JP5526257B1 - Post data processing apparatus and post data processing method - Google Patents

Abstract

In investigating the number of tweets for a television program, the tweet data for the television program is appropriately identified from all the acquired tweet data.
When a computer 1 investigates the number of tweets for a predetermined TV program based on the tweet data acquired from the tweet data management server 2, the tweet data satisfying the following (1) is included in the TV program. The number of tweets sent to the TV program using the tweets posted within the broadcast time of the TV program among the data posted in a predetermined time zone including the broadcast time and Tweet data satisfying the following (2) To investigate the.
(1) Data indicating a tweet including a hash tag for identifying the television program.
(2) Data indicating a tweet including a hash tag for identifying a broadcast station of the television program.
[Selection] Figure 7

Description

  The present invention relates to a posted data processing apparatus and a posted data processing method, and more particularly to a posted data processing apparatus and a posted data processing method for processing posted data acquired from a posted data management server that stores posted data.

  In recent years, an increasing number of people have posted comments on TV programs and radio programs through social network services and Twitter (registered trademark, hereinafter the same). Here, by investigating the number of postings for each program, it becomes possible to know public interest and interest in the program. On the other hand, when extracting post data relating to a predetermined program from data indicating a post comment (hereinafter referred to as post data), post data including a keyword related to the program in the comment is searched. However, since the words used in the comments are various, it is not possible to select an appropriate keyword, and as a result, there is a problem that post data for the program cannot be collected sufficiently.

  In response to the above problems, in recent years, a technique for searching for posted data using an appropriate keyword has been developed in collecting posted data for a program (see, for example, Patent Document 1). Patent Document 1 discloses a technique for searching for posted data using a hash tag as a keyword in order to extract posted data for a television program. Here, the hash tag is one of the functions of Twitter and is a tag given to a comment, and is configured by adding a predetermined character string after a code # (hash mark). Post comments tagged with a common hash tag can be viewed together.

  More specifically, for example, by searching a hashtag incorporating a program name or performer name of a TV program as key information, posted data for the TV program among the posted data existing on the network can be easily obtained. Can be extracted and collected. By utilizing such a feature, in Patent Document 1, it is possible to appropriately collect post data for a television program, and to investigate public interest in the television program.

JP2012-150662A

  However, since the hashtag is set arbitrarily by the Twitter user who is the contributor, even if you try to search for posted data using the hashtag as key information, you can identify which program is a tweet from the hashtag May become difficult. For this reason, for example, when trying to investigate the number of postings to a predetermined program, the posting data for the program is not properly specified from all the obtained posting data, and the number of postings cannot be correctly investigated. There is a fear.

  Therefore, the present invention has been made in view of the above-described problems, and the purpose of the present invention is to investigate post data for a predetermined program when investigating the number of postings to the predetermined program based on the acquired post data. It is an object of the present invention to provide a post data processing apparatus and a post data processing method that can be appropriately specified.

According to the posted data processing apparatus of the present invention, the subject is a posted data processing apparatus that processes the posted data acquired from the posted data management server that stores the posted data, and that is acquired from the posted data management server. A storage processing unit that executes processing for storing post data; and a survey processing unit that executes processing for investigating the number of postings to a predetermined program based on the post data stored by the storage processing unit. The survey processing unit includes, in the posted data satisfying at least the following condition (1), the posted data in a predetermined time zone including a broadcast time of the predetermined program, and the following (2) Investigating the number of postings based on the posting data in the posting data satisfying the conditions, the posting time being in a preset time zone for the predetermined program It is solved.
(1) The post data indicating a post comment including program identification information for identifying the predetermined program.
(2) The posting data indicating a posting comment including station identification information for identifying a broadcasting station that broadcasts the predetermined program.

  In the above configuration, post data for a predetermined program is extracted using program identification information for identifying a predetermined program and station identification information for identifying a broadcasting station that broadcasts the predetermined program as key information. To do. At this time, for the posting data indicating the posting comment including the program identification information, the posting data indicating the posting comment including the station identification information is the portion posted in a predetermined time zone including the broadcast time of the program. For the predetermined program, a portion posted in a preset time zone is extracted as post data for the predetermined program. If the number of postings to the predetermined program is investigated based on the posting data extracted by such a procedure, a reasonable investigation result can be obtained. In other words, the device of the present invention can appropriately specify the posting data for the predetermined program from the acquired posting data when investigating the number of postings to the predetermined program based on the posting data. .

In the posted data processing apparatus, the posted data that satisfies at least one of the following exclusion conditions (3) to (7) among the posted data for the predetermined program stored by the storage processing unit: The data processing unit may further include a data exclusion processing unit that executes processing for removing data from the posted data used in processing by the survey processing unit.
(3) The post data indicating the post comment posted under an account that includes a predetermined character string in the character information specifying the account.
(4) The posting data indicating the posting comment posted under an account in which the number of postings during a certain period is equal to or more than a predetermined number.
(5) The post data indicating the post comment including a plurality of at least one of the program identification information and the station identification information.
(6) The post data indicating the post comment including the program identification information and the station identification information indicating a broadcast station different from a broadcast station that broadcasts the program identified by the program identification information.
(7) The post data indicating the post comment posted under an account in which the number of other contributors who have followed the post comment within a predetermined period exceeds a predetermined number.
With the above configuration, posting data that is unsuitable for investigating the number of postings to a program, for example, posting data from an account (so-called bot) that automatically posts at a specific time, clearly not an individual It is possible to exclude post data from an account and post data that is not a comment on a program. As a result, the credibility of the survey result of the number of postings to the program is improved, and it is possible to appropriately capture the interests and interests of the public in the program.
In the present invention, when investigating the number of postings to a program, in addition to post comment data including program identification information for identifying the program, a broadcast station that broadcasts the predetermined program is identified. Since post comment data including station identification information is also extracted, the number of post data extraction is increased. As a result, there is a high probability that the extracted post data contains the data to be excluded, so it is more meaningful to set a condition and exclude post data corresponding to the condition It becomes.

In the posted data processing apparatus, the posted data is stored in the posted data management server in association with the attribute of the poster.
The investigation processing unit may perform a process of examining the number of postings based only on the posting data associated with the specific attribute. In particular, in the post data management server, the post data is stored in association with an area where the poster is located, and the survey processing unit is based only on the post data associated with the specific area. It is preferable to execute processing for checking the number of times.
If it is the above structure, it will become possible to total | post contribution data according to the attribute of a contributor, for example, the area where a contributor exists. As a result, for example, it is possible to deal with a case where programs broadcast by regions differ.

In the posted data processing apparatus, the program identification information and the station identification information may be hash tags. At this time, the hash tag corresponding to the program identification information is the hash designated by the party of the predetermined program or the user of the posted data processing apparatus as the hash tag for identifying the predetermined program A tag is preferred. Further, the hashtag designated by the user for identifying the predetermined program includes an approved hashtag notified from the user to an authorized person of the predetermined program and approved. More preferred.
In the above configuration, since the hash tag is used as key information when extracting post data for a predetermined program, post data for the program can be easily extracted. If a hash tag for identifying a program is specified by a related party of the program or a user of the posted data processing device, it is possible to more easily extract posted data for the program. Further, as a hash tag for program identification designated by the user of the posted data processing device, if a tag notified to the program concerned by the user of the posted data processing device and approved is used, for example, the above hash tag Thus, it is possible to easily determine a hash tag used as key information without having to find the program from the homepage of the program.

