JP5225418B2 - Information processing apparatus and method - Google Patents

Description

  Embodiments described herein relate generally to an information processing apparatus and method.

  The number of programs that can be viewed is increasing due to digital terrestrial and multi-channel. In addition, the advancement of hard disk recorders makes it possible to record a large number of programs and watch them at any time. However, due to the diversification of entertainment, the time available for viewing the program has decreased, and in parallel with “Tsumami” to see only the scene that interests you by fast-forwarding the video and doing something else in parallel. The number of users who “watch while watching” only the places of interest tends to increase. In addition, since a large amount of programs can be browsed, zapping tends to increase. Zapping indicates an action in which a user switches display programs one after another while viewing a program.

  As described above, user viewing forms are diversified, and in order to answer such user needs, a technique for appropriately navigating a specific program from a large number of programs has been proposed.

JP 2007-166363 A

  Patent Document 1 discloses that the entire content range is divided into chapters, and the score of each divided chapter is calculated. Patent Document 1 discloses that a chapter most recommended to the user is selected and sequentially reproduced based on the calculated score for each chapter. However, in the technique disclosed in Patent Document 1, it is necessary to divide the entire range of content into a plurality of chapters, that is, a plurality of scenes, and to calculate scores for all the divided scenes. Met.

  The information processing apparatus according to the embodiment includes a first storage unit, a second storage unit, a creation unit, a first calculation unit, a second calculation unit, an instruction unit, and a selection unit. A 1st memory | storage means matches and memorize | stores a program and the character information which shows the content of the said program. The second storage means stores viewing history information indicating the viewing history of the program. The creating means extracts a keyword related to the program from the character information, and creates a keyword index indicating the appearance position of the keyword in the program. The first calculation means calculates a keyword score indicating a degree of user interest in each of the keywords based on the viewing history information. The second calculating means sets a specific range in the program as a scene based on the keyword index and the keyword score, and calculates a scene score indicating a degree of interest of the user with respect to the scene based on the keyword score. . The instruction means receives a switching instruction indicating switching of the program or the scene from the user. The selection means, when receiving the switching instruction, selects the scenes in descending order of the scene score and displays them on the display unit.

1 is a block diagram illustrating an information processing apparatus according to an embodiment. The figure which shows an example of the metadata of embodiment. The figure which shows an example of the keyword index of embodiment. The figure which shows an example of the keyword score of embodiment. The figure which shows an example of the viewing history information of embodiment. The flowchart which shows an example of the zapping pre-process of embodiment. The flowchart which shows an example of the keyword index creation process of embodiment. The flowchart which shows an example of the keyword score calculation process of embodiment. The flowchart which shows an example of the scene score calculation process of embodiment. The figure which shows an example of the appearance position of the preference keyword of embodiment. The figure which shows an example of the scene range and scene score of embodiment. The flowchart which shows an example of the zapping process of embodiment. The figure of the hardware constitutions of an embodiment.

  Hereinafter, an example of an information processing apparatus and method according to the present embodiment will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing a functional configuration of the information processing apparatus according to the present embodiment. As shown in FIG. 1, the information processing apparatus 10 according to the present embodiment includes a receiving unit 12, a program storage unit 14 (first storage means), a metadata extraction unit 16, a program metadata storage unit 18, and a keyword index creation unit. 20 (creating unit), keyword index storage unit 22, scene selection unit 24 (selection unit), display unit 26, viewing history collection unit 28, viewing history storage unit 30 (second storage unit), keyword score calculation unit 32 (first 1 calculation means), a keyword score storage section 34, a scene score calculation section 36 (second calculation means), a scene score storage section 38, and an operation section 40 (instruction means).

  The receiving unit 12 receives content data. Examples of the receiving unit 12 include a TV tuner that receives broadcast radio waves from a broadcast station (not shown) as content data, a network interface that receives content data from the Internet, and the like.

  The content data is data including a program and metadata indicating the content of the program. Examples of programs include TV broadcast programs, recording media such as DVDs, movies and video clips distributed, sold and distributed in the form of VOD services, moving images distributed on the Web, cameras and mobile phones. Examples include moving images taken, video recorders, HDD recorders, DVD recorders, recorded programs recorded on a TV / PC with a recording function, and the like.

  Metadata is data indicating the content of a program, and in the present embodiment, is metadata including character information indicating at least the content corresponding to each position in the program. In the present embodiment, each position in the program metadata is indicated by time information. Specifically, in the present embodiment, the metadata includes a program ID for identifying a program, a broadcast date of the program (a date received by the receiving unit 12), a start time and an end time in the program. And character information indicating contents between the start time and the end time in the program.

  In the present embodiment, a case will be described in which the character information is closed captions (captions obtained by converting program speech into character information).

  In the present embodiment, the case where the receiving unit 12 receives content data including a program and metadata will be described, but the receiving unit 12 may receive the program and metadata at different timings. In this embodiment, the case where content data including a program and metadata is received by one receiving unit 12 will be described. However, the program and metadata may be received by separate receiving units. . In this case, for example, the information processing apparatus 10 is configured to include both a TV tuner and a network interface as the receiving unit 12. Then, the program may be received by the TV tuner and the metadata may be received by the network interface.

  As described above, the metadata may be data including at least character information indicating contents corresponding to each position in the program, and the character information is not limited to the closed caption as described above. For example, the character information may be EPG (Electronic Program Guide) or RSS (Rich Site Summary).

