JP2008003869A - Retrieval device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a retrieval device capable of performing retrieval that meets a variety of requirements of a user without making the user's input operation complicated. <P>SOLUTION: The retrieval device comprises a storage part for storing correspondence information showing the correspondence between a plurality of kinds of attribute values and retrieval targets for each of a plurality of retrieval targets; a reference attribute determination part for determining one of the attribute values as the attribute value of the reference attribute; a first attribute value varying part for varying the attribute value of the reference attribute according to the user's instruction; a second attribute value varying part that, when the attribute value of the reference attribute is varied, varies the attribute value according to the correlations between the reference attribute and at least one attribute other than the reference attribute determined based on the correspondence information, and the variation of the attribute value of the reference attribute; a display part for displaying information showing the attribute values before and after the variation on a display screen; and a retrieval part for performing retrieval to a plurality of retrieval targets using the respective attribute values after the variation. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a search device, and more particularly to a search device for searching for information desired by a user from a plurality of pieces of information.

  2. Description of the Related Art Conventionally, a search device has been proposed that searches for information desired by a user from a plurality of pieces of information in response to individual requests from the user (see, for example, Patent Document 1). Specifically, when the user operates the operation switch (single-axis slider bar) to change the parking fee, the search device searches for a parking lot corresponding to the changed parking fee. The parking fee is an attribute of the parking lot. In addition, the price of the parking fee, that is, the attribute value (hereinafter referred to as the attribute value) indicates the characteristics of the parking lot (low price, high price, etc.). The searched parking lot is displayed as a search result on the display unit of the search device.

  Here, there are various user requests. Therefore, it is not possible to perform a search satisfying various requests of a user only by performing a search using only one type of attribute. Therefore, a search device that realizes a search that satisfies various user requirements has been proposed (see, for example, Non-Patent Document 1). Hereinafter, a conventional search device that performs a search using a plurality of types of attributes will be described with reference to FIG. FIG. 16 is a diagram illustrating an input screen and a search result screen of a conventional search device that performs a search using a plurality of types of attributes. A regular pentagonal radar chart is displayed on the input screen shown in FIG. Specifically, axes 91 corresponding to the respective attributes 1 to 5 are arranged on the input screen. The user moves the icon 92 of each attribute displayed on the input screen using a mouse or the like on the axis, and changes each attribute value one by one for each attribute. For example, to change attribute 1 to 80% and attribute 2 to 50%, the user first changes attribute 1 to 80%, and then changes attribute 2 to 50%.

In the conventional search device shown in FIG. 16, when a user searches for a book or a CD, the user changes attribute values of attributes such as popularity and originality one by one for each attribute. Thereafter, the search device can search for a book or CD corresponding to each attribute value after the change. The searched book or CD is displayed on the search result screen. The image data shown on the search result screen is, for example, a cover image when a book is searched. When a CD is searched, the image data is a CD jacket image.
Japanese Patent No. 3291264 CBJ Corporation, “Press Release”, [online], May 18, 2005, Internet, <URL: http: // www. kbmj. com / press_release / pr20050518. html>

  However, in the conventional search device shown in FIG. 16, it is necessary to change each attribute value one by one for each attribute. For this reason, there existed a problem that a user's input operation will become complicated. Further, in order to perform a search that satisfies various user requirements, it is necessary to further increase the types of attributes. However, since it is necessary to change each attribute value one by one for each attribute, the user's input operation becomes complicated as the types of attributes are increased. As described above, in the conventional search device shown in FIG. 16, it is difficult to realize a search corresponding to various requests of the user without complicating the user's input operation.

  Therefore, an object of the present invention is to provide a search device that can perform a search that satisfies various user requirements without complicating the user's input operation.

  A first invention is a search device that performs a search for a plurality of search targets using a plurality of types of attribute values respectively indicating characteristics of the search target, and the correspondence between the plurality of types of attribute values and the search targets A storage unit that stores correspondence information for each of a plurality of search targets, a reference attribute determination unit that determines one attribute value of a plurality of types of attribute values as an attribute value of a reference attribute, and an attribute value of the reference attribute A first attribute value changing unit that changes in accordance with a user's instruction; and a reference attribute determined based on correspondence information and at least one other attribute other than the reference attribute when the attribute value of the reference attribute changes A second attribute value changing unit that changes the attribute value according to a correlation with other attributes and a change in the attribute value of the reference attribute, and each attribute before and after the change by the first and second attribute value changing units Display value information on the display screen Comprising a radical 113, and a search unit using the attribute values after variation by the first and second attribute value change unit to search for a plurality of search target.

  In a second aspect based on the first aspect, the second attribute value variation unit, when the correlation between the reference attribute and another attribute is a positive correlation, the variation direction of the attribute value of the reference attribute If the correlation between the reference attribute and the other attribute is negative, the other attribute is changed in the opposite direction to the change direction of the attribute value of the reference attribute. It is characterized by changing the attribute value of the attribute.

  A third invention further includes a correlation coefficient calculation unit that calculates a correlation coefficient indicating a correlation based on the correspondence information in the second invention, and the second attribute value changing unit includes the reference attribute and the like. The attribute value of the other attribute is varied by an amount corresponding to the absolute value of the correlation coefficient indicating the correlation with the attribute.

  In a fourth aspect based on the first aspect, the second attribute value variation unit further varies the attribute values of other attributes when the number of retrieval results by the retrieval unit is less than a predetermined number. Features.

  In a fifth aspect based on the first aspect, a radar chart having axes respectively corresponding to a plurality of types of attributes is displayed on the display screen, and the positional relationship of each axis of the radar chart to be displayed is correlated. An axis determining unit that determines based on the information is further provided.

  In a sixth aspect based on the fifth aspect, the axis determination unit determines the angle between the axis corresponding to the reference attribute and the axis corresponding to the other attribute, and the correlation between the reference attribute and the other attribute. It determines based on.

  In a seventh aspect based on the fifth aspect, if the correlation between the reference attribute and the other attribute is a positive correlation, the axis determining unit determines whether the axis corresponding to the reference attribute and the other attribute are When the angle between the corresponding axis is an acute angle and the correlation between the reference attribute and the other attribute is negative, the angle between the axis corresponding to the reference attribute and the axis corresponding to the other attribute Is an obtuse angle.

  In an eighth aspect based on the first aspect, the search target is music information, and the plurality of types of attribute values include attribute values representing physical characteristics of the music.

  In a ninth aspect based on the eighth aspect, the plurality of types of attribute values include spectral change, average number of pronunciations, pronunciation aperiodicity, beat period, beat period ratio, first night intensity, and second night intensity. And an attribute value indicating at least one of the beat intensity ratios.

  In a tenth aspect based on the first aspect, the search target is music information, and the plurality of types of attribute values include an attribute value indicating an impression value of the music.

  In an eleventh aspect based on the tenth aspect, the plurality of types of attribute values include attribute values indicating at least one of the intensity, dynamism, exhilaration, simplicity, and softness of the music It is characterized by that.

  In a twelfth aspect based on the first aspect, the plurality of types of attribute values include predetermined attribute values whose values fluctuate depending on the situation around the user, and the situation information indicating the situation around the user If the situation acquisition unit acquires status information, the attribute acquisition unit updates the predetermined attribute value included in the correspondence information based on the acquired situation information, and the correspondence information A correlation determining unit that determines a correlation based on the correlation;

  In a thirteenth aspect based on the twelfth aspect, the search target is information on a facility existing on the map, and an attribute value that varies depending on a situation around the user indicates a distance from the facility to the current position of the user. The situation acquisition unit acquires the current position of the user as the situation information.

  In a fourteenth aspect based on the first aspect, when the content of the correspondence information is updated, the fourteenth aspect further includes a correlation determination unit that newly determines a correlation based on the updated correspondence information, and the second attribute The value changing unit changes the attribute value according to the new correlation determined by the correlation determining unit when the content of the correspondence information is updated.

