JP5180639B2 - Content presentation device, content presentation method, and program - Google Patents

Description

  The present invention relates to an apparatus that presents content to a user, and more particularly, to a navigation apparatus that presents information on facilities such as restaurants and sightseeing spots as content.

  2. Description of the Related Art Conventionally, a device that presents content in a navigation device or the like according to a drive situation or a user attribute value has been known, for example, from Patent Document 1 and the like. Patent Document 1 is a patent application filed by the present inventors.

In the service candidate providing system described in Patent Literature 1, the in-vehicle device constructs a user model that characterizes the user's preference pattern and behavior pattern based on information such as the user's attribute value, preference, and history. And a service candidate provision server produces | generates the service candidate list information provided to a user using a user model, and transmits the produced | generated service candidate list information to a vehicle-mounted apparatus. The in-vehicle device updates the user model using the service selection result by the user as a learning sample. As a result, the user model is customized to a model that more accurately reflects the user's characteristics.
JP 2005-208943 A

  Since the conventional content presentation apparatus as described above learns the user model using the service selection result by the user or the like, there is an advantage that the user's preference is reflected and the content suitable for the user can be presented.

  On the other hand, if learning progresses and the degree of conformity to the user's preference increases, similar content may always be presented. Therefore, there is no possibility of unexpectedness or discovery in the presented content, and the user may get bored.

  In view of the above background, an object of the present invention is to provide a content presentation apparatus that can present appropriate content.

  The content presentation device of the present invention includes a content storage unit that stores content and attribute values thereof, and a user-adapted content selection that selects content from content stored in the content storage unit based on a history of content adopted by a user Selected from the content stored in the content storage unit, the content different in attribute value from the content selected in the user-adapted content selection unit, and selected in the user-adapted content selection unit A content presentation unit that presents the user-adapted content and the heterogeneous content selected by the heterogeneous content selection unit.

  In this way, by presenting different types of content having different attribute values from the user-adapted content together with the user-adapted content, it is possible to present diversified content in accordance with user preferences.

  In the content presentation device of the present invention, the user-adapted content selection unit has a user model representing a relationship between a situation when the user adopts the content and an attribute value of the adopted content, and uses the user model. An attribute value of content corresponding to the current situation may be obtained, and content having the attribute value may be selected as the user-adapted content.

  As described above, by using the user model, it is possible to appropriately select content corresponding to the current situation.

  In the content presentation device of the present invention, the heterogeneous content selection unit has a different model that is different from the user model that represents a relationship between a situation or a user attribute value and a content attribute value, and currently uses the different model. The content attribute value corresponding to the situation or the user attribute value may be obtained, and the content having the content attribute value may be selected as the heterogeneous content.

  In this way, by using a different model different from the user model, it is possible to select content having an attribute value different from the content selected by the user model. As another model, for example, a model reflecting general user preferences can be used. Different models may be different models depending on age, sex, consumption characteristics, personality, and the like. For example, when the user is in his twenties and men, a model reflecting the preferences of the twenties and men may be used.

  In the content presentation device of the present invention, the another model may be a model generated using content adoption history data obtained from a plurality of users.

  With this configuration, another model can be appropriately generated.

  In the content presentation device of the present invention, the different model may be a model generated using data of content adoption history in a specific situation.

  By limiting the situation (scene) in this way, different models can be generated according to the situation. The situation includes the situation of the season such as summer and graduation season, the situation concerning the user's behavior such as dating and at work, the situation concerning the weather such as a sunny day and the rainy day, the situation concerning the place along the coast, mountain road, etc. Various situations can be considered.

  In the content presentation device of the present invention, the user model may be a model generated using a history of content adopted by the user during a recent predetermined period, and the different model is adopted by the user for a period longer than the predetermined period. It may be a model generated using the history of the content.

  In this way, by generating another model using the content adoption history over a long period of time, it is possible to select content according to the preference that the user has as a base using the other model. On the other hand, by generating a user model using the content adoption history of a recent predetermined period, it is possible to select content according to recent preferences using the user model. Therefore, various contents can be presented according to the base preference and the recent preference.

  In the content presentation device of the present invention, the content presentation unit may present information related to another model in which the heterogeneous content is selected when the heterogeneous content is presented.

  In this way, by presenting information on another model in which different content is selected, the user can understand the reason why the different content different from his / her preference is presented.

  In the content presentation device of the present invention, the heterogeneous content selection unit may select a predetermined content.

  By selecting a predetermined content in this way and presenting it together with user-adapted content, trendy content, recommended content, and the like can be presented. The predetermined content data may be distributed from a server or the like. Thereby, the latest content can be presented.