Further, the above-described problem is a post data processing method for processing the post data acquired from a post data management server in which a computer stores post data, according to the post data processing method of the present invention, wherein the computer includes: Processing for storing the posting data acquired from the posting data management server and processing for investigating the number of postings to a predetermined program based on the stored posting data, and processing for investigating the number of postings In the above, the computer includes, in the post data satisfying at least the following condition (1), the post data in a predetermined time zone including a broadcast time of the predetermined program, and a condition of the following (2) In the posted data satisfying the above, the posted data having a posted time in a time zone set in advance for the predetermined program is converted into the predetermined program. Based on the posted data against, it is solved by executing the processing to investigate the post number.
(1) The post data indicating a post comment including program identification information for identifying the predetermined program.
(2) The posting data indicating a posting comment including station identification information for identifying a broadcasting station that broadcasts the predetermined program.
With the above processing method, it is possible to appropriately specify post data for the predetermined program from the acquired post data when investigating the number of postings to the predetermined program based on the post data. .

In the posted data processing apparatus and posted data processing method of the present invention, program identification information for identifying a predetermined program and station identification information for identifying a broadcasting station that broadcasts the predetermined program are used as key information. Then, post data for a predetermined program is extracted. At this time, for the posting data indicating the posting comment including the program identification information, the posting data indicating the posting comment including the station identification information is the portion posted in a predetermined time zone including the broadcast time of the program. For the predetermined program, a portion posted in a preset time zone is extracted as post data for the predetermined program. If the number of postings to the predetermined program is investigated based on the posting data extracted by such a procedure, a reasonable investigation result can be obtained.
As a result, the posting data processing apparatus and the posting data processing method can appropriately specify posting data for the predetermined program when investigating the number of postings to the predetermined program based on the posting data. .

It is a figure which shows the whole system containing the contribution data processing apparatus of this invention. It is a figure which shows an example of contribution data. It is a figure which shows the example of a display of the program information of a television program. It is a figure which shows an example of the input screen of a contribution comment. It is a conceptual diagram which shows a series of flows from acquisition of contribution data to investigation of the number of contributions. It is a figure which shows the example of a display regarding an investigation result. It is a figure which shows the structural example of the system containing the contribution data processing apparatus of this invention. It is a figure which shows the hardware constitutions of the contribution data processing apparatus of this invention. It is the figure which represented the structure of the contribution data processing apparatus of this invention from a functional surface. It is a figure which shows the list of hash tags. It is a schematic diagram which shows the posting data preserve | saved at the predetermined preservation | save destination. It is a figure which shows an example of the flow which determines a hash tag. It is a figure which shows an example of the flow which acquires contribution data. It is a figure which shows an example of the flow which investigates the frequency | count of posting. It is a figure which shows the 2nd example of a display regarding an investigation result.

<< Outline of Posted Data Processing Apparatus of the Present Invention >>
An outline of the posted data processing apparatus of the present invention will be described with reference to FIGS.
The posted data processing apparatus of the present invention corresponds to one of the devices constituting a system (hereinafter referred to as a survey system) S for surveying the number of tweets for a program, particularly a television program. Here, the program refers to audio information and video information broadcast from a predetermined broadcasting station according to a preset time schedule, and a program broadcast in a certain broadcast time zone includes only the main part of the program. In addition, advertisements (commercials) broadcast during the same broadcast time period are also included.

  Specifically, as shown in FIG. 1, the survey system S includes a broadcasting station (television station) that broadcasts a TV program, a viewer that tweetes about the TV program, and information that the viewer owns. A data management company that receives tweet data from a terminal and a data research company that investigates the number of tweets for a TV program are involved. Among these, the data research company has a tweet data processing apparatus 1 which is an example of the posted data processing apparatus of the present invention. That is, the data survey company is a user of the posted data processing device, and the tweet data processing device 1 is configured by a computer installed at the head office of the data survey company. Here, the user of the posted data processing device is not only a company that performs a survey of the number of tweets using the tweet data processing device 1, but also, for example, an affiliate company that supports the survey work of the company, a tweet to the company It is a concept that includes the company that commissions the survey of the number of cases and the parent company that determines the business policy of the company.

  Then, the data survey company uses the tweet data processing device 1 to acquire tweet data relating to each TV program from the data management company, and investigates the number of tweets for each TV program based on the acquired tweet data. Report the results to the broadcaster. Here, the report destination of the survey result of the number of tweets about the TV program is not limited to the broadcasting station of the TV program, but other companies that want to provide the survey result, for example, TV program sponsor companies and advertising agencies It may be a program performer or its affiliated office, a publisher that publishes the original TV program, a distributor of a theatrical movie broadcast as a TV program, or a record production company for the music played in the program.

  In the following, the contents and flow regarding the investigation of the number of tweets for a television program will be described. Note that a series of processes related to the investigation of the number of tweets for a television program is repeatedly executed every day, but in order to simplify the explanation, in the following, tweets for a television program broadcast on a certain day (hereinafter referred to as broadcast day) A case of investigating the number of cases will be described as a specific example.

  A tweet is a short sentence of 140 characters or less that can be posted using “Twitter”, an information service, and corresponds to an example of a posted comment. The tweet data is data indicating the tweet and corresponds to an example of post data. The tweet data will be described with reference to FIG. 2. The tweet data includes, in addition to the tweet text, the date and time (tweet date and time) when the tweet was made, and an ID (tweet ID) indicating the account acquired by the contributor to post the tweet. ), Information on the poster (poster information), latitude and longitude of the current location of the poster, and information (application) indicating application software activated to post a tweet.

  And the viewer of a television program can post a tweet through information terminals, such as a smart phone and a tablet. More specifically, the viewer activates an application for posting a tweet on the information terminal when posting a tweet to the television program being viewed. As one of the methods for starting this application, for example, a program information signal distributed from a broadcast station and superimposed on a video signal of a TV program is developed to display program information, and the program information is used to display the program information. The method of starting the application of can be considered.

More specifically, the displayed program information includes two-dimensional barcode information as shown in FIG. 3, and an application for posting a tweet is activated by reading the barcode information with the information terminal. Will come to do. When such an application is activated, a tweet input field as shown in FIG. 4 is displayed on the display of the information terminal. The viewer enters a tweet by inputting a comment (text) in the input field and then pressing a posting button (a button displayed as “tweet” in FIG. 4).
In addition, as a method of starting the application for posting a tweet, other than the method of reading the above-described two-dimensional barcode information, it can be considered. For example, digital watermark information is embedded in the video signal of a television program, and a tweet is posted by executing the above-described processing of expanding the digital watermark information on the information terminal while the poster is watching the television program It is also possible to start an application for use.

  As a result of the posting work by the contributor, data indicating a tweet is transmitted from the viewer's information terminal and received by the data management company. The received data is stored and managed as tweet data on the data management company side. The data transmitted from the viewer's information terminal is the original data of the tweet data, and predetermined additional information (for example, the number of followers and the number of followers) on the original data on the data management company side. The tweet data is configured by adding.

  On the other hand, as shown in FIG. 5, the data research company acquires tweet data for each television program from the data management company side for all television programs broadcast on the broadcast day using the tweet data processing device 1. The TV program broadcast on the broadcast day is a TV program broadcast from 5:00 AM on the broadcast day to 5:00 AM on the next day.

  Here, the data management company manages all the tweet data regardless of the content of the tweet. And the tweet data processing apparatus 1 which concerns on this embodiment acquires the data for the part posted on the broadcast day out of all the tweet data which the data management company manages.

  Moreover, the tweet data processing apparatus 1 investigates the number of tweets to the TV program broadcast on the broadcast day based on the tweet data acquired on the broadcast day. Here, the tweet data acquired by the tweet data processing device 1 on the day of the broadcast is data stored on the data management company side on the same day, and the tweet content is a cobblestone mixture. Therefore, the tweet data processing apparatus 1 extracts the tweet data for the television program from the tweet data acquired on the same day when investigating the number of tweets for the television program broadcast on the broadcast day. At this time, a hash tag is used as key information for extracting tweet data.