  FIG. 2 shows an example of the data structure of the metadata in the present embodiment. 2A to 2C, each row indicating a program ID, a broadcast date, a start time, an end time, and a closed caption corresponds to one metadata. That is, one or more metadata is associated with each program. For example, the metadata shown in the first line of FIG. 2A includes a program ID “500” and a program broadcast date of September 3, 2010 from the start time 6:00 to the end time 6 in the program. The closed caption, which is the text information indicating the contents of 4 seconds up to 4 minutes, indicates that the party leader's debate sponsored by the press club was held. The program ID is identification information for uniquely identifying a program.

  Further, the metadata shown in the first line of FIG. 2B includes the program ID “820” and the program broadcast date September 5, 2010 from the start time 52 minutes 30 seconds to the end time 52 in the program. The closed caption, which is text information indicating the content of 10 seconds up to 40 minutes, indicates that “Today's electronic book terminal will be introduced today”. Similarly, the metadata shown in the first line of FIG. 2C includes the program ID “1000” and the program broadcast date September 10, 2010 from the start time 1 hour 20 minutes 35 seconds in the program. The closed caption, which is text information indicating the content of 5 seconds until the end time 1 hour 20 minutes 40 seconds, indicates that “player B visits a village in India”.

  Returning to FIG. 1, the program storage unit 14 is a storage medium such as a hard disk drive (HDD), and includes a program included in the content data received by the receiving unit 12, a program ID for identifying the program, and metadata. Are stored in association with each other.

  Note that the content data received by the receiving unit 12 is sequentially stored in the program storage unit 14. For this reason, the program storage unit 14 stores a plurality of content data, that is, a plurality of programs and one or a plurality of metadata corresponding to each program.

  The metadata extraction unit 16 extracts metadata corresponding to each program stored in the program storage unit 14 and stores it in the program metadata storage unit 18. The program metadata storage unit 18 is a storage medium such as a hard disk drive (HDD). The program metadata storage unit 18 stores, for example, metadata as shown in FIG.

  Returning to FIG. 1, the keyword index creation unit 20 decomposes each closed caption included in each of the metadata corresponding to all the programs stored in the program metadata storage unit 18 into a plurality of keywords. Specifically, for example, the keyword index creating unit 20 performs a morphological analysis on each closed caption, thereby decomposing the closed caption into a column of morphemes (the smallest unit having meaning in the language). Then, the keyword index creation unit 20 extracts morphemes that are nouns or proper nouns from among the plurality of morphemes as keywords. As a result, the keyword index creating unit 20 decomposes the closed caption included in each metadata into a set of a plurality of keywords.

  By this morphological analysis, the keyword index creating unit 20, for example, from the closed caption “The Party Leader Discussion Hosted by the Press Club” of the program ID “500” shown in FIG. And “sales” are extracted as keywords. Similarly, the keyword index creating unit 20 extracts keywords from closed captions included in other metadata.

  Then, the keyword index creation unit 20 creates a keyword index based on each extracted keyword. The keyword index is a table in which keywords are associated with index information. In the present embodiment, the index information indicates a program ID and a keyword appearance position (start time) in the program. As this appearance position, in the present embodiment, a closed caption start time including each keyword is used. This start time may be read from the start time included in the metadata.

  FIG. 3 shows an example of the data structure of the keyword index. As shown in FIG. 3, index information corresponding to each keyword is registered in the keyword index.

  For example, the closed caption of certain metadata including the keyword “electronic book” is “electronic book sales are increasing.” The start time of this closed caption is 7 minutes 11 seconds of the program ID “500”. A case will be described (refer to the metadata in the ninth line in FIG. 2A). In this case, the keyword index creating unit 20 registers index information indicating that the keyword “electronic book” appears at 7 minutes and 11 seconds of the program ID “500” in the keyword index (in FIG. 3, ( 500, 0:07:11)).

  In addition, when there are a plurality of closed captions including the keyword “electronic book”, the keyword index creating unit 20 registers index information based on each of the plurality of closed captions in the keyword index. For example, in the example shown in FIGS. 2A to 2C, the keyword “electronic book” has a closed caption “Many terminals that read electronic books are mobile phones,” and closed caption “ , I will introduce the topic of electronic book terminals. ”(The above-mentioned metadata in the 9th and 10th lines in FIG. 2A and the meta data in the 1st line in FIG. 2B). Data reference). Therefore, in this case, the keyword index creation unit 20 determines that the keyword “electronic book” is the program ID “500” at 7 minutes 11 seconds, 7 minutes 15 seconds, and the program ID “820” is 52 minutes 30 seconds. Index information indicating that each of the eyes appears is registered in the keyword index ((500,0: 07: 11), (500,0: 07: 15), (820,0: 52: in FIG. 3). 30)).

  By creating such a keyword index, the information processing apparatus 10 according to the present embodiment is in a state where the appearance position of each keyword can be searched at high speed. For example, by using the keyword index shown in FIG. 3, the keyword “electronic book” has the program ID “500” at 7 minutes 11 seconds, 7 minutes 15 seconds, and the program ID “820” at 52 minutes 30 seconds. It will be in the state in which it can search at high speed that it will appear in a location. Also, by using the keyword index, the scene score calculator 36, which will be described in detail later, can calculate the scene score at high speed.

  Returning to FIG. 1, the keyword index storage unit 22 is a storage medium such as a hard disk drive (HDD), and stores the keyword index created by the keyword index creation unit 20.

  The keyword score calculation unit 32 calculates a keyword score for each keyword stored in the keyword index storage unit 22. The keyword score is a numerical value assigned to a keyword. The larger the value, the greater the user's interest and degree of interest in the keyword. The user indicates an operator who operates the information processing apparatus 10.