  In the content presentation device of the present invention, the number of content to be presented having the same attribute value may be set in the content presentation unit.

  With this configuration, it is possible to prevent a situation in which only a large number of contents having the same attribute value are presented and present various contents.

  In the content presentation device of the present invention, the content presentation unit may present the user-adapted content and the heterogeneous content in an identifiable manner.

  With this configuration, it is possible to identify whether or not the presented content is the content selected according to the user's preference, and it is possible to obtain hints when selecting desired content from the presented content. Various methods are conceivable as a method for presenting user-adapted content and heterogeneous content in an identifiable manner. For example, a character that recommends user-adapted content and a character that recommends different content may be displayed in the vicinity of the content. Further, the user-adapted content and the different content may be displayed in different colors and fonts. In the case of presenting content by sound, the sound may be changed depending on whether the user-adapted content is output as audio or when different types of content are output as audio.

  In the content presentation device of the present invention, the content presentation unit may display the user-adapted content and the heterogeneous content in no particular order.

  For example, a presentation order in which user-adapted content is arranged in order from the top and different types of content continue is advantageous in that it is easy to distinguish between user-adapted content and different types of content, but the user may tend to select content from the top. There is sex. With the configuration of the present invention, it is possible to reduce the influence of the presentation order of the user-adapted content and the different content on the content selection.

  The content presentation apparatus of the present invention may include a presentation ratio determination unit that determines a presentation ratio of the user-adapted content presented by the content presentation unit and the heterogeneous content.

  With this configuration, since the presentation ratio can be changed, the user-adapted content and the heterogeneous content can be presented at an appropriate ratio according to the situation or the like.

  The content presentation device of the present invention may change the presentation ratio based on the number of times of adoption of the user-adapted content and the number of times of adoption of the different type of content.

  With this configuration, it is possible to present user-adapted content and heterogeneous content at a rate suitable for the user. Therefore, it is possible to efficiently utilize resources (screen or audio output time) necessary for presenting content and present content that meets the user's needs.

  The content presentation apparatus of the present invention may change the presentation ratio based on a time-series history of content adopted by the user.

  With this configuration, an appropriate presentation ratio can be set using a content adoption pattern in time series.

  In the content presentation device of the present invention, the content is a facility, and the content storage unit stores data indicating the location of the facility as the attribute, and a search condition input unit that receives an input of a search condition; A condition search unit that searches the content storage unit for a plurality of contents that match the search condition input by the search condition input unit, and a content distribution calculation unit that calculates the distribution of the content searched by the condition search unit The user-adapted content selection unit and the heterogeneous content selection unit select the content searched by the condition search unit, and the presentation ratio determination unit calculates the content calculated by the content distribution calculation unit. The presentation ratio may be determined based on the distribution of.

  In this way, when presenting content selected from content that meets the search conditions, it is possible to appropriately set the presentation ratio using content distribution data corresponding to the search conditions input by the user. For example, when content is dispersed, there is a possibility that the search condition may be content with an unusual search condition, so the ratio of different types of content is set larger than the user-adapted content that matches the user's preference. The example shown here is an example, and the relationship between the distribution data and the presentation ratio can be set as appropriate according to the requirements of the system and the like.

  The content presentation device of the present invention includes a current position detection unit that detects a current position of a user, the content is a facility, and the content storage unit stores data indicating the location of the facility as the attribute. The user-adapted content selection unit and the heterogeneous content selection unit may select a facility using data on the distance from the current position of the user to the position of the content.

  With this configuration, the balance between distance and preference, such as content that is geographically close to the user's past employment history, content that is not so close to the content employment history, but geographically close, etc. You can select the appropriate content.

  The content presentation method of the present invention includes a step of selecting content suitable for a user as user-adapted content based on a history of content adopted by the user from a content storage unit storing content and attribute values thereof, and the content storage Selecting a content having an attribute value different from that of the user-adapted content from the section, and presenting the user-adapted content and the heterogeneous content.

  In this way, by presenting different types of content having different attribute values from the user-adapted content together with the user-adapted content, it is possible to present diversified content in accordance with the user's preference, similar to the content presentation device described above. Various configurations of the content presentation device of the present invention can be applied to the content presentation method of the present invention.

  The program of the present invention is a program for presenting content, and the content that matches the user is stored in the computer from the content storage unit that stores the content and its attribute value based on the history of the content adopted by the user. Selecting as user-adapted content, selecting from the content storage unit content having a different attribute value from the user-adapted content as heterogeneous content, and presenting the user-adapted content and the heterogeneous content Let

  In this way, by presenting different types of content having different attribute values from the user-adapted content together with the user-adapted content, it is possible to present diversified content in accordance with the user's preference, similar to the content presentation device described above. Various configurations of the content presentation apparatus of the present invention can be applied to the program of the present invention.

