JP2008306300A - Information processing device, method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To extract sentences expressing newsworthy and topical events from contents-related information in EPG data. <P>SOLUTION: In news programs to be broadcasted on a certain day, differences are found between candidate text data extracted from text data in detailed program information contained in EPG data delivered on the air date and those contained in EPG data delivered on the previous day. The text data updated in the EPG data delivered on the air date is determined as text data expressing newsy and topical events. This invention can be applied to digital recording equipment for obtaining EPG data. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an information processing apparatus, an information processing method, and a program, and in particular, an information processing apparatus and information that can accurately extract a keyword that is considered to represent an event with news or topicality The present invention relates to a processing method and a program.

  Conventionally provided ECG (Electric Contents Guide) data, and in particular EPG (Electric Program Guide) data provided as information about broadcast contents, are mainly television receivers as data for displaying program guides. And AV (Audio Visual) equipment such as digital recording equipment.

  The user can display a program guide by operating a remote controller or the like, and by viewing the program guide, a user can check a program that is being viewed or a program that is scheduled to be broadcast.

  EPG data is also used to extract predetermined keywords such as popular words. The detailed program information included in the EPG data includes text data representing the contents of each program, and keywords can be extracted from the text data. The extracted keyword is used, for example, to introduce a program in which the same keyword is included in the detailed program information to the user.

Japanese Patent Application Laid-Open No. 2004-151867 discloses a technique that allows a user to easily know information related to a specific topic by extracting keywords from a plurality of character string data and selecting information based on the extracted keywords. It is disclosed.
JP-A-8-166961

  When keywords are extracted from the entire text data included in the program details information, the keywords obtained as a result of extraction may include many words that are not popular words. Depending on the contents of the data, the accuracy of extracting the target keyword is lowered.

  The present invention has been made in view of such a situation, and makes it possible to accurately extract a keyword that is considered to represent an event with news or topicality.

  An information processing apparatus according to an aspect of the present invention obtains EPG data including information related to each content to be provided when a predetermined time comes, and obtains the first timing by the obtaining unit The text data representing the content of the content included in the EPG data thus obtained and the text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing are the same as the content of the content. The first extracting means for extracting the text data updated in the EPG data acquired at the second timing and comparing each text data to be represented, and the text data extracted by the first extracting means Second extraction means for extracting keywords.

  The second extraction unit causes the morpheme analysis to be performed on the text data extracted by the first extraction unit, and the keyword is extracted from the morpheme of a predetermined part of speech among the morphemes obtained as the analysis result. be able to.

  The first extraction means includes text data representing the content of at least one of a news program and a news program included in the EPG data acquired at the first timing, and at the second timing. The text data representing the content of at least one of the news program and the news program included in the acquired EPG data can be compared for each text data representing the content of the same program.

  The first extraction means starts providing content that represents the content of the same content as the text data representing the content of the content included in the EPG data provided on the day before the content is provided. The text data included in the EPG data provided on the day of the day to be provided can be compared, and the text data updated in the EPG data provided on the day of the day on which the provision of the content is started can be extracted.

  The first extraction means further starts providing the content as representing the same content as the text data included in the first EPG data provided on the day of the start of the content provision. Text data included in the second EPG data provided at a time later than the time when the first EPG data on the day of the selected day is provided, and the text data updated in the second EPG data Can be extracted.

  When the EPG data is provided by adding a predetermined symbol to the updated text data among the text data representing the contents of the content, the first extraction means obtains the first timing. The text data extracted based on the symbol from EPG data and the text data extracted based on the symbol from the EPG data acquired at the second timing are compared for each text data representing the content of the same content. Can do.

  An information processing method or program according to one aspect of the present invention acquires EPG data including information on each content to be provided when a predetermined time comes, and is included in the EPG data acquired at the first timing The text data representing the content of the content and the text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing are compared for each text data representing the content of the same content. And extracting the text data updated in the EPG data acquired at the second timing, and extracting a keyword for the extracted text data.

  In one aspect of the present invention, EPG data including information related to each content to be provided when a predetermined time comes is obtained. Further, text data representing the content of the content included in the EPG data acquired at the first timing, and text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing. Are compared for each piece of text data representing the content of the same content, and the text data updated in the EPG data acquired at the second timing is extracted. Further, keywords are extracted from the extracted text data.

  According to one aspect of the present invention, it is possible to accurately extract a keyword that is considered to represent an event with news or topicality.

  Embodiments of the present invention will be described below. Correspondences between the constituent elements of the present invention and the embodiments described in the specification or the drawings are exemplified as follows. This description is intended to confirm that the embodiments supporting the present invention are described in the specification or the drawings. Therefore, even if there is an embodiment which is described in the specification or the drawings but is not described here as an embodiment corresponding to the constituent elements of the present invention, that is not the case. It does not mean that the form does not correspond to the constituent requirements. On the contrary, even if an embodiment is described herein as corresponding to the invention, this does not mean that the embodiment does not correspond to other than the configuration requirements. .

  The information processing apparatus according to one aspect of the present invention (for example, the information processing apparatus 1 in FIG. 1) acquires an EPG data including information regarding each content that is to be provided when a predetermined time comes ( For example, the EPG data acquisition unit 31) in FIG. 7, the text data representing the content of the content included in the EPG data acquired at the first timing by the acquisition unit, and the second timing later than the first timing. The text data representing the content of the content included in the EPG data acquired in step 1 is compared for each text data representing the content of the same content, and the text data updated in the EPG data acquired at the second timing is extracted. First extracting means (for example, current affairs sentence determining unit 36 in FIG. 7) and text data extracted by the first extracting means. Second extracting means (e.g., topical sentence processing unit 38 of FIG. 7) for performing keywords extracted as target and a.

  An information processing method or program according to one aspect of the present invention acquires EPG data including information on each content to be provided when a predetermined time comes, and is included in the EPG data acquired at the first timing The text data representing the content of the content and the text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing are compared for each text data representing the content of the same content. , Extracting the text data updated in the EPG data acquired at the second timing, and extracting keywords for the extracted text data (for example, step S44 in FIG. 12).

  FIG. 1 is a diagram illustrating an example of an information processing apparatus 1 according to an embodiment of the present invention.

  The information processing apparatus 1 is a digital recording device with a built-in hard disk, and acquires data of programs provided by BS (Broadcasting Satellite) / CS (Communications Satellite) digital broadcasting, terrestrial digital broadcasting, or broadcasting performed via the Internet. Then, the acquired program data is recorded on the hard disk. That is, the information processing apparatus 1 is supplied with a signal from an antenna (not shown). The information processing apparatus 1 reproduces the data recorded on the hard disk in accordance with an instruction from the user, and causes the television receiver 2 to output program video and audio.

  The information processing apparatus 1 also acquires EPG data distributed by broadcast stations and the like via broadcast waves and the Internet. The EPG data includes information on the program such as the program title, broadcast date and time, a code representing the program genre, and detailed program information for each program. The program detailed information includes various text data such as the name of the performer, the name of the corner, and text data representing the contents of the program.

  EPG data including such information is distributed a plurality of times a day at a predetermined time in a form in which information on programs scheduled to be broadcast for 8 days including the distribution date is collected. For example, EPG data distributed on March 1 includes information on programs scheduled to be broadcast from March 1 to March 8, and EPG data distributed on March 2 includes 3 Information related to programs scheduled to be broadcast from March 2 to March 9 is included. Therefore, for a program broadcast on a certain day, information about the program is provided every day from the day seven days before the broadcast day to the day of the broadcast day.