  The keyword score storage unit 34 is a storage medium such as a hard disk drive (HDD), and stores a correspondence table that associates keyword scores with keywords.

  FIG. 4 shows an example of a correspondence table between keyword scores and keywords. Based on the keyword scores shown in the correspondence table shown in FIG. 4, the degree of user interest in keywords such as “electronic book”, “Company A”, “World Cup”, “Soccer” is higher than other keywords. I understand.

  As a keyword score calculation method in the keyword score calculation unit 32, various methods may be mentioned. In the present embodiment, the keyword score calculation unit 32 assigns a higher keyword score to a keyword that appears more frequently in the closed caption corresponding to the viewing range of the program. Details of the keyword score calculation method will be described later.

  Returning to FIG. 1, the operation unit 40 receives a signal indicating a program to be displayed next and a signal indicating a zapping instruction (switching instruction). The operation unit 40 is provided with a button (not shown) for instructing a program to be displayed next by the user and a button (not shown) for instructing zapping. The operation unit 40 receives a signal indicating a zapping instruction or a signal indicating a program to be displayed next in response to an operation instruction of these buttons on the operation unit 40 by the user. The operation unit 40 that has received these signals transmits these signals to the scene selection unit 24.

  Note that zapping refers to an action in which a user viewing a program switches display programs one after another. In the present embodiment, the zapping instruction indicates an instruction to switch the program or scene displayed on the display unit 26.

  When the scene selection unit 24 receives a signal indicating the next program to be displayed from the operation unit 40, the scene selection unit 24 reads the program from the program storage unit 14 based on the received signal and displays the program on the display unit 26. In addition, whenever the scene selection unit 24 receives a signal indicating a zapping instruction from the operation unit 40, the scene selection unit 24 displays the scenes in the descending order of the scene score stored in the scene score storage unit 38 (details will be described later). ).

  The display unit 26 is a device that displays a program, and examples thereof include an LCD (liquid crystal display).

  The viewing history collection unit 28 creates viewing history information of the program displayed on the display unit 26. The viewing history storage unit 30 is a storage medium such as a hard disk drive (HDD), and stores the viewing history information collected by the viewing history collection unit 28.

  The viewing history information is data indicating which viewing range of which program the user has viewed. Specifically, the viewing history information includes the program ID of the viewed program, the broadcast date of the program (the date received by the receiving unit 12), the start time as the viewing start time of the program, and the viewing of the program. And an end time as an end time. Note that the start time and end time correspond to the viewing range.

  FIG. 5 shows an example of the data structure of the viewing history information. For example, in the viewing history information shown in the first line of FIG. 5, the program ID “200” and the program broadcast on September 3, 2010, from 10 minutes 50 seconds to 28 minutes 20 seconds by the user. Indicates that it was watched.

  In the viewing history collection unit 28, the viewing history information is created by constantly monitoring the program displayed on the display unit 26 and stored in the viewing history storage unit 30.

  Returning to FIG. 1, the scene score calculation unit 36 stores the keyword score of each keyword stored in the keyword score storage unit 34, the metadata stored in the program metadata storage unit 18, and the keyword index storage unit 22. From the keyword index, one or more specific ranges of each program are defined as a scene, and the scene score of the determined scene is calculated. The scene score indicates the user's interest and degree of interest in a specific range defined as a scene in the program. Therefore, the larger the scene score value, the greater the user's interest and degree of interest in the scene. Details of the method for setting the “scene” of each program by the scene score calculation unit 36 and the method for calculating the scene score will be described later.

  The scene score storage unit 38 is a storage medium such as a hard disk drive (HDD) and stores the program scenes determined by the scene score calculation unit 36 and the scene scores corresponding to each scene in association with each other.

  Next, the zapping pre-processing executed by the information processing apparatus 10 according to the present embodiment will be described.

  FIG. 6 is a flowchart illustrating a procedure of pre-zapping processing executed by the information processing apparatus 10.

  In the present embodiment, before the zapping pre-process shown in FIG. 6 is executed, the program storage unit 14 includes a program included in each of the one or more pieces of content data received by the receiving unit 12. It is assumed that the metadata is stored.

  Each time the content data is received, the receiving unit 12 may store the received content data in the program storage unit 14, or may store only the content data reserved for recording in advance in the program storage unit 14. Good. When content data reserved for recording in advance is stored in the program storage unit 14, for example, information indicating channel information, broadcast start time, and broadcast end time of content data to be reserved for recording is stored in the memory of the reception unit 12 in advance. Remember. Then, the reception unit 12 may receive the stored channel information content data by the reception unit 12 and store it in the program storage unit 14 when the broadcast start time stored in advance is reached. For this recording reservation, for example, a button that is instructed by the user when making a recording reservation in advance on the operation unit 40 is prepared. Then, the recording reservation may be performed by operating this button by the user.

  Next, the metadata extraction unit 16 extracts metadata corresponding to each program stored in the program storage unit 14 (step S104). The program metadata storage unit 18 stores the metadata extracted by the metadata extraction unit 16 (step S106).

  By executing the processing of Step S104 and Step S106, the program metadata storage unit 18 stores, for example, the metadata shown in FIG.

  Next, the keyword index creation unit 20 executes keyword index creation processing (step S108). Then, after the keyword score calculation unit 32 executes the keyword score calculation process (step S110), the scene score calculation unit 36 executes the scene score calculation process (step S112) and ends this routine.

