“*3
`
`Ed
`
`TRANSPERFECT
`
`CERTIFICATION OF TRANSLATION
`
`I, Kayoko lmori, hereby certify that the attached Japanese to English translation
`has been translated by a qualified translator competent in both languages, and
`verified to be an accurate and complete rendering of the content of the original
`document to the best of our ability. The following documents are zncluded on this
`oer‘-.'*"ca: on.
`
`Japanese Unexamined Patent Application No. H11-7453
`
`Signature
`
`Sworn to before me this
`8"‘ day of October, 2014
`
`S
`
`
`No. 01DR6262048
`.
`' QUBIIIIOG in NEW YORK COUNIY
`
`
`" My Colssion W351’ 1 °6 _
`
`Stamp, Notary Public
`
`LANGUAGE AND TECHNOLOGY SOLUTIONS FOR GLOBAL BUSINESS
`
`THREE PARK AVENUE. 39TH FLOOR, NEW YORK. NY 10016 I T 212.689.5555 I F 212.689.1059 I WWW.TRANSPERFECT.COM
`OFFICES IN 80 CITIES WORLDWIDE
`APPIB 1005
`
`Apple 1005
`
`

`
`(19) Japan Patent Office (JP)
`
`(12) Japanese Unexamined Patent
`Application Publication (A)
`
`
`
`
` FI
`G06F 15/40
`H04N 5/44
`G06F 15/403
`H04N 5/782
`
`Identification codes
`
`
`
`(51) Int. Cl.6
`G06F 17/30
`H04N 5/44
` 5/7826
`
`(11) Japanese Unexamined Patent
`Application Publication Number
`H11-7453
`(43) Publication date: January 12, 1999
`
`
`370 G
` D
`340 A
` Z
`
`Request for examination: Not yet requested Number of claims: 23 OL (Total of 27 pages)
`
`(21) Application number
`
`
`Japanese Patent Application
`H10-67138
`
`(71) Applicant
`
`(22) Date of application
`
`(31) Priority claim number
`
`(32) Priority date
`(33) Priority country
`
`
`March 17, 1998
`
`Japanese Patent Application
`H9-104381
`April 22, 1997
`Japan (JP)
`
`
`(72) Inventor
`
`000006013
`Mitsubishi Electric Corporation
`2-2-3 Marunouchi, Chiyoda-ku, Tokyo-to
`Akira KOTANI
`℅ Mitsubishi Electric Corporation
`2-2-3 Marunouchi, Chiyoda-ku, Tokyo-to
`
`(74) Agent
`
`Patent Attorney Kenji YOSHIDA (and 2 others)
`
`
`
`
`
`
`MEDIA INFORMATION
`SOURCE 1
`
`NEWLY ARRIVED
`INFORMATION SOURCE 1
`
`MEDIA INFORMATION
`SOURCE 2
`
`NEWLY ARRIVED
`INFORMATION SOURCE 2
`
`ATTRIBUTE
`EXTRACTION
`PART 1
`
`ATTRIBUTE
`EXTRACTIO
`N PART 2
`
`MEDIA INFORMATION
`SOURCE n
`
`NEWLY ARRIVED
`INFORMATION SOURCE n
`
`ATTRIBUTE
`EXTRACTION
`PART n
`
`MEDIA
`INFORMATION
`EVALUATION
`PART
`
`MEDIA
`INFORMATION
`RECOMMENDATI
`ON PART
`
`INPUT/OUTPUT
`PROCESSING PART
`
`
`
`
`(54) (TITLE OF THE INVENTION) MEDIA INFORMATION RECOMMENDATION DEVICE
`
`(57) (ABSTRACT)
`(PROBLEM) To recommend information that coincides with a
`user’s preferences with high reliability by collectively
`managing an access history for a plurality of media
`information sources and more accurately evaluating the
`user’s degree of attention to the media information.
`SOLVING)
`The media
`information
`(MEANS
`FOR
`recommendation device comprises: respective attribute
`extraction parts 14 for automatically extracting keywords or
`the like from newly arrived information from newly arrived
`information sources 12 and creating attribute information;
`respective access monitoring parts 15
`for creating
`monitoring information based on attribute information
`corresponding to media information accessed by a user; a
`history information managing part 16 for updating history
`information with reference to a load value set for each media
`information source by a load information setting part 20
`based on monitoring information of the same format,
`regardless of the type of media information; and a media
`information evaluation part 18 for evaluating attribute
`information related to newly arrived information sent as a
`result of a comparison with the history information and
`displaying
`recommendation
`information;
`wherein
`information is recommended based on the user’s degree of
`attention to each piece of media information calculated
`based on the access history to a plurality of media
`information sources.