  The information processing apparatus 1 acquires and manages distributed EPG data each time. The EPG data managed by the information processing apparatus 1 is used when displaying a program guide.

  The EPG data managed by the information processing apparatus 1 is also used to determine a sentence representing an event with news or topicality. Here, events with news or topicality include events transmitted from performers to viewers in news programs and news programs.

  For example, the information processing apparatus 1 reads the text data included in the program detailed information of the news program for the day broadcast included in the EPG data distributed on that day and the next day broadcast included in the EPG data distributed on the previous day. The text data included in the detailed program information of the news program is compared for each text data representing the content of the same news program. As will be described in detail later, the information processing apparatus 1 actually compares text data extracted from the detailed program information.

  As a result of the comparison, if there is text data that has been updated, for example, newly added to the EPG data distributed on the day, the information processing apparatus 1 uses the text data to indicate a news or topical event. Determine as data.

  The events reported in the news program are thought to be events that are of interest to general viewers among the events that have occurred up until the start of the program broadcast. Although it was not the previous day, text data added to the program detailed information on the day of the broadcast represents news or topical events.

  As a result, the information processing apparatus 1 can determine, using EPG data, text data representing an event that has news or topicality at the time of the day.

  The text data determined by the information processing apparatus 1 is used for presenting it to the user as it is, or used as text data for extracting keywords.

  The determined text data is displayed on the television receiver 2 and presented, so that the user can know an event with news or topicality. Further, by performing a process of extracting a keyword using the determined text data as a target, the information processing apparatus 1 can perform news extraction as compared with a case where the keyword extraction process is performed for the entire text data described in the detailed program information. It is possible to accurately extract words related to events that are characteristic or topical as keywords.

  Here, the details of the process of extracting text data representing an event with news or topicality performed as described above will be described. Hereinafter, text data representing news and topical events will be referred to as current affairs sentences as appropriate.

  FIG. 2 is a diagram showing the concept of current affairs sentence determination processing.

  For example, EPG data including information on programs scheduled to be broadcast for 8 days including that day every day at 2:30 am (standards defined by ARIB STD-B10 and ARIB TR-B14 / 15) SI-EPG (Service Information-EPG) data (implemented and operated in accordance with technical data) is distributed by digital broadcasting. The current affair sentence determination process is started when EPG data is acquired.

For EPG data, the pre-filtering as shown in the arrow A ₁ in FIG. 2 is performed.

  The pre-filter process is a process for narrowing down the program detail information included in the EPG data to information that satisfies a predetermined condition. In the example of FIG. 2, target data, target broadcast date, and target genre are set as conditions, and the program is broadcast by terrestrial digital broadcast or BS digital broadcast on the day of the day when EPG data is distributed. It is narrowed down to information on news programs or news programs to be broadcast. A news program or a news program is identified by a code of “0x0”.

For detailed program information extracted by the pre-filtering, topical sentence candidate extraction process as shown in the arrowhead A ₁ is applied.

  The current affair sentence candidate extraction process is a process of extracting text data that is a candidate for a current affair sentence from text data included in the detailed program information extracted by the pre-filter process. From the text data extracted as candidates, the text data to be the current sentence is finally determined.

  FIG. 3 is a diagram showing an example of program detailed information of a certain news program.

  As described above, the program detail information includes various text data such as the name of the performer, the corner name, and text data representing the contents of the program. In the example of FIG. 3, “6-3 victory in Japan”, “consultation packed”, “Governor Tanaka elected horse”, “popular secret”, “tonight's menu”, “Hideo Yanagi, Megumi Ito, Yuko Akayama Other "text data is included.

  “6-3 victory in Japan”, “consultation packed” and “Tanaka Governor elected horse” are prefixed with white triangles (▽) pointing down, respectively, “Tanaka Governor elected horse” and “ Spaces are inserted between “Popular Secrets”, “Popular Secrets” and “Tonight's Menu”, “Tonight's Menu” and “Hideo Yanagi, Megumi Ito, Yuko Akayama and others”.

  For example, in such a current affair sentence candidate extraction process for program detail information, text data to which a downward white triangle mark is added is regarded as a current affair sentence candidate, and this downward white triangle mark Is used as a separator, and the current sentence candidate text data is cut out (extracted) from other text data.

  The downward-facing white triangle mark is that the text data to which it was added was not included in the EPG data that was first distributed 7 days before the broadcast date, and was added before the broadcast date. Indicates that the text data.

  EPG data is distributed every day, and since the broadcast content of news programs and news programs is often determined on the day of the broadcast, it is usually included in the detailed program information included in the EPG data distributed on the day of the broadcast. The text data is updated as compared with the text data in the detailed program information included in the EPG data distributed before the broadcast date.

  Other text data such as the program corner name and performer name included in the detailed program information is often determined from the first time EPG data including the detailed program information of the program is first delivered. Corner names and performer names are often not updated, and downward white triangle marks are often not added.

  If the process proceeds with attention starting from the first character of the program details information and there is no downward white triangle mark at the beginning of the program detail information, the first downward triangle mark from the first character The text data up to is considered as a current affair sentence candidate and is cut out.

  In the text data following the last downward white triangle mark included in the program detailed information, up to the next space is cut out as a current affair sentence candidate.

  In the example of FIG. 3, “6-3 victory in Japan” between the first downward white triangle mark at the top of the program details information and the second downward white triangle mark is the current event. It is regarded as a sentence candidate and cut out. Similarly, “consultation size” between the second downward white triangular mark and the third downward white triangular mark is regarded as a current topic sentence candidate and cut out.

  Since the third downward white triangle mark is the last of the downward white triangle marks included in the program details information, the next space from the third downward white triangle mark, In other words, the “Governor Tanaka Election Horse” up to the space inserted between “Tanaka Governor Election Horse” and “Popular Secret” is regarded as a current affair sentence candidate and cut out.

  "Popular secrets", "Tonight's menu", "Hideo Yanagi, Megumi Ito, Yuko Akayama, etc." Are not regarded as current affair sentence candidates, and become text data that is not subject to extraction.

  Such processing is performed on the program detail information of each news program and news program extracted by the pre-filter processing. The extracted current affair sentence candidates are managed in association with programs.

Returning to the description of FIG. 2, the post-filter process is performed on the current sentence sentence extracted by the current sentence sentence extraction process as indicated by an arrow A ₂ .

  The post filter process is included in the current affair sentence candidate extracted by the current affair sentence candidate extraction process for the information of the program broadcast on the day of the day when the EPG data is distributed, and the EPG data distributed the previous day. This is a process for obtaining, for each program, the difference between the current sentence sentences extracted by the current sentence sentence extraction process for the program information for the next day broadcast, and determining the current sentence sentence candidate obtained as the difference as the current sentence sentence. The current affair sentence candidate extraction processing for the program information for the next day broadcast included in the EPG data distributed on the previous day is performed in advance on the previous day, for example.

  FIG. 4 is a diagram showing EPG data from which a current affair sentence candidate from which a difference is to be obtained is extracted.