  Next, details of the keyword index creation process (step S108), the keyword score calculation process (step S110), and the scene score calculation process (step S112) will be described.

  First, details of the keyword index creation process in step S108 will be described. The keyword index creation unit 20 executes keyword index creation processing for each program based on one or more metadata corresponding to the program.

  FIG. 7 shows a flowchart showing the keyword index creation processing executed by the keyword index creation unit 20.

  First, the keyword index creating unit 20 determines whether there is a program that has not yet been indexed among the programs stored in the program metadata storage unit 18 (step S200). The determination in step S200 is performed, for example, by determining whether or not data indicating indexed is stored in association with the program ID stored in the program metadata storage unit 18. In this case, for example, each time the keyword index creating unit 20 executes the keyword index creating process (step S201 to step S204 described later) for each program, the index is associated with the program ID of the program for which the keyword index is performed. Information indicating conversion into a program is stored in the program metadata storage unit 18. Then, the keyword index creating unit 20 determines whether or not information indicating indexed corresponding to the program ID of the metadata stored in the program metadata storage unit 18 is stored in the program metadata storage unit 18. By determining, it may be determined whether or not there is an unindexed program.

  If it is determined that there is no program that has not been indexed (step S200: No), this routine is terminated (step S200: No).

  On the other hand, if it is determined that there is an unindexed program (step S200: Yes), the keyword index creating unit 20 selects all the programs corresponding to the program ID of one of the unindexed programs. The metadata is read (step S201). Specifically, the keyword index creating unit 20 selects one of one or a plurality of program IDs that are not associated with information indicating indexed, and programs all the metadata including the selected program IDs. Read from the metadata storage unit 18.

  For example, when the program ID that is not associated with the indexed information is “500”, the keyword index creating unit 20 displays the meta data including the program ID “500” shown in FIG. All of the data is read from the program metadata storage unit 18.

  Then, the keyword index creation unit 20 performs morphological analysis on the closed captions included in each of the metadata read in step S201 (step S202).

  By this morphological analysis, the keyword index creating unit 20 decomposes the closed caption included in the metadata read in step S201 into a set of keywords.

  Then, the keyword index creation unit 20 creates a keyword index based on each keyword obtained by the morpheme analysis and stores it in the keyword index storage unit 22 (step S204). Then, the keyword index creation unit 20 repeats and executes the processes of steps S201 to S204 until a negative determination is made in step S200 (step S200: No).

  Next, the details of the keyword score calculation process in step S110 will be described. This keyword score calculation process is executed by the keyword score calculation unit 32.

  The keyword score calculation unit 32 calculates a keyword score for each keyword stored in the keyword index storage unit 22. There are various methods for calculating the keyword score.

  The simplest keyword score calculation method is a method of counting the number of keywords in closed captions included in the viewing range of the program, and giving a higher keyword score to a keyword that appears more frequently in the viewing range. It is done. However, in this method, there are cases where the user's interest and keywords with low interest are also counted.

  Therefore, in the present embodiment, the keyword score calculation unit 32 executes the keyword score calculation process shown in FIG. 8 as the keyword score calculation process in step S110. The keyword score calculation process shown in FIG. 8 is a process of calculating the keyword viewing coverage as a keyword score. The viewing coverage is a value obtained by dividing the number of appearances in closed captions corresponding to the viewing range in the program by the number of appearances in closed captions in the entire range of the program for each keyword.

  Specifically, as shown in FIG. 8, the keyword score calculation unit 32 first has keywords whose keyword score has not been calculated among the keywords registered in the keyword index stored in the keyword index storage unit 22. Whether or not (step S300).

  The determination in step 300 can be made, for example, by determining whether or not there is a keyword whose keyword score is not registered for the keyword index in the keyword index storage unit 22.

  If the keyword score has been calculated for all keywords (step S300: No), the keyword score calculation unit 32 ends this routine.

  On the other hand, when there is a keyword whose keyword score has not been calculated (step S300: Yes), the keyword score calculation unit 32 sets one of the uncalculated keywords as a keyword score calculation target (step S302).

  Next, the keyword score calculation unit 32 counts the number of appearances of the keyword set in step S302 in the closed caption corresponding to the entire range of all the programs stored in the program storage unit 14 (step S304). Specifically, the keyword score calculation unit 32 is based on the keyword index stored in the keyword index storage unit 22, and index information associated with the keyword set in step S302 registered in the keyword index. The number of occurrences is counted by counting the number of.

  For example, assume that the keyword “electronic book” is set in step S302. Assume that the keyword index shown in FIG. 3 is stored in the keyword index storage unit 22. In this case, the index information corresponding to the keyword “electronic book” is three. Therefore, the keyword score calculation unit 32 counts “3” as the number of appearances of the keyword “electronic book” in the closed caption corresponding to the entire range of all programs of the keyword “electronic book” from the keyword index.

  Next, the keyword score calculation unit 32 counts the number of occurrences of the keyword set in step S302 in the closed caption corresponding to the viewing range (step S306).

  In step S306, the keyword score calculation unit 32 determines the keywords in the viewing range based on the viewing history information stored in the viewing history storage unit 30 and the keyword index stored in the keyword index storage unit 22. Count the number of occurrences of

  For example, assume that the keyword “electronic book” is set in step S302. Then, it is assumed that the keyword index storage unit 22 stores the keyword index shown in FIG. 3 and the viewing history storage unit 30 stores the viewing history information shown in FIG. In this case, the keyword score calculation unit 32 is the three index information of the keyword “electronic book” from the keyword index and the viewing history information (500, 0:07:11), (500, 0: 07:15) and (820,0: 52: 30), (500,0: 07: 11) and (500,0: 07: 15) are determined to be viewed, and (820,0: 52) : 30) is determined not to be viewed. The keyword score calculation unit 32 then counts “2”, which is the number of index information determined to be viewed, as the number of appearances in the closed caption corresponding to the viewing range of the keyword “electronic book”.