`
`MEDIA
`INFORMATION
`MONITORING
`PART 1
`
`MEDIA
`INFORMATION
`MONITORING PART
`2
`
`ACCESS
`MONITORING
`PART 1
`
`ACCESS
`MONITORING
`PART 2
`
`MEDIA
`INFORMATION
`MONITORING PART
`n
`
`ACCESS
`MONITORING
`PART n
`
`HISTORY
`INFORMATION
`MANAGING
`PART
`
`HISTORY
`INFORMATION
`STORAGE PART
`
`LOAD
`INFORMATION
`STORAGE PART
`
`LOAD SETTING
`PART
`
`MEDIA INFORMATION LOAD SETTING MEANS
`
`
`
`
`
`Apple 1005
`
`

`
`(SCOPE OF THE PATENT CLAIMS)
`(CLAIM 1) A media information recommendation device for
`estimating a user’s preferences based on the user’s history of
`accessing a plurality of media information sources in the past
`and recommending
`information coinciding with a user’s
`preferences from among newly arrived information sent from a
`newly arrived information source, the device comprising:
`an attribute extraction means for extracting and outputting
`attribute information from the newly arrived information;
`an access monitoring means for monitoring the status of the
`user’s access to each of the media information sources and
`outputting monitoring information related to media information
`accessed by the user based on the attribute information;
`a history
`information storage means for storing history
`information for the user’s access to the plurality of media
`information sources;
`a history information managing means for updating the history
`information based on the monitoring information outputted by
`the access monitoring means; and
`a media information evaluation means for evaluating whether
`the newly arrived information coincides with the user’s
`preferences by comparing the attribute information outputted by
`each attribute extraction means and the history information
`stored in the history information storage means;
`wherein history information based on the user’s access to the
`plurality of media information sources is collectively managed,
`and
`information
`is recommended based on
`the history
`information.
`(CLAIM 2) The media information recommendation device
`according to Claim 1, wherein a plurality of access monitoring
`means are provided so as to correspond to each of the media
`information sources.
`(CLAIM 3) The media information recommendation device
`according to Claim 1, wherein a plurality of attribute extraction
`means are provided so as to correspond to each newly arrived
`information source when there are a plurality of newly arrived
`information sources.
`(CLAIM 4) The media information recommendation device
`according to Claim 1, wherein the attribute extraction means
`automatically extracts keywords expressing the content of the
`newly arrived information or special notes from the newly
`arrived information.
`(CLAIM 5) The media information recommendation device
`according to Claim 1, wherein the attribute extraction means
`automatically extracts genre information specifying the genre
`from the newly arrived information.
`(CLAIM 6) The media information recommendation device
`according to Claim 2, wherein the access monitoring means
`outputs all monitoring information of the same format.
`(CLAIM 7) The media information recommendation device
`according to Claim 3, wherein the attribute extraction means
`outputs all attribute information of the same format.
`(CLAIM 8) The media information recommendation device
`according to Claim 1 having a media information source load
`setting means for setting a load value for each media
`information source;
`wherein the history information managing means updates the
`history information by assigning a load to the monitoring
`information sent from the access monitoring means based on the
`load value.
`(CLAIM 9) The media information recommendation device
`according to Claim 1, wherein when a television broadcasting
`
`information source which outputs a television broadcasting
`signal using radio waves or cables is at least one of the media
`sources and a program information source which outputs
`program information related to programs scheduled to be
`broadcast is the newly arrived information source,
`the attribute extraction means has a program attribute extraction
`part for extracting and outputting program attribute information
`from the program information; and
`the access monitoring means has a program viewing monitoring
`part which, when the user views a program, outputs program
`monitoring
`information based on
`the program attribute
`information of that program.
`(CLAIM 10) The media information recommendation device
`according to Claim 1, wherein an electronic news information
`source which outputs article information containing electronic
`news is at least one of the media information sources and the
`newly arrived information source,
`the attribute extraction means has an article attribute extraction
`means for extracting and outputting article attribute information
`from the article information; and
`the access monitoring means has an article reading monitoring
`part which, when the user reads electronic news, outputs article
`monitoring information based on the article attribute information
`of that article.