  The hash tag will be described. The hash tag is configured by adding a predetermined character string after the code # (hash mark). For example, “#Otaru Omoro ~ Family” in FIG. 4 corresponds to the hash tag. To do. In addition, it is possible to determine that the tweet to which the hash tag is attached relates to the content indicated by the character string after # in the hash tag (in the example of FIG. 4, the television program “Otata Omoro ~ Family”).

Therefore, if tweet data is searched using a hash tag incorporating a word related to a television program as key information, the tweet data for the television program can be extracted. On the other hand, as a hash tag set for extracting tweet data for a television program, for example, a tag for identifying the television program itself (hereinafter referred to as a program tag) and a broadcasting station that broadcasts the television program are identified. (Hereinafter referred to as a station tag). Here, the program tag corresponds to program identification information for identifying a television program, and the station tag corresponds to station identification information for identifying a broadcasting station that broadcasts the television program.
Of the two hash tags, the program tag is determined one day before the broadcast day (the day before the broadcast) as shown in FIG. 5, and the determined program tag is the user (the viewer of the TV program). ).

  Then, on the day of the broadcast, when the viewer posts a tweet with respect to the TV program being viewed, the viewer posts a tweet using the program tag that has been released for the program. On the other hand, the tweet data processing apparatus 1 acquires tweet data posted on the day of broadcasting from the data management company, and extracts the program tag as key information from all the acquired tweet data, so that the program tag The tweet data for the TV program indicated by is specified.

  On the other hand, when a tweet is posted on a TV program that is being viewed, the viewer broadcasts the program together with the program tag or instead of the program tag (strictly speaking, the station name or the equivalent) There is a case where a tweet is posted using a hash tag including a character string indicating (name), that is, a station tag. As with the program tag, this station tag is for a TV program (strictly, a TV program broadcast on the broadcast station indicated by the station tag) from the tweet data acquired by the tweet data processing apparatus 1 on the day of broadcasting. Used as key information when extracting tweet data.

  In the present embodiment, the number of tweets for each TV program broadcast on the broadcast day is checked by using the tweet data processing device 1 at a predetermined time on the next day of the broadcast (the day after the broadcast day), for example, during the daytime. Done.

  Then, the survey results are submitted to the broadcasting station in the form of a survey daily report shown in FIG. The survey daily report shows the program genre, the number of tweets per minute, the average number of tweets in the previous four weeks, the number of tweets at the end, the average number of tweets in the genre, and the average household viewing of the program. Rate and average household audience rating for the previous four weeks.

The number of tweets per minute is the total number of tweets for a television program divided by the number of minutes broadcast for that television program. Here, the total value of the number of tweets for a TV program is posted within a broadcast time and within a certain time before and after the broadcast time among tweets including a hash tag (program tag) for identifying the TV program. It is a value obtained by adding together the number of tweets and the number of tweets posted within the broadcast time among tweets including a hash tag (station tag) for identifying a broadcasting station that broadcasts the television program.
The average number of tweets for the previous four weeks is the sum of the calculation results for the past four weeks before the previous week of the number of tweets per minute, divided by 4. The calculation result of the number of tweets per minute in the past is the number of tweets per minute calculated in the past in the same time zone as the broadcast time of the TV program applied when this week's amount was calculated. It is.
The number of tweets at the end is the number of tweets at the end of broadcasting the television program (strictly speaking, one minute including the broadcast end time).
The genre average tweet count is obtained by adding up the total values of the tweet counts described above for TV programs of the same genre and dividing the total value by the number of TV programs of the same genre.

  Information on the average household audience rating of the program and the average household audience rating for the previous four weeks is available from the audience rating survey company, and the data survey company conducts an audience rating survey to investigate the number of tweets of TV programs broadcast on the broadcast day. Obtain the rating information from the company. Note that the data research company and the audience rating research company may be different companies or the same company.

  The tweet data processing apparatus 1 investigates the above survey items, particularly the items related to the number of tweets, by collecting the tweet data acquired as for the TV program broadcast on the day of the broadcast, and the result is shown in FIG. Report to the broadcasting station in daily report format.

<< Basic structure of survey system >>
Next, the basic configuration of the survey system S including the tweet data processing apparatus 1 will be described with reference to FIG.
The survey system S includes a tweet data processing device 1 managed by a data survey company, a tweet data management server 2 owned by a data management company, an information terminal (hereinafter referred to as a user terminal) 3 used by a viewer to post tweets, a broadcast A terminal (hereinafter referred to as a broadcasting station terminal 4) owned by a station and a parameter notification device 5 owned by a data survey company or its affiliated company are included as main components.

  The tweet data processing apparatus 1 can communicate with a device connected to a communication network such as the Internet. For example, the tweet data processing apparatus 1 communicates with the broadcasting station terminal 4 and shows data indicating the number of tweets for the TV program (survey result data). Is transmitted to the broadcasting station terminal 4.

  The tweet data management server 2 corresponds to a posting data management server, receives tweet data transmitted from the user terminal 3 through a communication network such as the Internet, adds predetermined information to the data, and saves it as tweet data To do. Then, the tweet data processing apparatus 1 acquires the tweet data stored in the server 2 by communicating with the tweet data management server 2 through a communication network such as the Internet.

  More specifically, the tweet data processing apparatus 1 uses an API (Application Programming Interface) provided by the tweet data management server 2 when acquiring tweet data related to a television program. This API is a streaming API, and once connected to the tweet data management server 2, it is an API for automatically continuing to distribute tweet data while maintaining the HTTP connection. By using such a streaming API, it becomes possible to acquire tweet data in real time (approximately the same time as when a tweet is posted).

  Then, the tweet data processing apparatus 1 starts to acquire tweet data by the streaming API by transmitting request data to the tweet data management server 2 at a predetermined time (for example, at 5 AM) on the broadcast day. In the present embodiment, as shown in FIG. 7, a relay device 6 composed of a PC is interposed between the tweet data processing apparatus 1 and the tweet data management server 2. The relay device 6 acquires tweet data by streaming API, temporarily stores the data in the internal memory, and then delivers it to the tweet data processing apparatus 1. That is, in this embodiment, the tweet data processing apparatus 1 acquires tweet data from the tweet data management server 2 via the relay device 6. However, the present invention is not limited to this, and the tweet data processing device 1 may directly acquire tweet data from the tweet data management server 2.

Further, in the present embodiment, among the tweet data acquired from the tweet data management server 2 by the relay device 6 by the streaming API, the data to be delivered to the tweet data processing device 1 is limited to data selected by a predetermined parameter. Here, as a parameter for selecting tweet data, a hash tag (program tag) for identifying a television program broadcast on the broadcast day or a hash tag (station) for identifying a broadcasting station that broadcasts the television program is used. Tag). In the present embodiment, the parameter is notified from the parameter notification device 5 to the relay device 6. That is, the parameter notification device 5 communicates with the relay device 6 through a communication network such as the Internet, and transmits data indicating parameters (parameter notification data). The relay device 6 that has received the parameter notification data analyzes the data to identify the parameter, and among the tweet data received from the tweet data management server 2 on the broadcast day, the tweet data processing device 1 To hand over.
The parameter notification data is distributed by the parameter notification device 5 every day at a predetermined time, and the relay device 6 identifies the above parameters every time the parameter notification data is received, that is, tweet data satisfying the parameters, that is, The tweet data to be delivered to the tweet data processing apparatus 1 is selected.

  In the present embodiment, another tweet data acquisition computer (hereinafter referred to as a spare device) 7 is provided as a backup for the relay device 6. The spare device 7 is installed at a location (for example, a branch office of a data research company) different from the tweet data processing device 1 and the relay device 6, while the tweet data management server 2 uses the streaming API like the relay device 6. Get tweet data in real time.