  Next, the keyword score calculation unit 32 calculates a keyword score by dividing the count calculated in step S306 by the count calculated in step S304 (step S308).

  For example, the keyword score as the viewing coverage of the keyword “electronic book” is 0.67, which is a value obtained by dividing the above “2” by “3”.

  Then, the keyword score calculation unit 32 stores the keyword score calculated in step S308 in the keyword score storage unit 34 in association with the corresponding keyword (step S310). Next, the keyword score calculation unit 32 repeatedly executes the processes in steps 302 to S310 until a negative determination is made in step 300.

  The keyword score calculation part 32 can calculate a keyword score about all the keywords registered in the keyword index memory | storage part 22 by performing the keyword score calculation process shown to said step S300-step S310.

  In addition, the keyword score calculation process performed in the keyword score calculation part 32 in this Embodiment is not restricted to the method of calculating viewing coverage as a keyword score as demonstrated above.

  For example, the keyword score calculation unit 32 may calculate the keyword score of each keyword using a method using machine learning. When calculating a keyword score using a method using machine learning, for example, data indicating user preferences may be collected and analyzed in advance, and a keyword score may be calculated for each keyword from the analysis result. . The data indicating the user's preference may be input by the user in advance according to an operation instruction of the operation unit 40, or may be collected from a web server or the like.

  Further, an arbitrary keyword score corresponding to each keyword may be registered manually by the user. Further, a method of calculating the viewing coverage by the keyword score calculation unit 32 as a keyword score may be combined with manual keyword score registration. When the method for calculating the viewing coverage as a keyword score and the method for manually registering an arbitrary keyword by the user are combined, the keyword score of the keyword manually registered by the user is obtained by the viewing coverage. A value higher than the obtained value (for example, 1.0 in the example shown in FIG. 4) may be used.

  Next, details of the scene score calculation process in step S112 will be described. This scene score calculation process is executed by the scene score calculation unit 36.

  FIG. 9 is a flowchart showing a scene score calculation process executed by the scene score calculation unit 36.

  In the scene score calculation unit 36, first, a keyword whose keyword score value is equal to or greater than a predetermined threshold is extracted from the keyword score storage unit 34 as a preference keyword (step S500).

  For example, it is assumed that the keyword score shown in FIG. 4 is stored in the keyword score storage unit 34 in association with each keyword. It is assumed that 0.8 is set in advance as the threshold value. In this case, the scene score calculation unit 36 performs “electronic book”, “Company A”, “World Cup”, “Soccer”, “Soccer” from among all the keywords shown in FIG. “Japan representative” is extracted as a preference keyword.

  Next, the scene score calculation unit 36 reads index information (program ID and appearance position) corresponding to the preference keyword extracted in step S500 from the keyword index storage unit 22 (step S502).

  For example, when the preference keyword is “electronic book” and the keyword index shown in FIG. 3 is stored in the keyword index storage unit 22, the scene score calculation unit 36 calculates (500, 0:07:11). , (500,0: 07: 15) and (820,0: 52: 30) are read from the keyword index.

  Note that in the process of step S502, there are a plurality of index information corresponding to the same preference keyword, and index information indicating the similar appearance position (for example, the time difference is within 30 seconds) with the same program ID is included in the plurality of index information. If it is included, it is preferable to read the index information with the earliest appearance position in the index information. For example, (500,0: 07: 11) and (500,0: 07: 15) of the three index information corresponding to the preference keyword “electronic book” have an appearance position close to 4 seconds. Therefore, in this case, the scene score calculation unit 36 reads (500, 0:07:11) and (820, 0:52:30) in step S502.

  In addition, in the process of step S502, for different preference keywords such as “World Cup” and “Soccer”, the index information indicating the same program ID and the close appearance position (for example, the time difference is within 30 seconds) It is preferable to read index information with the earliest appearance position.

  For example, in the example shown in FIG. 3, the index information of the preference keyword “soccer” (500, 0: 06:30) and the index information of the preference keyword “world cup” (500, 0: 06:30). And (500, 0:06:40), which is the index information of the preference keyword “Japan representative”, is index information indicating the same appearance ID and the nearest appearance position. In this case, the index information (500, 0:06:30) with the earliest appearance position and the preference keyword “soccer” are read out of the combination of the three preference keywords and the index information.

  Then, the scene score calculation unit 36 similarly extracts index information for other preference keywords.

  Through the processing in step S502, the scene score calculation unit 36 extracts, for all programs stored in the program storage unit 14, preference keywords related to each program and information indicating the appearance position of each preference keyword in the program. To do.

  FIG. 10 schematically shows the appearance position of the preference keyword in the entire range of the program. For example, by the processing in step S502, the scene score calculation unit 36, as shown in FIG. 10, the preference keyword “soccer” in the entire program ID 500 range (see the program ID time axis 50 in FIG. 10). And the appearance position of each “electronic book” (see the appearance position 50A and the appearance position 50B in FIG. 10). In addition, the scene score calculation unit 36 has an appearance position of the preference keyword “electronic book” in the entire range of the program with the program ID 820 (see the program ID time axis 52 in FIG. 10) (see the appearance position 52A in FIG. 10). ). In addition, the scene score calculation unit 36 shows the appearance position of the preference keyword “soccer” in the entire range of the program with the program ID 1000 (see the time axis 54 of the program ID in FIG. 10) (see the appearance position 54A in FIG. 10). Read.