  The present invention has an effect of presenting diversified content according to the user's preference by presenting different types of content having different attribute values from the user-adapted content together with the user-adapted content.

(First embodiment)
Hereinafter, a content presentation device according to an embodiment of the present invention will be described with reference to the drawings. Below, the example which applied the content presentation apparatus to the navigation apparatus is demonstrated. That is, the navigation device described below has a function of presenting information (contents) on facilities such as restaurants and sightseeing spots. However, the content presentation device of the present invention can be applied not only to the navigation device but also to other devices. For example, in the case of presenting music as content, it can be applied to an audio device having a music playback function.

  FIG. 1 is a diagram showing a configuration of a navigation apparatus 1 according to the first embodiment of the present invention. The navigation device 1 includes a position detector 10, a map database (hereinafter referred to as "map DB") 20, an operation switch group 22, a communication unit 26, a memory 28, a display device 30, a voice controller 32, A speaker 34, a voice recognition device 36, a microphone 38, a remote control sensor 40, a remote control terminal (hereinafter referred to as “remote control”) 42, and a control device 24 to which these devices are connected are provided.

  The control device 24 is a normal computer and includes a CPU, a ROM, a RAM, an I / O, and a bus line for connecting these components. A program to be executed by the control device 24 is written in the ROM, and the CPU executes arithmetic processing according to this program. A program for executing the processing of the control device 24 described below is also included in the scope of the present invention.

  The position detector 10 is based on a geomagnetic sensor 12 for detecting the absolute azimuth of the vehicle, a gyroscope 14 for detecting the relative azimuth of the vehicle, a distance sensor 16 for detecting the travel distance of the vehicle, and radio waves from the satellite. And a GPS receiver 18 for a global positioning system (GPS) for measuring the position of the vehicle. Each of these sensors 12, 14, 16, 18 has an error of a different nature. The plurality of sensors 12, 14, 16, 18 are used while complementing each other. Depending on the accuracy, the position detector 10 may be configured by a part of the above. Further, a steering rotation sensor (not shown), a vehicle speed sensor for each rolling wheel, or the like may be used.

  The map DB 20 is configured by a storage medium such as a DVD-ROM or a CD-ROM. The map DB 20 stores road map data, background data, character data, and facility data.

  FIG. 3 is a diagram illustrating an example of facility data. In the facility data, data for identifying a facility, a facility name, a facility attribute (genre, budget), and a facility position are associated with each other. Here, the genre of the restaurant is shown as “genre”, but in actuality, attribute values such as “tourist spot”, “hotel”, and “restaurant” are classified as large categories, and “sushi”, “tonkatsu”, and “tempura” as fine categories. ”And“ Kaishi ”are stored. FIG. 3 is an example, and the facility data may include data other than that shown in FIG. In the navigation device 1 of the present embodiment, a facility is presented as content, and therefore the facility is referred to as “content” in the following description.

  The operation switch group 22 includes, for example, a touch switch integrated with the display device 30 or a mechanical switch provided around the display device 30. The operation switch group 22 is used for various inputs such as scale change of the map displayed on the display device 30, menu display selection, destination setting, route search, route guidance start, current position correction, display screen change, and volume adjustment. The The remote controller 42 is provided with a plurality of operation switches (not shown), and the same input operation as that of the operation switch group 22 can be performed by operating the operation switches. A signal representing an input operation input to the remote control 42 is supplied to the control device 24 via the remote control sensor 40.

  The memory 28 as a storage device is, for example, a memory card or a hard disk, and includes a writable storage medium. The memory 28 stores a user model 60, a general model 62, and content adoption history data 64. The memory 28 may store various data such as a home position set by the user, text data, image data, audio data, and the like.

  The user model 60 is a model representing the relationship between the situation when a user adopts content and the attribute value of the adopted content. The general model 62 is a model representing the relationship between the situation or user attribute value and the content attribute value, and is a model different from the user model 60. Here, the general model 62 corresponds to “another model” in the claims. The content adoption history data 64 is data indicating content that the user has adopted in the past. The adoption history data 64 is associated with data such as a situation when the content is adopted.

  The display device 30 is configured by, for example, a liquid crystal display or an organic EL display. In a predetermined map display area of the display device 30, a vehicle position mark corresponding to the current position of the vehicle is superimposed and displayed on a road map around the vehicle generated by the map data. Further, the display device 30 displays the content selected by the navigation device 1.

  The communication unit 26 has a function of establishing communication connection with the outside, and performs communication with the content information center 44. The content information center 44 has a function of providing the latest content data and promotion data.