  Similarly to the relay device 6, the standby device 7 also delivers to the tweet data processing device 1 data satisfying a predetermined parameter among the tweet data acquired from the tweet data management server 2 on the broadcast day. Here, the parameters used by the spare device 7 and the parameters used by the relay device 6 need to be the same. For this reason, the relay device 6 transmits data indicating parameters used by itself, that is, parameter notification data, to the spare device 7.

  In this embodiment, as shown in FIG. 7, an FTP server 8 is interposed between the relay device 6 and the spare device 7, and the spare device 7 receives the parameter notification data transmitted from the relay device 6. , Via the FTP server 8. When the spare device 7 delivers the tweet data satisfying the above parameters to the tweet data processing device 1, the tweet data is temporarily stored in the FTP server 8 and then delivered to the tweet data processing device 1.

In the present embodiment, the tweet data for the past week is acquired from the tweet data management server 2 in case an abnormal situation occurs such that both the relay device 6 and the standby device 7 cannot acquire the tweet data. To do. Such data acquisition is executed by using a search API that is one of APIs provided by the tweet data management server 2. The search API is an API for retrieving tweet data that matches a specified condition from past tweet data stored in the tweet data management server 2 and acquiring the tweet data.
The tweet data acquired by the search API is also temporarily stored in the relay device 6, and only data satisfying a predetermined parameter is delivered to the tweet data processing device 1.
In this way, in this embodiment, when acquiring tweet data posted on the day of broadcasting, all measures are taken against abnormalities and emergencies to prevent forgetting to acquire the above data Is possible.

  In the survey system S configured as described above, acquisition of tweet data and survey of the number of tweets are performed. The basic flow will be briefly described. In obtaining tweet data for the day of broadcasting, first, a hash tag (program tag) for identifying each television program and a hash tag for identifying each broadcast station (Station tag) is set as key information for extracting tweet data. In the case of the present embodiment, among the hash tags, a program tag for identifying a television program is a broadcasting station (strictly speaking, a broadcasting station that broadcasts the television program) as an official hash tag of the television program. ) Is approved.

  On the day of broadcasting, while the tweet data posted on the same day is stored in the tweet data management server 2, the relay device 6 and the spare device 7 acquire the data by the streaming API. Then, the relay device 6 and the standby device 7 deliver to the tweet data processing device 1 data satisfying a predetermined parameter among the tweet data acquired on the broadcast day. Here, the parameter for selecting tweet data includes a program tag and a station tag, and tweet data including at least one of them is delivered to the tweet data processing apparatus 1. In this way, the tweet data processing apparatus 1 acquires data extracted from the tweet data posted on the broadcast day using the program tag and the station tag as key information. Thereby, the tweet data processing apparatus 1 comes to acquire the tweet data with respect to the television program broadcast on the same day from the tweet data posted on the broadcast day.

  Then, after the acquisition of tweet data is completed for all TV programs broadcast on the broadcast day, specifically, the number of tweets for each TV program is investigated on the next day of the broadcast. This survey of the number of tweets is conducted for the purpose of capturing public interest and interest in TV programs. Specifically, in the above survey, the number of tweets is counted for each TV program, and the total is calculated. From the result, a process of calculating the index value shown in FIG. 6 is executed.

  Here, when investigating the number of tweets for each TV program, the tweet data for each TV program is specified, but in this embodiment, the type of hash tag included in the tweet and the posting time of the tweet. , Which TV program the tweet data is for is specified. More specifically, with respect to tweet data including a program tag or both a program tag and a station tag, the tweet data whose posting time is in a predetermined time zone including the broadcast time of the TV program is described above. It is specified as tweet data for the program. On the other hand, for the tweet data including only the station tag, the tweet data for the television program is specified as the posting time in the time zone set in advance for the television program.

  On the other hand, the acquired tweet data includes, for example, those from accounts (so-called bots) that automatically post at a specific time, accounts from accounts that are clearly not individuals, and those that are not tweets about TV programs. Includes tweet data unsuitable for the above investigation. Therefore, in this embodiment, the data unsuitable for investigation as described above is excluded by the tweet data processing apparatus 1 in advance of the investigation. Details of the exclusion of the data will be described later.

  After removing data unsuitable for investigation as described above, the number of tweets is investigated for each TV program broadcast on the broadcast day, and the result of the investigation is finally reported to the broadcasting station.

<< Characteristics and Effectiveness of Post Data Processing Apparatus of Present Invention >>
Next, the characteristic configuration and significance of the tweet data processing apparatus 1 which is the posted data processing apparatus of the present invention will be described.

  The first feature of the tweet data processing apparatus 1 according to the present embodiment is that when extracting tweet data for a TV program broadcast on the same day from tweet data posted on a certain day, the TV program is selected. Not only a hash tag (program tag) for identification but also a hash tag (station tag) for identifying a broadcasting station that broadcasts the television program is used as key information. That is, when investigating the number of tweets for a certain TV program, in addition to tweets containing the program tag for the TV program, the tweets including the station tag for the broadcasting station that broadcasts the TV program are also counted. Accordingly, the number of tweets for the TV program increases. By increasing the number of tweets to be counted in this way, it becomes possible to more accurately grasp the public reaction to the program with respect to the television program.

  The second feature of the tweet data processing apparatus 1 is that the broadcast station approves a program tag for identifying a television program among hash tags used as key information for tweet data extraction. Is a point. With this configuration, post data for a television program can be easily extracted from the tweet data stored in the tweet data management server 2.

  More specifically, the hash tag (strictly speaking, a tag for a program) is usually set arbitrarily by a contributor, and therefore a plurality of hash tags may be set for a certain TV program. Therefore, when extracting the tweet data for a certain TV program, if the hash tag as key information is not set properly, the appropriate tweet data is collected in order to grasp the public interest and interest in the TV program. There is a risk that it will not be obtained.

  On the other hand, in this embodiment, the program tag as a hash tag used when specifying tweet data for a television program is an approved program tag approved by the broadcasting station. Then, among the tweet data posted on the broadcast day, data extracted using the approved program tag as key information is delivered to the tweet data processing apparatus 1. Thus, by using the approved program tag as key information for extracting tweet data, it becomes possible to collect sufficient tweet data for grasping public interest and interest in the television program. More specifically, since the approved program tag is recognized as an official hash tag of the television program, it is used as a unified program tag among viewers of the television program. As a result, if the approved program tag is used as key information for extracting tweet data, a considerable number of tweet data can be extracted for the television program.

  Here, a series of flows until the program tag is determined will be described. First, a data survey company or its affiliated company determines a program tag candidate by the day before the broadcast. Such work is performed, for example, by referring to a TV program guide or a TV program homepage by an employee of a data research company or an affiliated company. After the program tag candidate is determined, the candidate is notified from the data survey company (or affiliated company) to the broadcasting station as shown in FIG. More specifically, a notification mail indicating the determined program tag candidate is transmitted from the parameter notification device 5 to the broadcast station terminal 4. When the broadcast station terminal 4 receives the notification mail, the broadcast station confirms candidate program tags from the notification mail, and considers whether or not to adopt the candidate.

When the above candidate is approved as a program tag, a reply mail indicating that fact is transmitted from the broadcast station terminal 4 to the parameter notification device 5. When the parameter notification device 5 receives the reply mail, the data survey company (or affiliated company) confirms that the program tag candidate proposed by the company is approved by the broadcasting station. At such time, an official program tag, that is, an approved program tag is determined, and the data investigation company registers (stores) the approved program tag in the parameter notification device 5.
The flow from the determination of the program tag candidate described above to the registration of the approved program tag is merely an example. For example, data indicating the program tag candidate is stored in a predetermined site. The program information may be notified to the broadcasting station through the website. After that, the program tag candidates for which the broadcast station has been uploaded may be confirmed on the above-mentioned site, and whether or not each candidate can be adopted may be designated on the same side.