  A fifteenth aspect of the invention is a search method for performing a search on a plurality of search targets using a plurality of types of attribute values respectively indicating characteristics of the search target, and the search device includes a plurality of types of attribute values and Reference information determination step for storing correspondence information indicating correspondence with a search target for each of a plurality of search targets, and determining one attribute value of a plurality of types of attribute values as an attribute value of a reference attribute; A first attribute value changing step for changing the attribute value according to the user's instruction, and when the attribute value of the reference attribute changes, at least one other attribute other than the reference attribute is determined based on the correspondence information A second attribute value changing step for changing the attribute value in accordance with a correlation between the reference attribute and the other attribute and a change in the attribute value of the reference attribute, and a change in the first and second attribute value changing steps Before and after A display step for displaying information indicating each attribute value on the display screen, and a search step for performing a search for a plurality of search targets using each attribute value after the change in the first and second attribute value change steps. Including.

  According to a sixteenth aspect of the invention, there is provided a program for causing a search device computer to execute a search for a plurality of search targets using a plurality of types of attribute values respectively indicating characteristics of the search target. Stores correspondence information indicating correspondence between a plurality of types of attribute values and search targets for each of the plurality of search targets, and determines one attribute value of the plurality of types of attribute values as the attribute value of the reference attribute. A reference attribute determination step, a first attribute value change step for changing the attribute value of the reference attribute according to a user instruction, and when the attribute value of the reference attribute changes, at least one other attribute other than the reference attribute, A second attribute value changing step for changing the attribute value in accordance with the correlation between the reference attribute determined based on the correspondence information and the other attribute, and the change in the attribute value of the reference attribute; A display step for displaying information indicating each attribute value before and after the change in the second attribute value change step on the display screen, and a plurality of searches using each attribute value after the change in the first and second attribute value change steps A program for causing a computer to execute a search step for performing a search on a target.

  According to the first aspect, when the attribute value of the reference attribute changes according to the user's instruction, the attribute values of other attributes also change according to the correlation and the change of the attribute value of the reference attribute. Thereby, it is possible to perform a search without the user changing a plurality of types of attribute values one by one for each attribute. As a result, it is possible to provide a search device that can perform a search that satisfies various user requirements without complicating the user's input operation. Also, according to the present invention, information indicating each attribute value before and after the change is displayed, so that the user can know that a search according to his / her intention is being performed.

  According to the second aspect, the search device can change the attribute values of other attributes to the optimum attribute values when performing a search desired by the user. That is, the attribute value of another attribute can be set to an optimum value according to the attribute value of the reference attribute designated by the user, and a search according to the user's intention can be realized.

  According to the third aspect, attribute values of other attributes can be changed to more optimal attribute values. As a result, it is possible to realize a search more in line with the user's intention.

  According to the fourth aspect of the invention, it is possible to prevent the number of search results from being smaller than the predetermined number, and to make the number of search results easy for the user to select.

  According to the fifth aspect, the positional relationship between the axes corresponding to each attribute is a positional relationship based on the correlation, and the user can more intuitively refer to the positional relationship between the axes of the radar chart. Search can be performed.

  According to the sixth aspect, the user can grasp the correlation between the attributes by referring to the angle formed between the axis corresponding to the reference attribute and the axis corresponding to the other attribute. As a result, a more intuitive search can be provided to the user.

  According to the seventh aspect, by referring to the positional relationship of each axis of the radar chart, the user can grasp whether the correlation of each attribute is a positive correlation or a negative correlation. . As a result, a more intuitive search can be provided to the user.

  According to the eighth aspect, the user can search for music information using the attribute value indicating the physical characteristics of the music.

  According to the ninth aspect, the user can select at least one of spectrum change, average number of pronunciations, pronunciation aperiodicity, beat period, beat period ratio, first night intensity, second night intensity, and beat intensity ratio. By using an attribute value indicating one, the music information can be searched.

  According to the tenth aspect, the user can search for music information using the attribute value indicating the impression value of the music.

  According to the eleventh aspect, the user searches for music information using an attribute value indicating at least one of the intensity, vigor, refreshment, simplicity and softness of the music. Can do.

  According to the twelfth aspect of the present invention, the correlation can be dynamically changed according to changes in the situation around the user. As a result, it is possible to perform a search according to a change in the situation around the user.

  According to the thirteenth aspect, the correlation can be dynamically changed according to the current position of the user. As a result, a search according to the current position of the user can be performed.

  According to the fourteenth aspect, changes in attribute values of other attributes can be dynamically changed in accordance with changes in the contents of the correspondence information. Thus, for example, when searching for music information, even if information on some of the music information is replaced, a search according to the new correlation after the replacement is performed. Can do.

(First embodiment)
A search device 1 according to a first embodiment of the present invention will be described with reference to FIG. FIG. 1 is a block diagram illustrating a configuration example of a search device 1 according to the first embodiment. In FIG. 1, the search device 1 includes a control unit 11, an information storage unit 12, an input unit 13, and an output unit 14. The control unit 11 includes a correlation determination unit 111, an axis determination unit 112, an attribute value variation unit 113, a search unit 114, and a drawing unit 115. The output unit 14 includes a display unit 141 and an audio playback unit 142. Hereinafter, each component will be described.

  The control unit 11 includes, for example, a CPU or MPU, and a ROM (Read Only Memory) or a RAM (Random Access Memory). The control unit 11 controls processing of the entire search device 1. Note that the search device 1 has a working area unit (not shown) that functions as a separate memory (typically RAM), and the control unit 11 performs processing while reading and writing information necessary for processing in the working area unit. May be advanced. Each component of the control unit 11 will be described later.

  The information storage unit 12 is composed of, for example, an HDD or a DVD. The information storage unit 12 stores a plurality of search target information such as music, books, movies, or facilities. The information storage unit 12 stores the information table shown in FIG. FIG. 2 is a diagram illustrating an example of an information table stored in the information storage unit 12. The information table includes an information identification ID for identifying information to be searched. The information table includes a plurality of types of attribute values (attribute 1 to attribute 8) for each information identification ID. Here, for example, when the attribute 1 indicates “severity”, it can be seen that the search target whose information identification ID is 1 has a feature that the severity is “10%”. As described above, the attribute value indicates the characteristic of the information to be searched. The information identification ID is for identifying information to be searched. Therefore, the information table can be said to be correspondence information in which information to be searched is associated with a plurality of types of attribute values.

  Each attribute value is a value obtained by normalizing an actual value so as to be within a range of, for example, 0% to 100%. Each attribute value may be an actual value before normalization. In this case, normalization is performed at an appropriate timing according to the processing of the search device. When the information to be searched is, for example, music information, the audio data of the multimedia content corresponding to each information identification ID is stored in the information storage unit 12 together with the information table shown in FIG. In FIG. 2, eight types of attributes (attribute 1 to attribute 8) are shown, but the present invention is not limited to this. There may be eight or more types of attributes, or fewer than eight types. Further, the types of attributes may be different for each information identification ID. The information to be searched and the information table shown in FIG. 2 are appropriately downloaded from the center facility (not shown) to the information storage unit 12 by communication means (not shown) such as a mobile phone, PHS, or WiMAX. It may be.

  The input unit 13 receives user input. The input unit 13 is configured by arranging a predetermined number of push-type switches, for example. The input unit 13 may be configured with, for example, a touch panel, a remote control, or a mouse, or may be configured with, for example, a microphone or a voice recognition engine that recognizes a user's voice and converts it into input information to the search device 1. May be.

  The display unit 141 is configured by, for example, a liquid crystal display, a plasma display, or an organic EL display. The display unit 141 displays the screen drawn by the drawing unit 115. Specifically, the display unit 141 displays an input screen for the user to perform an input operation to change the attribute value and a search result screen for displaying the search result. When the input unit 13 is a touch panel, the display unit 141 and the input unit 13 are integrally configured.

  The audio reproducing unit 142 is configured by an amplifier and a speaker, for example. The audio playback unit 142 plays back the audio data of the multimedia content stored in the information storage unit 12. The audio reproducing unit 142 is not limited to an amplifier and a speaker, and may have any configuration as long as the audio data stored in the information storage unit 12 can be reproduced. For example, when audio data is stored in the linear PCM format in the information storage unit 12, the audio reproduction unit 142 may include a DA converter. In addition, when audio compression encoded data is stored in the information storage unit 12, the audio reproduction unit 142 may include a decompression unit that decompresses compression.