`(CLAIM 11) The media information recommendation device
`according to Claim 1, wherein a network information source
`which outputs page information is at least one of the media
`information sources and the newly arrived information source,
`the attribute extraction means has a page attribute extraction part
`for extracting and outputting page attribute information from the
`page information; and
`the access monitoring means has a page access monitoring part
`which, when the user reads page information, outputs page
`monitoring information based on the page attribute information
`of that page information.
`(CLAIM 12) A media information recommendation device
`comprising:
`an operation input means for receiving an input operation by a
`user;
`an information output means for outputting media information
`from a media information source selected by the user via the
`operation input means from among a plurality of media
`information sources;
`an information attention level evaluation means for evaluating
`an information attention level expressing the user’s degree of
`attention to media information based on the user’s behavior
`while the information output means is outputting the media
`information; and
`a media information recommendation means for recommending
`media information coinciding with the user’s preferences from
`among newly arrived information that is sent based on the
`information attention level.
`(CLAIM 13) A media information recommendation device
`comprising:
`an operation input means for receiving an input operation by a
`user;
`an information output means for outputting media information
`from a channel selected by the user via the operation input
`means from among media information sources having a plurality
`of channels;
`an information attention level evaluation means for evaluating
`an information attention level expressing the user’s degree of
`
`Apple 1005
`
`

`
`attention to media information based user’s behavior while the
`information output means is outputting the media information;
`and
`a media information recommendation means for recommending
`media information coinciding with the user’s preferences from
`among newly arrived information that is sent based on the
`information attention level.
`(CLAIM 14) The media information recommendation device
`according to Claim 13, wherein the information attention level
`evaluation means evaluates the information attention level for
`each program when the media information source provides each
`program that is broadcast in accordance with a broadcast
`schedule as each piece of media information via each channel;
`and
`the media information recommendation means recommends a
`program.
`(CLAIM 15) The media information recommendation device
`according to Claim 14, wherein the information attention level
`evaluation means evaluates the information attention level of
`each program by referencing the amount of time elapsed from
`when the user selects a channel broadcasting the program until
`the broadcast of the program is begun.
`(CLAIM 16) The media information recommendation device
`according to Claim 12 or 13, wherein the information attention
`level evaluation means evaluates the information attention level
`based on the operation frequency by the user from the operation
`input means.
`(CLAIM 17) The media information recommendation device
`according to Claim 16, wherein the information output means
`enables the output of information or some information by means
`of sound; and
`the information attention level evaluation means evaluates the
`information attention level based on a volume adjustment
`operation frequency.
`(CLAIM 18) The media information recommendation device
`according to Claim 16, wherein the information attention level
`evaluation means evaluates the information attention level based
`on a channel operation frequency.
`(CLAIM 19) The media information recommendation device
`according to Claim 12 or 13 having a distance measurement
`means for measuring a distance between the user and the
`information output means;
`wherein the information attention level evaluation means
`evaluates
`the
`information attention
`level based on
`the
`measurement value determined by the distance measurement
`means.
`(CLAIM 20) The media information recommendation device
`according to Claim 12 or 13, wherein the information output
`means enables the output of information or some information by
`means of sound; and
`the information attention level evaluation means evaluates the
`information attention level based on a volume corresponding to
`an adjustment operation by the user from the operation input
`means.
`(CLAIM 21) The media information recommendation device
`according to Claim 14, wherein the information attention level
`evaluation means evaluates the information attention level based
`on a degree of coincidence between the time slot during which
`the information output means outputs a program based on a
`channel selection operation by the user from the operation input
`means and the broadcast time slot of the program based on the
`broadcast schedule.
`(CLAIM 22) The media information recommendation device
`according to Claim 12 or 13 having an index value holding
`means in which, when the information attention level evaluation
`
`means indicates the information attention level as an evaluation
`value, an index value used to obtain the evaluation value is set in
`advance.
`(CLAIM 23) The media information recommendation device
`according to Claim 12 or 13, wherein the information attention
`level evaluation means indicates the information attention level
`with an evaluation value determined by a prescribed function.