  In this embodiment, the approved program tag is released to the viewer of the TV program, and when posting a tweet for the program, the approved program tag is automatically incorporated in the tweet. It is like that. 3 and 4, the user (viewer) who is viewing a television program (for example, a television program with the program name “Otata Omoro ~ Family”) will receive the program information of the television program. The user terminal 3 reads the two-dimensional barcode shown in FIG. 3 that is displayed and included in the information, thereby starting up an application for posting a tweet. When the application is activated, a tweet input field as shown in FIG. 4 is displayed on the display of the user terminal 3. At this point, an approved program tag (“#Otaku Omoro ~ Family” in FIG. 4) for the TV program being viewed is automatically entered in the input field. As a result, when a user posts a tweet about a TV program that is being viewed, the viewer can save the trouble of inputting the program tag for the TV program and can easily post it.

  As for the automatic input of the approved program tag, for example, when the application is started by reading the two-dimensional barcode included in the program information, the approved program tag is acquired by communicating with the parameter notification device 5. In this way, the program for controlling the user terminal 3 is realized by being incorporated in the application. Alternatively, the approved program tag information is also included in the two-dimensional barcode, and when the application is started by reading the two-dimensional barcode, the approved program tag information is acquired. The information may be delivered to the application.

The third feature of the tweet data processing apparatus 1 is that the tweet data posted on the day of broadcasting is sorted by TV program according to the hash tag and the posting time.
More specifically, in this embodiment, in addition to the approved program tag described above, both hash tags (station tags) for identifying broadcast stations are used as key information for extracting tweet data. That is, the tweet data processing device 1 sends tweet data satisfying at least one of the following conditions (1) and (2) from the tweet data management server 2 (strictly, via the relay device 6 and the standby device 7). Receive).
(1) Tweet data indicating a tweet including an approved program tag.
(2) Tweet data indicating a tweet including a station tag.

  Here, the tweet data indicating the tweet including the approved program tag can be specified as the tweet data for the television program identified by the approved program tag regardless of the posting time. On the other hand, when capturing the reaction of the world during the broadcast of the TV program and immediately before the broadcast, the tweet data for the TV program is transmitted to the tweet data management server 2 in a predetermined time zone including the broadcast time of the TV program. It is preferable to obtain the tweet data. For this reason, the tweet data processing apparatus 1 includes a tweet data indicating a tweet including an approved program tag, a tweet having a posting time within a certain time (for example, 30 minutes) before and after the broadcast time of the television program. The data is specified as tweet data for the television program.

  For tweet data indicating a tweet including a station tag (strictly, a tweet including a station tag but not an approved program tag), the tweet data for which TV program is specified by the posting time be able to. If it explains clearly, the tweet data transmitted toward the tweet data management server 2 within the broadcast time of a certain TV program among the tweet data indicating the tweet including the station tag is the tweet data for the certain TV program. become. For this reason, the tweet data processing device 1 is tweet data that indicates a tweet including a station tag, and tweet data whose posting time is set in a time zone set for a television program, more strictly within the broadcasting time. Certain data is specified as tweet data for the television program. Here, it is preferable that the time zone set for extracting the tweet data for a certain TV program from the tweet data indicating the tweet including the station tag coincides with the broadcast time of the certain TV program. There is no need to strictly match, and for example, a case where it deviates from the broadcast time of a certain TV program by several minutes is included.

  As described above, by sorting the tweet data as the data for the corresponding television program according to the hash tag and the posting time, it becomes possible to appropriately identify the tweet data for which television program. In the present embodiment, as tweet data for a TV program, not only data indicating a tweet including a program tag for the TV program but also data indicating a tweet including a station tag for a broadcasting station that broadcasts the TV program. Since it is included, it becomes possible to acquire more tweet data for the television program.

A fourth feature of the tweet data processing apparatus 1 is that, when the number of tweets is investigated, tweet data unsuitable for the investigation is excluded.
Specifically, as mentioned above, the above survey is conducted for the purpose of capturing public interest and interest in TV programs, so the number of tweets posted by actual viewers is investigated. There is a need. On the other hand, some of the tweet data are, for example, from accounts (so-called bots) that automatically post at a specific time, from accounts that are clearly not individuals, or from tweets that are not TV programs (for example, programs) Tweets for performers), and those intended for directing to specific accounts.

The tweet data of the above kind should be excluded when investigating the number of tweets in order to capture the public interest and interest in the TV program. Therefore, the tweet data processing apparatus 1 has the following (3) to ( The tweet data satisfying at least one of the exclusion conditions in 7) is excluded from the tweet data used for the investigation of the number of tweets.
(3) It is tweet data indicating a tweet posted under an account that includes a predetermined character string in character information (specifically, a display name in FIG. 2) specifying the account.
(4) Tweet data indicating tweets posted under an account in which the number of postings during a certain period is greater than or equal to a predetermined number.
(5) Tweet data indicating a tweet including a plurality of at least one of an approved program tag and a station tag.
(6) Tweet data indicating a tweet including an approved program tag and a station tag indicating a broadcast station different from the broadcast station that broadcasts the television program identified by the approved program tag.
(7) Data indicating a tweet posted under an account in which the number of other contributors who have followed a tweet within a predetermined period, that is, the so-called follower number exceeds a predetermined number.

More specifically, the tweet data processing apparatus 1 first searches the acquired tweet data for data corresponding to any of the following conditions (A) to (C), and excludes the data. .
(A) Tweet data having a plurality of station tags.
(B) Tweet data having a plurality of approved program tags.
(C) The tweet data includes one station tag and one approved program tag, and the broadcast station indicated by the station tag does not match the broadcast station broadcasting the TV program indicated by the approved program tag. .

Furthermore, the tweet data processing apparatus 1 searches for data that meets any of the following conditions (D) to (I) after searching for data that satisfies the above conditions (A) to (C), and the data Is excluded.
(D) Data indicating a tweet posted under an account including “bot” in “display name” in the tweet data.
(E) Data indicating tweets posted under an account posting L or more tweets per day (L is a natural number).
(F) Data indicating tweets posted under an account that appears more than a predetermined time per day and posts M tweets or more per day (M is a natural number).
(G) Data indicating tweets posted under an account posting N or more tweets within one hour (N is a natural number).
(H) Data indicating tweets posted under an account posting P tweets or more within 1 minute (P is a natural number).
(I) Data indicating a tweet posted under an account that posted Q or more tweets to a plurality of TV programs with different broadcasting stations within one minute (Q is a natural number).

  In addition to the above conditions (A) to (I), conditions for excluding data unsuitable for use in investigating the number of tweets may be set. For example, tweet data indicating tweets posted under an account acquired by each broadcasting station for publicity may be excluded. As a specific method of excluding such tweet data, it is conceivable to exclude data indicating tweets posted under an account in which the number of followers within a predetermined period exceeds a predetermined number. It should be noted that the work for identifying accounts having a predetermined number of followers or more is preferably performed on a predetermined day of each month.

As described above, in this embodiment, by removing data unsuitable for the survey from the tweet data used for the survey of the number of tweets, the credibility of the survey result of the number of tweets for each TV program is improved. It is possible to appropriately capture public interest and interest in the program.
In the present embodiment, when investigating the number of tweets for a TV program, in addition to tweet data including a program tag for identifying the TV program, a broadcast station that broadcasts the TV program is identified. Since the tweet data including the station tag is also extracted, the number of extracted tweet data is relatively large. As a result, the probability that the above-described data to be excluded will be included in the extracted tweet data is high, so it becomes more meaningful to set conditions and exclude tweet data corresponding to the conditions.

  A fifth feature of the tweet data processing apparatus 1 is that tweet data used when investigating the number of tweets for a television program can be specified based on the attribute of the poster. With this configuration, in this embodiment, for example, it is possible to cope with a case where television programs broadcast depending on regions are different.