  Next, the scene score calculation unit 36 sets a predetermined specific range including the appearance position of each preference keyword in each program as a scene for which a scene score is to be calculated (step S504).

  As the length of the predetermined specific range, for example, a time such as 30 seconds may be determined in advance, or it may be a range of one closed caption corresponding to the appearance position, or the closed caption. And a range of a plurality of closed captions (for example, five) continuous to the closed captions. However, from the viewpoint of improving the accuracy of the scene score of each scene, it is preferable to set a certain upper limit to the length of the specific range set as the scene score calculation target scene.

  The specific range set as the scene may be a predetermined specific range including the appearance position of each preference keyword. For example, the specific range set as a scene may be a specific range having the appearance position as a start position, or may be a specific range having the appearance position as an end position. In the present embodiment, a case where a specific range starting from the appearance position is set as a scene will be described.

  FIG. 10 shows a case where the specific range 51A starting from the appearance position 50A of the preference keyword “soccer” in the entire program ID 500 range (see the time axis 50) is set as a scene. FIG. 10 shows a case where the specific range 51B starting from the appearance position 50B of the preference keyword “electronic book” in the entire range of the program with the program ID 500 (see the time axis 50) is set as a scene. .

  Similarly, FIG. 10 shows the specific range 53A starting from the appearance position 52A of the preference keyword “electronic book” in the entire range of the program with the program ID 820 (see the time axis 52) and the entire range of the program with the program ID 1000. The case where the specific range 55A starting from the appearance position 54A of the preference keyword “soccer” (see the time axis 54) is set as a scene is shown.

  As described above, in the information processing apparatus 10 according to the present embodiment, the specific range in the program, not the entire range of the program, can be set as the scene score calculation scene by the processing in step S504.

  Next, the scene score calculation unit 36 associates the range of the scene set in step S504 with the program ID of the program including the scene, and stores it in the scene score storage unit 38 (step S506). In the present embodiment, the scene score calculation unit 36 stores information indicating the start position and end position of the scene as the range of the scene.

  Next, the scene score calculation unit 36 determines whether or not there is a scene for which no scene score has been calculated for all the scenes stored in step S506 (step S508). Then, the processes in steps S510 to S518 are repeated until it is determined that no scene score has not been calculated (step S508: No).

  If the scene score calculation unit 36 determines that there is a scene whose scene score has not been calculated (step S508: Yes), first, one of the scenes whose scene score has not been calculated is set as a scene score calculation target (step S510). .

  Next, the scene score calculation unit 36 extracts the closed caption of the scene set as a scene score calculation target in step S510, and decomposes the closed caption into a set of keywords by performing morphological analysis (step S512).

  In step S512, the closed caption is extracted from, for example, the program ID of the scene and the scene range (the start position and the end position of the scene in the program), and the closed caption corresponding to the program ID and the scene range is obtained. This may be done by reading from the program metadata storage unit 18.

  For example, when the range of 30 seconds starting from 7 minutes 11 seconds, which is the appearance position of the preference keyword “electronic book” of the program ID “500”, is set as the scene score calculation target scene, the scene score calculation unit 36 As closed captions, as shown in Figure 2 (A) above, “Electronic books are growing in sales.” “Many devices that read electronic books are mobile phones,” and “New products will appear one after another. And the market is expected to expand. ”

  Note that the morphological analysis in step S512 may be performed in the same manner as in step S202.

  Next, the scene score calculation unit 36 reads the keyword score corresponding to each keyword obtained in step S512 from the keyword score storage unit 34 (step S514). For keywords for which no keyword corresponding to the keyword score storage unit 34 is registered, “0” may be read as the keyword score value.

  Next, the scene score calculation unit 36 calculates the total value of the keyword scores of all the keywords included in the scene as a scene score (step S516).

  For example, in the above step S512, as described above, “Electronic books are increasing in sales” and “Many terminals that read electronic books are mobile phones,” as shown in FIG. , New products appear one after another, and the market is expected to expand. " In this case, the scene score calculation unit 36 includes keywords “electronic book”, “sales”, “electronic book”, “terminal”, “mobile phone”, “new product” included in each of these closed captions. ”And 3.0 of the total value (0.9 + 0.0 + 0.9 + 0.0 + 0.6 + 0.6 + 0.0) of the keyword scores corresponding to each of“ market ”are calculated as the scene score (see FIG. 10).

  As another example, the scene score calculation unit 36 sets a range of 30 seconds starting from 6 minutes 30 seconds as the appearance position of the preference keyword “soccer” of the program ID “500” as a scene score calculation target scene, As the closed caption corresponding to the scene, as shown in Fig. 2 (A) above, "Next is the breaking news of the Soccer World Cup", "Japan National Team has entered the final tournament before dawn on the 25th of Japan time." , And "I will fight with England." In this case, the scene score calculation unit 36 includes the keywords “soccer”, “world cup”, “breaking news”, “Japan national team”, “Japan time”, “final” that are keywords included in each of these closed captions. And 2.5 of the total value (0.8 + 0.9 + 0.0 + 0.8 + 0.0 + 0.0 + 0.0) of the keyword scores corresponding to each of the “tournament” is calculated as the scene score of the scene (see FIG. 10). ).