`(DETAILED DESCRIPTION OF THE INVENTION)
`(0001)
`(TECHNICAL FIELD OF THE INVENTION) The present invention
`relates to a media information recommendation device for
`recommending newly arrived media information and more
`particularly to the improvement of the reliability of estimating a
`user’s preferences.
`(0002)
`(PRIOR ART) Devices which estimate a user’s preferences based
`on the user’s past access history of media information and
`recommend newly arrived media information to the user based
`on
`these preferences (media
`information recommendation
`devices) have been proposed in the past.
`(0003) For example, at the time of channel selection with a
`video device such as a television in the “recording and channel
`selection method for a video device” disclosed in Japanese
`Unexamined Patent Application Publication H7-135621,
`keywords for programs selected by a user are extracted from a
`program guide expressed by character strings, and the obtained
`keywords and the frequencies of occurrence thereof are stored
`and accumulated. When a program guide is newly supplied
`thereafter, keywords are extracted from the descriptions of the
`program guide, and the frequency of occurrence stored for each
`extracted keyword
`is
`investigated. A program
`is
`then
`recommended by presenting the user with a program having a
`high total of the frequency of occurrence of the keywords.
`(0004) FIG. 17 is block configuration diagram of a conventional
`media information recommendation device. The conventional
`media information recommendation device comprises a media
`information reading part 1, an attribute extraction part 2, an
`access monitoring part 3, a history information managing part 4,
`a history information storage part 5, a media information
`evaluation part 6, a media information recommendation part 7,
`and a display part 8. The media information reading part 1 is a
`means for allowing a user to access a media information source
`9 which provides media information to the user. The attribute
`extraction part 2 extracts attribute information from newly
`arrived media information outputted from a newly arrived
`information source 10. The access monitoring part 3 monitors
`which media information the user has read using the media
`information reading part 1, selects information related to the
`media information read by the user from among attribute
`information from the attribute extraction part 2, and outputs
`monitoring information. The history information managing part
`4 updates the history information based on the monitoring
`information sent from the access monitoring part 3 and stores
`the information in the history information storage part 5. The
`media information evaluation part 6 evaluates the newly arrived
`information by comparing the attribute information from the
`attribute extraction part 2 based on
`the newly arrived
`information and the history information based on the access of
`the user. The media information recommendation part 7
`recommends newly arrived information to the user by displaying
`attribute information on the display part 8 in accordance with the
`evaluation results of the media information evaluation part 6.
`(0005) In the conventional media information recommendation
`device having the above configuration, history information is
`
`Apple 1005
`
`

`
`stored in the history information storage part 5 each time the
`user accesses media information. On the other hand, when the
`newly arrived information source 10 outputs newly arrived
`information and the media information evaluation part 6
`receives attribute information based on the newly arrived
`information from the attribute extraction part 2, the media
`information evaluation part 6 calculates an evaluation value for
`the newly arrived information by comparing the attribute
`information and the history information stored in the history
`information storage part 5. If this evaluation value is equal to or
`greater
`than a certain value,
`the media
`information
`recommendation part 7
`recommends
`the newly arrived
`information to users by displaying the attribute information of
`the newly arrived information on the display part 8 such as a
`CRT.
`(0006) In the case of the device disclosed in Japanese
`Unexamined Patent Application Publication H7-135621
`described above, the media information source 9 in FIG. 17
`corresponds to a television broadcast station; the media
`information reading part 1 corresponds to a television receiver;
`the attribute information corresponds to a program keyword; the
`history information corresponds to the frequency of occurrence
`of a keyword; and the evaluation value for newly arrived
`information corresponds to the total frequency of appearance of
`a keyword.
`(0007) Incidentally, in recent years, it has become possible to
`transmit information regarding a television broadcast program
`simultaneously with the broadcast as an EPG (Electronic
`Program Guide). In digital satellite broadcasting in Japan as
`well, a program broadcast schedule is sent together with a
`program name and other character information to a receiver
`simultaneously with a television broadcast.