  Hereinafter, an outline of a configuration for acquiring tweet data by designating a region where a contributor is present will be described. When a contributor posts a tweet using the user terminal 3, the data is transmitted to the tweet data management server 2. The tweet data management server 2 analyzes the received tweet data and specifies the region where the poster of the tweet data is present (strictly, the region at the time of posting). More specifically, the tweet data management server 2 reads longitude / latitude information included in the tweet data, and identifies the region where the poster is located from the information. And the tweet data management server 2 preserve | saves tweet data according to area. That is, in this embodiment, the tweet data is stored in the tweet data management server 2 in association with the area where the poster is located.

  On the other hand, as described above, when the tweet data processing device 1 receives the tweet data from the relay device 6 or the standby device 7, it receives only the data satisfying the predetermined parameter. In addition to hash tags (specifically, approved program tags and station tags), this parameter includes an area designated by the data survey company (hereinafter designated area). Here, the designated area corresponds to a specific area (specific attribute).

  That is, the tweet data processing apparatus 1 according to the present embodiment relays tweet data associated with the designated area, that is, tweet data in which the latitude / longitude in the tweet data matches the latitude / longitude of the designated area. Delivered from the device 6 or the spare device 7. And the tweet data processing apparatus 1 investigates the number of tweets to each television program based on the delivered tweet data, that is, the tweet data associated with the designated area.

<< Detailed Configuration of Post Data Processing Apparatus of the Present Invention >>
Next, a detailed configuration of the tweet data processing apparatus 1 which is the posted data processing apparatus of the present invention will be described with reference to FIGS.
As described above, the tweet data processing apparatus 1 is a computer owned by a data research company. As shown in FIG. 8, the CPU 1a, the memory 1b composed of ROM and RAM, the communication interface 1c (the communication I / O in FIG. 8). F), a hard disk 1d, an input device 1e such as a mouse and a keyboard, and an output device 1f such as a display and a printer. The hard disk 1d stores a program for the tweet data processing apparatus 1 to execute various processes to be described later.

  Here, the configuration of the tweet data processing device 1 will be described again from the functional aspect. As shown in FIG. 9, the tweet data processing device 1 includes a tweet data acquisition processing unit 11, a tweet data storage processing unit 12, and a data exclusion processing unit. 13 and a survey processing unit 14.

  The tweet data acquisition processing unit 11 executes processing for acquiring the tweet data posted on the broadcast day from the tweet data management server 2 via the relay device 6 and the standby device 7. The tweet data acquisition processing unit 11 includes a CPU 1a of the tweet data processing device 1, a memory 1b, a communication interface 1c, a hard disk 1d, and a tweet data reception processing program.

  In the present embodiment, the tweet data acquired by the tweet data acquisition processing unit 11 is data satisfying the parameter notified by the parameter notification device 5 among the tweet data posted on the broadcast day.

  More specifically, the relay device 6 receives the parameter notification data transmitted from the parameter notification device 5 by the day before the broadcast and analyzes the data to identify the parameter. Similarly, the standby device 7 receives parameter notification data transmitted from the relay device 6 via the FTP server 8, analyzes the data, and specifies the parameters. Here, the parameters indicated by the parameter notification data include an approved program tag for a TV program broadcast on the broadcast day, a station tag for a broadcasting station that broadcasts the TV program, and the specified area. . More specifically, the table data shown in FIG. 10 showing the correspondence between the TV program broadcast at each broadcasting station and the hash tag from 5 am to 5 am on the next day is transmitted from the parameter notification device 5. Come. The relay device 6 and the standby device 7 receive the above table data, and approve the program tag for each TV program (“# early morning news”, “# good morning TV” in FIG. )) And the station tag of each broadcasting station.

  Then, the relay device 6 and the standby device 7 extract tweet data including the identified approved program tag and station tag from the acquired tweet data for the broadcast day, and the tweet data processing device Deliver to 1. That is, the tweet data acquisition processing unit 11 acquires tweet data including at least one of the approved program tag and the station tag in the tweet data for the day of broadcasting.

  Further, the parameter notification data includes data indicating the designated area such as “Kanto”, “Kansai”, “Kyushu”, etc., and the relay device 6 and the spare device 7 identify the designated area by analyzing the data. To do. And the relay apparatus 6 and the preliminary | backup apparatus 7 extract the data of the tweet posted from the contributor who exists in the specified designated area from the tweet data for the acquired broadcast day, and the tweet data processing device 1 To hand over. That is, the tweet data acquisition processing unit 11 acquires tweet data (in other words, tweet data associated with the designated area) for a television program broadcast on the broadcast day in the area corresponding to the specified designated area.

  The tweet data storage processing unit 12 executes processing for storing the tweet data acquired by the tweet data acquisition processing unit 11 in a predetermined storage area. The tweet data storage processing unit 12 includes the CPU 1a of the tweet data processing device 1, the memory 1b, the hard disk 1d, and a program for data storage processing.

  Here, in this embodiment, among the tweet data stored by the tweet data storage processing unit 12, the posting time of data indicating a tweet including the approved program tag of the TV program broadcast on the broadcast day is described. Data within the broadcast time of the television program and within a certain time before and after the broadcast time are handled as tweet data for the television program. As for tweet data indicating a station tag of a broadcasting station that broadcasts the television program, data whose posting time is within the broadcasting time of the television program is handled as tweet data for the television program. Thus, in this embodiment, the tweet data stored by the tweet data storage processing unit 12 is sorted according to the hash tag and the posting time. As a result, as shown in FIG. 11, the tweet data stored by the tweet data storage processing unit 12 is sorted according to each television program broadcast by each broadcasting station on the day of broadcasting.

  The data exclusion processing unit 13 executes processing for excluding data unsuitable for the investigation processing of the number of tweets by the investigation processing unit 14 from the tweet data stored by the tweet data storage processing unit 12. The data exclusion processing unit 13 includes a CPU 1a, a memory 1b, a communication interface 1c, a hard disk 1d, and a data exclusion processing program of the tweet data processing apparatus 1.

  The data exclusion processing by the data exclusion processing unit 13 will be described. The tweet data stored by the tweet data storage processing unit 12 is read, and each data satisfies at least one of the above-described exclusion conditions (3) to (6). (Specifically, whether or not any of (A) to (I) described above is satisfied) is determined. And the data exclusion process part 13 performs a predetermined | prescribed exclusion process with respect to the tweet data which satisfy | filled the said exclusion condition. Here, regarding the exclusion process, a process of deleting tweet data satisfying the exclusion condition, a process of giving information indicating the fact to the tweet data satisfying the exclusion condition (for example, a process of switching a flag value), etc. What is necessary is just to become suitable content when excluding from the object utilized in the process by the part 14. FIG.

  The survey processing unit 14 aggregates the tweet data stored by the tweet data storage processing unit 12 for each television program, investigates the number of tweets for each television program, and reports the survey results to the broadcasting station. . The investigation processing unit 14 includes the CPU 1a, the memory 1b, the communication interface 1c, the hard disk 1d, and the investigation processing program of the tweet data processing apparatus 1.

  More specifically, the survey processing unit 14 executes a process of investigating the number of tweets for each of all TV programs broadcast on the broadcast day around noon on the next day of the broadcast. In such processing, the survey processing unit 14 reads the tweet data stored by the tweet data storage processing unit 12 and totals the number of tweet data for each television program.

Here, the tweet data used when investigating the number of tweets for a certain TV program is the tag for the approved program of the certain TV program among the tweet data posted on the day when the certain TV program is broadcast. And data indicating a tweet including a station tag of a broadcasting station that broadcasts the TV program. In this embodiment, for data indicating a tweet including an approved program tag for a certain TV program, the data in which the posting time is within a certain time before and after the broadcasting time of the certain TV program is investigated. It is supposed to be used as data for use. In addition, as for data indicating a tweet including a station tag of a broadcasting station that broadcasts a certain TV program, data having a posting time within the broadcasting time of the certain TV program is used as survey data.
Furthermore, among the tweet data narrowed down under the above conditions, the data remaining after the exclusion process by the data exclusion processing unit 13 is finally subjected to the investigation of the number of tweets, and the number of such data is tabulated for each television program. It will be.