  Assuming that the day of delivery of certain EPG data is N days and the previous day of N days is N−1 days, as shown in the lower part of FIG. Information on programs scheduled to be broadcast for 8 days from N days to N + 7 days is included. As shown in the upper part, EPG data distributed on the (N-1) th day includes information on programs scheduled to be broadcast for 8 days from the (N-1) th day to the (N + 6) th day. The EPG data shown in the upper row is distributed at an earlier timing than the EPG data shown in the lower row.

  In this case, in the post filter processing performed on the N day, a current affair sentence extracted from the program detailed information of news programs and news programs broadcast on the N day included in the EPG data distributed on the N day that is the current day. Candidates are extracted from the detailed program information of news programs and news programs broadcast on the Nth day of the next day, based on the N-1 day, included in the EPG data distributed on the N-1 day, the day before the Nth day. A difference between current topic sentences is obtained for each program.

  Specifically, in the post filter processing performed on March 2, news programs and news program programs broadcast on March 2 included in the EPG data distributed on March 2 that day. Candidates for current affairs sentence extracted from detailed information and included in EPG data distributed on March 1st, which is the day before March 2nd, broadcast on March 2nd of the following day based on March 1st The difference between current sentence candidates extracted from the detailed program information of news programs and news programs is obtained for each program.

  FIG. 5 is a diagram illustrating an example of program detail information included in EPG data distributed on the day of a broadcast day and program detail information included in EPG data distributed on the day before the broadcast day for a news program. .

  The program detailed information included in the EPG data distributed on the day of the broadcast date shown in the lower part of FIG. 5 is the same as the program detailed information of FIG. By performing current event sentence candidate extraction processing for this program detail information, as described above, “6-3 victory in Japan”, “consultation packed”, and “Governor Tanaka elected horse” are extracted as current event sentence candidates. Is done.

  The detailed program information included in the EPG data distributed on the day before the broadcast date shown in the upper part of FIG. 5 includes “Tomorrow's game”, “Six-country tomorrow”, “Frequent storm damage”, “Popular secrets” , “Tonight's menu”, “Hideo Yanagi, Megumi Ito, Yuko Akayama et al.” Are included.

  "Tomorrow's game", "Six-day talks tomorrow", and "storm damage frequent" are prefixed with white triangles pointing downwards, "storm damage frequent", "popular secret", "popular" Spaces are inserted between “Secret”, “Tonight's menu”, “Tonight's menu” and “Hideo Yanagi, Megumi Ito, Yuko Akayama and others”.

  If current event sentence candidate extraction processing was performed on the day before the broadcast date for such program detail information, “Tomorrow's game”, “Six-day talk tomorrow”, and “storm damage frequent occurrence” will be extracted as current event sentence candidates Is done.

  In the post filter processing that is performed on the day, "6-3 victory in Japan", "consultation big packed", "Governor Tanaka" that are current situation sentence candidates extracted from the program details information included in the EPG data distributed on the day “Selected horses” and “currently tomorrow game”, “six-day talks tomorrow”, “frequent storm damage” are extracted from the detailed program information included in the EPG data distributed on the previous day for the same program Three current affairs sentence candidates that are not included in the latter but included in the former, including “Victory in Japan in 6-3”, “Consultation” and “Electric horse selected by Mr. Tanaka” are determined as current affairs sentences. Is done.

  The above-described processing for determining current events is performed for all news programs and news programs broadcast on the day.

  The current affair sentence determined by the post filter processing is managed by the information processing apparatus 1 and is used for presenting to the user or used as a target of keyword extraction processing. In FIG. 2, the current affair sentence determined by the post filter processing as described above is distributed on the previous day with the program detail information included in the EPG data distributed on the day of the day on which the current affair sentence determination process was performed. It is shown as the current affairs difference on the day before that day in the sense that it represents the difference in the detailed program information included in the EPG data.

  A series of processing of the information processing apparatus 1 for determining the difference affair sentence on the day before the day will be described later with reference to a flowchart.

  FIG. 6 is a block diagram illustrating a hardware configuration example of the information processing apparatus 1.

  A CPU (Central Processing Unit) 11 executes various processes according to a program recorded in a ROM (Read Only Memory) 12 or a recording unit 19. A RAM (Random Access Memory) 13 appropriately stores programs executed by the CPU 11 and data. The CPU 11, ROM 12, and RAM 13 are connected to each other by a bus 14.

  An input / output interface 15 is also connected to the CPU 11 via the bus 14. The input / output interface 15 is connected to a receiving unit 16, an input unit 17, an output unit 18, a recording unit 19, a communication unit 20, and a drive 21.

  The receiving unit 16 receives and demodulates the broadcast wave signal from the antenna 16A, and acquires MPEG-TS (Moving Picture Experts Group-Transport Stream). The receiving unit 16 acquires program data and EPG data to be recorded from the MPEG-TS, and outputs the acquired data to the recording unit 19 via the input / output interface 15.

  The input unit 17 receives a signal from the remote controller, and outputs information representing the contents of the user's operation to the CPU 11 via the input / output interface 15 and the bus 14.

  The output unit 18 decodes the data of the program instructed to be reproduced, and displays the video of the program on the television receiver 2 based on the obtained video signal.

  The recording unit 19 includes, for example, a hard disk, and records various data such as a program executed by the CPU 11, program data supplied from the receiving unit 16 via the input / output interface 15, and EPG data.

  The communication unit 20 communicates with the server, and acquires program data distributed by broadcasting performed via the Internet and EPG data distributed by the server. The communication unit 20 outputs the acquired data to the recording unit 19 via the input / output interface 15 and records it.

  When a removable medium 22 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory is mounted, the drive 21 drives them to acquire programs and data recorded therein. The acquired program and data are transferred to the recording unit 19 and recorded as necessary.

  FIG. 7 is a block diagram illustrating a functional configuration example of the information processing apparatus 1. At least a part of the functional units illustrated in FIG. 7 is realized by the CPU 11 in FIG. 6 executing a predetermined program.

  As shown in FIG. 7, in the information processing apparatus 1, an EPG data acquisition unit 31, an EPG data storage unit 32, a processing target data extraction unit 33, current affair sentence candidate extraction units 34-1 and 34-2, current affair sentence candidates Storage units 35-1 and 35-2, a current affair sentence determination unit 36, a current affair sentence storage unit 37, and a current affair sentence processing unit 38 are realized.

  The EPG data acquisition unit 31 controls the reception unit 16 to acquire EPG data that is repeatedly distributed via digital broadcasting or the like. The EPG data acquisition unit 31 outputs the acquired EPG data to the EPG data storage unit 32 and stores it.

  The EPG data storage unit 32 stores the EPG data supplied from the EPG data acquisition unit 31. The EPG data stored in the EPG data storage unit 32 is rewritten every time the latest EPG data is supplied from the EPG data acquisition unit 31, for example. The EPG data storage unit 32 stores the latest information of programs scheduled to be broadcast for 8 days including the delivery date of EPG data.

  The processing target data extraction unit 33 narrows down the program information included in the EPG data stored in the EPG data storage unit 32 to the program information satisfying a predetermined condition by performing pre-filter processing as data cleaning. The information of the program that satisfies is output to the current affair sentence candidate extraction unit 34-1 or the current affair sentence candidate extraction unit 34-2.

  For example, the processing target data extraction unit 33 is broadcast on the current day (N day) by digital terrestrial broadcasting or BS digital broadcasting on the basis of the distribution date (N day) of EPG data stored in the EPG data storage unit 32. Information on each news program or news program is output to the current affair sentence candidate extraction unit 34-1.