`(0008) FIG. 45 is a block configuration diagram of a
`conventional media information recommendation device applied
`to an EPG. By performing a selection operation for media
`information from an operation input part 51, a user selects the
`media information from a media information source 52 and
`displays the information on an information display part 53. This
`configuration is included in the media information reading part 1
`of FIG. 17. An information attention level evaluation part 54
`detects
`the media
`information selected from
`the media
`information source 52 based on the input operation by the user
`from the operation input part 51, evaluates the degree to which
`the user was paying attention to the media information, and
`calculates an
`information attention
`level. Further,
`the
`information attention level evaluation part 54 stores a history of
`the input operations received by the operation input part 51
`together with the calculated information attention level as
`operation history
`information
`in an operation history
`information storage part 55. The selections of media information
`performed by the user in the past are accumulated in this
`operation history information, so the user’s interests are
`reflected in the information recommendation made based on this
`information. The operation history information storage part 55
`corresponds to the history information storage part 5 in FIG. 17.
`An
`information
`recommendation part 56 selects media
`information that appears to coincide with the user’s interests by
`comparing each piece of media information from the media
`information source 52 and the operation history information
`stored in the operation history information storage part 55 and
`recommends the media information to the user. There are
`various types of recommendation methods, examples of which
`include a method of displaying a title if a title is included in the
`media information or a method of displaying the entire media
`information. In addition, the information recommendation part
`
`56 corresponds to the media information recommendation part 7
`in FIG. 17.
`(0009) Several methods have been proposed as calculation
`methods for the information attention level determined by the
`information attention level evaluation part 54. For example,
`there is a method of directly setting the information attention
`level to a value of 1 if the user has accessed the media
`information and a value of 0 if the user has not accessed the
`media information. This is a method which uses the presence or
`absence of access to the media information in question as the
`information attention level.
`(0010) Another method of calculating the information attention
`level is a method of determining the information attention level
`based on the number of times that the user has accessed media
`information. That is, this is a method which sets the information
`attention level to N when the user has accessed the same media
`information N times.
`(0011) Yet another method for calculating the information
`attention level is a method of determining the information
`attention level based on the access time when the user accesses
`the media information. This is a method which sets the
`information attention level based on the value of 1 hour when
`the user has accessed the same media information for 1 hour, for
`example.
`(0012) The information attention level calculated as described
`above is stored in the operation history information storage part
`55 together with the operation history information based on the
`input operations of the user. There are also several possible
`methods for storing information in this operation history
`information storage part 55. First, there is a method of storing an
`identification number of media information selected by the user
`and the information attention level as a set. In addition, it is also
`possible to use a method of storing the title of the media
`information and the information attention level as a set when the
`media information is a broadcast program which is broadcast
`with different content for a program of the same title such as a
`news program providing new information every day or a drama
`program in which a story is divided and broadcast sequentially
`every week. More specifically, the title of the program and the
`number of times that the user has viewed the program having
`that title are stored as a set in the operation history information
`storage part 55.
`(0013) Further, there is also a method in which, when some sort
`of additional information is included in media information, the
`additional information is stored in the operation history
`information storage part 55 as a unit. For example, Japanese
`Unexamined Patent Application Publication H7-135621
`described above discloses a configuration in which a keyword of
`a program selected by a user is extracted from a program guide
`expressed by character strings, and the keywords obtained as a
`result and
`the
`frequencies of occurrence
`thereof are
`accumulated. In this case, the additional information of a
`keyword is provided to each program, and the information
`attention level is calculated as the viewing frequency of the
`program. The information attention level is stored in the
`operation history information storage part 55 as a set with the
`keyword.
`(0014)
`(PROBLEM TO BE SOLVED BY THE INVENTION) Incidentally, in
`recent years, various media information sources have come to be
`used in each household. Examples include compact discs,
`televisions, teletext broadcasting, electronic shopping, WWW
`(World Wide Web), electronic news, e-mail, and computer
`programs. Information obtained when the user views or accesses
`one of these media information sources may be useful in
`
`
`Apple 1005
`
`

`
`estimating the preferences of the user when viewing other media
`information sources. For example, if the user views many
`television programs in which a certain singer performs, it can be
`considered that the singer coincides with the user’s preferences.
`Accordingly, it can be considered highly likely that the user will
`purchase the work of that signer when purchasing a compact
`disc.
`(0015) However, in a conventional information recommendation
`device, it is only possible to handle a single media information
`source from among a wide variety of media information sources,
`so it is not possible to use such useful information or to apply it
`to the recommendation of other media information sources. That
`is, a device for recommending television programs, for example,
`is ultimately only able to recommend television programs.
