JP4776214B2 - Database search device, database search method, and database search program - Google Patents

Description

  The present invention relates to a database search apparatus for music, video and the like.

  A system for searching music by sensitivity words has been proposed ("Music database search system by sensitivity words using music sensitivity space", Information Processing Society of Japan Vol.42 No12 2001). In this conventional search system, a search space for mapping songs in a database is generated by the SD method and factor analysis. In addition, when a new song is registered in the database, mapping to a search space is performed by an automatic indexing system constituted by a GA and a neural network. At the time of search, by inputting the degree (1-7) of eight sensitivity word pairs to the neural network, the neural network outputs the coordinates in the sensitivity space corresponding to those inputs. The search candidates are presented to the user in order from the song with the closest Euclidean distance from the output coordinate value.

  The sensitivity word pairs are “bright-dark”, “heavy-light”, “hard-soft”, “stable-unstable”, “clear-cloudy”, “smooth-crisp”, “ Eight types are used: severe-mild and thick-thin.

  In a conventional search device, the degree is input for each of a plurality of types of sensitivity word pairs by the user, and a song corresponding to the input degree of each sensitivity word pair is selected. Incidentally, there is usually a correlation between the sensitivity word pairs. For example, when “bright-dark” and “fun-sad” are included as the types of sensitivity word pairs, they are not the same meaning but can be said to be similar expressions with high correlation. If the user inputs “bright” for the “light-dark” sensitivity word pair and “sad” for the “fun-sad” sensitivity word pair, the song that satisfies the condition “bright and sad” Since it usually does not exist, there is a high possibility that the corresponding song does not exist.

The more sensitive word pairs that can be specified, the more complicated the mutual correlation is, and it is possible that contradictory inputs are made. If it does so, appropriate music selection reflecting a user's intention will not be performed.
JP 2001-306580 A "Music database retrieval system by sensitivity word using music sensitivity space" IPSJ Journal Vol.42 No12

The present invention can automatically correct input values so that there is no contradiction between the input values of the sensitivity word pairs when the input is inconsistent with the correlation between the sensitivity word pairs. An object of the present invention is to provide a database search device that enables an appropriate search.
Another object of the present invention is to provide a database search apparatus that can prevent an input contradicting the correlation between the sensitivity word pairs from being made by the user and can perform an appropriate search.

  According to the first aspect of the present invention, for each content registered in the database, the Kansei space coordinate value obtained for the content is registered in the database as an index of the content, and is input by the user during the search. Based on the input value for each sensitivity word pair, the sensitivity space coordinate value corresponding to the input value for each input sensitivity word pair is calculated, and the spatial distance from the calculated sensitivity space coordinate value is the shortest. In the database search apparatus for searching for contents, an input value correction unit is provided that automatically corrects an input value for each sensitivity word pair input by a user based on a preset correlation value between each sensitivity word pair. The input value correcting means sequentially selects combinations of one sensitivity word pair having a higher priority and one sensitivity word pair having a lower priority from the highest priority. Each time two sensitivity word pairs are selected by the first means and the first means, the input value difference between the two sensitivity word pairs is selected based on the correlation value between the two sensitivity word pairs. The second means for calculating the permissible range of the two, and every time the permissible range of the input value difference between the two sensitivity word pairs is calculated by the second means, the difference between the input values for the two sensitivity word pairs It is determined whether or not the allowable range calculated by the above is exceeded, and if so, the difference between the input values for the two sensitivity word pairs is within the allowable range. And a third means for correcting the input value of the sensitivity word pair having the lower priority order.