  Further, the processing target data extraction unit 33 is broadcast on the next day (N + 1 day) by terrestrial digital broadcasting or BS digital broadcasting with reference to the distribution date (N day) of the EPG data stored in the EPG data storage unit 32. Information on each news program or news program is output to the current affair sentence candidate extraction unit 34-2. Information of each news program or news program broadcast on the next day after being narrowed down from the EPG data distributed on the Nth day is included in the EPG data distributed on the next day in the post filter processing performed on the next day ( It is used to obtain the difference from the information of each news program or news program broadcasted on the day N + 1).

  The current affair sentence candidate extraction unit 34-1 uses the downward white triangle mark as a separator from the text data described in the program detail information supplied from the processing target data extraction unit 33, for example, as a current affair sentence candidate. To extract. The current affair sentence candidate extraction unit 34-1 outputs the extracted current affair sentence candidate to the current affair sentence candidate storage unit 35-1, and stores it.

  Similarly, the current situation sentence candidate extraction unit 34-2 also uses a downward white triangle mark as a separator from the text data described in the detailed program information supplied from the processing target data extraction unit 33, for example. Sentence candidates are extracted. The current affair sentence candidate extraction unit 34-2 outputs the extracted current affair sentence candidate to the current affair sentence candidate storage unit 35-2 and stores it.

  The current affair sentence candidate storage unit 35-1 stores the current affair sentence candidates supplied from the current affair sentence candidate extraction unit 34-1 and the current affair sentence candidate storage unit 35-2 is supplied from the current affair sentence candidate extraction unit 34-2. Memorize current topic sentence candidates. The current affair sentence candidates stored in the current affair sentence candidate storage unit 35-2 are stored until they are used in the post filter processing on the next day.

  For example, at the time of March 2, the current affair sentence candidate storage unit 35-1 is extracted from the program detail information of the program for the March 2 broadcast included in the EPG data distributed on March 2 Current affair sentence candidates are stored, and the current affair sentence candidate storage unit 35-2 extracts the current affair sentence candidates extracted from the program detailed information of the program broadcast on March 3 included in the EPG data distributed on March 2 Will be memorized. The current affairs sentence candidate storage unit 35-2 is pre-extracted in advance on March 1st from the program detail information of the program broadcast on March 2nd included in the EPG data distributed on March 1st of the previous day. The current topic sentence candidate that had been stored is also remembered.

  The current affair sentence determination unit 36 performs post-filter processing based on the current affair sentence candidates stored in the current affair sentence candidate storage section 35-1 and the current affair sentence candidates stored in the current affair sentence candidate storage section 35-2 from the previous day. Determine current affairs sentence by doing.

  That is, the current affair sentence determination unit 36 obtains, for each program, a difference between the current affair sentence candidate stored in the current affair sentence candidate storage section 35-1 and the current affair sentence candidate stored in the current affair sentence candidate storage section 35-2. Then, the current affair sentence candidate obtained as the difference is determined as the current affair sentence. The current affair sentence determination unit 36 outputs the current affair sentence to the current affair sentence storage unit 37 and stores it.

  The current affair sentence storage unit 37 stores the current affair sentence supplied from the current affair sentence determination unit 36.

  The current affair sentence processing unit 38 appropriately reads out the current affair sentence stored in the current affair sentence storage unit 37, and performs various processes such as presentation of the current affair sentence and keyword extraction based on the read out current affair sentence.

  Next, processing of the information processing apparatus 1 having the above configuration will be described.

  First, the current sentence determination process of the information processing apparatus 1 will be described with reference to the flowchart of FIG.

  For example, when the EPG data is distributed every day at 2:30 am, this process is started in response to the acquisition. EPG data including information on the latest distributed program is acquired by the EPG data acquisition unit 31 and stored in the EPG data storage unit 32.

  In step S1, the processing target data extraction unit 33 performs a prefilter process. The detailed program information of the programs narrowed down by the prefiltering process is appropriately supplied to the current affair sentence candidate extraction units 34-1 and 34-2. Details of the prefiltering process will be described later with reference to the flowchart of FIG.

  In step S2, each of the current affair sentence candidate extraction unit 34-1 and the current affair sentence candidate extraction unit 34-2 performs a current affair sentence candidate extraction process. The current sentence sentence extracted by the current sentence sentence extraction unit 34-1 is stored in the current sentence sentence storage unit 35-1, and the current sentence sentence extracted by the current sentence sentence extraction unit 34-2 is stored in the current sentence sentence storage unit 35. -2. Details of the current affair sentence candidate extraction process will be described later with reference to the flowchart of FIG.

  In step S3, the current affair sentence determination unit 36 performs post filtering based on the current affair sentence candidates stored in the current affair sentence candidate storage section 35-1 and the current affair sentence candidates stored in the current affair sentence candidate storage section 35-2. Process. The current affair sentence determined by the post filter processing is stored in the current affair sentence storage unit 37, and the processing is terminated. Details of the post filter processing will be described later with reference to the flowchart of FIG.

  Next, the prefiltering process performed in step S1 of FIG. 8 will be described with reference to the flowchart of FIG.

  In step S11, the processing target data extraction unit 33 reads the EPG data stored in the EPG data storage unit 32, and includes the subtitle added to the title of each program in the program detailed information of the program.

  For example, when a theme handled in a program is added as a subtitle to the title of a certain news program, the theme is included in the program detailed information. Since a sentence representing news or topical events may be included in the whole or a part of the subtitle, current event sentence candidates are extracted from the subtitle text data.

  In step S12, the processing target data extraction unit 33 deletes characters and symbols to be excluded from the program detailed information of each program by referring to a dictionary prepared in advance. For example, a broadcast station name, a symbol indicating that the program sound is stereo, and the like are deleted.

  In step S13, the processing target data extraction unit 33 narrows down to information on programs that satisfy the conditions of channel, broadcast date, and genre, and selects the selected information to the current affair sentence candidate extraction unit 34-1 or the current affair sentence candidate extraction unit 34-2. Output. For example, terrestrial digital broadcasting and BS digital broadcasting channels are preset as channel conditions, EPG data delivery date and next day are preset as broadcast day conditions, and news programs and news programs are preset as genre conditions.

  The processing target data extraction unit 33 uses the delivery date (N day) of the EPG data read from the EPG data storage unit 32 as a reference to each news program broadcast on the current day (N day) by digital terrestrial broadcasting or BS digital broadcasting. Alternatively, the detailed program information of the news program is sent to the current affairs sentence candidate extraction unit 34-1 and the detailed program information of each news program or news program broadcasted on the next day (N + 1) by digital terrestrial broadcasting or BS digital broadcasting is selected as the current affair sentence candidate. Each is output to the extraction unit 34-2.

  After the detailed program information is output to the current affair sentence candidate extraction unit 34-1 and the current affair sentence candidate extraction unit 34-2, the process returns to step S1 in FIG. 8, and the subsequent processing is performed.

  Next, the current sentence sentence extraction process performed in step S2 of FIG. 8 will be described with reference to the flowchart of FIG.

  In step S21, the current affair sentence candidate extraction unit 34-1 has been described with reference to FIG. 3 from the program detailed information included in the EPG data distributed from the processing date data extraction unit 33 and distributed on the day of the broadcast. Thus, current affair sentence candidates are extracted.

