JP6536264B2 - Grouping processing method, grouping processing program and information processing apparatus - Google Patents

Description

  The present invention relates to a grouping processing method, a grouping processing program, and an information processing apparatus.

  Conventionally, ABC analysis may be used to manage products, services, etc. In the ABC analysis, for example, each factor is classified into a plurality of groups (hereinafter, also referred to as a rank) according to the accumulation of the numerical component ratio of the factor (hereinafter, also referred to as an element) representing a product or service. In ABC analysis, for example, the accumulation of the numerical component ratio of factors representing products, services etc. is classified as A rank up to 70%, B rank from 70% to 90%, C rank from 90% to 100%. Then, prioritize the scope of analysis and analyze. Also, using Pareto charts to represent ABC analysis, for example, determining factors important for problem solving has been performed.

Japanese Patent Application Publication No. 2004-010312 Unexamined-Japanese-Patent No. 5-081301

  However, for example, even if the numerical difference between the last factor of high rank A and the leading factor of B rank is very small, the accumulation of the numerical composition ratio is merely 70%, so Factors that are analyzed as A rank may be classified as B rank. That is, a factor that is originally analyzed as an A rank may be missed by a so-called cut-off. In addition, not only in the case of ABC analysis, but also in the case where a plurality of elements are classified into a plurality of groups, factors which are originally analyzed as one group may be classified into another group.

  In one aspect, the present invention is to provide a grouping processing method, a grouping processing program, and an information processing apparatus capable of suppressing variation in elements in a group.

  In one aspect, the grouping processing method is classified into at least one of the plurality of groups when the plurality of elements are classified into a plurality of groups based on an attribute value for each of a plurality of elements. The index value indicating the variation of the attribute value of the element is smaller than the index value indicating the variation of the attribute value of the element other than the element having the largest and / or the smallest attribute value among the elements classified into the group The computer executes the process to control.

  It is possible to suppress the variation of the elements in the group.

FIG. 1 is a block diagram showing an example of the configuration of the information processing apparatus of the embodiment. FIG. 2 is a diagram showing an example of the primary result table. FIG. 3 is a diagram showing an example of the secondary result table. FIG. 4 is a diagram showing an example of the secondary result table after replacement. FIG. 5 is a diagram showing an example of the standard deviation table. FIG. 6 is a diagram showing an example of the tertiary result table. FIG. 7 is a diagram showing an example of a Pareto chart. FIG. 8 is a diagram showing an example of determination of rank boundaries. FIG. 9 is a diagram showing another example of determination of rank boundaries. FIG. 10 is a diagram illustrating another example of determination of rank boundaries. FIG. 11 is a diagram showing another example of determination of rank boundaries. FIG. 12 is a diagram showing another example of determination of rank boundaries. FIG. 13 is a diagram illustrating an example of the addition of sub ranks. FIG. 14 is a flowchart showing an example of grouping processing of the embodiment. FIG. 15 is a flowchart showing an example of classification processing. FIG. 16 is a flowchart showing an example of the BA incorporation process. FIG. 17 is a flowchart showing an example of the BA or AB incorporation process. FIG. 18 is a flowchart showing an example of the AB process. FIG. 19 is a flowchart illustrating an example of the BA process. FIG. 20 is a flowchart showing an example of the sub rank addition process. FIG. 21 is a flowchart showing an example of the sub rank addition process. FIG. 22 is a flowchart showing an example of the sub rank addition process. FIG. 23 is a diagram illustrating an example of a computer that executes a grouping processing program.

  Hereinafter, embodiments of a grouping processing method, a grouping processing program, and an information processing apparatus disclosed in the present application will be described in detail based on the drawings. The disclosed technology is not limited by the present embodiment. In addition, the following embodiments may be combined as appropriate as long as no contradiction arises.

  FIG. 1 is a block diagram showing an example of the configuration of the information processing apparatus of the embodiment. When the analysis target data is input, the information processing apparatus 100 illustrated in FIG. 1 classifies the plurality of elements into a plurality of groups based on the attribute value of each of the plurality of elements of the analysis target data. For example, the information processing apparatus 100 arranges a plurality of elements in descending order of attribute value, and the accumulation of the numerical component ratio of the elements reaches A rank of 70%, A rank of 70% to 90%, B rank of 90% to 100%. Is classified into C rank. The information processing apparatus 100 calculates an index value indicating a variation in attribute value of an element classified into at least one group among a plurality of groups. The information processing apparatus 100 calculates, for example, the difference between attribute values of elements classified into A rank, and calculates the standard deviation of the calculated difference.

  According to the information processing apparatus 100, the index value indicating the variation of the attribute value is the index value indicating the variation of the attribute value of the elements other than the element having the largest and / or the smallest attribute value among the elements classified into the group. Control to be smaller too. That is, in the information processing apparatus 100, for example, the standard deviation of the difference in attribute value from the first factor A (1) to the final factor A (n) of the A rank is The boundary between A rank and B rank is moved so as to be smaller than the standard deviation up to 1). Here, n is a serial number in A rank. Thereby, the information processing apparatus 100 can suppress the variation of the elements in the group.

  Next, the configuration of the information processing apparatus 100 will be described. As shown in FIG. 1, the information processing apparatus 100 includes an input unit 111, a display unit 112, an operation unit 113, a storage unit 120, and a control unit 130. The information processing apparatus 100 may have various functional units of a known computer, for example, various functional units such as various input devices and voice output devices, in addition to the functional units shown in FIG.

  The input unit 111 is, for example, a medium reader. The input unit 111 reads data to be analyzed stored in a portable recording medium such as a CD-ROM, a DVD disk, or a USB (Universal Serial Bus) memory. That is, analysis target data is input to the input unit 111 from the recording medium. The input unit 111 outputs the input analysis target data to the control unit 130. Further, the input unit 111 may be a communication unit capable of communicating with another information processing apparatus, and may receive data to be analyzed from another information processing apparatus.

  The display unit 112 is a display device for displaying various information. The display unit 112 is realized by, for example, a liquid crystal display as a display device. The display unit 112 displays various screens including the primary result table, the secondary result table, the tertiary result table, the Pareto chart, and the like input from the control unit 130.

  The operation unit 113 is an input device that receives various operations from the user. The operation unit 113 is realized by, for example, a keyboard, a mouse or the like as an input device. The operation unit 113 outputs the operation input by the user to the control unit 130 as operation information. The operation unit 113 may be realized by a touch panel or the like as an input device, and the display device of the display unit 112 and the input device of the operation unit 113 may be integrated.

  The storage unit 120 is realized by, for example, a semiconductor memory device such as a random access memory (RAM) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 has a primary result table 121, a secondary result table 122, a post-replacement secondary result table 123, a standard deviation table 124, and a tertiary result table 125. Further, the storage unit 120 stores information used for processing in the control unit 130.

  The primary result table 121 stores the primary result which is the ABC analysis result of the analysis target data. FIG. 2 is a diagram showing an example of the primary result table. As shown in FIG. 2, the primary result table 121 includes “sequence number”, “factor name”, “factor value”, “factor cumulative value”, “factor cumulative ratio”, “rank”, and “serial number within rank”. Items such as The primary result table 121 stores, for example, one record for each factor.

  "Sequence number" is a number assigned to factors of data to be analyzed, and is an identifier for identifying each factor. “Factor name” is information indicating the name of a factor. The “factor numerical value” is information indicating an attribute value of a factor. “Factor cumulative value” is information indicating a cumulative value obtained by adding factor numerical values in ascending order of serial numbers. “Factor cumulative ratio” is information indicating the ratio of factor cumulative value to the whole. "Rank" is information indicating each rank of the ABC analysis. The “in-rank serial number” is a number assigned to a factor in each rank, and is an identifier for identifying each factor in each rank.

  Returning to the explanation of FIG. 1, the secondary result table 122 stores secondary results obtained by adding a factor numerical difference to primary results. FIG. 3 is a diagram showing an example of the secondary result table. As shown in FIG. 3, the secondary result table 122 includes “sequence number”, “factor name”, “factor value”, “factor cumulative value”, “factor cumulative ratio”, “rank”, and “serial number within rank”. Items such as “factor numerical difference”. The secondary result table 122 stores, for example, one record for each factor.

  "Sequence number", "Factor name", "Factor value", "Factor total value", "Factor total ratio", "Rank" and "Rank within serial number" are the same as the items with the same name in primary result table 121 Therefore, the explanation is omitted. The "factor numerical difference" is the difference between the factor numerical value of one factor and the factor numerical value of the next factor. In the example of the first line of FIG. 3, the factor numerical difference of serial number "1" is the difference "44" between the factor numerical value "678" of serial number "1" and the factor numerical value "634" of serial number "2". It becomes ".

  Returning to the explanation of FIG. 1, the post-replacement secondary result table 123 moves the boundary between ranks with respect to the secondary result, that is, stores the secondary result after moving the factor of the rank boundary. . FIG. 4 is a diagram showing an example of the secondary result table after replacement. As shown in FIG. 4, the secondary result table after replacement 123 includes “sequence number”, “factor name”, “factor value”, “factor cumulative value”, “factor cumulative ratio”, “rank”, “within rank” It has items such as "serial number" and "factor numerical difference". The post-replacement secondary result table 123 stores, for example, one record for each factor.