  As another example, the scene score calculation unit 36 sets a range for 30 seconds, which is the appearance position of the preference keyword “soccer” with the program ID “1000”, as a scene score calculation target scene, and corresponds to the scene. As a closed caption to be read, it is assumed that “the children have been given a soccer ball and enjoyed soccer together” shown in FIG. In this case, the scene score calculation unit 36 adds the keyword score total values (0) corresponding to the keywords “children”, “soccer ball”, “present”, and “soccer” included in this closed caption. .0 + 0.0 + 0.0 + 0.8) is calculated as the scene score of the scene (see FIG. 10).

  Next, the scene score calculation unit 36 stores the calculated scene score in the scene score storage unit 38 in association with the scene range of the scene score and the program ID (step S518).

  By the processing in step S518, the scene score storage unit 38 stores a program score and a scene score corresponding to the scene range. FIG. 11 shows an example of the data configuration of the program ID, scene range, and scene score stored in the scene score storage unit 38.

  As shown in FIGS. 10 and 11, the 30-second scene (specific range 55A) starting from the preference keyword “soccer” of the program ID “1000” is related to the preference keyword “soccer” but related to sports such as the World Cup. There is no content of (see also FIG. 2C). Therefore, the scene score of the scene with the program ID “1000” (specific range 55A) is lower than the scene with the program ID “500” (specific range 51A) starting from the same preference keyword “soccer”. For this reason, in this Embodiment, it can be said that a user's interest and a scene with high interest can be provided correctly at the time of the zapping process mentioned later.

  In the present embodiment, the scene score calculation unit 36 calculates the total value of the keyword scores of all keywords included in each scene as a scene score, but using another calculation method using the keyword score. A scene score may be calculated.

  Next, a zapping process when a zapping instruction is input by operating the operation unit 40 will be described.

  FIG. 12 shows a flowchart showing the zapping process performed in the information processing apparatus 10.

  The scene selection unit 24 determines whether a zapping instruction has been input (step S600). The determination in step S600 can be made by determining whether or not a signal indicating a zapping instruction is received from the operation unit 40. The signal indicating the zapping instruction is output from the operation unit 40 to the scene selection unit 24 when the user operates a button for instructing zapping provided on the operation unit 40. If it is determined that a zapping instruction has been input (step S600: Yes), the process proceeds to step S602. on the other hand. If no zapping instruction has been input (step S600: No), this routine ends.

  Next, the scene selection unit 24 reads the scene range and program ID corresponding to the highest scene score among the scene scores stored in the scene score storage unit 38 (step S602). Then, the scene selection unit 24 reads the program corresponding to the program ID read in step S602 from the program storage unit 14 (step S604), and moves from the position corresponding to the start time of the scene range read in step S602 to the display unit 26. Display processing to be displayed is started (step S606).

  By executing the processing of step S602 to step S606, the display unit 26 displays the scene with the highest scene score on the display unit 26.

  Next, the scene selection unit 24 determines whether or not a zapping instruction has been input in the same manner as in step S600 (step S608). When the scene selection unit 24 determines that a zapping instruction has been input (step S608: Yes), the scene score next to the previously displayed scene among the scene scores stored in the scene score storage unit 38 is displayed. The high scene range and program ID are read (step S610). Then, the scene selection unit 24 reads the program corresponding to the program ID read in step S610 from the program storage unit 14 (step S612), and moves from the position corresponding to the start time of the scene range read in step S610 to the display unit 26. Display processing to be displayed is started (step S614).

  Next, the scene selection unit 24 determines whether or not an end instruction indicating the end of display on the display unit 26 has been input (step S616). The determination in step S616 can be made by determining whether or not a signal indicating display end has been received from the operation unit 40. The signal indicating the end of display is output from the operation unit 40 to the scene selection unit 24 when the user operates a button for instructing display end provided on the operation unit 40. If the scene selection unit 24 does not determine the input of the end instruction (step S616: No), the processes in steps S608 to S614 are repeated until the input of the end instruction is determined (step S616: Yes). Run.

  For this reason, every time a zapping instruction is input by operating the operation unit 40, scenes are displayed on the display unit 26 in order from the scene with the highest scene score.

  As described above, in the information processing apparatus 10 according to the present embodiment, the keyword index creation unit 20 extracts keywords from the closed captions, and creates a keyword index indicating the keywords and the appearance positions of the keywords in the program. Then, the keyword score calculation unit 32 calculates a keyword score for each keyword based on the viewing history information. Then, the scene score calculation unit 36 sets a specific range in each program as a scene based on the keyword index and the keyword score, and calculates the scene score of the set scene based on the keyword score.

  As described above, the information processing apparatus 10 according to the present embodiment creates a keyword index. The keyword index is used when setting a scene. Therefore, the scene can be set with a small amount of calculation. In addition, in the information processing apparatus 10 according to the present embodiment, a specific range of high interest of the user is selectively set as a scene instead of the entire program range, and the scene score of the set scene is calculated. For this reason, the calculation amount at the time of calculating the scene score can be reduced as compared with the conventional case.

  Therefore, the information processing apparatus 10 according to the present embodiment can sequentially provide scenes according to the user's interest and degree of interest with a small amount of calculation.

  Next, the hardware configuration of the information processing apparatus 10 according to the present embodiment will be described. FIG. 13 is a block diagram illustrating a hardware configuration example of the information processing apparatus 10 according to the present embodiment.

