and storage processing is started. An information providing method in an information providing apparatus for providing information that occurs in association with a predetermined event in which the situation changes,
A first storage step of storing in the storage means a main related keyword and a sub related keyword that are considered to be highly related to the predetermined event;
At every predetermined timing, among the information generated within a predetermined time period, the extract information including at least one of the stored in the storage unit main related keywords and the secondary related keywords, among the extracted information, An extraction step of repeating extraction and storage processing for storing the keywords included in the information of the first predetermined number or more as sub-related keywords in the storage means;
Of the information generated within the predetermined period, the sub-related keyword is satisfied when the number of extracted information including at least one of the main related keyword and the sub-related keyword is less than a second predetermined number. An extraction end control step for performing processing for excluding the information from the application target at the time of extraction of information in the extraction step;
An information providing method comprising: Computer
Storage means for storing a main related keyword and a sub related keyword that are considered highly relevant to a predetermined event whose situation changes;
At each predetermined timing, out of information generated within a predetermined period, information including at least one of the main related keyword and the sub related keyword stored in the storage unit is extracted, and among the extracted information , Extraction means for repeating the extraction and storage processing for storing the keywords included in the information of the first predetermined number or more as sub-related keywords in the storage means;
Of the information generated within the predetermined period, the sub-related keyword is satisfied when the number of extracted information including at least one of the main related keyword and the sub-related keyword is less than a second predetermined number. Extraction end control means for performing processing for excluding information from the application target when extracting information by the extraction means,
An information providing program characterized by functioning as

Images