  In step S22, the current affair sentence candidate extraction unit 34-1 outputs the extracted current affair sentence candidate to the current affair sentence candidate storage unit 35-1, and stores it.

  In step S23, the current affairs sentence candidate extraction unit 34-2 obtains the current affair sentence as described with reference to FIG. 3 from the program detailed information of the program broadcast on the next day supplied from the processing target data extraction unit 33. Extract candidates.

  In step S24, the current affair sentence candidate extraction unit 34-2 outputs the extracted current affair sentence candidate to the current affair sentence candidate storage unit 35-2 and stores it. Thereafter, the process returns to step S2 in FIG. 8, and the subsequent processing is performed.

  Next, the post filter processing performed in step S3 of FIG. 8 will be described with reference to the flowchart of FIG.

  In step S31, the current affair sentence determination unit 36 reads the current affair sentence candidates stored in the current affair sentence candidate storage section 35-1, and the current affair sentence candidates stored in the current affair sentence candidate storage section 35-2 from the previous day. Of the programs whose contents are represented by the detailed program information from which each current affair sentence candidate is extracted, attention is paid to one news program or news program. The current affair sentence candidate storage unit 35-1 and the current affair sentence candidate storage unit 35-2 store current affair sentence candidates extracted from the detailed program information of a news program or a news program broadcast on the day when the processing of FIG. 11 is performed. Has been.

  In step S32, the current affairs sentence determination unit 36 extracts the current affair sentence candidates stored in the current affair sentence candidate storage unit 35-1 as extracted from the program detail information of the program of interest, and the current affair sentence candidate storage unit 35. -2 current event sentence candidates stored in the program, that is, current event sentence candidates extracted from the detailed program information included in the EPG data distributed on the day of the broadcast date of the program, and EPG data distributed the day before the broadcast date The difference between current sentence sentences extracted from the detailed program information included in the current information is obtained.

  In step S33, the current affairs sentence determination unit 36 is not included in the data extracted from the EPG data distributed on the day before the broadcast date, but is extracted from the EPG data distributed on the day of the broadcast date of the program. A certain current affair sentence candidate is determined as a current affair sentence and stored in the current affair sentence storage unit 37.

  In step S34, the current affairs sentence determination unit 36 determines whether or not all news programs or news programs have been noticed.

  If it is determined in step S34 that all the news programs or news programs are not focused, the process returns to step S31, and the current affairs sentence determination unit 36 repeats the same processing while focusing on other programs.

  If it is determined in step S34 that all news programs or news programs have been noticed, the process returns to step S3 in FIG. 8, and the current affair sentence determination process is terminated.

  As described above, the information processing apparatus 1 can extract text data that is considered to represent an event with news or topicality from EPG data.

  In the above, the current affair sentence candidate used for obtaining the difference from the current affair sentence candidate extracted from the program detailed information of the program for the current day broadcast included in the EPG data distributed on the N day is the previous day's N-1 In the case where EPG data distributed for several days such as two days is stored in the EPG data storage unit 32, the extraction is performed on a day and stored until the next day. When processing for extracting current affairs candidates from program detailed information included in EPG data distributed on N days is performed, current affairs sentence candidates are extracted from program detailed information included in EPG data distributed on N-1 days The process to perform may also be performed.

  For example, when the EPG data distributed on March 1 and the EPG data distributed on March 2 are stored in the EPG data storage unit 32, in the process performed on March 2, Extraction of current affairs sentence candidates for news programs broadcast on March 2nd, which is the next day of the EPG data distribution date, and detailed information on news programs included in the EPG data distributed on March 1st Extraction of current affair sentence candidates for news program and broadcast program details information broadcast on March 2, which is the day of EPG data distribution date, included in EPG data distributed on March 2 Is done.

  In this way, the information processing apparatus 1 also distributes the current program candidate extracted from the detailed program information included in the EPG data distributed on the day of the program broadcast and the day before the program broadcast date for the same program. It is possible to obtain a difference between current sentence candidates extracted from the detailed program information included in the EPG data.

  Next, the process performed by the current affairs sentence processing unit 38 using the current affair sentence determined as described above will be described.

  First, the trend keyword determination process of the current affairs sentence processing unit 38 will be described with reference to the flowchart of FIG. A trend keyword is a word that is considered to be related to a news or topical event, and is determined from words included in a current affair sentence.

  In step S41, the current affair sentence processing unit 38 performs morphological analysis on each current affair sentence stored in the current affair sentence storage unit 37, and extracts each morpheme of the analysis result as a keyword.

  In step S42, the current affair sentence processing unit 38 deletes the keyword of the part of speech to be deleted from the extracted keywords. For example, only noun keywords are left.

  In step S43, the current affair sentence processing unit 38 calculates an appearance frequency in the current affair sentence for each of the remaining keywords.

  In step S44, the current affair sentence processing unit 38 determines a predetermined number of keywords such as 50 as trend keywords in descending order of appearance frequency, and ends the processing. The trend keyword is used, for example, to automatically set a recording reservation for a program that includes the same word in the program detailed information without depending on a user operation.

  As a result, the information processing apparatus 1 can accurately extract words related to news events and topical events as keywords as compared to the case where keyword extraction processing is performed on the entire text data described in the detailed program information. Can do.

  Next, with reference to the flowchart of FIG. 13, the current affairs sentence display process of the current affairs sentence processing unit 38 will be described.

  In step S51, the current affair sentence processing unit 38 pays attention to one current affair sentence stored in the current affair sentence storage unit 37.

  In step S52, the current affairs sentence processing unit 38 causes the television receiver 2 to display the current affair sentence of interest by controlling the output unit 18, for example.

  FIG. 14 is a diagram illustrating a display example of current affair sentences.

  In the example of FIG. 14, “6-3 victory in Japan” is displayed on the lower left of the screen of the television receiver 2. In the screen of the television receiver 2 other than the part where the current affair sentence is displayed, the video of the program being viewed by the user, the screen of the website displayed by the user using a web browser, etc. Is displayed.

  Returning to the description of FIG. 13, in step S <b> 53, the current affair sentence processing unit 38 determines whether or not attention has been paid to all current affair sentences stored in the current affair sentence storage unit 37.

  If it is determined in step S53 that not all the current sentence sentences are noticed, the process returns to step S51, and the current sentence sentence processing unit 38 pays attention to other current sentence sentences and repeats the same processing. By performing the same processing, the current affair sentence displayed on the television receiver 2 is sequentially switched.

  If it is determined in step S53 that attention has been paid to all current affair sentences, the processing is terminated.

  Through the above processing, the user can confirm the event that is currently being discussed even while viewing a program. When a user performs a predetermined operation using a remote controller while a certain current event sentence is displayed, detailed program information from which the current event sentence is extracted is displayed, and the user can display the current event sentence using the current event sentence. It may be possible to confirm information on a program to be reported about an event to be performed.

  In the above, based on the EPG data distributed at 2:30 am on the day of the broadcast date of the program and the EPG data distributed at 2:30 am on the previous day (current difference sentence on the day before the current day) However, the current affair sentence may be determined based on EPG data distributed at other timings.

  For example, for a news program broadcasted in the evening, the content of EPG data distributed on the same broadcast day may differ depending on whether it is distributed in the morning or distributed in the evening. Therefore, it is also possible to obtain the difference between the current topic sentence extracted from the program detail information included in the EPG data distributed in the morning and the current event sentence candidate extracted from the program detail information included in the EPG data distributed in the evening. It becomes possible to determine current affairs sentences.