  The speaker 34 outputs a predetermined sound (sound for guidance, explanation of screen operation, sound recognition result, etc.) to the outside based on the sound output signal input from the sound controller 32.

  The microphone 38 inputs the voice uttered by the operator to the voice recognition device 36 as an electrical signal. The voice recognition device 36 collates the operator's input voice input from the microphone 38 with lexical data (comparison reference pattern) in a recognition dictionary (not shown) stored therein, and obtains the one with the highest degree of matching. The recognition result is input to the voice controller 32.

  The voice controller 32 controls the voice recognition device 36 and performs talkback output control (voice output) through the speaker 34 for voice input. Moreover, the process which inputs the recognition result of the speech recognition apparatus 36 to the control apparatus 24 is also performed.

  Based on the information from the voice recognition device 36, the control device 24 performs predetermined processing for the utterance of the operator and predetermined processing for input operation of the operation switch group 22 or the remote control 42 (for example, storage processing of map data in the memory 28). Map scale change processing, menu display selection processing, destination setting processing, route search execution processing, route guidance start processing, current position correction processing, display screen change processing, volume adjustment processing, etc.). Moreover, the control apparatus 24 of this Embodiment has a function which selects a content according to a condition etc. and presents the selected content so that it may demonstrate in detail below.

  FIG. 2 is a block diagram illustrating functions of the control device 24. The control device 24 includes a user-adapted content selection unit 50, a heterogeneous content selection unit 52, a content presentation unit 54, a presentation ratio determination unit 56, and a model learning unit 58.

  The user-adapted content selection unit 50 has a function of selecting content that matches the user's preference from the content stored in the map DB 20. First, the user-adapted content selection unit 50 uses the user model 60 stored in the memory 28 to obtain an attribute value of content that conforms to the current situation. Here, the user-adapted content selection unit 50 may obtain a plurality of attribute values as the attribute values of the content suitable for the current situation. Next, the user-adapted content selection unit 50 selects content corresponding to the obtained attribute value from among the content within a predetermined range from the current position detected by the position detector 10.

  The heterogeneous content selection unit 52 has a function of selecting content having an attribute value different from the user-adapted content from the content stored in the map DB 20. The heterogeneous content selection unit 52 first uses the general model 62 stored in the memory 28 to obtain the attribute value of the content that matches the current situation. Here, the heterogeneous content selection unit 52 may obtain a plurality of attribute values as the attribute values of the content suitable for the current situation. Next, the heterogeneous content selection unit 52 selects content corresponding to the obtained attribute value from among the content within a predetermined range from the current position detected by the position detector 10.

  Here, the user model 60 and the general model 62 will be described. In the present embodiment, a Bayesian network model is used as the user model 60 and the general model 62. The Bayesian network model is a model for obtaining the likelihood of content attribute values from observed events (for example, situations and user attribute values). In the present embodiment, a Bayesian network model is used, but other models may be used as long as the attribute value of the content can be obtained based on the past content adoption history. For example, a neural network model may be used.

  FIG. 4A to FIG. 4C are diagrams illustrating a Bayesian network model. As shown in FIG. 4A, the Bayesian network model expresses a probability that another event occurs when a certain event occurs by connecting nodes indicating the event with a directed link. The conditional probability of each link is expressed by a conditional probability table. In the Bayesian network model, values are set in the observation nodes (in this example, the node N1 for weekdays / holidays and the node N2 for presence / absence of passengers), so that an unobserved node (in this example, the node N3 of the content attribute) ).

  FIG. 4B is a diagram illustrating an example of a conditional probability table. From this table, it can be seen that, for example, when there is a holiday and there are passengers, the probability of adopting 3000 yen content of Japanese food is high (50%). Note that FIG. 4B shows an example in which there are few conditions for convenience of explanation, but an actual model includes various conditions such as time zone, weather, traffic conditions, seasons, and the like. The conditional probability table is generated based on the content adoption history by the user. FIG. 4C is a table showing a history that the user has adopted the content. In this table, a value obtained by dividing the number of times of adopting each content by the total number of times of adopting the content is used as the conditional probability.

  The user model 60 is a model that reflects the history of content adoption by users of the navigation device 1, while the general model 62 is a model that reflects the history of content adoption by many users. The general model 62 can be acquired as follows, for example. First, the content information center 44 collects content adoption histories of a plurality of users, and generates a general model 62 from the collected history data. Then, the content information center 44 distributes the generated general model 62 to the navigation device 1.

  The model learning unit 58 updates the user model 60 using the content adoption history by the user. That is, when the content is adopted, a condition such as a situation where the content is adopted is obtained, and the conditional probability table is updated by counting up the number of times the content attribute value is adopted corresponding to the condition. When generating the user model 60, the general model 62 may be used as the initial model.