  And the investigation process part 14 specifies the number of tweets to each television program from the total result for every television program. Further, the survey processing unit 14 calculates an index value such as the number of tweets per minute and the average number of tweets for the previous four weeks for each television program based on the number of tweets specified. Finally, the survey processing unit 14 creates a daily report shown in FIG. 6 in which the above index values are posted in a predetermined format for each TV program broadcast on the day of the broadcast, and data indicating the daily report (survey result data) ) To the broadcasting station terminal 4.

<< Regarding Posting Data Processing Method of the Present Invention >>
Next, a tweet data processing method as a post data processing method of the present invention will be described with reference to FIGS.
The tweet data processing method according to the present embodiment (hereinafter, this processing method) is applied in the survey process in the survey system S described above. More specifically, this processing method proceeds according to a tweet data acquisition flow and a survey flow during the survey process. In other words, the processing executed in each flow of the tweet data acquisition flow and the survey flow corresponds to a component of this processing method. In this processing method, the above processing is executed by a computer, more specifically, by a computer constituting the tweet data processing apparatus 1 described above. The flow of the investigation process will be described below.

  The survey process starts with a step of determining a hash tag as key information for extracting tweet data, particularly a program tag for specifying a television program. The flow of this process (hash tag determination flow) will be described with reference to FIG. 12. First, the data research company or its affiliated company determines a program tag candidate by the day before the broadcast, and then a notification indicating the candidate. The parameter notification device 5 transmits the mail to the broadcasting station (S001). The broadcast station receives the notification mail (S002), confirms the program tag candidate from the notification mail, and considers whether or not to adopt the candidate.

  When the candidate is approved as an official program tag, a reply mail indicating that is sent from the broadcast station to the parameter notification device 5 (S003). The parameter notification device 5 receives the reply mail (S004), and confirms from the mail that the notified program tag candidate has been approved by the broadcast station. At this time, an official program tag, that is, an approved program tag is determined, and the data investigation company registers the approved program tag in the parameter notification device 5 (S005).

  After the hash tag (strictly, the approved program tag) is determined by the above procedure, the tweet data processing apparatus 1 as a computer performs a step of acquiring tweet data for the television program. The flow of this process (tweet data acquisition flow) will be described with reference to FIG. 13. In this flow, first, parameters including the approved program tag determined in the above-described hash tag determination flow are notified. The process starts when the relay device 6 or the standby device 7 receives the parameters (S011). This process is performed, for example, at a predetermined time on the broadcast day (for example, at 5 AM).

  On the other hand, the tweet data processing device 1 requests the relay device 6 and the standby device 7 to start acquiring tweet data at the same time (that is, at 5 AM) (S012, S013). Receiving this request, the relay device 6 and the spare device 7 transmit the request data to the server 2 in order to acquire the tweet data posted on the broadcast day from the tweet data management server 2 by the streaming API (S014). ). On the tweet data management server 2 side, when request data is received (S015), tweet data for the day of the broadcast is sequentially transmitted in substantially real time (S016), and the relay device 6 and the spare device 7 are transmitted by the streaming API. The incoming tweet data is sequentially received (S017).

  Thereafter, the relay device 6 and the standby device 7 deliver the received tweet data to the tweet data processing device 1, but in the data delivery, the data is extracted using the parameters received in the previous step S011, and the extracted tweet data Is delivered to the tweet data processing apparatus 1 (S018). Here, the above parameters include an approved program tag for each television program broadcast on the broadcast day and a station tag for a broadcasting station that broadcasts each program, and further includes a designated area.

  Then, the relay device 6 and the standby device 7 deliver the tweet data that matches the above parameters to the tweet data processing device 1. The tweet data matching the parameter is data of a tweet including at least one of an approved program tag of each TV program broadcast on the broadcast day and a station tag of a broadcasting station that broadcasts each program, Furthermore, it is data associated with the designated area.

  When the tweet data processing apparatus 1 acquires the tweet data delivered from the relay device 6 or the standby apparatus 7 (S019), the tweet data processing apparatus 1 executes a process of storing the data in the apparatus (S020). Then, a series of processes S016 to S020 from when the tweet data management server 2 transmits the tweet data until the tweet data processing apparatus 1 acquires the tweet data is repeatedly executed until AM 5 of the next day, and the next day AM 5 o'clock. (S021), the acquisition of tweet data for the day of broadcasting is completed.

After the acquisition of the tweet data for the day of the broadcast is completed, the tweet data processing device 1 as a computer performs a step of investigating the number of tweets for the television program. This process is usually performed around noon on the day after the broadcast.
The flow of the process of investigating the number of tweets (survey flow) will be described with reference to FIG. 14. This flow shows that the tweet data for the broadcast day is included in the tweet data stored in the tweet data processing apparatus 1. The process starts from reading (S031). Here, the tweet data to be read out is data indicating a tweet including an approved program tag of a television program broadcast on the broadcast day, and a data indicating a tweet including a station tag of a broadcasting station that broadcasts the television program. is there.

  And the tweet data processing apparatus 1 investigates the tweet number of each television program broadcast on the broadcast day based on the read tweet data. On the other hand, the tweet data read out by the tweet data processing apparatus 1 is classified by program according to a hash tag and a posting time included in the tweet. More specifically, for data indicating a tweet including an approved program tag for a television program, data that has a posting time within the broadcast time of the television program and within a certain time before and after the broadcast time is transferred to the television program. It is sorted as data used when investigating the number of tweets. In addition, for data indicating a tweet including a station tag of a broadcasting station that broadcasts a television program, data whose posting time is within the broadcasting time of the television program is used when investigating the number of tweets for the television program. Sorted as data.

  Further, the tweet data processing apparatus 1 executes a process of excluding data unsuitable for the tweet count investigation process from the tweet data for each television program sorted as described above. More specifically, the tweet data processing apparatus 1 executes a process for determining whether or not the read data satisfies at least one of the above-described exclusion conditions (3) to (6). More specifically, the tweet data processing apparatus 1 searches and excludes data satisfying any of the above-described (A) to (C) from the read tweet data, and further, the tweet data of the remaining tweet data. The data satisfying any of (D) to (I) described above is searched and excluded from the inside. In this way, the tweet data processing apparatus 1 searches for data corresponding to the exclusion condition from the read tweet data (S032). In addition, the tweet data processing apparatus 1 executes a process of excluding tweet data corresponding to the exclusion condition (S033).

  After completion of the exclusion process, the tweet data processing apparatus 1 aggregates the data remaining after the exclusion process among the read tweet data for each television program, and identifies the number of tweets for each television program from the aggregation result. Furthermore, the tweet data processing apparatus 1 calculates index values such as the number of tweets per minute and the average number of tweets for the previous four weeks for each television program based on the number of tweets specified and the like (S034). Finally, the tweet data processing apparatus 1 creates a daily report that summarizes the above index values for each of the TV programs broadcast on the broadcast day, and directs the data (survey result data) indicating the daily report to the broadcast station. Transmit (S035). Then, when the survey result data is received on the broadcasting station side (S036), the survey process is completed.

<< Other Embodiments >>
In the above embodiment, the post data processing apparatus of the present invention and the post data processing method realized by a computer have been mainly described. However, the above embodiment is for facilitating the understanding of the present invention, and does not limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes the equivalents thereof.

  In the above embodiment, the user terminal 3 reads the two-dimensional barcode included in the program information of the television program for the approved program tag approved by the broadcasting station as a hash tag for identifying the television program. It was decided that it would be automatically entered in the tweet entry field. However, the present invention is not limited to this, and the poster may manually input the approved program tag.