  Next, based on the difference between the current topic sentence extracted from the program detail information included in the EPG data distributed in the morning and the current event sentence candidate extracted from the program detail information included in the EPG data distributed in the evening Processing for determining a sentence will be described.

  Here, for convenience of explanation, EPG data distributed at morning time such as 2:30 am is called morning edition EPG data, and EPG data distributed at evening time such as 3:30 pm is evening EPG data. That's it. The current affair sentence obtained from the difference between the current affair sentence candidate extracted from the program detail information included in the morning edition EPG data and the current affair sentence candidate extracted from the program detail information included in the evening edition EPG data is the difference between the previous day and the previous day. For the current affairs sentence, it becomes the evening edition morning difference diffrent current affair sentence.

  FIG. 15 is a diagram illustrating the concept of evening morning difference differential current affairs sentence determination processing.

  For example, when evening EPG data relating to a program scheduled to be broadcast for eight days including the day is distributed every day at 3:30 pm, the evening EPG data is acquired. Will start when.

  In the information processing device 1, current affairs sentence candidates extracted from the detailed program information included in the morning edition EPG data are managed when the difference current affair sentence on the day before is calculated at the morning time such as 2:30 am on the same day. Yes.

The evening EPG data is pre-filtered as shown by an arrow A ₁₁ in FIG. 15 and is broadcast on the day of the evening EPG data distribution date broadcast by terrestrial digital broadcasting or BS digital broadcasting. Detailed program information of a news program or a news program is extracted.

For a news program or program detailed information of news programs that have been extracted by the pre-filtering, topical sentence candidate extraction process as shown in the arrowhead A ₁₁ is applied, current events sentence candidates are extracted. The current affair sentence candidate extraction processing for the detailed program information included in the evening EPG data is also performed in the same manner as the processing described with reference to FIG. That is, text data to which downward white triangle marks are added is regarded as a current affair sentence candidate, and is sequentially cut out from the program detailed information.

For topical sentence candidates extracted by the topical sentence candidate extraction processing, the post-filter process, as indicated by arrow A ₁₂ is performed.

  In the post-filter process, current sentence sentences extracted by the current sentence sentence extraction process performed on the detailed program information included in the evening EPG data, and the detailed program information included in the morning EPG data that has been obtained in advance. The difference between the current sentence sentences extracted from the program is obtained for each program, and the updated current sentence sentence in the program detailed information included in the evening EPG data is determined as the evening morning difference current topic sentence.

  The evening morning difference differential current affair sentence is used to extract a keyword or to be displayed as it is, like the current difference sentence on the day before the day.

  In news programs, etc., the broadcast content is determined several hours before the broadcast start time of the program and may be reflected in the detailed program information included in the EPG data that is distributed after that. As well as EPG data distributed on the day of the broadcast day and EPG data distributed on the morning of the broadcast day, the current affairs sentence is also based on EPG data distributed at a time closer to the broadcast start time of the program on the broadcast day. This makes it possible to determine a newer current sentence related to news-related or topical events.

  FIG. 16 and FIG. 17 are diagrams showing processing for determining the evening morning difference differential current affair sentence. The horizontal direction in FIGS. 16 and 17 represents the time direction.

  Each of the blocks # 1 to # 9 shown on the left side of FIG. 16 is extracted from the program detailed information included in the morning edition EPG data distributed at 2:30 am on the day before the broadcast date of the program for a certain program. Represents a current topic sentence candidate.

  On the other hand, each of the blocks # 11 to # 19 shown on the right side of the blocks # 1 to # 9 is included in the morning edition EPG data distributed at 2:30 am on the day of the program broadcast day for the same program. Represents current affair sentence candidates extracted from the detailed program information.

  In this case, the current topic sentence candidates extracted from the program detail information included in the morning edition EPG data distributed on the day of the broadcast date of the program are each noted, and the same current topic sentence candidate is included in the morning edition EPG data distributed the previous day By determining whether or not there is a current sentence sentence extracted from the detailed program information, the difference between the current sentence sentences included in the two EPG data is obtained. If there is the same current topic sentence candidate among the current topic sentence candidates extracted from the detailed program information included in the morning edition EPG data distributed on the previous day, the current topic sentence candidate was text data representing the corner name of the program, etc. If it is determined that there is no candidate for the current topic sentence, the candidate for the current topic sentence is determined as the current topic sentence.

  In the example of FIG. 16, current affair sentence candidates represented by blocks # 11 to # 14 are determined as differential current affairs on the day before the current day.

  Also, the current sentence candidates represented by the blocks # 15 to # 19 are the same as the current sentence candidates represented by the blocks # 5 to # 9 extracted from the program detail information included in the morning edition EPG data distributed on the previous day. These current topic sentences are determined as text data representing the corner names of the program.

  In this way, when the current topic sentence candidates represented by the blocks # 11 to # 19 are extracted from the detailed program information included in the morning edition EPG data, the evening edition EPG distributed at 3:30 pm on the day of the broadcast date of the program. When current situation sentence candidates represented by blocks # 21 to # 25 are extracted as shown in FIG. 17 by the current situation sentence candidate extraction process performed on data, the current situation sentence candidates are represented by blocks # 21 to # 25. Whether or not the same current sentence candidate is among the current sentence candidates represented by blocks # 11 to # 19 extracted from the detailed program information included in the morning edition EPG data. By determining, the difference between the current sentence sentences extracted from the detailed program information included in the two EPG data is obtained.

  In the example of FIG. 17, the same thing as the current news sentence candidate represented by blocks # 22 and # 23 is not included in the current news sentence candidate extracted from the program detail information included in the morning edition EPG data, but the evening edition morning news difference current news sentence. As determined.

  The current situation sentence candidate represented by block # 21 is the same as the current situation sentence candidate represented by block # 13 extracted from the detailed program information included in the morning edition EPG data. Further, the current topic sentence candidates represented by the blocks # 24 and # 25 are the same as the current topic sentence candidates represented by the blocks # 18 and # 19 extracted from the detailed program information included in the morning edition EPG data. .

  It can be said that the text data represented by blocks # 22 and # 23 determined as the evening morning differential difference current affairs sentence represents an event that is more news and topical when the evening EPG data is distributed.

  On the other hand, although it was decided as the difference current topic sentence on the day before the day, it is not decided as the evening edition morning difference difference current sentence because the same thing is included in the current situation sentence candidates extracted from the program details information included in the evening EPG data. The text data represented by block # 21 represents an event such as news from the delivery of the morning EPG data of the previous day until the morning EPG data of the day was delivered. From the delivery of the morning edition EPG data to the delivery of the evening edition EPG data of the day, it is text data with a slightly reduced freshness in terms of news and topicality.

  In FIG. 17, the text data represented by the dotted block is included in the program detailed information of the EPG data distributed immediately before, but is included in the program detailed information of the EPG data distributed next. The text data is no longer available.

  The text data represented by blocks # 22 and # 23 that are finally determined as evening morning difference current topic sentences and the block # 22 that has been determined as difference current topic sentences the day before the day, but not determined as evening morning difference current topic sentences Different weights may be set for the text data represented by 21 and managed by the information processing apparatus 1.