  FIG. 5 is a diagram illustrating an example in which the general model 62 and the user model 60 are arranged on the feature space. FIG. 5 shows attribute values of contents selected under specific conditions (for example, weekdays, presence of passengers, etc.). The general model 62 indicates that content of about 4000 yen is adopted with a slightly Western taste. By using the general model 62 as an initial value and counting up the number of times the user adopts content, the tendency of user adoption is strongly reflected. When the user adopts Japanese-style content centered on 2000 yen, the user model 60 as shown in FIG. 5 is customized.

  Note that the user model 60 may be learned using not only the content adoption history but also the content non-recruitment history. For example, if the presented content is not adopted, the content adoption count may be counted down assuming that the attribute value of the content does not match the condition. However, in this case, the minimum value of the number of times of adoption is set to 0 and no more is counted down.

  The content presentation unit 54 has a function of displaying the user-adapted content selected by the user-adapted content selection unit 50 and the heterogeneous content selected by the heterogeneous content selection unit 52 on the display device 30 and presenting the content to the user. Have. The content presentation unit 54 displays the user-adapted content and the heterogeneous content according to the ratio determined by the presentation ratio determination unit 56.

  The presentation ratio determination unit 56 has a function of determining the ratio between the number of user-adapted contents to be displayed and the number of different types of content, and notifying the content presentation unit 54 of the determined presentation ratio. For example, the presentation ratio determination unit 56 may use a predetermined presentation ratio such as “8: 2”, or may update the presentation ratio according to the content adoption history by the user. For example, it is possible to obtain the adoption ratio of the user-adapted content and the different content from the content adoption history, and determine the adoption ratio as the presentation ratio. In this case, the adoption ratio may be obtained for each situation when the content is adopted, and the presentation ratio may be changed according to the current situation.

  FIG. 6 is a diagram illustrating an example of a screen presented by the content presentation unit 54. As shown in FIG. 6, the content presentation unit 54 displays the user-adapted content and the heterogeneous content in no particular order. As a result, the different types of content are arranged in a moderately scattered manner, so that the different types of content are also easily noticed by the user. In addition, a picture showing one face is assigned to the user-adapted content, and a picture showing multiple faces is assigned to the different content. As a result, the user can easily determine whether the content is user-adapted content or heterogeneous content.

  FIG. 7 is a flowchart showing a content presentation operation by the navigation device 1 of the present embodiment. The navigation device 1 first acquires situation data (S10). As situation data, weekday / holiday data is acquired using a calendar function (not shown), or a time zone is detected using a clock function. Further, the presence or absence of a passenger may be detected using a weight sensor in the passenger seat (not shown).

  Next, the user-adapted content selection unit 50 of the navigation device 1 uses the user model 60 to select user-adapted content corresponding to the acquired situation data from the map DB 20 (S12). Subsequently, the heterogeneous content selection unit 52 of the navigation device 1 uses the general model 62 to select heterogeneous content corresponding to the acquired situation data from the map DB 20 (S14).

  Next, the presentation ratio determination unit 56 of the navigation device 1 determines the presentation ratio of the user-adapted content and the heterogeneous content (S16). For example, the presentation ratio determination unit 56 determines the presentation ratio based on the past content adoption history in the same situation. Subsequently, the navigation device 1 presents the user-adapted content and the heterogeneous content in no particular order (S18). The above is the content presentation operation.

  The navigation device 1 receives an operation by the user (S20), and detects the content adopted by the user from the presented content. The navigation device 1 stores the adopted content data in the memory as the content adoption history data 64 (S22). Subsequently, the navigation device 1 learns the user model 60 using the content adoption history data 64 (S24). As a result, the user model 60 is customized to a model suitable for the user's preference. Although FIG. 7 illustrates an example in which user model learning (S20 to S24) is performed following content presentation (S18), user model learning is not necessarily performed following content presentation. Good. The learning of the user model may be performed at a timing when a certain amount of content adoption history data is accumulated. The configuration and operation of the navigation device 1 according to the present embodiment have been described above.

  Since the navigation apparatus 1 according to the present embodiment presents content different from the user-adapted content selected by the heterogeneous content selection unit 52 together with the user-adapted content, the content adapted to the user's preference is presented by learning the user model 60. Even when it becomes possible, it is possible to prevent a bias in the content presented. Thus, there is an effect that content different from the preference is presented together with the content suitable for the preference, and the user is not bored with the presented content.

(Second Embodiment)
Next, a navigation device according to a second embodiment of the present invention will be described. The basic configuration of the navigation device of the second embodiment is the same as that of the navigation device 1 of the first embodiment, but the navigation device of the second embodiment is a model for obtaining heterogeneous content. The difference is that an expert model 66 is provided in addition to the general model 62.