  Since each item of the secondary result table 123 after replacement is the same as each item of the secondary result table 122, the description thereof will be omitted. In the post-replacement secondary result table 123, compared with the secondary result table 122, the factor of serial number "12" is moved from the A rank to the B rank. That is, in the post-replacement secondary result table 123, the boundary between the A rank and the B rank is on the A rank side, that is, between the serial numbers "12" and "13" compared to the secondary result table 122, It is moved between serial numbers "11" and "12".

  Returning to the explanation of FIG. 1, the standard deviation table 124 stores the standard deviations before and after the exchange of factors in each rank. FIG. 5 is a diagram showing an example of the standard deviation table. As shown in FIG. 5, the standard deviation table 124 has items such as “pre-replacement standard deviation” and “post-replacement standard deviation”. Also, “standard deviation before replacement” and “standard deviation after replacement” have items such as “A rank”, “B rank”, and “C rank”, respectively.

  The “pre-replacement standard deviation” is information indicating the standard deviation of factor numerical differences before replacing between ranks of factors, that is, in each rank of the secondary result table 122. The “post-replacement standard deviation” is information indicating the standard deviation of the factor numerical difference in each rank of the post-replacement secondary result table 123 after inter-place ranks of factors. “A rank” is information indicating the standard deviation of factor numerical differences of factors classified into A rank. “B rank” is information indicating the standard deviation of factor numerical differences of factors classified into B rank. “C rank” is information indicating the standard deviation of factor numerical differences of factors classified into C rank.

  Returning to the explanation of FIG. 1, the cubic result table 125 stores the cubic result obtained by adding the sub rank in each rank to the quadratic result after moving the factor of the rank boundary. FIG. 6 is a diagram showing an example of the tertiary result table. As shown in FIG. 6, the third-order result table 125 includes “sequence number”, “factor name”, “factor value”, “factor cumulative value”, “factor cumulative ratio”, “rank”, and “serial number within rank”. , "Factor numerical difference", "sub rank", etc. The tertiary result table 125 stores, for example, one record for each factor.

  “Serial number”, “Factor name”, “Factor value”, “Factor total value”, “Factor total ratio”, “Rank”, “Rank within serial number” and “Factor numerical difference” are secondary result table 122 Since it is the same as the item of the same name, its explanation is omitted. “Sub rank” is information indicating sub ranks further grouped within each rank. For example, if the sub rank is within the A rank, sub ranks such as “a1” and “a2” are added.

  Returning to the description of FIG. 1, the control unit 130 executes a program stored in an internal storage device using a RAM as a work area, for example, by a central processing unit (CPU) or a micro processing unit (MPU). Is realized by Further, the control unit 130 may be realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The control unit 130 includes a calculation unit 131, a determination unit 132, and a classification unit 133, and implements or executes a function or an action of the information processing described below. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 1, and may be another configuration as long as it performs the information processing described later.

  When classifying the plurality of elements into the plurality of groups based on the attribute values of each of the plurality of elements, the control unit 130 detects variations in the attribute values of the elements classified into at least one of the plurality of groups. Control based on the indicated index value. The control unit 130 sets the index value indicating the variation of the attribute value to the index value indicating the variation of the attribute value of the elements other than the element having the maximum and / or the minimum attribute value among the elements classified into the group. Control to be smaller too.

  When the analysis target data is input from the input unit 111, the calculation unit 131 performs ABC analysis on the analysis target data to generate a primary result, and stores the generated primary result in the primary result table 121. . That is, the calculation unit 131 generates the primary result table 121. The calculation unit 131 sorts, for example, the attribute value of each element (factor) of the analysis target data, that is, the factor having the largest factor numerical value at the top. The calculation unit 131, for the factor to be analyzed, for example, an A rank of up to 70%, a B rank of 70% to 90%, and a rank of 90% to 100% for the factor accumulation ratio which is the accumulation of numerical component ratio Classify into C rank. In other words, the calculation unit 131 sorts the plurality of elements into a plurality of groups by arranging in order from the element having the largest attribute value of each of the plurality of elements.

  Here, a Pareto chart based on a primary result which is a result of the ABC analysis will be described. FIG. 7 is a diagram showing an example of a Pareto chart. As shown in FIG. 7, a graph 5 which is a Pareto chart is obtained by arranging each factor on the horizontal axis, and plotting the factor numerical value, factor total value, and factor total ratio of each factor on the vertical axis. As shown in the graph 5, the results of the ABC analysis can distinguish each rank of A to C at a glance. In addition, the Pareto chart may be generated based not only on the primary result but also on the secondary result and the tertiary result. In the present embodiment, the classification of factors at the boundary between the A rank and the B rank and the boundary between the B rank and the C rank in the Pareto chart is optimized.

  Returning to the explanation of FIG. 1, the calculation unit 131 refers to the primary result table 121 to calculate the difference between each factor. The calculation unit 131 stores the factor numerical value difference, which is the calculated difference, in the secondary result table 122 together with the primary result. That is, the calculation unit 131 generates the secondary result table 122. For example, the calculation unit 131 calculates a difference A (1) −A (2) between the factor value of the first factor A (1) of rank A and the factor value of the second factor A (2) Calculated as factor numerical difference of 1). In the following description, the factor numerical difference is simply expressed as a difference.

  When generating the secondary result table 122, the calculation unit 131 executes AB rank classification processing which is classification processing of A rank and B rank. The AB rank classification process is shared between the calculation unit 131 and the determination unit 132 and executed. Also, in the following description, the classification process of A rank and B rank will be described for the classification process, but the BC rank classification process which is a classification process of B rank and C rank is also the same, so the description thereof will be omitted. . In the BC rank classification process, the A rank and the B rank in the AB rank classification process are replaced with the B rank and the C rank.

  The calculation unit 131 calculates the standard deviation of the factor numerical value difference of each rank based on the secondary result table 122, and stores the standard deviation in the pre-replacement standard deviation column of the standard deviation table 124. In addition, the calculation unit 131 copies the content of the secondary result table 122 to the secondary result table 123 after replacement and stores it. The calculation unit 131 outputs a determination instruction to the determination unit 132 when copying of the content of the secondary result table 122 to the secondary result table 123 after replacement is completed.

  When a calculation instruction is input from determination unit 132, calculation unit 131 calculates the standard deviation of the rank according to the calculation instruction, and stores it in the corresponding rank column of the standard deviation column after replacement of standard deviation table 124. When the calculation of the standard deviation of the rank according to the calculation instruction is completed, the calculation unit 131 outputs the calculation completion information to the determination unit 132.

  When the determination unit 132 receives the determination instruction from the calculation unit 131, the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the final factor of the A rank and the factor one higher The differences A (n-1) -A (n) are compared. Here, n is a serial number within rank A. The determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) as a result of the comparison. If the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n), the determination unit 132 determines that the boundary factor difference A (n)-A rank and B rank. B (1) and the difference B (1) -B (2) between the leading factor of B rank and the factor one lower than B rank are compared. As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2). If the difference A (n) −B (1) is equal to or greater than the difference B (1) −B (2), the determination unit 132 determines that the boundary between the A rank and the B rank of the primary result is valid. judge. That is, the determination unit 132 determines that the boundary between the A rank and the B rank is not moved.

  Here, using FIG. 8, the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n), and the difference A (n) -B (1) is the difference B (B). 1) An example in the case of more than -B (2) will be described. FIG. 8 is a diagram showing an example of determination of rank boundaries. As shown in FIG. 8, in the secondary result table 122a, the A rank is serial numbers "1" to "12", and the B rank is serial numbers "13" to "26". At this time, the boundary factor difference A (n) -B (1) between the A rank and the B rank is A (12) -B (1) = 272-241 = 31. Also, the difference A (n-1) -A (n) is A (11) -A (12) = 298-272 = 26. Further, the difference B (1) -B (2) is B (1) -B (2) = 241-222 = 20. That is, the condition that the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) is satisfied because 31 ≧ 26. Further, the condition that the difference A (n) −B (1) is equal to or greater than the difference B (1) −B (2) is satisfied because 313120. Therefore, in the example of FIG. 8, it is determined that the boundary between the A rank and the B rank is not moved.

  Returning to the description of FIG. 1, the determination unit 132 determines that the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n), but the difference A (n) -B (1). If the difference B) is not more than the difference B (1) -B (2), the BA incorporation process is executed. The BA incorporation process is a process of determining whether or not a factor of B rank is incorporated into A rank. As the BA incorporation process, the determination unit 132 first sets an initial value “1” to the position parameter m for confirming in order from the head +1 of the factor of B rank. Further, the determination unit 132 refers to the post-replacement secondary result table 123 and sets the in-rank serial number of the B rank final factor in the parameter p.

  The determination unit 132 increments m and determines whether p is equal to or greater than m. If p is not equal to or larger than m, the determination unit 132 ends the BA incorporation process. If p is m or more, the determination unit 132 determines that the boundary factor difference A (n) −B (1) between A rank and B rank, and B rank factors B (m) and B (m + 1). And the difference B (m) -B (m + 1) with. The determination unit 132 determines whether the difference A (n) −B (1) is equal to or greater than B (m) −B (m + 1) as a result of the comparison. If the difference A (n) -B (1) is not B (m) -B (m + 1) or more, the determination unit 132 increments m and returns to the determination of whether p is m or more. . If the difference A (n) -B (1) is B (m) -B (m + 1) or more, the determination unit 132 incorporates up to the factor B (m-1) of the B rank into the A rank, End the BA incorporation process.