  For example, when displaying the current affairs sentence and presenting it to the user, the text data finally determined as the evening morning difference current affairs sentence was determined as the difference current affairs sentence on the day before, but it was determined as the evening morning difference affairs current sentence. It is possible to display the text data in preference to the missing text data.

  FIG. 18 is a block diagram illustrating an example of a functional configuration of the information processing apparatus 1 that determines the evening morning / difference current affair sentence as described above.

  The configuration shown in FIG. 18 is basically the same as the configuration shown in FIG. 7 except that a current affair sentence candidate extraction unit 34-3 and a current affair sentence candidate storage unit 35-3 are added. The same code | symbol is attached | subjected to the same structure. The overlapping description will be omitted as appropriate.

  When the morning edition EPG data is stored in the EPG data storage unit 32, the processing target data extraction unit 33 in FIG. 18 uses the delivery date (N day) of the morning edition EPG data stored in the EPG data storage unit 32 as a reference. Information of each news program or news program broadcast on the current day (N day) by terrestrial digital broadcasting or BS digital broadcasting is output to the current affair sentence candidate extraction unit 34-1.

  Further, the processing target data extraction unit 33 is broadcast on the next day (N + 1 day) by digital terrestrial broadcasting or BS digital broadcasting on the basis of the distribution date (N day) of the morning edition EPG data stored in the EPG data storage unit 32. The information of each news program or news program is output to the current affair sentence candidate extraction unit 34-2.

  Further, when the evening EPG data is stored in the EPG data storage unit 32, the processing object data extraction unit 33 uses the distribution date (N days) of the evening EPG data stored in the EPG data storage unit 32 as a reference. Information of each news program or news program broadcast on the current day (N day) by digital broadcasting or BS digital broadcasting is output to the current affair sentence candidate extraction unit 34-3.

  Current affair sentence candidate extraction units 34-1 to 34-3 extract current affair sentence candidates from the detailed program information supplied from the processing target data extraction unit 33 by using a downward white triangle mark as a separator, for example. To do.

  The current affair sentence candidate extraction unit 34-1 stores the extracted current affair sentence candidate in the current affair sentence candidate storage unit 35-1, and the current affair sentence candidate extraction unit 34-2 stores the extracted current affair sentence candidate in the current affair sentence candidate storage unit 35. -2 is stored. Moreover, the current affair sentence candidate extraction unit 34-3 stores the extracted current affair sentence candidate in the current affair sentence candidate storage unit 35-3.

  The current affair sentence determination unit 36 performs post-filter processing based on the current affair sentence candidates stored in the current affair sentence candidate storage section 35-1 and the current affair sentence candidates stored in the current affair sentence candidate storage section 35-2. By the above, the difference current affair sentence the day before the day is determined.

  Moreover, the current affair sentence determination unit 36 stores the current affair sentence candidates extracted from the program detailed information included in the morning edition EPG data stored in the current affair sentence candidate storage section 35-1, and the current affair sentence candidate storage section 35-3. The difference between the current topic sentence candidates extracted from the stored program detail information included in the evening EPG data is obtained, and the evening morning morning difference current topic sentence is determined.

  The current affairs sentence determination unit 36 causes the current affairs sentence storage unit 37 to store the determined difference current sentence on the day before the current day and the evening edition morning difference differential current affair sentence.

  Next, with reference to the flowchart of FIG. 19, the evening morning difference differential current affairs sentence determination process of the information processing apparatus 1 having the configuration of FIG. 18 will be described.

  The process for determining the evening morning difference differential current affair sentence is performed after the difference current sentence on the day before is determined by the same process as described with reference to the flowchart of FIG. The processes in steps S101 to S103 in FIG. 19 are the same as the processes in steps S1 to S3 in FIG.

  That is, in step S101, the processing target data extraction unit 33 performs pre-filter processing, and terrestrial digital broadcasting or BS digital broadcasting is based on the delivery date (N days) of the morning edition EPG data stored in the EPG data storage unit 32. The information of each news program or news program broadcast on the current day (N day) is output to the current affair sentence candidate extraction unit 34-1. Further, the processing target data extraction unit 33 is broadcast on the next day (N + 1 day) by digital terrestrial broadcasting or BS digital broadcasting on the basis of the distribution date (N day) of the morning edition EPG data stored in the EPG data storage unit 32. The information of each news program or news program is output to the current affair sentence candidate extraction unit 34-2.

  In step S102, the current affair sentence candidate extraction unit 34-1 and the current affair sentence candidate extraction unit 34-2 perform current affair sentence candidate extraction processing on the program detailed information supplied from the processing target data extraction unit 33, respectively. Extract current affairs sentence candidates. The current affair sentence candidate extraction unit 34-1 stores the extracted current affair sentence candidate in the current affair sentence candidate storage unit 35-1, and the current affair sentence candidate extraction unit 34-2 stores the extracted current affair sentence candidate in the current affair sentence candidate storage unit 35. -2 is stored.

  In step S103, the current affair sentence determination unit 36 performs post-filter processing, and is stored from the previous day in the current affair sentence candidate stored in the current affair sentence candidate storage unit 35-1 and the current affair sentence candidate storage unit 35-2. The text data representing the difference between current affair sentence candidates is determined as the difference current affair sentence on the day before the current day. The current affair sentence determination unit 36 manages the determined current affair sentence on the day before that day until evening EPG data is distributed.

  In step S104, the processing target data extraction unit 33 determines whether evening EPG data has been stored in the EPG data storage unit 32, and waits until it is determined that it has been stored.

  If it is determined in step S104 that evening EPG data has been stored in the EPG data storage unit 32, the processing target data extraction unit 33 performs prefiltering on the evening EPG data stored in the EPG data storage unit 32 in step S105. I do.

  The prefilter process performed in step S105 is the same process as the prefilter process performed in step S101. In other words, the processing target data extraction unit 33 is broadcast on the current day (N days) by digital terrestrial broadcasting or BS digital broadcasting on the basis of the distribution date (N days) of the evening EPG data stored in the EPG data storage unit 32. The information of each news program or news program is output to the current affair sentence candidate extraction unit 34-3.

  In step S106, the current affair sentence candidate extraction unit 34-3 performs a current affair sentence candidate extraction process. The current sentence sentence extraction process performed in step S106 will be described later with reference to the flowchart of FIG.

  In step S107, the current affairs sentence determination unit 36 performs post-filter processing, and is extracted from the current affair sentence candidates extracted from the program detailed information included in the morning edition EPG data and the program detailed information included in the evening edition EPG data in step S106. Based on the current affairs sentence candidates, the evening morning difference affairs current affair sentence is determined. The post filter processing performed in step S107 will be described later with reference to the flowchart of FIG.

  Next, with reference to the flowchart of FIG. 20, the current sentence sentence extraction process performed in step S106 of FIG. 19 will be described.

  In step S111, the current topic sentence candidate extraction unit 34-3 adds a downward white triangle mark from the program detailed information of the program for the day broadcast included in the evening EPG data supplied from the processing target data extraction unit 33. Extract current affairs sentence candidates such as separators.

  In step S112, the current affair sentence candidate extraction unit 34-3 outputs the extracted current affair sentence candidate to the current affair sentence candidate storage unit 35-3 and stores it. Thereafter, the process returns to step S106 in FIG. 19, and the subsequent processing is performed.