  FIG. 8 is a diagram illustrating a configuration of the control device 24 and the memory 28 in the second embodiment. In the second embodiment, the memory 28 stores an expert model 66 in addition to the general model 62. The expert model 66 is a model that reflects the history of content adoption by an expert who has a taste different from general tastes. In the present embodiment, it is shown what attribute value content is adopted by a user (expert) who likes South American cuisine. Although an example using one expert model 66 is given here, a plurality of expert models 66 may be used.

  The heterogeneous content selection unit 52 selects the heterogeneous content using either the general model 62 or the expert model 66. The heterogeneous content selection unit 52 selects the heterogeneous content using a model having a larger difference from the user model 60 among the general model 62 and the expert model 66.

  FIGS. 9A and 9B are diagrams showing the user model 60, the general model 62, and the expert model 66 arranged in the feature space. FIG. 9A shows the arrangement when the conditions are weekdays and no passengers, and FIG. 9B shows the arrangement when the conditions are holidays and passengers are present. The heterogeneous content selection unit 52 selects heterogeneous content using a model having a large distance from the user model 60 in the feature space. For example, when the condition is a weekday and there is no passenger (see FIG. 9A), the expert model 66 is used. When the condition is a holiday and a passenger is present (see FIG. 9B), the general model is used. 62 is used. As a result, it is possible to select different types of content whose attribute values are significantly different from the user-adapted content selected using the user model 60. The content presentation unit 54 displays the user-adapted content and the heterogeneous content in no particular order.

  FIG. 10 is a diagram illustrating an example of a screen displayed by the content presentation unit 54. The basic configuration of the screen displayed by the content presentation unit 54 is the same as the screen displayed by the navigation device 1 of the first embodiment, but for content selected by a model other than the user model 60. , Display information about that model. In the example shown in FIG. 10, “* South American” information is added to the restaurant D and restaurant K selected by the expert model 66, and the reason why the restaurant D and restaurant K are recommended is notified.

  Similarly to the navigation device 1 of the first embodiment, the navigation device of the second embodiment can prevent a bias in the content presented even when the learning of the user model 60 proceeds. Has an effect. In the navigation device according to the second embodiment, the different content selection unit 52 selects the different content using a model having a large distance from the user model 60 in the feature space. Different heterogeneous content can be presented. Thereby, it is possible to provide a new viewpoint to the user.

  As mentioned above, although the content presentation apparatus of this invention was demonstrated in detail, giving embodiment which made a navigation apparatus an example, this invention is not limited to above-described embodiment.

  In the above-described embodiment, the general model 62 and the expert model 66 are examples of models for obtaining attribute values of different types of content under conditions, but the general model 62 and the expert model 66 may not include conditions as conditions. Good. For example, a model that calculates content having the same attribute value regardless of the situation, that is, a model that does not include a situation node may be used. Thereby, it is possible to reduce the process of selecting different contents. Further, the general model 62 may be a model in which the situation is limited. For example, using a model for each situation, such as a model created based on content adoption history data obtained while driving along the coast, a model created based on content adoption history data obtained during the graduation season, etc. Also good. Thereby, an appropriate heterogeneous content can be obtained.

  In the above-described embodiment, the example in which the general model 62 or the expert model 66 is used as a model for obtaining different contents has been described. However, other models may be used. For example, a past user model may be generated using a history of user adoption of content during a predetermined period in the past, and the past model may be used as a model for obtaining different types of content. Since the user's preference is considered to change with time, it is possible to obtain content having an attribute value different from that obtained using the current user model 60 by using the past model. By presenting the heterogeneous content thus obtained, the user can remember and enjoy the old tastes.

  In the above-described embodiment, the example in which the different content selected by the general model 62 or the expert model 66 is displayed together with the user-adapted content has been described. However, instead of the content selected by the models 62 and 66, a predetermined content is set. Content may be displayed. For example, content in a campaign or content of a newly opened store may be presented together with user-adapted content. Thereby, it is possible to prevent the bias of the presented content and to advertise the predetermined content.

  In the above-described embodiment, an example in which content is selected using a model such as the user model 60 and the general model 62 has been described. However, the model is not necessarily used. For example, for user-adapted content, content having an attribute value that has been adopted in the same situation may be selected, or for different content, content that has a different attribute value from the user-adapted content may be selected at random. .

  In the first embodiment described above, as an example of changing the presentation ratio of the user-adapted content and the different type of content, the example of matching the ratio of the respective number of past adoptions has been described. However, the presentation ratio is changed by another method. You can also For example, when recent adoption results are analyzed in chronological order and adoption from user-adapted content continues, processing such as increasing the presentation ratio of different types of content may be performed.