  Here, although the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) using FIG. 9, the difference A (n) -B (1) is different from the difference B An example in the case where it is not more than (1) -B (2) is explained. FIG. 9 is a diagram showing another example of determination of rank boundaries. As shown in FIG. 9, in the secondary result table 122b, the A rank is serial numbers "1" to "12", and the B rank is serial numbers "13" to "26". At this time, the boundary factor difference A (n) −B (1) between the A rank and the B rank is A (12) −B (1) = 278−244 = 34. Further, the difference A (n-1) -A (n) is A (11) -A (12) = 308-278 = 30. Also, the difference B (1) -B (2) is B (1) -B (2) = 244-201 = 43. That is, the condition that the difference A (n) -B (1) is equal to or more than the difference A (n-1) -A (n) is satisfied because 34 ≧ 30. Further, the condition that the difference A (n) -B (1) is equal to or more than the difference B (1) -B (2) is not satisfied because 34 <43. Therefore, in the example of FIG. 8, the BA incorporation process is executed.

  In the BA incorporation process, first, the determination unit 132 sets parameters m = 1 and p = 14. The determination unit 132 increments m to m = 2. When p and m are compared, 14 ≧ 2, and thus the difference A (n) −B (1) is compared with B (m) −B (m + 1). The difference A (n) -B (1) is A (12) -B (1) = 278-244 = 34. B (m) -B (m + 1) becomes B (2) -B (3) = 201-198 = 3. That is, the condition that the difference A (n) -B (1) is B (m) -B (m + 1) or more is satisfied because 34 ≧ 3. Therefore, the determination unit 132 incorporates up to B (m-1) = B (1) into the A rank. That is, the new boundary factor difference is "34" which is a factor numerical difference of serial number "13".

  It returns to the explanation of FIG. If the difference A (n) -B (1) is not the difference A (n-1) -A (n) or more, the determination unit 132 determines that the boundary factor difference A (n) -B between the A rank and the B rank. (1) And, the difference B (1) -B (2) between the leading factor of the B rank and the factor one lower than the rank is compared. As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2). When the difference A (n) -B (1) is equal to or more than the difference B (1) -B (2), the determination unit 132 determines that the final factor A (n) of the A rank is suitable for the B rank judge. In addition, the determination unit 132 removes A (n) of the secondary result table 123 after replacement from the A rank, shifts it to the top of the B rank, and moves the boundary. Furthermore, the determination unit 132 renumbers the in-rank serial numbers as the boundary moves.

  After moving the boundary, the determination unit 132 outputs, to the calculation unit 131, a calculation instruction to calculate the standard deviation of the factor numerical difference of the A rank in the secondary result table 123 after replacement. When the determination unit 132 receives the calculation completion information of the standard deviation of the factor numerical value difference of rank A in the secondary result table 123 after replacement from the calculation unit 131, the determination unit 132 determines the rank A of the standard deviation after replacement in the standard deviation table 124. refer. The determination unit 132 determines whether the post-replacement standard deviation of the A rank is equal to or less than the pre-replacement standard deviation. If the post-replacement standard deviation is less than or equal to the pre-replacement standard deviation, the determination unit 132 determines that the replacement result is valid, and repeats the comparison of the post-movement boundary. If the post-replacement standard deviation is not equal to or less than the pre-replacement standard deviation, the determination unit 132 makes the replacement result unsuitable and restores the boundary.

  Here, using FIG. 10, the difference A (n) -B (1) is not more than the difference A (n-1) -A (n), and the difference A (n) -B (1) is a difference An example in the case of B (1) -B (2) or more will be described. FIG. 10 is a diagram illustrating another example of determination of rank boundaries. As shown in FIG. 10, in the secondary result table 122c, the A rank is serial numbers "1" to "12", and the B rank is serial numbers "13" to "26". At this time, the boundary factor difference A (n) -B (1) between the A rank and the B rank is A (12) -B (1) = 271-244 = 27. Further, the difference A (n-1) -A (n) is A (11) -A (12) = 308-271 = 37. Further, the difference B (1) -B (2) is B (1) -B (2) = 244-235 = 9. That is, the condition that the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) is not satisfied because 27 <37. The condition that the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2) is satisfied because 27 で 9. Therefore, in the example of FIG. 10, A (12) is moved to B rank.

  First, since the determination unit 132 satisfies the condition that the difference A (n) −B (1) is equal to or more than the difference B (1) −B (2), the final factor A (12) of A rank is B rank. Determined to be suitable. The determination unit 132 removes A (12) of the secondary result table 123a after replacement from A rank, shifts it to B (1) which is the head of B rank, and moves the boundary. The post-replacement secondary result table 123a in FIG. 10 is in a state in which A (12) is removed from A rank and shifted to B (1) which is the head of B rank to move the boundary.

  The determination unit 132 calculates the standard deviation of the factor numerical value differences corresponding to A (1) to A (10), and uses it as the standard deviation after replacement. The factor numerical difference corresponding to A (11) is excluded since it is the difference from the leading factor of B rank. The determination unit 132 determines whether the post-replacement standard deviation is less than or equal to the pre-replacement standard deviation that is the standard deviation of the factor numerical differences corresponding to A (1) to A (11). If the standard deviation after replacement is equal to or less than the standard deviation before replacement, the determination unit 132 determines that the replacement result is valid. If the post-replacement standard deviation is not equal to or less than the pre-replacement standard deviation, the determination unit 132 makes the replacement result unsuitable and restores the boundary.

  Returning to the explanation of FIG. 1, the determination unit 132 determines that the difference A (n) -B (1) is not more than the difference A (n-1) -A (n) and the difference A (n) -B ( If 1) is not more than difference B (1) -B (2), execute BA or AB incorporation process. The BA or AB incorporation process is a process of determining whether or not a factor of B rank is incorporated into A rank, or whether a factor of A rank is incorporated into B rank. The determination unit 132 sets the in-rank serial number of the A rank final factor in the parameter n with reference to the post-replacement secondary result table 123 as the BA or AB incorporation process.

  The determination unit 132 compares the difference A (n-1) -A (n) of the A rank final two factors with the difference B (1) -B (2) of the B rank first two factors. The determination unit 132 determines whether the difference A (n-1) -A (n) is equal to or greater than the difference B (1) -B (2) as a result of the comparison. When the difference A (n-1) -A (n) is equal to or greater than the difference B (1) -B (2), the determination unit 132 determines that the difference A (n-1) -A (n) is the difference B. (1) It is determined whether it is larger than -B (2). The determination unit 132 determines that the difference A (n-1) -A (n) is not larger than the difference B (1) -B (2), that is, the difference A (n-1) -A (n) and the difference B If (1) −B (2) is equal, it is determined that the final factor A (n) of A rank is close to B rank. In addition, the determination unit 132 removes A (n) of the secondary result table 123 after replacement from the A rank, shifts it to the top of the B rank, and moves the boundary. Furthermore, as the boundary moves, the determination unit 132 renumbers the in-rank serial numbers and ends the BA or AB incorporation process.

  Here, an example in which the difference A (n-1) -A (n) and the difference B (1) -B (2) are equal will be described using FIG. FIG. 11 is a diagram showing another example of determination of rank boundaries. As shown in FIG. 11, in the secondary result table 122d, the A rank is serial numbers "1" to "12", and the B rank is serial numbers "13" to "26". At this time, the boundary factor difference A (n) −B (1) between the A rank and the B rank is A (12) −B (1) = 256−244 = 12. Also, the difference A (n-1) -A (n) is A (11) -A (12) = 298-256 = 42. Also, the difference B (1) -B (2) is B (1) -B (2) = 244-202 = 42. That is, the difference A (n-1) -A (n) and the difference B (1) -B (2) are equal to 42 = 42. Therefore, in the example of FIG. 11, it is determined that the final factor A (12) of the A rank is close to the B rank.

  Returning to the description of FIG. 1, when A (n−1) −A (n) is larger than the difference B (1) −B (2), the determination unit 132 executes AB processing. The AB process is a process of determining whether or not a factor of A rank is incorporated into B rank. The determination unit 132 first calculates, as the AB process, the standard deviation of the A rank before incorporating the factor A (n) into the B rank. In the calculation of the standard deviation of rank A, the standard deviation of rank A is calculated excluding the difference between the boundary factor A (n) of rank A and the leading factor B (1) of rank B. The determination unit 132 temporarily stores the standard deviation of the A rank before incorporating the calculated factor A (n) into the B rank as a prestandard deviation in the storage unit 120. The calculation of the standard deviation may be performed by outputting a calculation instruction from the determination unit 132 to the calculation unit 131, and temporarily storing the standard deviation calculated by the calculation unit 131 in the storage unit 120.

  The determination unit 132 decrements the parameter n and determines whether n is zero. If n is zero, the determination unit 132 ends the AB process. If n is not zero, the determination unit 132 determines that the boundary factor difference A (n) −B (1) between A rank and B rank, and A rank factors A (n−1) and A (n). And the difference A (n-1) -A (n) with. The determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) as a result of the comparison. If the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n), the determination unit 132 incorporates up to a factor A (n + 1) of the A rank into the B rank. , AB processing ends.