According to the second aspect of the present invention, for each content registered in the database, the Kansei space coordinate value obtained for the content is registered in the database as an index of the content, and is input by the user at the time of search. Based on the input value for each sensitivity word pair, the sensitivity space coordinate value corresponding to the input value for each input sensitivity word pair is calculated, and the spatial distance from the calculated sensitivity space coordinate value is the shortest. In the database search device for searching for content, an input recommended range display means for displaying an input recommended range for each sensitivity word pair based on a preset correlation value between each sensitivity word pair is provided. For each combination of a sensitivity word pair with a higher priority and a sensitivity word pair with a lower priority, the display means is based on the correlation value between the two sensitivity word pairs. A first means for calculating an input allowable range of a sensitivity word pair having a low priority with respect to a current input value of a sensitivity word pair having a high rank, and for each sensitivity word pair, the sensitivity word pair and a sensitivity having a higher priority than that. A range common to all the input allowable ranges calculated by the first means for each combination with each word pair is calculated, and the obtained range is determined as an input recommended range for the sensitivity word pair. The second means and the third means for displaying the recommended input range determined for each sensitivity word pair by the second means are provided.
According to the third aspect of the present invention, when a plurality of contents are registered and a database in which the emotional space coordinate values obtained for the contents are registered as an index of the contents is searched for each content. Based on the input value for each sensitivity word pair input by the user, the emotional space coordinate value corresponding to the input value for each input sensitivity word pair is calculated, and the spatial relationship with the calculated sensitivity space coordinate value is calculated. In a database search method executed by a database search device that searches for content with the shortest distance, an input value for each input sensitivity word pair is automatically set based on a preset correlation value between each sensitivity word pair. The input value correction step includes a sensitivity word pair having a high priority and a sensitivity word having a lower priority than the sensitivity word pair. Are selected in order from the highest priority, and each time two sensitivity word pairs are selected in the first step, based on the correlation value between the two sensitivity word pairs. The second step of calculating the allowable range of the input value difference between the two sensitivity word pairs, and whenever the allowable range of the input value difference of the two sensitivity word pairs is calculated in the second step, the two sensitivity values It is determined whether or not the difference between the input values for the word pair exceeds the allowable range calculated in the second step, and if so, the difference between the input values for the two sensitivity word pairs is equal to the allowable range. The third step of correcting the input value of the sensitivity word pair having the lower priority of the two sensitivity word pairs is provided.
According to a fourth aspect of the invention, a plurality of contents are registered, for each content, if the sensitive space coordinate values determined for the content search a database that has been registered as an index of the content Based on the input value for each sensitivity word pair input by the user, the emotional space coordinate value corresponding to the input value for each input sensitivity word pair is calculated, and the spatial relationship with the calculated sensitivity space coordinate value is calculated. In the database search method executed by the database search device for searching for the content with the shortest distance, input recommendation for displaying a recommended input range for each sensitivity word pair based on a preset correlation value between each sensitivity word pair A range display step is provided, and the recommended input range display step includes one sensitivity word pair having a higher priority and one sensitivity word having a lower priority. The first step of calculating the input allowable range of the sensitivity word pair with low priority with respect to the current input value of the sensitivity word pair with high priority based on the correlation value between the two sensitivity word pairs for each combination with For each sensitivity word pair, a range that is common to all of the input allowable ranges calculated in the first step for each combination of the sensitivity word pair and each of the sensitivity word pairs with higher priority than the sensitivity word pair. A second step of calculating and determining the obtained range as the recommended input range for the sensitivity word pair, and a third step of displaying the recommended input range determined for each sensitivity word pair in the second step are provided. It is characterized by that.
The invention according to claim 5 searches a database in which a plurality of contents are registered in the computer, and for each content, Kansei space coordinate values obtained for the contents are registered as an index of the contents. When calculating the sensitivity space coordinate value corresponding to the input value of each sensitivity word pair input based on the input value of each sensitivity word pair input by the user, and the calculated sensitivity space coordinate value In the database search program for executing the database search method for searching for the content having the shortest spatial distance, the database search method is input by the user based on a correlation value between each sensitivity word pair set in advance. The input value correction step for automatically correcting the input value for each sensitivity word pair is provided. In the first step and the first step, two sensibilities are selected in order of the combination of one sensitivity word pair with a higher priority and one sensitivity word pair with a lower priority in order from the highest priority. Each time a word pair is selected, based on the correlation value between the two sensitivity word pairs, the second step of calculating the allowable range of the input value difference between the two sensitivity word pairs, and two in the second step Each time the permissible range of the input value difference between the sensitivity word pairs is calculated, it is determined whether or not the difference between the input values for the two sensitivity word pairs exceeds the allowable range calculated in the second step. If so, the input value of the lower sensitivity word pair of the two sensitivity word pairs is corrected so that the difference between the input values for the two sensitivity word pairs is within the allowable range. A third step is provided. To.
The invention according to claim 6 searches a database in which a plurality of contents are registered in the computer, and for each content, Kansei space coordinate values obtained for the contents are registered as an index of the contents. When calculating the sensitivity space coordinate value corresponding to the input value of each sensitivity word pair input based on the input value of each sensitivity word pair input by the user, and the calculated sensitivity space coordinate value In the database search program for executing the database search method for searching for the content having the shortest spatial distance, the database search method is based on a correlation value between each sensitivity word pair set in advance. There is a recommended input range display step that displays the recommended input range for each input, and the recommended input range display step has a high priority. For each combination of one sensitivity word pair and one sensitivity word pair having a lower priority, the priority is given to the current input value of the sensitivity word pair having a higher priority based on the correlation value between the two sensitivity word pairs. A first step of calculating an input allowable range of a sensitivity word pair having a lower rank, and a first step for each combination of the sensitivity word pair and a sensitivity word pair having a higher priority for each sensitivity word pair. The second step of calculating a range common to all of the calculated input allowable ranges and determining the obtained range as an input recommended range for the sensitivity word pair, and for each sensitivity word pair in the second step A third step of displaying the determined recommended input range is provided.

According to the present invention, when an input contradicting the correlation between the sensitivity word pairs is made by the user, the input values are automatically corrected so that the input values of the sensitivity word pairs are not inconsistent. And an appropriate search becomes possible.
Furthermore, according to the present invention, it is possible to prevent the user from making an input that contradicts the correlation between the sensitivity word pairs, and an appropriate search is possible.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention applied to a music database search system will be described below with reference to the drawings.