In the above embodiment, a data survey company or an affiliated company proposes candidates for hash tags (program tags) for identifying television programs, and a broadcasting station that broadcasts the television programs accepts the candidates. It was decided to consider. When the broadcast station adopts the program tag candidate, the candidate is approved as the official program tag of the television program. However, the program tag may be decided not by the data research company or its affiliated company but by the broadcasting station that broadcasts the television program. When the broadcasting station determines the program tag in this way, the data investigation company finds the program tag from, for example, the homepage of the broadcasting station or the homepage dedicated to the television program. On the other hand, in the above-described embodiment, the program tag can be easily determined as much as it does not require such labor, so that it is more desirable.
When a broadcast station determines a program tag, the program tag naturally corresponds to an approved program tag approved as an official hash tag of the corresponding television program.

  In the above embodiment, program tags (more specifically, approved program tags) are set for all TV programs. However, program tags are not necessarily used for all TV programs. A tag does not need to be set, and a program tag may not be set for some television programs. For example, in a case where a broadcast station determines a program tag, when a new program is broadcast, it is conceivable that the new program is broadcast with the program tag being undetermined. In such a case, as the tweet data for the new program, data having a posting time within the broadcast time of the new program is selected from the tweet data extracted using the station tag of the broadcasting station that broadcasts the new program as key information. do it. In addition, in the case where the broadcast station determines a program tag, the program tag may not be set by the broadcast day as described above, so that a predetermined day of each month (for example, the third week) On the first Friday, it is desirable to perform a monthly process for examining program tags for all TV programs broadcast (or scheduled to be broadcast) in the month.

  Further, in the above embodiment, among the hash tags used as key information at the time of data aggregation, the program tag is an approved program tag, but the present invention is not limited to this. The program tag for data aggregation may be a program tag specified by a person concerned with the TV program or a data research company (including its affiliated company, parent company, or commissioned research company), and not all program tags. May not be an approved program tag. That is, as the program tag for data aggregation, a hash tag designated as an official tag by a person concerned with a TV program, a hash tag designated by a data research company and notified to a person concerned with a TV program, and approved It is also possible to use what has been notified but has not yet been approved by the person concerned with the television program. A TV program related person is a person related to TV program broadcasting. For example, a TV program broadcasting station, a TV program production company, a TV program sponsoring company, or in response to a request from these companies. It is a concept that includes companies that set official program tags.

  In the above embodiment, the daily survey shown in FIG. 6 has been described as an example of the result of the survey of the number of tweets for each television program. In addition, as shown in FIG. 6, the survey daily report should include the program genre, the number of tweets per minute, the average number of tweets for the previous four weeks, the number of tweets at the end, and the average number of tweets for each TV program. It was. However, regarding the survey results, the posted items and display formats can be arbitrarily set. For example, individual survey results of each television program as shown in FIG. 15 may be displayed. Also, as shown in FIG. 15, the broadcast time of each TV program and the transition of the number of tweets (“number of tweets per minute” in the figure) in a certain time (for example, 30 minutes) before and after the broadcast time may be posted. Good. In addition, the total number of tweets for each TV program, the number of tweets per person, the number of contributors (number of unique users), the number of followers, the gender composition and age composition of the poster, the number of retweets, etc. are also formed It is good to do.

  In the above embodiment, the number of tweets for a TV program is checked on the day after the broadcast of the TV program. However, there is no particular restriction on the time of the survey, for example, after the end of the broadcast of the TV broadcast. It may be performed at a time when a predetermined time has passed.

  In the above-described embodiment, a TV program is cited as an example of a program, and the configuration of counting tweet data for the TV program has been described. On the other hand, as described above, a program is audio information or video information broadcast from a predetermined broadcasting station according to a preset time schedule, and a program other than a television program corresponding to such a definition, for example, a radio program The present invention can also be applied to the case of collecting the tweet data for. Alternatively, the present invention can also be applied to the case of collecting tweet data for video and audio streamed and distributed via the Internet during a preset broadcast time zone.

DESCRIPTION OF SYMBOLS 1 Tweet data processing apparatus 1a CPU, 1b Memory, 1c Communication interface 1d Hard disk, 1e Input device, 1f Output device 2 Tweet data management server 3 User terminal, 4 Broadcast station terminal 5 Parameter notification device 6 Relay device, 7 Spare device 8 FTP server 11 Tweet data acquisition processing unit 12 Tweet data storage processing unit 13 Data exclusion processing unit 14 Survey processing unit S Survey system

Claims (8)

A post data processing apparatus for processing the post data acquired from a post data management server for storing post data,
A storage processing unit for executing processing for storing the posted data acquired from the posted data management server;
A survey processing unit that executes a process of investigating the number of postings to a predetermined program based on the posted data stored by the storage processing unit;
The investigation processing unit
In the post data that satisfies at least the following condition (1), the post data is in a predetermined time zone including a broadcast time of the predetermined program;
The posting count is investigated based on the posting data in the posting time set in advance for the predetermined program in the posting data satisfying the following condition (2): Post data processing device.
(1) The post data indicating a post comment including program identification information for identifying the predetermined program.
(2) The posting data indicating a posting comment including station identification information for identifying a broadcasting station that broadcasts the predetermined program. Of the posted data for the predetermined program stored by the storage processing unit, the posted data satisfying at least one of the following exclusion conditions (3) to (7) is used in the processing by the survey processing unit. The posted data processing apparatus according to claim 1, further comprising: a data exclusion processing unit that executes a process of removing the posted data from the posted data.
(3) The post data indicating the post comment posted under an account that includes a predetermined character string in the character information specifying the account.
(4) The posting data indicating the posting comment posted under an account in which the number of postings during a certain period is equal to or more than a predetermined number.
(5) The post data indicating the post comment including a plurality of at least one of the program identification information and the station identification information.
(6) The post data indicating the post comment including the program identification information and the station identification information indicating a broadcast station different from a broadcast station that broadcasts the program identified by the program identification information.
(7) The post data indicating the post comment posted under an account in which the number of other contributors who have followed the post comment within a predetermined period exceeds a predetermined number. In the post data management server, the post data is stored in association with the attribute of the poster,
The posted data processing apparatus according to claim 2, wherein the survey processing unit executes a process of surveying the number of postings based only on the posted data associated with the specific attribute. In the post data management server, the post data is stored in association with the region where the poster is located,
The posted data processing apparatus according to claim 3, wherein the survey processing unit executes a process of surveying the number of postings based only on the posted data associated with the specific region.   The posted data processing apparatus according to any one of claims 1 to 4, wherein the program identification information and the station identification information are hash tags.   The hash tag corresponding to the program identification information is the hash tag designated by a related party of the predetermined program or a user of the posted data processing apparatus as the hash tag for identifying the predetermined program. The posted data processing apparatus according to claim 5.   The hash tag designated by the user for identifying the predetermined program includes an approved hash tag notified from the user to an authorized person of the predetermined program and approved. The posted data processing apparatus according to claim 6. A post data processing method for processing the post data acquired from a post data management server in which a computer stores post data,
The computer is
A process for storing the post data acquired from the post data management server;
A process of investigating the number of postings to a predetermined program based on the stored posting data;
In the process of investigating the number of postings, the computer
In the post data that satisfies at least the following condition (1), the post data is in a predetermined time zone including a broadcast time of the predetermined program;
Executing a process of investigating the number of postings based on the posting data in which the posting time is set in advance for the predetermined program in the posting data satisfying the following condition (2) A post data processing method characterized by the above.
(1) The post data indicating a post comment including program identification information for identifying the predetermined program.
(2) The posting data indicating a posting comment including station identification information for identifying a broadcasting station that broadcasts the predetermined program.

Images