  If the difference A (n) -B (1) is not equal to or greater than the difference A (n-1) -A (n), the determination unit 132 determines that the A rank of the factor A (n) before incorporating it into the B rank. Calculate the standard deviation. That is, the determination unit 132 calculates the standard deviation of the A rank in a state where the factor of the A rank is smaller by one than the standard deviation of the A rank calculated before the decrement of n. That is, the calculated standard deviation is the standard deviation after incorporating A (n) before decrement of n into the B rank. The determination unit 132 temporarily stores in the storage unit 120 a standard deviation after incorporating the calculated A (n) before decrement in n into the B rank as a standard deviation after incorporation. Note that the calculation unit 131 outputs a calculation instruction to the calculation unit 131 from the determination unit 132 to calculate the standard deviation in the AB process and the BA process described later, and temporarily stores the standard deviation calculated by the calculation unit 131 in the storage unit 120. May be

  The determination unit 132 reads out the pre-embedded standard deviation and the post-embedded standard deviation temporarily stored in the storage unit 120, and determines whether the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation. If the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation, the determination unit 132 incorporates the factor A (n + 1) of the A rank into the B rank to calculate the standard deviation of the new A rank. The determination unit 132 temporarily stores the calculated standard deviation of the new A rank in the storage unit 120 as a pre-embedded standard deviation. After calculating the standard deviation of the new A rank, the determination unit 132 decrements n and returns to the determination of whether n is zero.

  If the pre-embedded standard deviation is not greater than or equal to the post-embedded standard deviation, the determining unit 132 determines that the incorporation of the factor A (n) into the B rank is nonconforming, and ranks up to the factor A (n + 1) of the A rank To complete the AB process. When the determination unit 132 ends the AB process, the determination unit 132 ends the BA or AB incorporation process.

  The determination unit 132 compares the difference A (n-1) -A (n) of the A rank final two factors with the difference B (1) -B (2) of the B rank top two factors as A (n). -1) If A-(n) is not equal to or greater than difference B (1)-B (2), BA processing is executed. The BA process is a process of determining whether or not the factor of B rank is incorporated into A rank. As a BA process, the determination unit 132 first sets an initial value “1” to a position parameter m for checking in order from the top +1 of the factor of B rank. Further, the determination unit 132 refers to the post-replacement secondary result table 123 and sets the in-rank serial number of the B rank final factor in the parameter p. The determination unit 132 calculates the standard deviation of the A rank before incorporating the factor B (m) into the A rank. The determination unit 132 temporarily stores the standard deviation of the A rank before incorporating the calculated factor B (m) into the A rank as a prestandard deviation in the storage unit 120.

  The determination unit 132 increments the parameter m and determines whether m is equal to or less than p. If m is not equal to or smaller than p, the determination unit 132 ends the BA process. If m is equal to or less than p, the determination unit 132 determines that the boundary factor difference A (n) −B (1) between A rank and B rank, and B rank factors B (m) and B (m + 1). And the difference B (m) -B (m + 1) with. The determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference B (m) -B (m + 1) as a result of the comparison. If the difference A (n) -B (1) is equal to or greater than the difference B (m) -B (m + 1), the determination unit 132 incorporates up to the factor B (m-1) of the B rank into the A rank , BA process ends.

  If the difference A (n) −B (1) is not equal to or more than the difference B (m) −B (m + 1), the determination unit 132 determines the standard deviation of the A rank before incorporating the factor B (m) into the A rank. Calculate That is, the determination unit 132 calculates the standard deviation of the A rank in a state where the factor of the A rank is one more than the standard deviation of the A rank calculated before the increment of m. That is, the calculated standard deviation is the standard deviation after incorporating B (m) before incrementing m into the A rank. The determination unit 132 temporarily stores the standard deviation after incorporating the calculated B (m) before incrementing m into the A rank as a standard deviation after incorporating in the storage unit 120.

  The determination unit 132 reads out the pre-embedded standard deviation and the post-embedded standard deviation temporarily stored in the storage unit 120, and determines whether the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation. If the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation, the determination unit 132 incorporates the factor B (m-1) of the B rank into the A rank to calculate the standard deviation of the new A rank. The determination unit 132 temporarily stores the calculated standard deviation of the new A rank in the storage unit 120 as a pre-embedded standard deviation. After calculating the standard deviation of the new A rank, the determination unit 132 increments m and returns to the determination of whether m is equal to or less than p.

  If the pre-embedded standard deviation is not greater than or equal to the post-embedded standard deviation, the determining unit 132 determines that the incorporation of the factor B (m) into the A rank is nonconforming, up to the factor B (m-1) of the B rank. Incorporate into the A rank and complete the BA process. When the determination unit 132 ends the BA process, the determination unit 132 ends the BA or AB incorporation process. Further, at the end of the BA or AB incorporation process, the determination unit 132 reads the standard deviation after incorporation temporarily stored in the storage unit 120, and the A rank column of the standard deviation column after replacement of the standard deviation table 124. Remember to

  Here, an example of the BA process will be described with reference to FIG. FIG. 12 is a diagram showing another example of determination of rank boundaries. As shown in FIG. 12, in the secondary result table 122e, the A rank is a serial number "1" to "12", and the B rank is a serial number "13" to "26". At this time, the boundary factor difference A (n) −B (1) between the A rank and the B rank is A (12) −B (1) = 256−244 = 12. Also, the difference A (n-1) -A (n) is A (11) -A (12) = 298-256 = 42. Also, the difference B (1) -B (2) is B (1) -B (2) = 244-185 = 59. That is, the condition that the difference A (n-1) -A (n) is equal to or more than the difference A (n) -B (1) is not satisfied because 42 <59. Therefore, in the example of FIG. 12, the BA process is performed.

  In the BA process, the determination unit 132 first sets parameters m = 1 and p = 14. The determination unit 132 calculates the standard deviation of the A rank based on the differences corresponding to the factors A (1) to A (11). Here, it is assumed that the standard deviation of the rank A of the difference corresponding to the factors A (1) to A (11) is, for example, 36.63. The determination unit 132 increments m to m = 2. When p and m are compared, 14 ≧ 2, and thus the difference A (n) −B (1) is compared with B (m) −B (m + 1). The difference A (n) -B (1) is A (12) -B (1) = 256-244 = 12. B (m) -B (m + 1) becomes B (2) -B (3) = 185-169 = 16. That is, the condition that the difference A (n) -B (1) is B (m) -B (m + 1) or more is not satisfied because 12 <16.

  The determination unit 132 calculates the standard deviation of the A rank before incorporating the B (2) into the A rank. That is, the determination unit 132 calculates the standard deviation of the A rank after incorporating the B (1) into the A rank. That is, the determination unit 132 calculates the standard deviation of the A rank based on the differences corresponding to A (1) to A (12). Here, it is assumed that the standard deviation of the A rank after incorporating B (1) into the A rank is 35.82. The determination unit 132 incorporates the factor B (1) of the B rank into the A rank and calculates the standard deviation of the new A rank because the pre-embedded standard deviation = 36.63 is equal to or greater than the post-embedded standard deviation = 35.82. Do. The determination unit 132 repeats the BA process with the calculated standard deviation of the new A rank as the incorporated prestandard deviation, and adds the B rank factor to the A rank when the incorporated standard deviation becomes larger than the incorporated prestandard deviation. Finish.

  Returning to the description of FIG. 1, when the AB rank classification process and the BC rank classification process are completed, the determination unit 132 outputs the determination completion information to the classification unit 133. Here, in the post-replacement secondary result table 123, each factor is classified as a determination result of the boundary between A rank and B rank and the boundary between B rank and C rank by the processing of the determination unit 132. The rank is stored. That is, in the AB rank classification process and the BC rank classification process, when the process is completed, the boundary between the first group and the second group adjacent to the first group among the plurality of groups is It will be completed about determination of whether to move based on the difference between the attribute value of each element. That is, the AB rank classification process and the BC rank classification process classify factors into each rank in order from the A rank of high importance in ABC analysis. Moreover, although the rank to which each factor is classified is stored in the secondary result table 123 after replacement, since each of these factors is before the determination of the rank, it is considered as an element candidate to be classified into each group. It can be said.

  When the determination completion information is input from the determination unit 132, the classification unit 133 copies and stores the content of the secondary result table 123 after replacement into the tertiary result table 125. That is, the classification unit 133 determines the determination result of the boundary between the A rank and the B rank and the boundary between the B rank and the C rank in the determination unit 132. In other words, the classification unit 133 classifies each element into each group based on the determination result of the determination unit 132.

  The classification unit 133 refers to the tertiary result table 125 and executes the sub rank addition process. In sub rank addition processing, the part where the numerical difference of the factors in the same rank is large (thin), that is, the part where the relation between the factors is thin and the part where the numerical difference between the factors is small (dark), that is, the relation between the factors is thick Divide the parts into groups. While this analyzes the place which should be analyzed originally from another viewpoint as the factor in the same rank, the more accurate and practical ABC analysis is realized by further dividing the rank in the sub rank. it can.

  The classification unit 133 sets “0” to the parameter q, and sets the in-rank serial number of the final factor of A rank to the parameter r. The classification unit 133 increments q and determines whether q and r are equal. If q and r are not equal, the classification unit 133 compares the factor numerical difference of the q-th factor A (q) of the A rank with the A rank of the standard deviation after replacement of the standard deviation table 124.

  As a result of the comparison, the classification unit 133 determines whether the factor numerical difference of the factor A (q) is smaller than the value of the A rank of the standard deviation after replacement. If the factor numerical difference of the factor A (q) is smaller than the value of the rank A of the standard deviation after replacement, the classification unit 133 sets the factor A (q) and the factor A (q + 1) as the same sub rank. If the factor numerical difference of the factor A (q) is equal to or more than the value of the A rank of the standard deviation after replacement, the classification unit 133 does not set the factor A (q) and the factor A (q + 1) as the same sub rank. The classification unit 133 similarly determines other factors in the A rank. The classification unit 133 determines whether q and r are equal, and if q and r are equal, performs sub-rank addition of B rank.