  Next, each component of the control unit 11 will be described. The correlation determination unit 111 determines the correlation of each attribute based on the information table stored in the information storage unit 12. Specifically, the correlation determining unit 111 sets an attribute that the user changes among the attributes as a reference attribute (hereinafter referred to as a reference attribute). Then, the correlation determination unit 111 calculates a correlation coefficient between the reference attribute and other attributes. Here, the range of values of the correlation coefficient is a range from −1 to 1. The correlation coefficient is calculated based on the attribute value included in the information table. After calculating the correlation coefficient, the correlation determining unit 111 determines the correlation for each attribute according to the correspondence shown in FIG. FIG. 3 is a diagram showing an example of the correspondence between the correlation coefficient and the correlation. As shown in FIG. 3, the correlation of attributes whose correlation coefficient is negative (a range of −1 or more and less than 0) is a negative correlation. Further, the correlation of attributes whose correlation coefficient is positive (range of 0 or more and 1 or less) is a positive correlation. Furthermore, these positive and negative correlations are classified into four types according to the value of the correlation coefficient: “strong correlation”, “medium correlation”, “weak correlation”, and “almost no correlation”. It is divided. Correlation determining section 111 determines a total of eight types of correlation classified into the above four types for each of positive and negative correlations from attributes other than the reference attribute from the values of correlation coefficients. FIG. 4 is a diagram showing the correlation between attribute 2 to attribute 8 included in the information table of FIG. In FIG. 4, attribute 1 is the reference attribute. In FIG. 4, for example, the correlation of attribute 2 is “medium positive correlation”.

  The axis determination unit 112 determines the positional relationship of the axes corresponding to the plurality of types of attributes based on the correlation. Here, the axis is an axis that the radar chart has. The radar chart is one of display formats for displaying each attribute value. FIG. 5 is a diagram illustrating an example of an input screen and a search result screen displayed on the display unit 141. As shown in the input screen of FIG. 5, here, the radar chart has eight axes 15a to 15h. Attributes 1 to 8 are associated with each axis. Specifically, the axis 15a is attribute 1, the axis 15b is attribute 2, the axis 15c is attribute 6, the axis 15d is attribute 4, the axis 15e is attribute 8, the axis 15f is attribute 5, the axis 15g is attribute 7, and the axis 15h is Each corresponds to attribute 3. The position on the axis is associated with the attribute value. Specifically, the attribute value at the center 16 is set to 0%, and the attribute value at the vertex constituting the regular octagon is set to 100%. A method for determining the positional relationship between the shafts 15a to 15h will be specifically described later.

  On the input screen shown in FIG. 5, icons 17 indicating attribute names and attribute values for each axis are displayed. The icon 17 is arranged so as to be movable on the axis. The user can change the attribute value by moving the icon 17 on the axis by the input unit 13. As described above, the input unit 13 varies the attribute value according to the user's instruction, and corresponds to the first attribute value variation unit according to the present invention.

  When the attribute value of the reference attribute changes, the attribute value changing unit 113 changes the attribute values of attributes other than the reference attribute according to the change of the attribute value of the reference attribute and the correlation. How the attribute values of other attributes change will be described in detail later. This is the end of the description of each component of the control unit 11.

  The search unit 114 performs a search using each attribute value after the change from the information to be searched stored in the information storage unit 12. As an example of a search method, the search unit 114 calculates, for each attribute, the difference between the attribute value shown on the input screen and the attribute value of the information table, and obtains information whose sum of absolute values of the difference is a predetermined value or less. Search for. In other words, the search unit 114 calculates the degree of similarity to each attribute value on the input screen (hereinafter referred to as similarity) by the sum of the absolute values of the differences, and the similarity is higher than a predetermined value (the difference The information whose value is the sum of the absolute values of these is smaller than a predetermined value is taken as the search result. The search is performed every time the user changes the attribute value of the reference attribute on the input screen, and the search result screen is updated each time the search is performed. In this way, the search result is dynamically displayed on the search result screen.

  As illustrated in FIG. 5, the drawing unit 115 draws an input screen for the user to change the attribute value of the reference attribute by an input operation and a search result screen for displaying the search result. In addition to the radar chart, the drawing unit 115 draws an UP / DOWN button 18 for designating the upper limit (10 in FIG. 5) of the number of search results displayed on the search result screen in addition to the radar chart. . The drawing unit 115 draws the number of search results (50 in FIG. 5) and 10 pieces of information arranged in descending order of similarity from the search result screen. Each of the ten pieces of information is assigned a ranking corresponding to the degree of similarity. In addition, image data is attached to each of the 10 pieces of information. The image data shown on the search result screen is, for example, a cover image when a book is searched. When a CD is searched, the image data is a CD jacket image. If the number of search results is less than 10 (for example, 5), information for 5 cases is drawn on the search result screen. Note that the upper limit of the display number of search results may be set in advance on the search device 1 side in addition to being designated by the user using the UP / DOWN button 18. In this case, the UP / DOWN button 18 need not be displayed on the input screen. Further, all the search results may be displayed on the search result screen without providing an upper limit on the number of display items.

  Next, with reference to FIG. 6, the flow of processing of the search device 1 according to the first embodiment will be described. FIG. 6 is a flowchart showing the flow of processing of the search device 1 according to the first embodiment. In the following description, it is assumed that the information to be searched is information managed by the information table in FIG. 2 and the reference attribute is attribute 1. Assume that the correlation between attribute 2 and attribute 8 is the correlation shown in FIG. 4, the correlation coefficient of attribute 2 is 0.5, and the correlation coefficient of attribute 8 is -1. Further, it is assumed that the input screen shown in FIG. 5 is displayed on the input screen.

  In FIG. 6, the correlation determining unit 111 sets the attribute that the user changes among the attributes of the information table illustrated in FIG. 2 as a reference attribute (step S <b> 101). That is, the attribute value that is changed by the user is determined as the attribute value of the reference attribute. As a method of determining the reference attribute, for example, the user himself / herself may be determined by designating the reference attribute. Further, for example, the search device 1 itself may automatically determine an attribute with a high operation frequency as a reference attribute with reference to a user's past operation history. Thus, the process of step S101 corresponds to the process of the reference attribute determination unit in the present invention.

  Following step S101, the correlation determining unit 111 calculates a correlation coefficient between the attribute 1 that is the reference attribute and the attributes 2 to 8 using each attribute value included in the information table shown in FIG. To do. Then, the correlation determining unit 111 determines the correlation between attribute 2 and attribute 8 (step S102). Of the information identified by the information identification ID shown in FIG. 2, for information consisting of attribute values whose values do not vary (hereinafter referred to as static attribute values), a search request from the user is generated. It is desirable to calculate the correlation coefficient and determine the correlation in advance. If the information consisting of static attribute values is, for example, music information, the correlation coefficient of each attribute is calculated and the correlation is determined when the music information is ripped from a CD or the like and stored in the information storage unit 12. It is desirable to keep it.

  After step S102, the axis determination unit 112 determines the positional relationship of the axes corresponding to attribute 1 to attribute 8 based on the correlation (step S103). In step S103, it is assumed that the positional relationship between the eight axes 15a to 15h corresponding to each of the attributes 1 to 8 is determined as shown in the input screen of FIG. Here, as an example, it is assumed that the angle formed between the axes of the radar chart is an angle that forms a regular octagon. Specifically, the axis determination unit 112 determines the axis 15a connecting the top vertex of the regular octagonal vertices and the center 16 as the axis of attribute 1 which is the reference attribute. The correlation between attribute 2 and attribute 3 is “medium positive correlation”. Therefore, the axis determination unit 112 determines the axes of attribute 2 and attribute 3 as the axes 15b and 15h at positions closest to the axis 15a of attribute 1. The correlation of attribute 8 is “There is a strong negative correlation”. Therefore, the axis determination unit 112 determines the axis of attribute 8 as the axis 15e farthest from the axis 15a of attribute 1. That is, the axis 15e with the attribute 8 is arranged at a position symmetrical to the axis 15a with the attribute 1 with respect to the center 16, as shown in the input screen of FIG. The correlation between attribute 4 and attribute 5 is “there is a moderate negative correlation”. Therefore, the axis determination unit 112 determines the axes of the attribute 4 and the attribute 5 as the axes 15d and 15f at positions far from the axis 15e of the attribute 8 with respect to the axis 15a of the attribute 1. The correlation between attribute 6 and attribute 7 is “almost no correlation”. Therefore, the axis determination unit 112 determines the axis of attribute 6 to be the axis 15c between the axes 15b and 15d of attribute 2 and attribute 4. Similarly, the axis of attribute 7 is determined to be the axis 15g between the axes 15h and 15f of attribute 3 and attribute 5.