[1] Explanation of the configuration of the music database search system

  FIG. 1 shows the configuration of a music database search system.

  The music database search system includes a music sensitivity space generation unit 100 and a search system unit 200. The music sensitivity space generation unit 100 generates a music sensitivity space as a search space. The search system unit 200 searches the music database for a song that matches the input song impression, and outputs the song.

[2] Description of the music sensitivity space generation unit 100

  As shown in FIG. 1, the music sensitivity space generation unit 100 automatically calculates an initial music sensitivity space generation unit 101 that generates a music sensitivity space by a listening experiment, and music sensitivity space coordinate values of a song to be added to the database. And an automatic indexing unit 102.

[2-1] Description of the initial music sensitivity space generation unit 101

  The initial music sensitivity space generation unit 101 includes a song impression evaluation unit 111 that calculates a song impression value by a listening experiment, a factor analysis unit 112 that performs factor analysis using the song impression value calculated by the song impression evaluation unit 111, and And a database unit 113.

  The song impression evaluation unit 111 will be described. For 10 subjects, 100 sample pop songs were presented, and each sensitivity word pair was evaluated on a 7-point scale.

The following four types of sensitivity word pairs were used.
(1) Bright-dark (2) clear-crammed (3) smooth-crisp (4) mild-intense

  From this experimental result, the music impression value for each sensitivity word pair was calculated for each sample music piece. The music impression value for each sensitivity word pair is represented by an average value of the evaluation values of the sensitivity word pairs.

  The factor analysis unit 112 performs a factor analysis with the music impression value for each sensitivity word pair of each sample music as a multivariate, and for each sample music, the psychological correlation of the music impression with each factor as an axis. The coordinate data of the space (music sensitivity space) to represent is obtained. Thus, the music sensitivity space is a space having a plurality of factor axes. In this example, it is assumed that there are three types of factors and three types of factor axes.

  In the database unit 113, music data of each sample music is registered, and coordinate data in the sensitivity space of each sample music is registered as index information of each sample music.

[2-2] Description of the automatic indexing unit 102

  The automatic indexing unit 102 analyzes a newly registered song and extracts physical feature amounts such as tempo and rhythm, and a physical feature amount extracted by the physical feature amount extraction unit 121. Is converted into a song impression value for each sensitivity word pair, and a music impression value for each sensitivity word pair obtained by the physical feature value / music impression value conversion unit 122. Is converted into a music sensitivity space coordinate value, and a music impression space / music sensitivity space coordinate conversion unit 123 is provided.

  The music sensitivity space coordinate value of the newly registered song obtained by the automatic indexing unit 102 is registered in the database unit 113 together with the song data of the newly registered song.

  FIG. 2 shows the configuration of the automatic indexing unit 102 more specifically.

  The physical feature quantity / musical impression value conversion unit 122 includes regression analysis units 122a to 122d corresponding to four types of sensitivity word pairs. The music impression value / musical sensitivity space coordinate conversion unit 123 includes a neural network.

  A plurality of physical feature quantities related to the corresponding sensitivity word pairs are input to the regression analysis units 122 a to 122 d in the physical feature quantity / musical impression value conversion unit 122. The types of physical feature values input to the regression analysis units 122a to 122d are determined in advance.

  Each regression analysis part 122a-122d converts a physical feature-value into the music impression value of the corresponding sensitivity word pair based on following Formula (1).

  In the above equation (1), OUT is an output value of the regression analysis unit, and represents an estimated value of the music impression value related to the sensitivity word pair corresponding to the regression analysis unit. In [k] represents the k-th physical feature amount input to the regression analysis unit. W [k] is a weighting coefficient for In [k]. W [0] is a weighting coefficient.

  The weighting factors W [k] and W [0] in the above equation (1) are obtained in advance based on the above-described listening experiment results and the physical feature values of the sample music pieces used for the above. The weighting factors W [k] and W [0] are determined as follows. The result of the listening experiment, that is, the song impression value for each sensitivity word pair obtained for each sample song is used as a teacher signal (output value OUT). Further, the physical feature amount extracted by the physical feature amount extraction unit 121 for each sample music piece is set as the input value In. Then, the weighting factors W [k] and W [0] are calculated so that the regression analysis results for all the sample music pieces have values closest to the teacher signal for the sample music piece.

  Learning of the neural network constituting the music impression value / musical sensitivity space coordinate conversion unit 123 will be described. The input signal at the time of learning is the result of the above-described listening experiment, that is, the song impression value for each sensitivity word pair obtained for each sample song. The teacher signal at the time of learning is a music sensitivity space coordinate value (coordinate value of each factor axis) of each song obtained by the factor analysis unit 112.

[3] Description of the search system unit 200

  As shown in FIG. 1, the search system unit 200 includes a user input unit 201, a song impression value / musical sensitivity space coordinate conversion unit 202, a search unit 203, and a search result output unit 204.