`Therefore, in order to obtain recommendation information for a
`plurality of media information sources with a conventional
`device, it is necessary to use separate media information
`recommendation devices, and recommendation information
`which accurately reflects the user’s preferences cannot be
`obtained without substantially using each of
`the media
`information recommendation devices.
`(0016) In addition, in today’s age of diversified media
`information sources, and in the future as the types of new media
`information sources are expected to continue to increase due to
`technological advances, it becomes structurally redundant to
`separately use media information recommendation devices
`specified for each of the respective media information sources,
`and there is a large amount of waste from the perspective of
`cost.
`(0017) On the other hand, in a conventional media information
`recommendation device applied to an EPG, the information
`attention level is calculated based on values such as the number
`of times that a user has viewed media information or the
`viewing time, but it is not necessarily possible to always
`accurately assess the degree to which the user had been paying
`attention to the media information in question from values such
`as the number of times the information was viewed. This will be
`described using television viewing as an example.
`(0018) Conventionally, when the power of a television is ON
`and a certain program is displayed on the screen, the amount of
`time that the program is displayed is considered the amount of
`time that the user actually views the program and is defined as
`the viewing time. The user may indeed be viewing the program
`being broadcast, but the user may have also left his or her seat.
`In addition, even if the user is close to the television, there are
`also cases in which the user is not actually viewing the program,
`such as cases in which the user is engrossed in other activities
`such as eating or reading or cases in which the user has fallen
`asleep. That is, conventionally, there are cases in which such
`situations are also determined as the viewing time, so the
`reliability of the information attention level is lacking. Such a
`problem is not limited to television and is also present for
`general media information.
`(0019) Accordingly, when the information attention level is
`calculated conventionally, the evaluation of the degree to which
`the user is paying attention to media information being
`displayed is inaccurate, which results in the problem that the
`content of recommendation information provided by the media
`information recommendation means does not necessarily
`coincide with the actual interests of the user.
`(0020) The present invention was conceived in order to solve the
`problem described above, and an object of the present invention
`
`is to provide a media information recommendation device which
`recommends
`information
`that coincides with a user’s
`preferences with high reliability after use for a short period of
`time by collectively managing access history information of the
`user for a plurality of media information sources.
`(0021) Another object of the present invention is to provide a
`media information recommendation device capable of more
`accurately evaluating a user’s degree of attention to media
`information.
`(0022)
`(MEANS FOR SOLVING THE PROBLEM) In order to achieve the
`objects described above, the media information recommendation
`device of
`the
`first
`invention
`is a media
`information
`recommendation device for estimating a user’s preferences
`based on the user’s history of accessing a plurality of media
`information sources in the past and recommending information
`coinciding with a user’s preferences from among newly arrived
`information sent from a newly arrived information source, the
`device comprising: an attribute extraction means for extracting
`and outputting attribute information from the newly arrived
`information; an access monitoring means for monitoring the
`status of the user’s access to each of the media information
`sources and outputting monitoring information related to media
`information accessed by the user based on the attribute
`information; a history information storage means for storing
`history information for the user’s access to the plurality of media
`information sources; a history information managing means for
`updating the history information based on the monitoring
`information outputted by the access monitoring means; and a
`media information evaluation means for evaluating whether the
`newly arrived information coincides with the user’s preferences
`by comparing the attribute information outputted by each
`attribute extraction means and the history information stored in
`the history
`information storage means; wherein history
`information based on the user’s access to the plurality of media
`information sources is collectively managed, and information is
`recommended based on the history information.
`(0023) The media information recommendation device of the
`second invention is a device according to the first invention,
`wherein a plurality of access monitoring means are provided so
`as to correspond to each of the media information sources.
`(0024) The media information recommendation device of the
`third invention is a device according to the first invention,
`wherein a plurality of attribute extraction means are provided so
`as to correspond to each newly arrived information source when
`there are a plurality of newly arrived information sources.
`(0025) The media information recommendation device of the
`fourth invention is a device according to the first invention,
`wherein the attribute extraction means automatically extracts
`keywords expressing
`the content of
`the newly arrived
`information or special notes from the newly arrived information.
`(0026) The media information recommendation device of the
`fifth invention is a device according to the first invention,
`wherein the attribute extraction means automatically extracts
`genre information specifying the genre from the newly arrived
`information.
`(0027) The media information recommendation device of the
`sixth invention is a device according to the second invention,
`wherein the access monitoring me