  According to the positional relationship between the axes 15a to 15h shown in FIG. 5, the positions of the axes 15b and 15h having the positive correlation are positions where the angle formed with the axis 15a of the reference attribute is an acute angle. ing. Also, the positions of the axes 15d and 15f of the attribute having a negative correlation are positions where the angle formed with the axis 15a of the reference attribute is an obtuse angle. By determining each of the axes 15a to 15h in such a positional relationship, the user can refer to the positional relationship of each axis included in the radar chart so that the correlation between the attributes is positive or negative. It is possible to intuitively grasp whether the correlation is present. As a result, a more intuitive search can be provided to the user.

  Following step S103, the drawing unit 115 draws an initial screen of the input screen and the search result screen, and displays it on the display unit 141 (step S104). Here, nothing is displayed on the search result screen because no search has been performed yet. On the other hand, the axis of each attribute whose positional relationship has been determined in step S103 is displayed on the input screen. Here, the initial values of the attributes are all set to 50%, for example.

  Next to step S104, when the user changes the attribute value of attribute 1, which is the reference attribute, in the input unit 13, the attribute value changing unit 113 determines whether the attribute value of attribute 1 has changed (step S105). ). When the attribute value of attribute 1 varies, the attribute value variation unit 113 detects the variation direction of the attribute value of attribute 1 and the variation amount from the initial value, and other attributes based on the detected variation direction and variation amount. The attribute values of 2 to 8 are changed (step S106).

  Here, the change direction of the attribute value of the other attribute is a direction based on the positive or negative direction of the correlation. Specifically, when the attribute value of attribute 1 increases, the attribute value of the attribute having a positive correlation varies in the increasing direction, and the attribute value of the attribute having the negative correlation varies in the decreasing direction. That is, the change direction of the attribute that is a positive correlation is the same direction as the change direction of attribute 1 that is the reference attribute, and the change direction of the attribute that is negative correlation is the opposite direction to the change direction of attribute 1 that is the reference attribute. .

The variation amount of the attribute value of the other attribute is an amount corresponding to the absolute value of the correlation coefficient of the other attribute. Here, as an example, the variation amount is defined as a variation amount X represented by the following equation.
(Variation amount X) = (variation amount of reference attribute) × (absolute value of correlation coefficient)
Here, for example, it is assumed that the attribute 1 has changed from an initial value of 50% to 40%. At this time, since the correlation coefficient of the attribute 8 is set to −1, the variation amount of the attribute 8 is 10% × | −1 | = 10%. | -1 | indicates the absolute value of -1. The attribute 8 has a negative correlation, and the direction of change is opposite to the direction of change of the reference attribute as described above. Therefore, the attribute value of attribute 8 varies from the initial value of 50% to 60%. Next, attribute 2 is considered. Since the correlation coefficient of attribute 2 is 0.5, the variation amount of attribute 2 is 10% × | 0.5 | = 5% from the above equation. Also, attribute 2 is a positive correlation, and its fluctuation direction is the same as the fluctuation direction of the reference attribute as described above. Therefore, the attribute value of attribute 2 varies from the initial value of 50% to 55%. Attribute 3 to attribute 5 also vary according to the variation direction described above and the variation amount corresponding to the absolute value of the correlation coefficient. Finally, consider attribute 6 and attribute 7. The correlation between attribute 6 and attribute 7 is “almost no correlation”. Therefore, even if the attribute 1 as the reference attribute changes, the attribute values of the attribute 6 and the attribute 7 do not change. Note that only the attribute having no correlation (correlation coefficient is 0) may not be changed. When the attribute value of attribute 1 that is the reference attribute varies in this manner, the attribute values of other attributes vary based on the variation of the attribute 1 and the correlation. That is, the attribute values of the other attributes are uniquely determined with respect to the attribute value of attribute 1 that is the reference attribute.

  After step S106, the search unit 114 performs a search using each attribute value after the change from the information to be searched stored in the information storage unit 12 (step S107). The drawing unit 115 draws the result searched in step S107 on the search result screen (step S108). Note that the attribute values before and after the change are drawn on the input screen. After step S108, it is determined whether or not to end the process (step S109). If the search is further performed with the attribute 1 as the reference attribute having a different value without ending the process, the process returns to step S105. By such processing in steps S105 to S109, the search is performed every time the user changes the attribute value of the reference attribute on the input screen. The search result screen is updated every time a search is performed.

  As described above, in the search device 1 according to the present embodiment, when the attribute value of the reference attribute changes due to a user input operation, the attribute value of the other attribute changes based on the change and correlation of the reference attribute. To do. Thus, the search can be performed without the user changing the attribute value one by one for each of a plurality of types of attributes. As a result, it is possible to provide a search device that can perform a search that satisfies various user requirements without complicating the user's input operation.

  In the search device 1 according to the present embodiment, the attribute values of other attributes vary based on the correlation with the reference attribute. As a result, the attribute value of the other attribute can be changed to the optimum attribute value when performing the search desired by the user. That is, the attribute value of another attribute can be set to an optimum value according to the attribute value of the reference attribute designated by the user, and a search according to the user's intention can be realized. In addition, the user's satisfaction with the search can be improved.

  In the search device 1 according to the present embodiment, the positional relationship between the axes of the attributes is determined based on the correlation. In other words, the positional relationship between the axes of each attribute indicates the correlation between the respective attributes, and the user tries to search for information on what trend is currently by referring to the positional relationship between the axes of each attribute that the radar chart has. It is possible to grasp intuitively whether there is. As a result, an intuitive search can be provided to the user.

  The positional relationship between the axes determined by the axis determining unit 112 is not limited to the positional relationship shown in FIG. The axis determination unit 112 may determine the positional relationship as shown in FIG. FIG. 7 is a diagram schematically showing another example of the positional relationship between the axes. In FIG. 7, the axis determination unit 112 first determines the axis of the reference attribute (for example, attribute 1) at the position A shown in FIG. The axis of attribute 1 as the reference attribute is determined at a position directly above the center 16 with the center 16 as a starting point. Next, the axis determination unit 112 determines an angle between the attribute axis to be determined and the attribute 1 axis in accordance with the correlation coefficient. Specifically, if the correlation coefficient is a positive value of 0 or more, the angle between the axis of attribute 1 is alternately determined in the range of “0 degrees to 90 degrees” and “0 degrees to −90 degrees”. To do. Note that the closer the correlation coefficient is to 1, the smaller the angle. Further, a clockwise angle from the attribute 1 axis is a positive angle. If the correlation coefficient is a negative value smaller than 0, the angle between the axis of attribute 1 is alternately determined in the range of “90 degrees to 180 degrees” and “−90 degrees to −180 degrees”. Note that, as the correlation coefficient is closer to -1, the angle is set to a value closer to 180 degrees or -180 degrees.