  The classification unit 133 sets “0” to the parameter s, and sets an in-rank serial number of the final factor of B rank to the parameter t. The classification unit 133 increments s and determines whether s and t are equal. If s and t are not equal, the classification unit 133 compares the factor numerical difference of the s-th factor B (s) of B rank with the B rank of the standard deviation after replacement of the standard deviation table 124.

  As a result of the comparison, the classification unit 133 determines whether the factor numerical difference of the factor B (s) is smaller than the value of the B rank of the standard deviation after replacement. If the factor numerical difference of the factor B (s) is smaller than the value of the B rank of the standard deviation after replacement, the classification unit 133 sets the factor B (s) and the factor B (s + 1) as the same sub rank. If the factor numerical difference of the factor B (s) is equal to or greater than the B rank value of the standard deviation after replacement, the classification unit 133 does not consider the factor B (s) and the factor B (s + 1) as the same sub rank. The classification unit 133 similarly determines other factors in the B rank. The classification unit 133 determines whether s and t are equal. If s and t are equal, the classification unit 133 adds a sub rank of C rank.

  The classification unit 133 sets “0” to the parameter u, and sets the in-rank serial number of the final factor of C rank to the parameter v. The classification unit 133 increments u and determines whether u and v are equal. If u and v are not equal, the classification unit 133 compares the factor numerical difference of the u-th factor C (u) of the C rank with the C rank of the standard deviation after replacement of the standard deviation table 124.

  As a result of the comparison, the classification unit 133 determines whether the factor numerical difference of the factor C (u) is smaller than the value of the C rank of the standard deviation after replacement. If the factor numerical difference of the factor C (u) is smaller than the value of the C rank of the standard deviation after replacement, the classification unit 133 sets the factor C (u) and the factor C (u + 1) as the same subrank. The classification unit 133 does not set the factor C (u) and the factor C (u + 1) as the same subrank, when the factor numerical difference of the factor C (u) is equal to or more than the C rank value of the standard deviation after replacement. The classification unit 133 similarly determines other factors in the C rank. If u and v are equal, the classification unit 133 stores the added sub rank in each rank of A to C in the tertiary result table 125 if u and v are equal. The sub rank addition process is ended.

  Here, an example of sub rank addition will be described with reference to FIG. FIG. 13 is a diagram illustrating an example of the addition of sub ranks. As shown in FIG. 13, in the tertiary result table 125a, the A rank is a serial number "1" to "12", the B rank is a serial number "13" to "26", and the C rank is a serial number "27" to " 52 ”. Within the A rank, serial numbers "3" to "8" are classified as sub rank a1, and serial numbers "10" and "11" are classified as sub rank a2. Moreover, the standard deviation of A rank used for the classification of sub rank is 36.63.

  Similarly, in the B rank, serial numbers “14” to “16” are classified into sub rank b1, and serial numbers “17” to “25” are classified into sub rank b2. Moreover, the standard deviation of B rank used for the classification | category of a sub rank is 15.78. Within rank C, serial numbers “27” to “29” are classified into sub rank c1, and serial numbers “33” and “34” are classified into sub rank c2. Although the sub rank of C rank is classified up to sub rank c5, it is omitted in FIG. Moreover, the standard deviation of C rank used for the classification of sub rank is 2.36.

  Returning to the explanation of FIG. 1, when the classification unit 133 ends the sub rank addition process, the classification unit 133 refers to the tertiary result table 125 and generates a display screen on which the tertiary result table is displayed. The classification unit 133 causes the display unit 112 to display a display screen on which the generated tertiary result table is displayed.

  Next, the operation of the information processing apparatus 100 of the embodiment will be described. FIG. 14 is a flowchart showing an example of grouping processing of the embodiment.

  When the analysis target data is input from the input unit 111, the calculation unit 131 performs ABC analysis on the analysis target data to generate a primary result table 121 (step S1). The calculation unit 131 refers to the primary result table 121 to calculate the difference between each factor. The calculation unit 131 generates the secondary result table 122 based on the factor numerical value difference which is the calculated difference and the primary result (step S2). After generating the secondary result table 122, the calculation unit 131 executes AB rank classification processing, which is classification processing of A rank and B rank (step S3).

  Here, the AB rank classification process will be described with reference to FIG. FIG. 15 is a flowchart showing an example of classification processing. The calculation unit 131 calculates the standard deviation of the factor numerical value difference of each rank based on the secondary result table 122, and stores the standard deviation in the pre-replacement standard deviation column of the standard deviation table 124 (step S11). In addition, the calculation unit 131 copies and stores the content of the secondary result table 122 in the secondary result table 123 after replacement (step S12). The calculation unit 131 outputs a determination instruction to the determination unit 132 when copying of the content of the secondary result table 122 to the secondary result table 123 after replacement is completed.

  When the determination unit 132 receives the determination instruction from the calculation unit 131, the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the final factor of the A rank and the factor one higher The differences A (n-1) -A (n) are compared (step S13). As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is greater than or equal to the difference A (n-1) -A (n) (step S14). If the difference A (n) -B (1) is greater than or equal to the difference A (n-1) -A (n) (Yes at step S14), the determination unit 132 determines whether the difference A (n) -B (1) ) And the difference B (1) -B (2). That is, the determination unit 132 sets the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the difference B (1) −B (2) between the leading factor of the B rank and the factor one lower. ) Are compared (step S15).

  The determination unit 132 determines whether or not the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2) as a result of the comparison (step S16). If the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2) (Step S16: Yes), the determination unit 132 determines that the A result rank and the B rank result. It is determined that the boundary is valid (step S17), the AB rank classification process is finished, and the process returns to the original process.

  If the difference A (n) -B (1) is not equal to or more than the difference B (1) -B (2) (step S16: negative), the determination unit 132 executes the BA incorporation process (step S18).

  Here, the BA incorporation process will be described with reference to FIG. FIG. 16 is a flowchart showing an example of the BA incorporation process. The determination unit 132 sets an initial value “1” to the position parameter m for checking in order from the head +1 of the factor of B rank. Further, the determination unit 132 refers to the post-replacement secondary result table 123, and sets the in-rank serial number of the B rank final factor in the parameter p (step S181).

  The determination unit 132 increments m (step S182), and determines whether p is equal to or greater than m (step S183). If p is not equal to or larger than m (No at Step S183), the determination unit 132 ends the BA incorporation process and returns to the original process. If p is equal to or greater than m (step S183: affirmative), the determination unit 132 determines that the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the B factor B (m). ) And B (m + 1) are compared (step S184). As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than B (m) -B (m + 1) (step S185). If the difference A (n) -B (1) is not equal to or more than B (m) -B (m + 1) (step S185: negative), the determination unit 132 returns to step S182. If the difference A (n) -B (1) is equal to or greater than B (m) -B (m + 1) (step S185: affirmative), the determination unit 132 determines up to the factor B (m-1) of the B rank. After being incorporated into the A rank (step S186), the BA incorporation processing is ended, and the processing returns to the original processing. This allows the factor of B rank to be incorporated into A rank.

  Returning to the description of the AB rank classification process of FIG. 15, when the BA incorporation process of step S18 ends, the determination unit 132 ends the AB rank classification process, and returns to the original process.

  If the difference A (n) -B (1) is not equal to or greater than the difference A (n-1) -A (n) (No at step S14), the determination unit 132 determines that the difference A (n) -B (1) And the difference B (1) -B (2). That is, the determination unit 132 sets the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the difference B (1) −B (2) between the leading factor of the B rank and the factor one lower. ) Are compared (step S19).

  As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2) (step S20). If the difference A (n) -B (1) is equal to or greater than the difference B (1) -B (2) (step S20: affirmative), the determination unit 132 determines that the final factor A (n) of the A rank is B It determines that it is suitable for the rank. Further, the determination unit 132 removes A (n) of the secondary result table 123 after replacement from the A rank, shifts it to the top of the B rank, and moves the boundary (step S21).

  After moving the boundary, the determination unit 132 outputs, to the calculation unit 131, a calculation instruction to calculate the standard deviation of the factor numerical difference of the A rank in the secondary result table 123 after replacement. When the calculation instruction is input from the determination unit 132, the calculation unit 131 calculates the standard deviation of the factor numerical value difference of the A rank, and stores it in the A rank of the standard deviation after replacement of the standard deviation table 124 (step S22). When the calculation of the standard deviation of the factor numerical value difference of rank A is completed, the calculation unit 131 outputs calculation completion information to the determination unit 132. When the calculation completion information is input from the calculation unit 131, the determination unit 132 refers to the A rank of the standard deviation after replacement of the standard deviation table 124. The determination unit 132 determines whether the post-replacement standard deviation of the A rank is less than or equal to the pre-replacement standard deviation (step S23).

  If the post-replacement standard deviation is less than or equal to the pre-replacement standard deviation (Step S23: Yes), the determination unit 132 determines that the replacement result is valid (Step S24), and returns to Step S13. If the post-replacement standard deviation is not less than the pre-replacement standard deviation (step S23: negative), the determiner 132 makes the replacement result unsuitable, restores the boundary back (step S25), and ends the AB rank classification process. , Return to the original processing.