  Next, the post filter processing performed in step S107 of FIG. 19 will be described with reference to the flowchart of FIG.

  In step S121, the current affair sentence determination unit 36 reads out the current affair sentence candidates stored in the current affair sentence candidate storage section 35-3, and the contents are represented by the detailed program information from which the respective current affair sentence candidates are extracted. Focus on one news program or news program among the programs.

  In step S122, the current affairs sentence determination unit 36 extracts the current affairs sentence candidates extracted from the detailed program information included in the morning edition EPG data and stored in the current affair sentence candidate storage unit 35-1 for the program of interest, and the evening publication. A difference between current sentence sentences extracted from the detailed program information included in the EPG data and stored in the current current sentence candidate storage unit 35-3 is obtained, and the current sentence candidate extracted from the detailed program information included in the evening EPG data is obtained. Among these, current-language sentence candidates that are not included in the current-language sentence candidates extracted from the detailed program information included in the morning edition EPG data are determined as evening-day morning-difference current topic sentences.

  In step S <b> 123, the current affairs sentence determination unit 36 stores the determined evening newspaper morning difference differential current affair sentence in the current affairs sentence storage unit 37.

  In step S124, the current affairs sentence determination unit 36 determines whether or not all news programs or news programs have been noticed.

  If it is determined in step S124 that all news programs or news programs are not focused, the process returns to step S121, and the current affairs sentence determination unit 36 repeats the same processing while focusing on other programs.

  If it is determined in step S124 that all news programs or news programs have been noticed, the process returns to step S107 in FIG. 19 and the evening morning difference differential current affairs sentence determination process ends.

  With the above processing, the information processing apparatus 1 can extract text data that represents an event with news or topicality at a time closer to the broadcast start time of the program.

  In the above, the current situation sentence is required only twice a day when the morning edition EPG data is acquired and when the evening edition EPG data is acquired. However, the current situation sentence is requested at other timings. Also good. It is also possible to require current affairs sentences every time EPG data is acquired.

  Also, in the above, the current situation sentence candidates are extracted from the detailed program information of the news program and the news program, and the current situation sentence is obtained from them, but the detailed program information of one of the programs is obtained. The current situation sentence candidate may be extracted as a target, and the current situation sentence may be obtained, or the current situation sentence candidate is extracted from the program detail information of programs of other genres so that the current situation sentence is obtained. It may be.

  Further, in the above, the current situation sentence is required using the program detailed information of the broadcast program, but the current situation sentence is obtained using information related to other contents such as a program distributed from the server via the Internet. It may be required.

  The series of processes described above can be executed by hardware or can be executed by software. When a series of processing is executed by software, a program constituting the software executes various functions by installing a computer incorporated in dedicated hardware or various programs. For example, it is installed from a program recording medium in a general-purpose personal computer or the like.

  The program executed by the computer may be a program that is processed in time series in the order described in this specification, or in parallel or at a necessary timing such as when a call is made. It may be a program for processing.

  The embodiments of the present invention are not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

It is a figure which shows the example of the information processing apparatus which concerns on one Embodiment of this invention. It is a figure which shows the concept of a current affair sentence determination process. It is a figure which shows the example of the program detailed information of a news program. It is a figure shown about the current affair sentence candidate used as the object which calculates | requires a difference. It is another figure which shows the example of the program detailed information of a news program. It is a block diagram which shows the hardware structural example of information processing apparatus. It is a block diagram which shows the function structural example of information processing apparatus. It is a flowchart explaining the current affairs sentence determination process of information processing apparatus. It is a flowchart explaining the pre filter process performed in step S1 of FIG. It is a flowchart explaining the current affair sentence candidate extraction process performed in step S2 of FIG. It is a flowchart explaining the post filter process performed in step S3 of FIG. It is a flowchart explaining a trend keyword determination process. It is a flowchart explaining a current affair sentence display process. It is a figure which shows the example of a display of current affair sentence. It is a figure which shows the concept of evening edition morning edition difference current affair sentence determination processing. It is a figure shown about the process which determines evening edition morning edition difference current affair sentence. It is a figure shown about the process which determines evening edition morning edition difference current affair sentence. It is a block diagram which shows the other function structural example of information processing apparatus. It is a flowchart explaining the evening edition morning difference differential current affair sentence determination process of the information processing apparatus of FIG. It is a flowchart explaining the current affair sentence candidate extraction process performed in step S106 of FIG. It is a flowchart explaining the post filter process performed in step S107 of FIG.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Information processing apparatus, 31 EPG data acquisition part, 32 EPG data storage part, 33 Process target data extraction part, 34-1 and 34-2 Current affair sentence candidate extraction part, 35-1, 35-2 Current affair sentence candidate storage part, 36 current affairs sentence determination part, 37 current affair sentence storage part, 38 current affair sentence processing part

Claims (8)

Obtaining means for obtaining EPG data including information on each content to be provided at a predetermined time;
The text data representing the content of the content included in the EPG data acquired at the first timing by the acquisition means, and the content of the content included in the EPG data acquired at the second timing later than the first timing. First extracting means for comparing the text data to be represented for each text data representing the content of the same content, and extracting the text data updated in the EPG data acquired at the second timing;
An information processing apparatus comprising: second extraction means for extracting a keyword for text data extracted by the first extraction means. The second extraction unit performs morphological analysis on the text data extracted by the first extraction unit, and extracts a keyword from morphemes of predetermined parts of speech among morphemes obtained as an analysis result. The information processing apparatus according to 1. The first extracting means acquires text data representing the content of at least one of a news program and a news program included in the EPG data acquired at the first timing, and acquires at the second timing The information processing apparatus according to claim 1, wherein text data representing the content of at least one of a news program and a news program included in the EPG data is compared for each text data representing the content of the same program. The first extraction means starts to provide content that represents the content of the same content as the text data representing the content of the content included in the EPG data provided on the day before the content is provided. 2. The text data included in the EPG data provided on the day of the day of comparison is compared, and the text data updated in the EPG data provided on the day of the day on which the provision of the content is started is extracted. Information processing device. The first extraction means further starts providing the content as representing the same content as the text data included in the first EPG data provided on the day of the start of the content provision. The text data included in the second EPG data provided at a time later than the time when the first EPG data on the day of the day is provided is compared, and the text data updated in the second EPG data is compared. The information processing apparatus according to claim 4, wherein the information processing apparatus is extracted. When the EPG data is provided by adding a predetermined symbol to the updated text data among the text data representing the content content,
The first extraction means extracts the text data extracted from the EPG data acquired at the first timing based on the symbol and the EPG data acquired at the second timing based on the symbol. The information processing apparatus according to claim 1, wherein the text data is compared for each text data representing details of the same content. Obtain EPG data that contains information about each content that will be provided when the specified time arrives,
The same text data representing the content of the content included in the EPG data acquired at the first timing and the text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing. Compared to each text data representing the content content, the text data updated in the EPG data acquired at the second timing is extracted,
An information processing method including a step of extracting keywords from extracted text data. Obtain EPG data that contains information about each content that will be provided when the specified time arrives,
The same text data representing the content of the content included in the EPG data acquired at the first timing and the text data representing the content of the content included in the EPG data acquired at the second timing later than the first timing. Compared to each text data representing the content content, extract the text data updated in the EPG data acquired at the second timing,
A program that causes a computer to execute processing including a step of extracting keywords from extracted text data.

Images