Hereinafter, FIG. 7 will be described more specifically. In FIG. 7, with respect to attribute 1, which is the reference attribute, the correlation coefficient of attribute 2 is r2, the correlation coefficient of attribute 3 is r3, the correlation coefficient of attribute 4 is r4, and the correlation coefficient of attribute 5 is r5. To do. Also, the angle between the attribute 1 axis and the attribute 2 axis is θ ₂ , the angle between the attribute 1 axis and the attribute 3 axis is θ ₃ , and the angle between the attribute 1 axis and the attribute 4 axis is The angle between θ ₄ , the attribute 1 axis and the attribute 5 axis is θ ₅ . Here, when a relationship of 0 ≦ r3 <r2 <1 between r2 and r3, theta _2, the _{θ 3 | θ 2 | <|} θ 3 | relation holds. Further, if a relationship of -1 <r4 <r5 <0 between r4 and r5, theta _4, the _{θ 5 | θ 5 | <|} θ 4 | relation holds. In addition, | θ ₂ |, | θ ₃ |, | θ ₄ |, and | θ ₅ | are values indicating absolute values. As shown in FIG. 7, each attribute axis has an attribute axis having a positive correlation coefficient above the horizontal axis passing through the center 16, and a correlation coefficient below the horizontal axis. An axis with an attribute that is negative will be placed. An attribute axis arranged at a position closer to the horizontal axis is an attribute having no correlation with the attribute 1 that is the reference attribute.

  Here, when there are six types of attributes (attribute 1 to attribute 6), FIG. 8 shows the positional relationship between the axes arranged by the determination method shown in FIG. According to FIG. 8, attribute 3 and attribute 4 have a stronger positive correlation than attribute 5 and attribute 6, and attribute 5 and attribute 6 are weaker than attribute 3 and attribute 4, with respect to attribute 1, which is the reference attribute. It can be seen that there is a positive correlation. It can also be seen that attribute 2 has a strong negative correlation. As described above, the axis determination unit 112 determines the positional relationship between the axes according to the value of the correlation coefficient. Accordingly, the user can intuitively grasp the positive / negative relationship and the magnitude of the correlation of the correlation of each attribute by referring to the angle formed by the axis of the reference attribute and the axis of the other attribute. As a result, a more intuitive search can be provided to the user. The angle formed between the axes of the radar chart is an angle corresponding to the value of the correlation coefficient, as shown in FIGS. That is, the angle formed between the axes of the radar chart does not have to be a regular polygon as shown in FIG.

  In the above description, the axis determining unit 112 arranges the axes of attribute 6 and attribute 7 that have little correlation on the input screen, but the present invention is not limited to this. The attribute values of attribute 6 and attribute 7 that have almost no correlation do not change even if the attribute value of attribute 1, which is the reference attribute, is input. Therefore, the axes of attribute 6 and attribute 7 may not be arranged on the input screen. Good. Alternatively, attributes that have a negative correlation with each other may be paired, and the axis of the attribute that is the pair may be determined to be symmetric with respect to the center 16. Also, attributes having a positive correlation and a negative correlation that are equal to or greater than a certain value may be grouped together to constitute a radar chart in units of groups. For example, each attribute is classified into a group having a correlation coefficient of 0.5 or more and a group having a correlation coefficient of −0.5 or less. At this time, a button or the like that allows the user to instruct group switching is provided on the input screen, and when the user performs a button operation, the group displayed on the input screen is switched.

  In addition, in step S104 of FIG. 6 described above, the initial values of the attributes are all set to 50%, but the present invention is not limited to this. For example, the attribute value of the reference attribute may be the maximum value, and the attribute values of the other attributes may be values proportional to the correlation coefficient with respect to the maximum value. A specific description will be given using attributes 1 to 8 shown on the input screen of FIG. Here, the correlation coefficient of attribute 2 and attribute 3 is 0.5, the correlation coefficient of attribute 6 and attribute 7 is 0, the correlation coefficient of attribute 4 and attribute 5 is -0.5, and the correlation coefficient of attribute 8 Is -1. Further, it is assumed that the correlation coefficient -1 corresponds to the attribute value 0%, and the correlation coefficient 1 corresponds to the attribute value 100%. Attribute 1 is a reference attribute, and an initial value is set to a maximum value of 100%. On the other hand, the initial value of the attribute 8 whose correlation coefficient is −1 is 0%. The initial values of attribute 6 and attribute 7 whose correlation coefficient is 0 are at the center of the range of the correlation coefficient (−1 to 1), and are 50%. Similarly, the initial values of attribute 2 and attribute 3 with a correlation coefficient of 0.5 are 75%. The initial value of attribute 4 and attribute 5 with a correlation coefficient of −0.5 is 25%.

  Although the above-described search device 1 according to the present embodiment includes the correlation determination unit 111, the present invention is not limited to this. In the configuration in which information stored in the information storage unit 12 is downloaded from a center facility (not shown) such as a server as appropriate by the communication unit, a configuration in which a correlation coefficient and a correlation of each attribute are downloaded may be used. In this case, the search device 1 transmits information on attribute values changed by the user at the input unit 13 to the server, and receives information such as correlation coefficients and correlations corresponding to the changed attribute values from the server. You just have to do it.

  In addition, although the search device 1 which concerns on this embodiment mentioned above was provided with the axis | shaft determination part 112, it is not limited to this. In the configuration in which information stored in the information storage unit 12 is downloaded from a center facility (not shown) such as a server as appropriate by communication means, the positional relationship information of each axis determined in the center facility is downloaded. May be.

  In the process of FIG. 6 described above, only the case where the reference attribute is attribute 1 has been described, but the present invention is not limited to this. For example, after determining the correlation between attribute 2 and attribute 8 using attribute 1 as a reference attribute in steps S101 and S102, the processing of steps S101 and S102 is further performed using attribute 2 as a reference attribute. To determine the correlation. Thereby, the attribute value changing unit 113 can change the attribute values of other attributes using the attribute 2 as a reference attribute. That is, the user can change not only attribute 1 but also any attribute value of attribute 1 and attribute 2. Further, by repeating the processes of steps S101 and S102 for attribute 3 to attribute 8, respectively, the user can perform a search by freely changing the attribute values of attribute 3 to attribute 8.

In the above description, the attribute value changing unit 113 has been described with respect to a changing method in which the attribute value of another attribute is uniquely changed by the attribute value of the reference attribute. However, the invention is not limited to this changing method. The attribute value changing unit 113 may change the attribute values of the other attributes 2 to 8 so that the number of search results searched by the search unit 114 is equal to or greater than the predetermined number or the predetermined number. In this case, in step S106 of FIG. 6, the attribute value changing unit 113 changes the other attributes 2 to 8 by the amount of change indicated by the following expression. Note that the amount of variation is Y.
(Variation Y) = (predetermined amount) × (absolute value of correlation coefficient)
The search unit 114 performs a search every time the attribute value of another attribute changes. While referring to the number of search results searched by the search unit 114, the attribute value changing unit 113 sets the other attribute 2 to attribute 8 for each change amount Y until the number of search results reaches or exceeds the predetermined number. Fluctuate. The changing direction is the same as the direction described in step S106. That is, the change direction of the attribute that is a positive correlation is the same direction as the change direction of attribute 1 that is the reference attribute, and the change direction of the attribute that is negative correlation is the opposite direction to the change direction of attribute 1 that is the reference attribute. .

  Here, the variation method for unambiguously varying the attribute value of the other attribute depending on the attribute value of the reference attribute is a method of varying so as to maintain the relationship between the attribute value of the reference attribute and the attribute value of the other attribute. . That is, even if the user changes the attribute value of the reference attribute, the attribute values of other attributes are uniquely determined according to the changed attribute value. Therefore, when each attribute value fluctuates, the number of search results does not fluctuate very rapidly. On the other hand, in the variation method described above, the attribute values of other attributes are varied so as to break the relationship between the attribute value of the reference attribute and the attribute value of the other attribute. Accordingly, the variation in the number of search results is larger than the variation method in which the attribute values of the other attributes are uniquely determined by the attribute values of the reference attributes. That is, the variation method for varying the variation amount Y is a variation method suitable for a case where the number of search results is to be greater than or equal to a predetermined number. As described above, by changing the attribute values of the other attributes 2 to 8 so that the number of search results searched in the search unit 114 is equal to or greater than the predetermined number or the predetermined number, the number of search results is extremely small. The number of search results can be easily selected by the user.

  Further, the attribute value changing unit 113 may change the attribute values of other attributes so that the area of the polygon having each attribute value as a vertex is constant. Accordingly, the attribute value changing unit 113 can change the attribute values of other attributes so that the number of search results is equal to or greater than a predetermined value or a predetermined value. The changing direction of the attribute values of the other attributes is the same as the direction described in step S106 above.