  If the difference A (n) -B (1) is not equal to or more than the difference B (1) -B (2) (step S20: negative), the determination unit 132 executes the BA or AB incorporation process (step S26). .

  Here, the BA or AB incorporation process will be described with reference to FIG. FIG. 17 is a flowchart showing an example of the BA or AB incorporation process. The determination unit 132 refers to the post-replacement secondary result table 123, and sets the in-rank serial number of the A rank final factor in the parameter n (step S261).

  The determination unit 132 compares the difference A (n-1) -A (n) of the A rank final two factors with the difference B (1) -B (2) of the B rank top two factors (step S262). As a result of comparison, the determination unit 132 determines whether the difference A (n-1) -A (n) is equal to or more than the difference B (1) -B (2) (step S263). If the difference A (n-1) -A (n) is greater than or equal to the difference B (1) -B (2) (Yes at step S263), the determination unit 132 determines whether the difference A (n-1) -A It is determined whether (n) is larger than the difference B (1) -B (2) (step S264). If the difference A (n-1) -A (n) is not larger than the difference B (1) -B (2), that is, they are equal (step S264: negative), the determination unit 132 The factor A (n) is determined to be close to the B rank. Further, the determination unit 132 removes A (n) of the secondary result table 123 after replacement from the A rank, shifts it to the top of the B rank, moves the boundary (step S265), and returns to the original processing.

  If A (n-1) -A (n) is larger than the difference B (1) -B (2) (Yes at Step S264), the determination unit 132 executes AB processing (Step S266).

  Here, the AB process will be described with reference to FIG. FIG. 18 is a flowchart showing an example of the AB process. The determination unit 132 calculates the standard deviation of the A rank before incorporating the factor A (n) into the B rank (step S2661). The determination unit 132 temporarily stores the standard deviation of the A rank before incorporating the calculated factor A (n) into the B rank as a prestandard deviation in the storage unit 120. The determination unit 132 decrements the parameter n (step S2662), and determines whether n is zero (step S2663). If n is zero (step S2663: YES), the determination unit 132 returns to the original process.

  If n is not zero (step S2663: negative), the determination unit 132 determines that the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the factor A (n−1) of the A rank ) And A (n) are compared (step S2664). As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) (step S2665). If the difference A (n) -B (1) is equal to or greater than the difference A (n-1) -A (n) (Yes at step S2665), the determination unit 132 continues until the factor A (n + 1) of the A rank. Are incorporated into B rank (step S2666), and the process returns to the original process.

  If the difference A (n) -B (1) is not greater than or equal to the difference A (n-1) -A (n) (No at step S2665), the determination unit 132 assigns the factor A (n) to the B rank. The standard deviation of rank A before installation is calculated (step S2667). That is, the determination unit 132 calculates the standard deviation after incorporating A (n) before decrement of n into the B rank. The determination unit 132 temporarily stores in the storage unit 120 a standard deviation after incorporating the calculated A (n) before decrement in n into the B rank as a standard deviation after incorporation.

  Determination unit 132 reads out the pre-embedded standard deviation and the post-embedded standard deviation temporarily stored in storage unit 120, and determines whether or not the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation (step S2668). ). If the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation (Yes at step S2668), determination unit 132 incorporates factor A (n + 1) of A rank into B rank and calculates the standard deviation of new A rank (Step S2669), and returns to step S2662. If the pre-embedded standard deviation is not equal to or greater than the post-embedded standard deviation (No at step S2668), the decision unit 132 determines that the factor A (n) is not included in the B rank (step S2670). The determination unit 132 incorporates up to the factor A (n + 1) of the A rank into the B rank (step S2666), and returns to the original processing. This allows the factor of A rank to be incorporated into B rank.

  Returning to the description of the BA or AB incorporation process of FIG. 17, the determination unit 132 determines that A (n-1) -A (n) is not greater than or equal to difference B (1) -B (2) (step S263: No), the BA process is executed (step S 267).

  Here, the BA process will be described with reference to FIG. FIG. 19 is a flowchart illustrating an example of the BA process. The determination unit 132 sets an initial value “1” to the position parameter m for checking in order from the head +1 of the factor of B rank. In addition, the determination unit 132 refers to the post-replacement secondary result table 123, and sets the in-rank serial number of the B rank final factor in the parameter p (step S2671). The determination unit 132 calculates the standard deviation of the A rank before incorporating the factor B (m) into the A rank (step S2672). The determination unit 132 temporarily stores the standard deviation of the A rank before incorporating the calculated factor B (m) into the A rank as a prestandard deviation in the storage unit 120.

  The determination unit 132 increments the parameter m (step S 2673), and determines whether m is equal to or less than p (step S 2674). If m is not equal to or smaller than p (step S2674: negative), the determination unit 132 returns to the original process. If m is equal to or smaller than p (step S2674: affirmative), the determination unit 132 determines that the boundary factor difference A (n) −B (1) between the A rank and the B rank, and the B rank factor B (m ) And B (m + 1) are compared (step S2675). As a result of comparison, the determination unit 132 determines whether the difference A (n) -B (1) is equal to or greater than the difference B (m) -B (m + 1) (step S2676). If the difference A (n) -B (1) is equal to or greater than the difference B (m) -B (m + 1) (step S2676: affirmative), the determination unit 132 continues until the factor B (m-1) of the B rank. Are incorporated into the A rank (step S2677), and the process returns to the original process.

  If difference A (n) -B (1) is not equal to or more than difference B (m) -B (m + 1) (step S2676: negative), determination unit 132 determines whether factor B (m) is incorporated into A rank The standard deviation of the A rank of is calculated (step S2678). That is, the determination unit 132 calculates the standard deviation after incorporating B (m) before increment of m into the A rank. The determination unit 132 temporarily stores the standard deviation after incorporating the calculated B (m) before incrementing m into the A rank as a standard deviation after incorporating in the storage unit 120.

  The determination unit 132 reads out the pre-embedded standard deviation and the post-embedded standard deviation temporarily stored in the storage unit 120, and determines whether the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation (step S2679). ). If the pre-embedded standard deviation is equal to or greater than the post-embedded standard deviation (Yes at step S2679), the determination unit 132 incorporates the factor B (m-1) of the B rank into the A rank and the standard deviation of the new A rank Is calculated (step S2680), and the process returns to step S2673.

  If the pre-embedded standard deviation is not equal to or greater than the post-embedded standard deviation (No at step S2679), the decision unit 132 determines that the factor B (m) is not included in the A rank (step S2681). The determination unit 132 incorporates up to the factor B (m-1) of the B rank into the A rank (step S2677), and returns to the original processing. This allows the factor of B rank to be incorporated into A rank.

  Returning to the description of the BA or AB incorporation process of FIG. 17, when the process of step S265, S266 or S267 is completed, the determination unit 132 ends the BA or AB incorporation process, and returns to the original process. Thereby, the factor of B rank can be incorporated into A rank, or the factor of A rank can be incorporated into B rank.

  Returning to the description of the AB rank classification process of FIG. 15, when the BA or AB incorporation process of step S26 ends, the determination unit 132 ends the AB rank classification process, and returns to the original process. Thereby, each factor can be classified between A rank and B rank.

  Returning to the grouping process of FIG. 14, when the AB rank classification process of step S3 ends, the determination unit 132 executes BC rank classification process (step S4). The BC rank classification process is the one in which the A rank and the B rank are replaced with the B rank and the C rank, respectively, in the AB rank classification process described with reference to FIGS. In addition, each factor can be classified between B rank and C rank by BC rank classification processing.

  When the AB rank classification process and the BC rank classification process are completed, the determination unit 132 outputs the determination completion information to the classification unit 133. When the determination completion information is input from the determination unit 132, the classification unit 133 copies and stores the content of the secondary result table 123 after replacement into the tertiary result table 125. The classification unit 133 refers to the tertiary result table 125 and executes the sub rank addition process (step S5).

  Here, the sub rank addition process will be described using FIG. 20 to FIG. FIGS. 20 to 22 are flowcharts showing an example of the sub rank addition process. The classification unit 133 sets “0” to the parameter q, and sets the in-rank serial number of the final factor of A rank to the parameter r (step S51). The classification unit 133 increments q (step S52), and determines whether q and r are equal (step S53). If q and r are not equal (step S53: negative), the classification unit 133 determines that the factor numerical difference of the qth factor A (q) of A rank and the standard deviation A after replacement of the standard deviation table 124 The rank is compared (step S54).

  As a result of the comparison, the classification unit 133 determines whether the factor numerical difference of the factor A (q) is smaller than the value of the A rank of the standard deviation after replacement (step S55). If the factor numerical difference of the factor A (q) is smaller than the value of the rank A of the standard deviation after replacement (step S55: affirmative), the classification unit 133 determines the factor A (q) and the factor A (q + 1). The same subrank is set (step S56), and the process returns to step S52. If the factor numerical difference of factor A (q) is equal to or greater than the value of rank A of the standard deviation after replacement (step S55: negative), the classification unit 133 makes factor A (q) and factor A (q + 1) identical. The process returns to step S52 without making it a subrank of (step S57).

  If q and r are equal (Step S53: Yes), the classification unit 133 sets “0” to the parameter s, and sets the in-rank serial number of the final factor of B rank to the parameter t (Step S58). ). The classification unit 133 increments s (step S59), and determines whether s and t are equal (step S60). If s and t are not equal (step S60: negative), the classification unit 133 determines that the factor numerical difference of the sth factor B (s) of B rank and the standard deviation B after replacement of the standard deviation table 124 The rank is compared (step S61).