  The user inputs a song impression value (seven levels) for each sensitivity word pair related to the song to be searched to the user input unit 201. The music impression value for each sensitivity word pair input to the user input unit 201 is given to the music impression value / music sensitivity space coordinate conversion unit 202. As the music impression value / musical sensitivity space coordinate conversion unit 202, a neural network constituting the music impression value / musical sensitivity space coordinate conversion unit 123 in the automatic indexing unit 102 is used as it is.

  The music impression value / musical sensitivity space coordinate conversion unit 202 inputs the music impression value for each sensitivity word pair given from the user input unit 201 to the neural network, and acquires the music sensitivity space coordinate value corresponding thereto. . The music impression value / music sensitivity space coordinate conversion unit 202 gives the acquired music sensitivity space coordinate value to the search unit 203.

  Processing of the search unit 203 will be described. The music sensitivity space coordinate values of the music registered in the database unit 113 are (X1 [i], X2 [i], X3 [i]). i is a song number. Further, the values obtained by converting the music impression value for each sensitivity word pair input by the user into music sensitivity space coordinates by the music impression value / music sensitivity space coordinate conversion unit 202 are (a1, a2, a3). And

  The search unit 203 calculates a spatial distance L for each registered song based on the following formula (2), and selects a song having the smallest spatial distance L. The search result output unit 204 reads the song data of the song selected by the search unit 203 from the database unit 113 and outputs it.

L = (X1 [i] -a1) ² + (X2 [i] -a2) ² + (X3 [i] -a3) ² (2)

[4] Description of the User Input Unit 201 Although not shown, each of the user input unit 201 includes “bright-dark”, “clear-crimped”, “smooth-crisp”, and “gentle-severe”. For each sensitivity word pair, a dial is provided for inputting the degree of the sensitivity word pair (seven levels). Then, the degree of input by each dial is displayed on the display as shown in FIG.

  Sensitive word pair “bright-dark” is “0”, “clear” -sensitive word pair is “1”, “smooth-crisp” sensitivity word pair is “2”, “gentle-severe” Assuming that the sensitivity word pair of “3” is represented by “3”, it was found from the questionnaire results that there is a correlation as shown in Table 1 between these sensitivity word pairs.

The user input unit 201 includes a correction function (correction unit) that automatically corrects an input value. The concept of this correction function will be described.

The correlation value between the sensitivity word pair i and the sensitivity word pair j is represented by H _ij . Also, the input value of the sensitivity word pair _i is represented by D _i , and the input value of the sensitivity word pair _j is represented by D _j .

According to the correlation value between the two sensitivity word pairs i and j, an allowable range S _ij of the input difference | D _i −D _j | between the two sensitivity word pairs i and j is calculated. Then, it is determined whether or not the input difference | D _i −D _j | between the two sensitivity word pairs i and j exceeds the allowable range S _ij . If it exceeds, either one of the two sensitivity word pairs i and j so that the input difference | D _i −D _j | falls within the allowable range S _ij (in this example, the one with the lower priority) ) The sensitivity word pair input value is corrected.
An allowable range S _ij of an input difference between two sensitivity word pairs i and j is obtained by the following equation (3).

S _ij = INT [{(1.0−H _ij ) × (width of input value)} + K] (3)

  In the above formula (3), INT [A] means that the decimal part of A is truncated. The width of the input value is the difference between the minimum value and the maximum value of the input value. Since the input value has 7 levels from 0 to 6, if the difference between each level is 1, the width of the input value is 6. K is a constant and is 0.9 in this example.

  For example, a case is assumed where the input values for each sensitivity word pair are as shown in FIG. “Smooth-crisp” and “gentle-severe” have a very high correlation value of 0.8, but the input values are contradictory values. The input value of “smooth-crisp” is 6.0, the input value of “gentle-severe” is 0, and the input difference between these sensitivity word pairs is 6.0. The allowable range of the input difference between these sensitivity word pairs is obtained by the following equation (4).

S _ij = INT [{(1.0−0.8) × 6)} + 0.9]
= INT [2.1]
= 2.0 (4)

  While the allowable range of the input difference of these sensitivity word pairs is 2.0, the input difference is 6, and the input difference exceeds the allowable range. Assuming that the priority for the “smooth-crisp” sensitivity word pair is higher than the priority for the “mild-severe” sensitivity word pair, the input values of the “slow-severe” sensitivity word pair are used as the input values of these sensitivity words. Correction is made so that the input difference between word pairs becomes 2.0. That is, the input value for the sensitivity word pair of ““ gentle-severe ”is corrected to 4.0. Therefore, the input value for each sensitivity word pair after correction is as shown in FIG.

  Such processing is performed for each combination of sensitivity word pairs in order from the sensitivity word pair having the highest priority.