  In the above description, the attribute value changing unit 113 changes the attribute values of the other attributes 2 to 8 (excluding attributes 6 and 7), but is not limited thereto. If the attribute value of at least one of the other attributes varies according to the variation of the attribute value of the reference attribute, it is possible to reduce the complexity of the user's input operation.

  In the above description, when the user changes the attribute value of the reference attribute, there may be a situation in which the user does not move any more due to the relationship with the attribute value of another attribute that changes according to the value. For example, consider a case where the user changes the attribute value of attribute 2 from 70% to 80% when the attribute value of attribute 1 is 100%. At this time, assuming that the correlation of attribute 1 is a positive correlation with attribute 2 as a reference attribute, the attribute value of attribute 1 is originally changed in the positive direction by the attribute value changing unit 113. However, since the attribute value of attribute 1 is already the maximum value of 100%, the attribute value of attribute 1 does not increase any more. Therefore, when such a state is reached, the display mode of attribute 1 is changed. Thereby, the convenience of a user's input operation improves.

  In the process described with reference to FIG. 6, the correlation determination unit 111 determines the correlation based on the information table. Here, new music information may be stored in the information storage unit 12. In this case, a plurality of types of attribute values that the new music information has are added to the information table. That is, the contents of the information table are updated according to changes in the music information to be searched. At this time, the correlation determination unit 111 determines a new correlation based on the updated information table. The attribute value changing unit 113 changes the attribute values of other attributes based on the new correlation. As described above, the search device 1 can dynamically change the attribute values of other attributes in accordance with changes in the contents of the information table. As a result, for example, when searching for music information, even if the information of some of the music information is replaced, it is possible to perform a search according to the new correlation after the replacement. .

  Hereinafter, an example in which the search device 1 according to this embodiment described above is applied to a music search device will be described. The configuration of the music search device is the same as the configuration shown in FIG. FIG. 9 is a diagram illustrating an example of a music information table stored in the information storage unit 12. FIG. 10 is a diagram showing another example of the music information table stored in the information storage unit 12. The music information table shown in FIG. 9 includes a music identification ID for identifying music to be searched and eight types of attribute values. The attributes are the degree of spectral change (degree of spectral change between frames with a predetermined time length), average number of pronunciations (pronunciation frequency of sounds produced in music), and pronunciation aperiodicity (sounded in music) (Degree of sound non-periodicity), beat period (length of time corresponding to a quarter note of music), beat period ratio (period ratio of a sound corresponding to a quarter note of music and a sound pronounced predominantly) ), 1st night intensity (intensity of sound sounded with a period corresponding to approximately half of the beat period), 2nd night intensity (intensity of sound sounded with a period corresponding to approximately 1/4 of the beat period) ), And the beat intensity ratio (the ratio of the first beat intensity to the second night intensity). These attributes are attributes indicated by physical feature amounts extracted from the sound data of the music. That is, each attribute value shown in FIG. 9 is the feature amount, and is a value indicating a physical feature of the music information. As a feature amount extraction method, for example, a method described in Japanese Patent Laid-Open No. 2002-278547 is used, and detailed description thereof is omitted here. The range from which the feature amount is extracted may be the entire music, a part of the music, or a range having a plurality of parts of the music. In addition, the range from which the feature amount is extracted may be a range obtained by arbitrarily combining these ranges.

  In FIG. 9, bibliographic information that can be identified by a music identification ID such as a song name, a singer name, and a genre name is omitted, but in actuality, the bibliographic information is associated with the music identification ID and is stored in the information storage unit 12. Stored in Therefore, bibliographic information is acquired by specifying the music identification ID.

  The music information table stored in the information storage unit 12 may be the music information table shown in FIG. The music information table shown in FIG. 10 includes a music identification ID and five types of attribute values. The attributes shown in FIG. 10 are attributes indicated by the impression value of the music such as intenseness, dynamic feeling, refreshing feeling, simplicity, and softness. That is, each attribute value shown in FIG. 9 is a value indicating the impression value. Here, the impression value is an index value representing a feeling or emotion that a person feels when listening to music. The impression value can be expressed as a linear or non-linear function of the attribute shown in FIG. Specifically, these impression values are calculated by adding predetermined weights to the attribute values that are physical feature amounts shown in FIG.

  In addition to the music information table, the information storage unit 12 stores audio data by linear PCM. The information storage unit 12 also stores compressed audio data such as AAC, MP3, and WMA. The audio data stored in the information storage unit 12 is reproduced by the audio reproduction unit 142 in response to the user's music search. Such reproduction is performed, for example, when the user selects music information displayed on the search result screen.

  Hereinafter, a case where a music search is performed using the music information table shown in FIG. 10 will be described with reference to FIG. FIG. 11 is a diagram illustrating an example of an input screen and a search result screen displayed on the display unit 141 when performing a music search. In FIG. 11, a pentagonal radar chart is displayed on the input screen. Icons 17 indicating attributes 1 to 5 are arranged so as to be movable on the axes. The search result is dynamically updated on the search result screen in accordance with the movement of the icon 17. By selecting the search result displayed on the search result screen via the input unit 13, the user can not only search for music but also perform playback. The axis relationship between the attributes 1 to 5 is determined by the axis determination unit 112. In FIG. 11, “simpleness” is a reference attribute. Here, “softness” and “exhilaration” have a positive correlation with “simpleness”. There is a negative correlation between “strength” and “dynamic feeling” versus “simpleness”. As described above, it is possible to search for desired music information without the user operating multiple types of attribute values. As a result, it is possible to reduce the complexity of the user input operation.

(Second Embodiment)
With reference to FIG. 12, a search device 2 according to a second embodiment of the present invention will be described. FIG. 12 is a block diagram illustrating a configuration example of the search device 2 according to the second embodiment. In FIG. 12, the search device 2 includes a control unit 11, an information storage unit 12, an input unit 13, an output unit 14, and a situation acquisition unit 21. The control unit 11 includes a correlation determination unit 111, an axis determination unit 112, an attribute value change unit 113, a search unit 114, a drawing unit 115, and an attribute value update unit 121. The output unit 14 includes a display unit 141 and an audio playback unit 142.

  The search device 2 according to the present embodiment differs from the search device 1 according to the first embodiment described above in that an attribute value update unit 121 and a situation acquisition unit 21 are newly provided. Therefore, other components are the same as those of the search device 1 according to the first embodiment, and the same reference numerals are given and detailed description thereof is omitted. In the present embodiment, a case where the search device 2 is applied to a navigation device will be described. This navigation device is installed in a moving body such as a vehicle, an aircraft, a ship, or a train.

  The information storage unit 12 stores facility information (such as restaurants and parking lots) as information to be searched. In addition, as shown in FIG. 13, the information storage unit 12 stores a facility information table composed of a facility identification ID for identifying facility information and a plurality of types of attribute values. FIG. 13 is a diagram illustrating an example of a facility information table stored in the information storage unit 12. The facility information table shown in FIG. 13 shows a facility information table related to a restaurant as an example. In addition to the facility information table related to the restaurant, the information storage unit 12 stores facility information related to the restaurant. Note that the search device 2 according to the present embodiment is applied to a navigation device. Accordingly, not only such facility information but also map data such as road network data (node data and link data) and background data such as rivers and green spaces are stored in the information storage unit 12.

  The attributes included in the facility information table of FIG. 13 are distance, number of seats (number of seats in a restaurant), average price, popularity, number of parking lots, atmosphere, view, and tranquility. Among these attributes, distance means the distance between the current location of the vehicle and the restaurant. This distance value is a value that varies with the movement of the vehicle, and the distance value is a dynamic attribute value. On the other hand, attributes other than distance (number of seats, average price, popularity, number of parking lots, atmosphere, view, and quietness) are values that do not change even when the vehicle moves, and attribute values other than distance are static. Attribute value.