  In the above-described embodiment, when content is presented by the content presentation unit 54, it may be configured such that a predetermined number or more of content having the same attribute value is not displayed. As described above, in order to prevent a predetermined number or more of contents having the same attribute value from being displayed, the number of contents to be selected having the same attribute value is selected when the user-adapted content selection unit 50 selects the contents. You may restrict | limit, and when presenting a content in the content presentation part 54, you may restrict | limit the number of the content which has the same attribute value. Here, the former example will be described as an example.

  The user-adapted content selection unit 50 first obtains an attribute value of content corresponding to a situation or the like. For example, as shown in FIG. 11, it is assumed that a combination of four types of content attribute values corresponding to the current situation is obtained. When selecting eight contents from a large number of contents corresponding to combinations of four types of content attribute values, the user-adapted content selection unit 50 selects “Japanese” or “2000 yen” with the highest likelihood. Rather than selecting eight contents corresponding to the attribute value, two contents corresponding to each of the four types of combinations are selected. As a result, it is possible to provide diversity to content that matches the user's preference.

  Further, in the above-described embodiment, an example in which content is presented by displaying on the display device 30 has been described. However, the content of the content may be output from the speaker 34 as audio. In this case, different audio (for example, male and female audio) may be output so that the user-adapted content and the different content can be identified.

  In the above-described embodiment, the example in which the user model 60 is learned based on whether the presented content is adopted after the content is presented has been described. However, the learning of the user model 60 based on other information is described. May be performed. For example, when a destination is set by the navigation device 1, it may be determined that the destination has been adopted, and an attribute value of the destination and a situation at the time of setting may be obtained and reflected in the user model 60. Even when the destination is not set, when the traveling vehicle stops at the facility, it is determined that the facility has been adopted, the attribute value of the facility and the situation at the time of the stop are obtained, and the user model 60 May be reflected.

  In the above-described embodiment, an example in which content is presented in accordance with the situation has been described. However, the content presentation device of the present invention presents content corresponding to a search condition when a user performs a search. It is also possible to apply it. For example, when a search is performed using “Japanese restaurant” as a search condition, the present invention can also be applied to presenting a restaurant that suits the user's preference among the restaurants included in the category of the Japanese restaurant. In this case, as the attribute value of the content, a genre subdivided from “Japanese style” such as “sushi”, “tonkatsu”, “tempura”, “kaiseki” or the like is used.

  Furthermore, when presenting content in response to an input of a search condition, it is possible to control to change the presentation ratio of the user-adapted content and the heterogeneous content using the data of the search condition. For example, when there are many contents corresponding to the search conditions, control is performed such as reducing the presentation ratio of different contents, and increasing the presentation ratio of different contents when there are few contents corresponding to the search conditions. May be. In this example, when the corresponding content is small, it is considered that there is a need for an unusual content, and the presentation ratio of the different content is increased. Whether there are many contents corresponding to the search condition can be quantified by using, for example, the size of an area to be searched for finding a predetermined number of contents. For example, the distribution of content can be quantified by the radius of an area where 100 contents corresponding to the search condition exist.

  As described above, the present invention has the effect of presenting diversified content according to the user's preference by presenting different types of content having different attribute values from the user-adapted content together with the user-adapted content. For example, it is useful as a navigation device or the like that presents a restaurant, a sightseeing spot or the like as content.

It is a figure which shows the structure of the navigation apparatus of 1st Embodiment. It is a figure which shows the structure of the control apparatus in 1st Embodiment. It is a figure which shows the example of the facility data memorize | stored in map DB. (A) It is a figure which shows the example of the model of a Bayesian network. (B) It is a figure which shows the example of the conditional probability table | surface in a Bayesian network. (C) It is a figure which shows the frequency table for calculating | requiring a conditional probability table | surface. It is a figure which shows the example which has arrange | positioned the general model and the user model on the feature space. It is a figure which shows the example of the screen shown in the content presentation part. It is a figure which shows operation | movement of the navigation apparatus of 1st Embodiment. It is a figure which shows the structure of the control apparatus in 2nd Embodiment. (A) It is a figure which arrange | positions and shows a user model, a general model, and an expert model in the feature space. (B) It is a figure which arrange | positions and shows a user model, a general model, and an expert model in the feature space. It is a figure which shows the example of the screen shown in the content presentation part. It is a figure which shows the example which selects the content corresponding to the value of a content attribute.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Navigation apparatus 10 Position detector 12 Geomagnetic sensor 14 Gyroscope 16 Distance sensor 18 GPS receiver 20 Map DB
22 Operation switch group 24 Control device 26 Communication unit 28 Memory 30 Display device 32 Audio controller 34 Speaker 36 Voice recognition device 38 Microphone 40 Remote control sensor 42 Remote control 44 Content information center 50 User-adapted content selection unit 52 Heterogeneous content selection unit 54 Content presentation unit 56 Presentation ratio determining unit 58 Model learning unit 60 User model 62 General model 64 Content adoption history data 66 Expert model