FIG. 5 shows an input value correction processing procedure.
Here, the order of priority of each sensitivity word pair is “0” (bright-dark)> “1” (clear—cracked)> “2” (smooth—crisp)> “3” (mild—strong ).

i = 0 and j = 1 are set (step S1). An allowable range S _ij of an input difference between the sensitivity word pairs i and j is calculated based on the following equation (5) (step S2).

S _ij = INT [(1.0−H _ij ) × 6 + 0.9] (5)

Next, it is determined whether or not the input difference | D _i −D _j | between the sensitivity word pairs i and j is larger than the allowable range S _ij of the input difference between the sensitivity word pairs i and j (step S3). If | D _i −D _j | ≦ S _ij , the process proceeds to step S7.

If | D _i −D _j |> S _ij , it is determined whether or not the input value D _j of the sensitivity word pair _j is larger than the input value D _i of the sensitivity word pair i (step S4). If D _j > D _i , the input value D _j of the sensitivity word pair _j is corrected by D _j = D _i + S _ij (step S5). Then, the process proceeds to step S7.

If D _j <D _i in step S4, the input value D _j of the sensitivity word pair _j is corrected by D _j = D _i −S _ij (step S6). Proceed to step S7.

  In step S7, it is determined whether j is less than 3 (j <3). In the first step S7, since j = 1 and j <3, j is incremented by 1 (j = j + 1) (step S8), and the process returns to step S2. And the process of step S2-step S7 is performed with respect to i = 0 and j = 2. In the second step S7, since j = 2 and j <3, j is incremented by 1 (j = j + 1) (step S8), and the process returns to step S2. And the process of step S2-step S7 is performed with respect to i = 0 and j = 3.

  When the processing of step S2 to step S7 is performed for i = 0 and j = 3, NO is determined in step S7, and i is incremented by 1 (i = i + 1) (step S9). Then, it is determined whether i is less than 3 (i <3) (step S10). In the first step S9, since i = 1 and i <3, j is updated by j = i + 1 (step S11), and then the process returns to step S2. Then, the process of steps S2 to S8 is performed for i = 1 and j = 2, and the process of steps S2 to S7 is performed for i = 1 and j = 3.

  When the processing of steps S2 to S7 is performed for i = 1 and j = 3, NO is determined in step S7, and i is incremented by 1 (i = i + 1) (step S9). Then, it is determined whether i is less than 3 (i <3) (step S10). In the second step S10, since i = 2 and i <3, j is updated by j = i + 1 (step S11), and then the process returns to step S2. And the process of step S2-step S7 is performed with respect to i = 2 and j = 3.

  When the processes of steps S2 to S7 are performed for i = 2 and j = 3, NO is determined in step S7, and i is incremented by 1 (i = i + 1) (step S9). Then, it is determined whether i is less than 3 (i <3) (step S10). In step S10 for the third time, since i = 3, the determination is NO and the correction process is terminated.

  In the first embodiment, the input value input by the user is automatically corrected. However, the input allowable range is obtained by calculating the input allowable range for each sensitivity word pair without automatically correcting the input value. As shown in FIG. 6, you may make it show clearly with a specific color on a display.

  The input allowable range of each sensitivity word pair is determined as follows, for example. That is, for each combination of a sensitivity word pair having a higher input priority and a sensitivity word pair having a lower priority, a sensitivity word pair having a higher priority is based on the correlation value between the two sensitivity word pairs. The input allowable range of a sensitivity word pair having a low priority with respect to the current input value is calculated. For each sensitivity word pair, a range that is common to all of the input allowable ranges calculated by the first means for each combination of the sensitivity word pair and each of the sensitivity word pairs with higher priority. And the obtained range is determined as the recommended input range for the sensitivity word pair.

  FIG. 7 shows an input allowable range display processing procedure.

  In this process, the input allowable range display process is performed every time the input value is changed.

  Sensitive word pair “bright-dark” is “0”, “clear” -sensitive word pair is “1”, “smooth-crisp” sensitivity word pair is “2”, “gentle-severe” The sensitivity word pair “” is represented by “3”. Further, it is assumed that correlations as shown in Table 1 exist between the sensitivity word pairs as described above from the questionnaire results.

  In addition, the priority order of each sensitivity word pair is “0” (bright-dark)> “1” (clear-congested)> “2” (smooth-crisp)> “3” (mild-violent) It shall be.

The correlation value between the sensitivity word pair i and the sensitivity word pair j is represented by H _ij . Also, the input value of the sensitivity word pair _i is represented by Di. The recommended input upper limit value of the sensitivity word pair j is represented by To _j and the recommended input lower limit value of the sensitivity word pair _j is represented by Bo _j . Also, represents the input tolerance limit of sensitivity word pair j which is determined by the input value D _i of sensitivity word pair i at MT _ij, allowable input sensibility word pair j which is determined by the input value D _i of sensitivity word pairs i The lower limit value is expressed by MB _ij .

  First, for the 0th sensitivity word pair “0” having the highest priority, the range from the scale 0 to the scale 6 (all input ranges) is displayed in a specific color as the recommended input range (step S21).