  The facility information table shown in FIG. 13 shows a restaurant, and the type and number of attributes change if the genre (parking lot, supermarket, gas station, etc.) is different. For example, in the case of a gas station, there are attributes such as the price of gasoline per liter and the number of vehicles that can be refueled simultaneously. In FIG. 13, information that can be identified by a facility identification ID such as a facility name, an address, a telephone number, and a position (latitude, longitude) is omitted, but these information are actually associated with the facility identification ID. And stored in the information storage unit 12. Therefore, these pieces of information are acquired by specifying the facility identification ID. Further, the facility information and the facility information table shown in FIG. 13 are appropriately downloaded from the center facility (not shown) to the information storage unit 12 by communication means (not shown) (for example, a mobile phone, PHS, or WiMAX). There may be.

  The situation acquisition unit 21 acquires situation information indicating the situation around the user. The situation information includes information such as the current position, speed, and direction of the vehicle. The situation acquisition unit 21 includes, for example, a GNSS (Global Navigation Satellite System) receiver, a vehicle speed sensor, a gyro (angular velocity) sensor, and an acceleration sensor. The GNSS receiver is, for example, a GPS receiver. The GNSS receiver receives radio waves from a plurality of satellites and demodulates them to measure the absolute position of the receiver itself. The current position, speed, and direction of the vehicle are measured by the GNSS receiver and various sensors. In addition to the above, the status information includes the number of passengers, temperature, weather information, traffic congestion status, driving operation status (operation status of steering wheel, accelerator, brake, direction indicator, etc.), user's physical status (Such as face direction, line of sight, and pulse), and other information such as the situation that changes from moment to moment as the vehicle moves.

  The attribute value update unit 121 updates the dynamic attribute value included in the facility information table based on the situation information indicating the situation around the user acquired by the situation acquisition unit 21. For example, it is assumed that the facility information table includes a dynamic attribute called distance. In this case, the attribute value update unit 121 updates the distance from the current position of the vehicle acquired by the situation acquisition unit 21 to each facility information specified by the facility identification ID as a new distance.

  Next, with reference to FIG. 14, the flow of processing of the search device 2 according to the second embodiment will be described. FIG. 14 is a flowchart showing a flow of processing of the search device 2 according to the second embodiment. In the following description, it is assumed that the information to be searched is facility information managed by the facility information table of FIG. 13, and the reference attribute is distance. The correlation between the number of seats and the number of parking lots is assumed to be “medium positive correlation”. The correlation between the view and the degree of popularity is assumed to have “medium negative correlation”. The correlation between the average prices is “There is a strong negative correlation”. The correlation between atmosphere and tranquility is “almost no correlation”. The processes in steps S201 to S208 in FIG. 14 are substantially the same as the processes in steps S101 to S108 in FIG. 6 described above, and different processes will be mainly described.

  In FIG. 14, the correlation determining unit 111 sets the attribute that the user changes among the attributes of the facility information table illustrated in FIG. 13 as a reference attribute (step S <b> 201). Following step S201, the correlation determining unit 111 calculates a correlation coefficient between the distance that is the reference attribute and other attributes, and determines the correlation between the other attributes (step S202). Following step S202, the axis determination unit 112 determines the positional relationship between the axes corresponding to the attributes based on the correlation (step S203). In step S203, it is assumed that the positional relationship between the eight axes corresponding to each attribute is determined as the positional relationship shown in the input screen of FIG. FIG. 15 is a diagram illustrating an example of an input screen and a search result screen displayed on the display unit 141.

  Hereinafter, the positional relationship of each axis determined by the axis determination unit 112 will be described with reference to the input screen shown in FIG. The axis determination unit 112 determines the axis 15a connecting the topmost vertex of the regular octagonal vertices and the center 16 as the axis of distance as a reference attribute. The correlation between the number of parking lots and the number of seats is “medium positive correlation”. Therefore, the axis determination unit 112 determines the axes of the number of parking lots and the number of seats as the axes 15b and 15h at positions closest to the distance axis 15a. The correlation of the average price is “There is a strong negative correlation”. Accordingly, the axis determination unit 112 determines the axis of the average price as the axis 15e farthest from the distance axis 15a. That is, the average price axis 15e is arranged at a position symmetrical to the distance axis 15a with respect to the center 16, as shown in the input screen of FIG. The correlation between popularity and view is “There is a moderate negative correlation”. Therefore, the axis determination unit 112 determines the popularity and view axes as the axes 15d and 15f at positions farthest after the average price axis 15e with respect to the distance axis 15a. The correlation between atmosphere and tranquility is “little correlation”. Therefore, the axis determination unit 112 determines the axis of quietness as the axis 15c between the parking lot number and the popularity axes 15b and 15d. Similarly, the axis of the atmosphere is determined to be the axis 15g between the number of seats and the axes 15h and 15f of the view.

  Following step S203, the drawing unit 115 draws the input screen and the initial screen of the search result screen, and displays them on the display unit 141 (step S204). Here, the initial values of the attributes are all set to 50%, for example. After step S204, when the user changes the distance value as the reference attribute in the input unit 13, the attribute value changing unit 113 determines whether or not the distance value has changed (step S205). When the distance value fluctuates, the attribute value fluctuation unit 113 detects the fluctuation direction of the distance value and the fluctuation amount from the initial value, and determines the attribute values of other attributes based on the detected fluctuation direction and fluctuation amount. Fluctuate (step S206). Next to step S206, the search unit 114 performs a search using the changed attribute values from the facility information that is the search target stored in the information storage unit 12 (step S207). The drawing unit 115 draws the result searched in step S207 on the search result screen (step S208). A map based on the map data is displayed on the search result screen shown in FIG. A black triangle on the map indicates the current position of the vehicle. In addition, “R” on the map indicates the restaurant 19 as a search result, and ranking by search is given to each. In addition, an input button for “Destination setting” is displayed on the search result screen. The drawing unit 115 can display the search result screen by dynamically changing the scale so that all facilities corresponding to the number of search results are included in the map. Moreover, the retrieved facility information may be displayed in a list format. In addition, the user selects any of the restaurants 19 displayed on the search result screen and presses the “destination setting” input button, thereby not only searching for facilities but also setting destinations (or registering points) (Route setting).

  Following step S208, the situation acquisition unit 21 acquires situation information indicating the situation around the user (step S209). Here, it is assumed that the current position of the vehicle is acquired. Next to step S209, the attribute value update unit 121 determines whether or not the current position of the vehicle acquired by the situation acquisition unit 21 has changed, and when changed, the distance value of each facility information included in the facility information table. Each value is updated (step S210). If the attribute value is updated in step S210, the process proceeds to step S202. In step S202, the correlation determining unit 111 calculates a correlation coefficient between the distance whose value is updated and another attribute, and determines the correlation of the other attribute, respectively (step S202). In step S <b> 203, the axis determination unit 112 determines the positional relationship of the axes corresponding to each attribute using the correlation based on the updated distance value. In this way, the situation acquisition unit 21 and the attribute value update unit 121 allow the correlation (step S202), the positional relationship of each axis (step S203), and how other attributes change (step S206) Optimized for changes in the situation. In step S209, the status acquisition unit 21 acquires status information every time a search is performed. However, the status acquisition unit 21 may acquire the status information at a timing other than the search, for example, every predetermined time.

  If the attribute value is not updated in step S210, the process proceeds to step S211 and it is determined whether or not to end the process (step S211). When the search is further performed by changing the distance as the reference attribute to a different value without ending the process, the process returns to step S205. By such processing in steps S205 to S210, the search is performed every time the user changes the attribute value of the reference attribute on the input screen. The search result screen is updated every time a search is performed.

  As described above, in the search device 2 according to the present embodiment, when applied to a navigation device, it is possible to perform a search that satisfies various user requirements without complicating the user's input operation. To do.

  In the search device 1 according to the present embodiment, the attribute value of the reference attribute is updated by the situation acquisition unit 21 and the attribute value update unit 121. As a result, the correlation between the attributes, the positional relationship between the axes arranged on the input screen, and the manner in which the attribute values of other attributes are changed can be dynamically changed according to changes in the situation around the user. it can. As a result, it is possible to further reduce the complexity of the user input operation.

  In the first and second embodiments described above, the radar chart is displayed on the input screen. However, the present invention is not limited to this. Any display format may be used as long as each attribute value can be changed. For example, a display format may be used in which numbers representing each attribute value are displayed on the input screen. The information indicating each attribute value is not limited to a number, and any information that can recognize the size of each attribute value may be used. Therefore, the information is not limited to numbers, and may be information that expresses the size of the attribute value by the display position, for example.