  As a result of comparison, the classification unit 133 determines whether the factor numerical difference of the factor B (s) is smaller than the value of the B rank of the standard deviation after replacement (step S62). If the factor numerical difference of the factor B (s) is smaller than the value of the rank B of the standard deviation after replacement (step S62: affirmative), the classification unit 133 determines the factor B (s) and the factor B (s + 1). The same subrank is set (step S63), and the process returns to step S59. If the factor numerical difference of the factor B (s) is equal to or greater than the B rank value of the standard deviation after replacement (step S62: negative), the classification unit 133 makes the factor B (s) and the factor B (s + 1) the same. The process returns to step S59 without making it a subrank of (step S64).

  If s and t are equal (Step S60: Yes), the classification unit 133 sets “0” to the parameter u, and sets the in-rank serial number of the final factor of the C rank to the parameter v (Step S65). ). The classification unit 133 increments u (step S66), and determines whether u and v are equal (step S67). If u and v are not equal (step S67: negative), the classification unit 133 determines the factor numerical difference of the u-th factor C (u) of the C rank and the C of the standard deviation after replacement of the standard deviation table 124. The rank is compared (step S68).

  As a result of the comparison, the classification unit 133 determines whether the factor numerical difference of the factor C (u) is smaller than the value of the C rank of the standard deviation after replacement (step S69). If the factor numerical difference of the factor C (u) is smaller than the value of the C rank of the standard deviation after replacement (step S69: affirmative), the classification unit 133 determines the factor C (u) and the factor C (u + 1). The same subrank is set (step S70), and the process returns to step S66. If the factor numerical difference of the factor C (u) is equal to or greater than the C rank value of the standard deviation after replacement (step S69: negative), the classification unit 133 makes the factor C (u) and the factor C (u + 1) the same. The process returns to step S66 without making it a subrank of (step S71).

  If u and v are equal (Step S67: Yes), the classification unit 133 stores the added sub rank in each rank of A to C in the tertiary result table 125, and ends the sub rank addition process, Return to the original process. Thereby, the classification unit 133 can further classify the factors in each rank of A to C into subgroups.

  Returning to the grouping process of FIG. 14, when the sub-rank addition process is ended, the classification unit 133 refers to the tertiary result table 125 and generates a display screen displaying the tertiary result table. The classification unit 133 causes the display unit 112 to display a display screen on which the generated tertiary result table is displayed (step S6). Thereby, the information processing apparatus 100 can suppress the variation of the factor in each rank of A to C. That is, the information processing apparatus 100 can suppress the variation of the elements in the group.

  As described above, when the information processing apparatus 100 classifies the plurality of elements into the plurality of groups based on the attribute value of each of the plurality of elements, the information processing apparatus 100 can identify the elements classified into at least one of the plurality of groups. Control is performed such that the index value indicating the variation of the attribute value is smaller than the index value indicating the variation of the attribute value for the elements other than the element having the attribute value of maximum and / or minimum among the elements classified into the group Do. As a result, it is possible to suppress the variation of the elements in the group.

  Further, when the information processing apparatus 100 classifies the plurality of elements into the plurality of groups based on the attribute values of the plurality of elements, the index value indicating the variation of the attribute values of the element candidates included in the first group is If it is smaller than the index value indicating the variation of the attribute value of the element candidate excluding the element candidate having the smallest attribute value from the element candidate, the element candidate having the smallest attribute value is classified into the second group. As a result, it is possible to suppress the variation of the elements in the group.

  Further, the information processing apparatus 100 arranges the plurality of elements in order from the element having the largest attribute value of each of the plurality of elements, classifies them into a plurality of groups, and calculates the difference between the attribute values of each element. Further, the information processing apparatus 100 determines whether to move the boundary between the first group and the second group adjacent to the first group among the plurality of groups based on the difference. In addition, the information processing apparatus 100 classifies each element into each group based on the result of the determination. As a result, it is possible to suppress the variation of the elements in the group.

  Further, the information processing apparatus 100 sets the difference between the attribute value of the element of the first group in contact with the boundary and the attribute value of the element of the second group in contact with the boundary as the reference difference. Move the boundary to the previous difference or later difference side that is larger than the reference difference if the previous difference or later difference is greater than the reference difference, comparing the previous difference and the later difference. Make a decision. As a result, the elements at the boundary of the group can be incorporated into the group that can suppress the variation of the elements in the group. Note that the previous difference is the difference between the attribute value of the element of the first group adjacent to the boundary and the attribute value of the element of the first group adjacent to the element of the first group adjacent to the boundary. Also, the difference behind is the difference between the attribute value of the element of the second group adjacent to the boundary and the attribute value of the element of the second group adjacent to the element of the second group adjacent to the boundary.

  In addition, when the previous difference and the subsequent difference are equal, the information processing apparatus 100 determines that the boundary is moved so as to classify the elements of the first group in contact with the boundary into the second group. Do. As a result, it is possible to suppress the variation in the elements in the first group.

  Further, the information processing apparatus 100 calculates an index value indicating variation in difference for each of a plurality of groups before and after the movement of the boundary. Further, when the index value indicating the variation of the difference after the movement of the boundary is smaller than the index value indicating the variation of the difference before the movement of the boundary, the information processing apparatus 100 determines to move the boundary. As a result, for the group on the side where the index value indicating the variation of the difference becomes smaller, the variation of the elements in the group can be suppressed.

  Further, in the information processing apparatus 100, the index value indicating the variation of the difference is a standard deviation of the differences of the plurality of groups. As a result, based on the standard deviation of each group, it is possible to suppress the variation of the elements in the group.

  Further, the information processing apparatus 100 further compares the difference and the standard deviation for each of a plurality of groups, and when the difference is smaller than the standard deviation, each element for which the difference to be compared is obtained is the same. Classify into subgroups. As a result, the elements in each group can be further classified into subgroups.

  Although the standard deviation of the factor numerical value difference is used as the index value indicating the variation in the above embodiment, the present invention is not limited to this. For example, the standard deviation of the factor value may be used as an index value indicating variation. Further, the index value indicating the variation may be a value used for not only the standard deviation but also other statistics such as an average value and a median value.

  Moreover, in the said Example, although the subrank was provided 1 step of rank lower rank, it is not limited to this. For example, the rank of the lower hierarchy may be provided below the sub rank.

  Further, each component of each unit shown in the drawings does not necessarily have to be physically configured as shown in the drawings. That is, the specific form of the dispersion and integration of each part is not limited to the illustrated one, and all or a part thereof is functionally or physically dispersed or integrated in any unit according to various loads, usage conditions, etc. Can be configured. For example, the calculation unit 131 and the determination unit 132 may be integrated. The illustrated processes are not limited to the above-described order, and may be performed at the same time as long as the process contents do not contradict each other, or the order may be changed.

  Furthermore, all or any part of various processing functions performed by each device may be executed on a CPU (or a microcomputer such as an MPU or an MCU (Micro Controller Unit)). In addition, various processing functions may be executed in whole or any part on a program analyzed and executed by a CPU (or a microcomputer such as an MPU or MCU) or on hardware by wired logic. It goes without saying that it is good.

  The various processes described in the above embodiments can be realized by executing a prepared program on a computer. So, below, an example of a computer which runs a program which has the same function as the above-mentioned example is explained. FIG. 23 is a diagram illustrating an example of a computer that executes a grouping processing program.

  As shown in FIG. 23, the computer 200 has a CPU 201 that executes various arithmetic processing, an input device 202 that receives data input, and a monitor 203. The computer 200 also includes a medium reading device 204 reading programs and the like from a storage medium, an interface device 205 for connecting with various devices, and a communication device 206 for connecting with other information processing devices and the like by wire or wirelessly. Have. The computer 200 also has a RAM 207 for temporarily storing various information, and a hard disk drive 208. Each of the devices 201 to 208 is connected to the bus 209.

  The hard disk drive 208 stores a grouping processing program having the same function as each processing unit of the calculation unit 131, the determination unit 132, and the classification unit 133 shown in FIG. Further, the hard disk drive 208 includes a primary result table 121, a secondary result table 122, a secondary result table after replacement 123, a standard deviation table 124, a tertiary result table 125, and various types for realizing a grouping processing program. Data is stored. The input device 202 receives, for example, an input of various information such as analysis target data and management information from the manager of the computer 200. The monitor 203 displays, for example, various screens such as a screen of a tertiary result table and a screen of management information to an administrator of the computer 200. For example, a printing device or the like is connected to the interface device 205. The communication device 206 has, for example, the same function as the communication unit, which is an example of the input unit 111 illustrated in FIG. 1, is connected to a network (not shown), and receives analysis target data from other information processing devices. Exchange information.

  The CPU 201 reads out each program stored in the hard disk device 208, develops the program in the RAM 207, and executes the program to perform various processes. Also, these programs can cause the computer 200 to function as the calculation unit 131, the determination unit 132, and the classification unit 133 illustrated in FIG.

  The above-described grouping processing program does not necessarily have to be stored in the hard disk device 208. For example, the computer 200 may read out and execute a program stored in a storage medium readable by the computer 200. The storage medium readable by the computer 200 corresponds to, for example, a portable storage medium such as a CD-ROM or a DVD disk, a USB memory, a semiconductor memory such as a flash memory, a hard disk drive, or the like. Alternatively, the grouping processing program may be stored in a device connected to a public line, the Internet, a LAN or the like, and the computer 200 may read out and execute the grouping processing program therefrom.