As an initial setting, with respect to j = 1, 2, 3, and sets the input recommended lower limit Bo _j to the minimum value 0, sets the input recommended limits the To _j the maximum value 6 (step S22). That is, Bo ₁ = Bo ₂ = Bo ₃ = 0 and To ₁ = To ₂ = To ₃ = 6 are set.

Next, i = 0 and j = 1 are set (step S23). An allowable range S _ij of an input difference between the sensitivity word pairs i and j is calculated based on the following equation (6) (step S24).

S _ij = INT [(1.0−H _ij ) × 6 + 0.9] (6)

The input allowable upper limit value MT _{ij of the} sensitivity word pair j and the input of the sensitivity word pair _i determined by the input value D _i of the sensitivity word pair i based on the allowable range S _ij of the input difference between the sensitivity word pair i and j. An input allowable lower limit MB _ij of the sensitivity word pair j determined by the value D _i is calculated (step S25). Specifically, the input allowable upper limit value MT _ij and the input allowable lower limit value MB _ij of the sensitivity word pair j are calculated based on the following equation (7).

MT _ij = D _i + S _ij
MB _ij = D _i −S _ij (7)

MT _ij calculated by the above equation (7) may exceed 6. Also, MB _ij calculated by the above equation (7) may be smaller than 0.

Therefore, it is determined whether MT _ij exceeds 6 (step S26). If MT _ij is 6 or less (MT _ij ≦ 6), the process proceeds to step S28. If MT _ij > 6, after MT _ij = 6 (step S27), the process proceeds to step S28.

In step S28, it is determined whether MB _ij is smaller than zero. If MB _ij is 0 or more (MB _ij ≧ 0), the process proceeds to step S30. If MB _ij <06, after _setting MB _ij = 0 (step S29), the process proceeds to step S30.

In step S30, it is determined whether MT _ij is smaller than the recommended input upper limit value To _j . If MT _ij is input recommended limits the To _j or more, the process proceeds to step S32 without changing the input recommended limits the To _j. If MT _ij is smaller than the recommended input upper limit value To _j , the recommended input upper limit value To _j is changed to MT _ij (step S31), and the process proceeds to step S32.

In step S32, it is determined whether or not MB _ij is larger than the recommended input lower limit value Bo _j . If MB _ij is less than the input recommended limit value Bo _j, the process proceeds to step S34 without changing the input recommended lower limit Bo _j. If MB _ij is larger than the recommended input lower limit Bo _j , the recommended input lower limit Bo _j is changed to MB _ij (step S33), and the process proceeds to step S34.

  In step S34, it is determined whether j is less than 3 (j <3). In the first step S34, since j = 1 and j <3, j is incremented by 1 (j = j + 1) (step S35), and the process returns to step S24. Then, the processes of steps S24 to S34 are performed for i = 0 and j = 2. In the second step S34, since j = 2 and j <3, j is incremented by 1 (j = j + 1) (step S35), and the process returns to step S24. Then, the processing of step S24 to step S34 is performed for i = 0 and j = 3.

When the processing of step S24 to step S34 is performed for i = 0 and j = 3, NO is determined in step S34, and i is incremented by 1 (i = i + 1) (step S36). Further, j = i is set (step S37). In the first step S37, j = 1. Then, it is determined whether or not the recommended input upper limit value To _j is larger than the recommended input lower limit value Bo _j (step S38). If the input is greater than recommended limits the To _j input recommended lower limit Bo _j, relative sensitivity word pairs j, in particular color as input recommended range range of input recommended limits the To _j from the input recommended lower limit Bo _j Displayed (step S39). Thus, the recommended input range is displayed for the sensitivity word pair “1”. Then, the process proceeds to step S40. In step S38, when the recommended input upper limit value To _j is equal to or smaller than the recommended input lower limit value Bo _j , there is no recommended input range for the sensitivity word pair j, so that the recommended input range is not displayed in a specific color. The process proceeds to step S40.

  In step S40, it is determined whether i is less than 3 (i <3). In the first step S40, since i = 1 and i <3, j is updated by j = i + 1 (step S41), and then the process returns to step S24. Then, the processes of steps S24 to S35 are performed for i = 1 and j = 2, and the processes of steps S24 to S34 are performed for i = 1 and j = 3.

When the processing of step S24 to step S34 is performed for i = 1 and j = 3, NO is determined in step S34, and i is incremented by 1 (i = i + 1) (step S36). Further, j = i is set (step S37). In the second step S37, j = 2. Then, it is determined whether or not the recommended input upper limit value To _j is larger than the recommended input lower limit value Bo _j (step S38). If the input is greater than recommended limits the To _j input recommended lower limit Bo _j, relative sensitivity word pairs j, in particular color as input recommended range range of input recommended limits the To _j from the input recommended lower limit Bo _j Displayed (step S39). Thus, the recommended input range is displayed for the sensitivity word pair “2”. Then, the process proceeds to step S40. In step S38, when the recommended input upper limit value To _j is equal to or smaller than the recommended input lower limit value Bo _j , there is no recommended input range for the sensitivity word pair j, so that the recommended input range is not displayed in a specific color. The process proceeds to step S40.