  Although the present invention has been described in detail above, the above description is merely illustrative of the present invention in all respects and is not intended to limit the scope thereof. It goes without saying that various improvements and modifications can be made without departing from the scope of the present invention.

  A search device according to the present invention is a search device capable of performing a search satisfying various requests of a user without complicating a user's input operation, and can search music, books, movies, facilities, and the like. The present invention is applied to a search device for searching, a navigation device, an application that operates on a television, an application that operates on a mobile phone, and an application that operates on a personal computer.

The block diagram which shows the structural example of the search device 1 which concerns on 1st Embodiment. The figure which showed an example of the information table stored in the information storage part 12 Diagram showing an example of the correspondence between correlation coefficient and correlation The figure which showed the correlation of the attribute 2-the attribute 8 in the information table of FIG. The figure which showed an example of the input screen and search result screen which were displayed in the display part 141 The flowchart which showed the flow of the process of the search device 1 which concerns on 1st Embodiment. The figure which showed other examples of the positional relationship of each axis typically The figure which shows the positional relationship of each axis | shaft arrange | positioned by the determination method shown in FIG. 7 when there are six types of attributes (attribute 1 to attribute 6). The figure which showed an example of the music information table stored in the information storage part 12 The figure which showed the other example of the music information table stored in the information storage part 12 The figure which showed an example of the input screen and search result screen which were displayed in the display part 141 when performing a music search The block diagram which shows the structural example of the search device 2 which concerns on 2nd Embodiment. The figure which showed an example of the facility information table stored in the information storage part 12 The flowchart which showed the flow of the process of the search device 2 which concerns on 2nd Embodiment. The figure which showed an example of the input screen and search result screen which were displayed in the display part 141 The figure which showed the input screen and search result screen of the conventional search device which searches using multiple kinds of attributes

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 2 Search apparatus 11 Control part 12 Information storage part 13 Input part 14 Output part 21 Situation acquisition part 111 Correlation determination part 112 Axis determination part 113 Attribute value change part 114 Search part 115 Drawing part 121 Attribute value update part 141 Display part 142 Audio playback unit

Claims (16)

A search device that performs a search for a plurality of search targets using a plurality of types of attribute values each indicating a characteristic of the search target,
A storage unit for storing correspondence information indicating correspondence between the plurality of types of attribute values and the search target for each of the plurality of search targets;
A reference attribute determination unit that determines one attribute value of the plurality of types of attribute values as an attribute value of a reference attribute;
A first attribute value changing unit that changes the attribute value of the reference attribute according to a user instruction;
When the attribute value of the reference attribute changes, for at least one other attribute other than the reference attribute, the correlation between the reference attribute determined based on the correspondence information and the other attribute, and the attribute of the reference attribute A second attribute value changing unit that changes the attribute value according to the value change;
A display unit for displaying information indicating each attribute value before and after the change by the first and second attribute value change units on a display screen;
A search device, comprising: a search unit that searches the plurality of search targets using each attribute value after the change by the first and second attribute value change units.   When the correlation between the reference attribute and the other attribute is a positive correlation, the second attribute value changing unit has the attribute value of the other attribute in the same direction as the change direction of the attribute value of the reference attribute. When the correlation between the reference attribute and the other attribute is a negative correlation, the attribute value of the other attribute is changed in the opposite direction to the change direction of the attribute value of the reference attribute. The search device according to claim 1. A correlation coefficient calculating unit that calculates a correlation coefficient indicating the correlation based on the correspondence information;
The second attribute value variation unit varies the attribute value of the other attribute by an amount corresponding to the absolute value of the correlation coefficient indicating the correlation between the reference attribute and the other attribute. The search device according to claim 2.   The said 2nd attribute value fluctuation | variation part further fluctuates the attribute value of said another attribute, when the search result by the said search part becomes a number of cases smaller than predetermined number of items, The attribute value of Claim 1 characterized by the above-mentioned. Search device. The display screen displays a radar chart having axes corresponding to a plurality of types of the attributes,
The search device according to claim 1, further comprising an axis determination unit that determines a positional relationship of each axis of the radar chart to be displayed based on the correlation.   The axis determination unit determines an angle formed by an axis corresponding to the reference attribute and an axis corresponding to another attribute based on a correlation between the reference attribute and the other attribute, The search device according to claim 5.   When the correlation between the reference attribute and the other attribute is a positive correlation, the axis determination unit sets the angle formed by the axis corresponding to the reference attribute and the axis corresponding to the other attribute as an acute angle, When the correlation between the reference attribute and the other attribute is a negative correlation, the angle formed by the axis corresponding to the reference attribute and the axis corresponding to the other attribute is an obtuse angle, The search device according to claim 5. The search target is music information,
The search device according to claim 1, wherein the plurality of types of attribute values include attribute values indicating physical characteristics of the music piece.   The plurality of types of attribute values include at least one of spectral change, average pronunciation number, pronunciation aperiodicity, beat period, beat period ratio, first night intensity, second night intensity, and beat intensity ratio. The search device according to claim 8, wherein an attribute value to be displayed is included. The search target is music information,
The search device according to claim 1, wherein the plurality of types of attribute values include attribute values indicating impression values of music.   11. The attribute value according to claim 10, wherein the plurality of types of attribute values include an attribute value indicating at least one of the intensity, dynamic feeling, exhilaration, simplicity, and softness of the music. Search device. The plurality of types of attribute values include predetermined attribute values whose values vary depending on the situation around the user,
A situation acquisition unit that obtains situation information indicating a situation around the user;
When the situation information is acquired by the situation acquisition unit, an attribute value update unit that updates the predetermined attribute value included in the correspondence information based on the acquired situation information;
The search device according to claim 1, further comprising a correlation determination unit that determines the correlation based on the correspondence information. The search target is information on facilities existing on a map,
The attribute value that varies depending on the situation around the user indicates a distance from the facility to the current position of the user,
The search device according to claim 12, wherein the situation acquisition unit acquires a current position of a user as the situation information. When the content of the correspondence information is updated, the information processing apparatus further includes a correlation determination unit that newly determines the correlation based on the updated correspondence information,
The second attribute value variation unit varies the attribute value according to a new correlation determined by the correlation determination unit when the content of the correspondence information is updated. The search device according to 1. A search method for performing a search on a plurality of search targets using a plurality of types of attribute values respectively indicating characteristics of the search target,
The search device stores correspondence information indicating the correspondence between the plurality of types of attribute values and the search target for each of the plurality of search targets,
A reference attribute determination step for determining one attribute value of the plurality of types of attribute values as an attribute value of a reference attribute;
A first attribute value changing step for changing the attribute value of the reference attribute according to a user instruction;
When the attribute value of the reference attribute changes, for at least one other attribute other than the reference attribute, the correlation between the reference attribute determined based on the correspondence information and the other attribute, and the attribute of the reference attribute A second attribute value variation step for varying the attribute value according to the value variation;
A display step of displaying information indicating each attribute value before and after the change in the first and second attribute value change steps on a display screen;
And a search step for performing a search for the plurality of search objects using the attribute values after the change in the first and second attribute value change steps. A program for causing a computer of a search device to perform a search for a plurality of search targets using a plurality of types of attribute values respectively indicating features of the search target,
The search device stores correspondence information indicating the correspondence between the plurality of types of attribute values and the search target for each of the plurality of search targets,
A reference attribute determination step for determining one attribute value of the plurality of types of attribute values as an attribute value of a reference attribute;
A first attribute value changing step for changing the attribute value of the reference attribute according to a user instruction;
When the attribute value of the reference attribute changes, for at least one other attribute other than the reference attribute, the correlation between the reference attribute determined based on the correspondence information and the other attribute, and the attribute of the reference attribute A second attribute value variation step for varying the attribute value according to the value variation;
A display step of displaying information indicating each attribute value before and after the change in the first and second attribute value change steps on a display screen;
A program that causes the computer to execute a search step of performing a search on the plurality of search targets using each attribute value after the change in the first and second attribute value change steps.

Images