Claims (13)

A content storage unit that stores data indicating a facility and its attribute value and position as content;
A search condition input unit that accepts input of search conditions;
A condition search unit that searches the content storage unit for a plurality of contents that match the search condition input in the search condition input unit;
A content distribution calculation unit for obtaining a distribution of content searched by the condition search unit;
A user-adapted content selection unit that selects content suitable for the user from the content searched by the condition search unit based on the history of the content adopted by the user;
A heterogeneous content selection unit that selects content having a different attribute value from the content selected by the user-adapted content selection unit from the content stored in the content storage unit;
A presentation ratio determination unit that determines a presentation ratio of the user-adapted content and the different type content, and when the content distribution calculated by the content distribution calculation unit is distributed, the content distribution is distributed. A presentation ratio determination unit that sets a ratio of different kinds of content larger than the case where
A content presentation unit that presents the user-adapted content selected by the user-adapted content selection unit and the heterogeneous content selected by the heterogeneous content selection unit according to the presentation ratio ;
A content presentation device comprising: The user-adapted content selection unit has a user model representing a relationship between a situation when a user adopts content and an attribute value of the adopted content,
The content presentation apparatus according to claim 1, wherein an attribute value of content corresponding to a current situation is obtained using the user model, and content having the attribute value is selected as the user-adapted content. The heterogeneous content selection unit has another model different from the user model representing the relationship between the situation or user attribute value and the content attribute value,
The content presentation apparatus according to claim 2, wherein an attribute value of content corresponding to a current situation or a user attribute value is obtained using the different model, and content having the content attribute value is selected as the heterogeneous content.   The content presentation apparatus according to claim 3, wherein the different model is a model generated using content adoption history data obtained from a plurality of users.   The content presentation apparatus according to claim 4, wherein the another model is a model generated using data of content adoption history in a specific situation. The user model is a model generated using a history of content adopted by a user during a recent predetermined period,
The content presentation apparatus according to claim 3, wherein the different model is a model generated using a history of content adopted by a user during a period longer than the predetermined period.   The content presentation device according to claim 3, wherein, when presenting the heterogeneous content, the content presenting unit presents information related to another model in which the heterogeneous content is selected.   The content presentation apparatus according to claim 1, wherein the heterogeneous content selection unit selects a predetermined content.   The content presentation apparatus according to any one of claims 1 to 8, wherein the number of contents to be presented having the same attribute value is set in the content presentation unit.   The content presentation device according to claim 1, wherein the content presentation unit presents the user-adapted content and the heterogeneous content in an identifiable manner.   The content presentation device according to claim 1, wherein the content presentation unit displays the user-adapted content and the heterogeneous content in random order. Content presentation device, comprising: providing a content storage unit which stores data indicating the facility and its attribute value and position of the content,
The content presentation device accepting an input of a search condition;
The content presentation device searching the content storage unit for a plurality of contents that match the search condition;
The content presentation device obtaining a distribution of searched content;
The content presentation device selecting content suitable for the user as user-adapted content from the searched content based on the history of the content adopted by the user;
The content presentation device selecting, as the heterogeneous content, content having a different attribute value from the user-conforming content from the content storage unit;
The content presentation device is a step of determining a presentation ratio of the user-adapted content and the heterogeneous content, and when the content distribution is distributed, the content distribution is different from the case where the content distribution is not distributed. A step to increase the content percentage,
The content presentation device presenting the user-adapted content and the heterogeneous content according to the presentation ratio ;
A content presentation method comprising: A program for presenting content, and a computer having a content storage unit that stores data indicating a facility as a content and its attribute value and position ,
A step of accepting input of search conditions;
Searching a plurality of contents satisfying the search condition from the content storage unit;
Determining the distribution of searched content;
Selecting content suitable for the user from the searched content based on the history of the content adopted by the user ;
From the content storage unit, a step of, based on the history of content that the user employs to select a content matching the user as matching content,
Selecting from the content storage unit content different in attribute value from the user-adapted content as heterogeneous content;
A step of determining a presentation ratio of the user-adapted content and the dissimilar content, wherein when the distribution of the content is distributed, the ratio of the dissimilar content is set larger than when the distribution of the content is not distributed And steps to
Presenting the user-adapted content and the heterogeneous content according to the presentation ratio ;
A program that executes