  In step S40, it is determined whether i is less than 3 (i <3). In the second step S40, since i = 2 and i <3, j is updated by j = i + 1 (step S41), and then the process returns to step S24. Then, the processing of step S24 to step S34 is performed for i = 2 and j = 3.

When the processes of steps S24 to S34 are performed for i = 2 and j = 3, NO is determined in step S34, and i is incremented by 1 (i = i + 1) (step S36). Further, j = i is set (step S37). In the third step S37, j = 3. Then, it is determined whether or not the recommended input upper limit value To _j is larger than the recommended input lower limit value Bo _j (step S38). If the input is greater than recommended limits the To _j input recommended lower limit Bo _j, relative sensitivity word pairs j, in particular color as input recommended range range of input recommended limits the To _j from the input recommended lower limit Bo _j Displayed (step S39). Thus, the recommended input range is displayed for the sensitivity word pair “3”. Then, the process proceeds to step S40. In step S38, when the recommended input upper limit value To _j is equal to or smaller than the recommended input lower limit value Bo _j , there is no recommended input range for the sensitivity word pair j, so that the recommended input range is not displayed in a specific color. The process proceeds to step S40.

  In step S40, it is determined whether i is less than 3 (i <3). In step S40 for the third time, since i = 3, the determination is NO, and the input allowable range display process is terminated.

It is a block diagram which shows the structure of a music database search system. 2 is a block diagram showing more specifically the configuration of an automatic indexing unit 102. FIG. It is a schematic diagram which shows the example of a display of the input value input with respect to each sensitivity word pair. It is a schematic diagram which shows the input value after correction | amendment with respect to FIG. It is a flowchart which shows the correction process procedure of an input value. It is a schematic diagram which shows the example which specified the input tolerance | permissible_range with respect to each sensitivity word pair with the specific color on the display. It is a flowchart which shows an input permissible range display processing procedure.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Music sensitivity space generation part 200 Search system part 101 Initial music sensitivity space generation part 102 Automatic indexing part 111 Music impression evaluation part 112 Factor analysis part 113 Database part 121 Physical feature-value extraction part 122 Physical feature-value / music impression value conversion Unit 123 song impression value / music sensitivity space coordinate conversion unit 201 user input unit 202 song impression value / music sensitivity space coordinate conversion unit 203 search unit 204 search result output unit

Claims (6)