  The information processing apparatus 10 according to the embodiment includes a display unit 68 such as a display device controlled by the display unit 26, a receiving unit 70 such as a tuner, an operation unit 72, a CPU 60 (Central Processing Unit), and a ROM (Read Only Memory) 62. A RAM (Random Access Memory) 64, an HDD 66, and the like are connected to each other by a bus 67, and have a hardware configuration using a normal computer.

  The CPU 60 is an arithmetic device that controls the overall processing of the information processing apparatus 10. The RAM 64 stores data necessary for various processes by the CPU 60. The ROM 62 stores a program that realizes various processes by the CPU 60. The HDD 66 stores data stored in each storage unit described above. The receiving unit 70 is an interface for connecting to an external device or an external terminal via a communication line or the like and transmitting / receiving data to / from the connected external device or external terminal. The operation unit 72 corresponds to the operation unit 40 described above.

  A program for executing the various processes executed by the information processing apparatus 10 of the present embodiment is provided by being incorporated in advance in the ROM 62 or the like.

  The program executed by the information processing apparatus 10 according to the present embodiment is a file in a format that can be installed in the information processing apparatus 10 or an executable format, and is a CD-ROM, a flexible disk (FD), a CD-R, or a DVD. (Digital Versatile Disk) or the like may be provided by being recorded on a computer-readable recording medium.

  In addition, the program executed by the information processing apparatus 10 according to the present embodiment may be provided by being stored on a computer connected to a network such as the Internet and downloaded via the network. Further, a program for executing the above-described processes in the information processing apparatus 10 according to the present embodiment may be configured to be provided or distributed via a network such as the Internet.

  The program executed by the information processing apparatus 10 according to the present embodiment includes the above-described units (metadata extraction unit 16, keyword index creation unit 20, keyword score calculation unit 32, scene score calculation unit 36, scene selection unit 24, The module configuration includes a viewing history collection unit 28). As actual hardware, the CPU 60 reads out and executes various programs from the ROM 62 and the like, whereby the above-described units are loaded onto the main storage device, and the above-described functional configuration is generated on the main storage device. Yes.

  In the present embodiment, the configuration including the display unit 26 (display unit 68) is described as the information processing apparatus 10, but the configuration including the display unit 26 (display unit 68) as a separate body from the apparatus main body. It may be. In other words, the information processing apparatus 10 may be applied to a recording / reproducing apparatus that records contents (programs) in a readable and writable manner using a large-capacity storage medium such as an HDD, or may be applied to a television apparatus or a PC (Personal Computer). Also good.

  In addition, although several embodiment of this invention was described above, these embodiment is shown as an example and is not intending limiting the range of invention. These novel embodiments can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the scope of the invention. These embodiments and modifications thereof are included in the scope and gist of the invention, and are included in the invention described in the claims and the equivalents thereof.

DESCRIPTION OF SYMBOLS 10 Information processing apparatus 12 Receiving part 14 Program storage part 16 Metadata extraction part 18 Program metadata storage part 20 Keyword index preparation part 22 Keyword index storage part 24 Scene selection part 26 Display part 28 Viewing history collection part 30 Viewing history storage part 32 Keyword score calculation unit 34 Keyword score storage unit 36 Scene score calculation unit 38 Scene score storage unit 40 Operation unit

Claims (6)

First storage means for storing a program and character information indicating the content of the program in association with each other;
Second storage means for storing viewing history information indicating the viewing history of the program;
Creating means for extracting a keyword related to the program from the character information and creating a keyword index indicating an appearance position of the keyword in the program;
First calculation means for calculating a keyword score indicating a degree of user interest in each of the keywords based on the viewing history information;
Second calculation means for setting a specific range in the program as a scene based on the keyword index and the keyword score, and calculating a scene score indicating a degree of user interest in the scene based on the keyword score;
Instruction means for accepting a switching instruction indicating switching of the program or the scene from a user;
Selecting means for selecting the scenes in descending order of the scene score and displaying them on the display unit when the switching instruction is received;
An information processing apparatus comprising: The viewing history information includes viewing range information indicating a viewing range in the program,
The first calculation means calculates a value obtained by dividing the number of appearances of the keyword in the character information corresponding to the viewing range by the number of appearances of the keyword in the character information corresponding to the entire range of the program, The information processing apparatus according to claim 1, wherein the information processing apparatus calculates the keyword score.   The second calculation means determines a keyword corresponding to a keyword score equal to or higher than a predetermined threshold among the keywords as a preference keyword, and sets a specific range including the appearance position of the preference keyword in the program as the scene. The information processing apparatus according to claim 1 or 2.   The information processing apparatus according to claim 3, wherein the second calculation unit calculates the scene score based on the keyword in the character information corresponding to the scene.   The information processing apparatus according to claim 1, wherein the character information is closed captioning, EPG (Electronic Program Guide), or RSS (Rich Site Summary). An information processing method executed by an information processing apparatus,
The information processing apparatus includes a first storage unit that stores a program and character information indicating the content of the program, and a second storage unit that stores viewing history information indicating a viewing history of the program.
Creating means for extracting a keyword related to the program from the character information and creating a keyword index indicating the appearance position of the keyword in the program;
A first calculating means calculating a keyword score indicating a degree of user interest in each of the keywords based on the viewing history information;
The second calculation means sets a specific range in the program as a scene based on the keyword index and the keyword score, and calculates a scene score indicating the degree of user interest in the scene based on the keyword score. Steps,
When the selection unit receives a switching instruction indicating switching of the program or the scene, the step of selecting the scene in descending order of the scene score and displaying it on the display unit;
An information processing method comprising:

Images