For each content registered in the database, the sensitivity space coordinate value obtained for the content is registered in the database as an index of the content, and the input value for each sensitivity word pair input by the user at the time of search In the database search device for searching the content having the shortest spatial distance from the calculated sensitivity space coordinate value, the sensitivity space coordinate value corresponding to the input value for each input sensitivity word pair is calculated based on
Based on a correlation value between each sensitivity word pair set in advance, the input value correction means for automatically correcting the input value for each sensitivity word pair input by the user,
Input value correction means
A first means for sequentially selecting a combination of one sensitivity word pair having a higher priority and one sensitivity word pair having a lower priority in descending order of priority;
A second means for calculating an allowable range of an input value difference between the two sensitivity word pairs based on a correlation value between the two sensitivity word pairs each time two sensitivity word pairs are selected by the first means; In addition, every time the allowable range of the input value difference between two sensitivity word pairs is calculated by the second means, does the difference between the input values for the two sensitivity word pairs exceed the allowable range calculated by the second means? If not, and if it exceeds, the lower sensitivity word of the two sensitivity word pairs so that the difference between the input values of the two sensitivity word pairs is within the allowable range. A third means for correcting the input value of the pair;
A database search device characterized by comprising: For each content registered in the database, the sensitivity space coordinate value obtained for the content is registered in the database as an index of the content, and the input value for each sensitivity word pair input by the user at the time of search In the database search device for searching the content having the shortest spatial distance from the calculated sensitivity space coordinate value, the sensitivity space coordinate value corresponding to the input value for each input sensitivity word pair is calculated based on
Based on a preset correlation value between each sensitivity word pair, it is provided with a recommended input range display means for displaying a recommended input range for each sensitivity word pair,
The recommended input range display means is
For each combination of one sensitive word pair having a higher priority and one sensitive word pair having a lower priority, the current sensitivity word pair having a higher priority is determined based on the correlation value between the two sensitive word pairs. A first means for calculating an input allowable range of a sensitivity word pair having a low priority order with respect to an input value;
For each sensitivity word pair, a range common to all the input allowable ranges calculated by the first means for each combination of the sensitivity word pair and each of the sensitivity word pairs with higher priority is calculated. Second means for determining the obtained range as the recommended input range for the sensitivity word pair, and third means for displaying the recommended input range determined for each sensitivity word pair by the second means,
A database search device characterized by comprising: When a plurality of contents are registered and each database is searched for a database in which the sensitivity space coordinate values obtained for the contents are registered as an index of the contents, each sensitivity word input by the user A database that calculates the emotional space coordinate value corresponding to the input value for each input sensitivity word pair based on the input value for each pair and searches for the content having the shortest spatial distance from the calculated sensitivity space coordinate value In a database search method executed by a search device ,
An input value correction step for automatically correcting the input value for each input sensitivity word pair based on the correlation value between each sensitivity word pair set in advance;
The input value correction step
A first step of sequentially selecting a combination of one sensitivity word pair having a higher priority and one sensitivity word pair having a lower priority in order from the highest priority;
A second step of calculating an allowable range of an input value difference between the two sensitivity word pairs based on a correlation value between the two sensitivity word pairs each time two sensitivity word pairs are selected in the first step; In addition, every time the allowable range of the input value difference between the two sensitivity word pairs is calculated in the second step, does the difference of the input value for the two sensitivity word pairs exceed the allowable range calculated in the second step? If not, and if it exceeds, the lower sensitivity word of the two sensitivity word pairs so that the difference between the input values of the two sensitivity word pairs is within the allowable range. A third step of correcting the input value of the pair;
A database search method characterized by comprising: When a plurality of contents are registered and each database is searched for a database in which the sensitivity space coordinate values obtained for the contents are registered as an index of the contents, each sensitivity word input by the user A database that calculates the emotional space coordinate value corresponding to the input value for each input sensitivity word pair based on the input value for each pair and searches for the content having the shortest spatial distance from the calculated sensitivity space coordinate value In a database search method executed by a search device ,
A recommended input range display step for displaying a recommended input range for each sensitivity word pair based on a preset correlation value between each sensitivity word pair,
The recommended input range display step is
For each combination of one sensitive word pair having a higher priority and one sensitive word pair having a lower priority, the current sensitivity word pair having a higher priority is determined based on the correlation value between the two sensitive word pairs. A first step of calculating an input allowable range of a sensitivity word pair having a low priority with respect to an input value;
For each sensitivity word pair, a range common to all of the input allowable ranges calculated in the first step is calculated for each combination of the sensitivity word pair and each of the sensitivity word pairs having higher priority. A second step of determining the obtained range as the recommended input range for the sensitivity word pair, and a third step of displaying the recommended input range determined for each sensitivity word pair in the second step;
A database search method characterized by comprising: On the computer,
When a plurality of contents are registered and each database is searched for a database in which the sensitivity space coordinate values obtained for the contents are registered as an index of the contents, each sensitivity word input by the user A database that calculates the emotional space coordinate value corresponding to the input value for each input sensitivity word pair based on the input value for each pair and searches for the content having the shortest spatial distance from the calculated sensitivity space coordinate value In a database search program for executing a search method,
The database search method includes:
An input value correcting step for automatically correcting an input value for each sensitivity word pair input by the user based on a correlation value between the sensitivity word pairs set in advance;
The input value correction step
A first step of sequentially selecting a combination of one sensitivity word pair having a higher priority and one sensitivity word pair having a lower priority in order from the highest priority;
A second step of calculating an allowable range of an input value difference between the two sensitivity word pairs based on a correlation value between the two sensitivity word pairs each time two sensitivity word pairs are selected in the first step; In addition, every time the allowable range of the input value difference between the two sensitivity word pairs is calculated in the second step, does the difference of the input value for the two sensitivity word pairs exceed the allowable range calculated in the second step? If not, and if it exceeds, the lower sensitivity word of the two sensitivity word pairs so that the difference between the input values of the two sensitivity word pairs is within the allowable range. A third step of correcting the input value of the pair;
A database search program characterized by comprising: On the computer,
When a plurality of contents are registered and each database is searched for a database in which the sensitivity space coordinate values obtained for the contents are registered as an index of the contents, each sensitivity word input by the user A database that calculates the emotional space coordinate value corresponding to the input value for each input sensitivity word pair based on the input value for each pair and searches for the content having the shortest spatial distance from the calculated sensitivity space coordinate value In a database search program for executing a search method,
The database search method includes:
A recommended input range display step for displaying a recommended input range for each sensitivity word pair based on a preset correlation value between each sensitivity word pair,
The recommended input range display step is
For each combination of one sensitive word pair having a higher priority and one sensitive word pair having a lower priority, the current sensitivity word pair having a higher priority is determined based on the correlation value between the two sensitive word pairs. A first step of calculating an input allowable range of a sensitivity word pair having a low priority with respect to an input value;
For each sensitivity word pair, a range common to all of the input allowable ranges calculated in the first step is calculated for each combination of the sensitivity word pair and each of the sensitivity word pairs having higher priority. A second step of determining the obtained range as the recommended input range for the sensitivity word pair, and a third step of displaying the recommended input range determined for each sensitivity word pair in the second step;
A database search program characterized by comprising:

Images