  The following appendices are further disclosed regarding the embodiment including the present embodiment.

(Supplementary Note 1) When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, variation in the attribute values of elements classified into at least one of the plurality of groups Is controlled so that the index value indicating is smaller than the index value indicating the variation of the attribute value for the elements other than the element having the attribute value of maximum and / or minimum among the elements classified into the group
And a computer executes the processing.

(Supplementary Note 2) When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, an index value indicating a variation of the attribute value of the element candidate included in the first group is the element candidate The element candidate with the smallest attribute value is classified into the second group if the element value is smaller than the index value indicating the variation in the attribute value of the element candidate excluding the element candidate with the smallest attribute value from.
And a computer executes the processing.

(Supplementary Note 3) A plurality of elements are arranged in order from the element with the largest attribute value of each of the plurality of elements and classified into a plurality of groups, and the difference between the attribute values of the respective elements is calculated
It is determined whether to move the boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference.
The elements are classified into the respective groups based on the result of the determination.
And a computer executes the processing.

(Supplementary Note 4) In the determination processing, a difference between an attribute value of the element of the first group in contact with the boundary and an attribute value of the element of the second group in contact with the boundary is used as a reference difference Comparing the difference of the reference with the difference before and after the difference of the reference, and when the previous difference or the later difference is larger than the difference of the reference, the difference of the reference is greater than the difference of the reference It is determined to move the boundary to the larger previous difference or the later difference side,
The grouping processing method according to appendix 3, characterized in that

(Supplementary Note 5) In the determination process, the elements of the first group in contact with the boundary are classified into the second group when the previous difference and the later difference are equal. To make a decision to move the boundary,
The grouping processing method according to appendix 4, characterized in that

(Supplementary Note 6) The calculating process calculates an index value indicating the variation of the difference for each of the plurality of groups before and after the movement of the boundary.
The determination processing performs the determination to move the boundary when the index value indicating the variation of the difference after the movement of the boundary is smaller than the index value indicating the variation of the difference before the movement of the boundary. ,
The grouping processing method according to appendix 3, characterized in that

(Supplementary Note 7) The grouping processing method according to Supplementary note 6, wherein the index value indicating the variation of the difference is a standard deviation of the difference for each of the plurality of groups.

(Supplementary Note 8) The classification processing further compares the difference with the standard deviation for each of the plurality of groups, and when the difference is smaller than the standard deviation, the difference to be compared is compared. Classify the determined elements into the same sub-group,
The grouping processing method according to appendix 7, characterized in that

(Supplementary Note 9) When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, variation in the attribute values of elements classified into at least one of the plurality of groups Is controlled so that the index value indicating is smaller than the index value indicating the variation of the attribute value for the elements other than the element having the attribute value of maximum and / or minimum among the elements classified into the group
A grouping processing program that causes a computer to execute the processing.

(Supplementary Note 10) When classifying the plurality of elements into a plurality of groups based on the attribute value of each of a plurality of elements, an index value indicating a variation in the attribute value of the element candidate included in the first group is the element candidate The element candidate with the smallest attribute value is classified into the second group if the element value is smaller than the index value indicating the variation in the attribute value of the element candidate excluding the element candidate with the smallest attribute value from.
A grouping processing program that causes a computer to execute the processing.

(Supplementary Note 11) A plurality of elements are arranged in order from the element with the largest attribute value of each of the plurality of elements and classified into a plurality of groups, and the difference between the attribute values of the respective elements is calculated
It is determined whether to move the boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference.
The elements are classified into the respective groups based on the result of the determination.
A grouping processing program that causes a computer to execute the processing.

(Supplementary Note 12) The determination process uses a difference between an attribute value of the element of the first group in contact with the boundary and an attribute value of the element of the second group in contact with the boundary as a reference difference. Comparing the difference of the reference with the difference before and after the difference of the reference, and when the previous difference or the later difference is larger than the difference of the reference, the difference of the reference is greater than the difference of the reference It is determined to move the boundary to the larger previous difference or the later difference side,
The grouping processing program according to appendix 11, characterized in that

(Supplementary Note 13) In the determination process, the elements of the first group in contact with the boundary are classified into the second group when the previous difference and the subsequent difference are equal. To make a decision to move the boundary,
The grouping processing program according to appendix 12, characterized in that

(Supplementary Note 14) The calculating process calculates an index value indicating variation in the difference for each of the plurality of groups before and after movement of the boundary,
The determination processing performs the determination to move the boundary when the index value indicating the variation of the difference after the movement of the boundary is smaller than the index value indicating the variation of the difference before the movement of the boundary. ,
The grouping processing program according to appendix 11, characterized in that

(Supplementary note 15) The grouping processing program according to supplementary note 14, wherein the index value indicating the variation of the difference is a standard deviation of the difference for each of the plurality of groups.

(Supplementary Note 16) The classification processing further compares the difference with the standard deviation for each of the plurality of groups, and when the difference is smaller than the standard deviation, the difference to be compared is compared. Classify the determined elements into the same sub-group,
The grouping processing program according to appendix 15, characterized in that

(Supplementary Note 17) When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, variation in the attribute values of elements classified into at least one of the plurality of groups A control unit configured to control an index value indicating an index value to be smaller than an index value indicating an attribute value variation of an element other than the element having the attribute value of maximum and / or minimum among the elements classified into the group;
An information processing apparatus comprising:

(Supplementary Note 18) When classifying the plurality of elements into a plurality of groups based on the attribute value of each of a plurality of elements, an index value indicating a variation in the attribute value of the element candidate included in the first group is the element candidate A classification unit that classifies the element candidate with the smallest attribute value into the second group if the index value is smaller than the index value indicating the variation in the attribute value of the element candidate excluding the element candidate with the smallest attribute value from
An information processing apparatus comprising:

(Supplementary note 19) A calculation unit that arranges a plurality of elements in order from the element with the largest attribute value of each of the plurality of elements and classifies them into a plurality of groups, and calculates differences between the attribute values of the respective elements;
A determination unit that determines whether to move a boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference;
A classification unit that classifies each of the elements into the respective groups based on the result of the determination;
An information processing apparatus comprising:

(Supplementary Note 20) The determination unit sets a difference between an attribute value of the element of the first group in contact with the boundary and an attribute value of the element of the second group in contact with the boundary as a reference difference. The difference between the reference and the difference before and after the difference between the reference is compared, and if the difference before or after the difference is larger than the difference between the reference, the difference between the reference and the difference It is determined to move the boundary to the larger previous difference or the later difference side,
The information processing apparatus according to appendix 19, characterized in that

(Supplementary Note 21) The determination unit may classify the elements of the first group in contact with the boundary into the second group when the previous difference and the subsequent difference are equal. , Determine to move the boundary,
20. The information processing apparatus according to appendix 20, characterized in that

(Supplementary note 22) The calculation unit calculates an index value indicating variation in the difference for each of the plurality of groups before and after movement of the boundary,
The determination unit determines to move the boundary when the index value indicating the variation of the difference after the movement of the boundary is smaller than the index value indicating the variation of the difference before the movement of the boundary.
The information processing apparatus according to appendix 19, characterized in that

(Supplementary Note 23) The information processing apparatus according to Supplementary note 22, wherein the index value indicating the variation of the difference is a standard deviation of the difference for each of the plurality of groups.

(Supplementary Note 24) The classification unit further compares the difference with the standard deviation for each of the plurality of groups, and determines the difference to be compared when the difference is smaller than the standard deviation. Group each element into the same sub-group,
24. An information processing apparatus according to appendix 23, characterized in that

100 information processing apparatus 111 input unit 112 display unit 113 operation unit 120 storage unit 121 primary result table 122 secondary result table 123 secondary result table after replacement 124 standard deviation table 125 tertiary result table 130 control unit 131 calculation unit 132 determination Part 133 Classification Department

Claims (14)

When classifying the plurality of elements into a plurality of groups based on the attribute value of each of a plurality of elements, an index value indicating a variation of the attribute value of an element classified into at least one group among the plurality of groups Is controlled to be smaller than the index value indicating the variation of the attribute value for the elements other than the element having the attribute value of the maximum and / or the minimum among the elements classified into the group,
And a computer executes the processing. When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, an index value indicating a variation in attribute values for the element candidates included in the first group is an attribute value from the element candidate The element candidate with the smallest attribute value is classified into the second group if it is smaller than the index value indicating the dispersion of the attribute value of the element candidate excluding the smallest element candidate.
And a computer executes the processing. The plurality of elements are arranged in order from the element having the largest attribute value of each of the plurality of elements and classified into a plurality of groups, and the difference between the attribute values of the respective elements is calculated,
It is determined whether to move the boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference.
The elements are classified into the respective groups based on the result of the determination.
And a computer executes the processing. In the determination process, the difference between the attribute value of the element of the first group in contact with the boundary and the attribute value of the element of the second group in contact with the boundary is used as a reference difference. The difference is compared with the previous difference and the subsequent difference of the reference difference, and the previous difference or the subsequent difference is greater than the reference difference, and the previous difference is greater than the reference difference. Making a decision to move the boundary to the difference or the later difference side,
The grouping processing method according to claim 3, characterized in that: In the determination processing, when the previous difference and the subsequent difference are equal to each other, the element of the first group in contact with the boundary is classified into the second group. Make a decision to move the
The grouping processing method according to claim 4, characterized in that: The process of calculating calculates an index value indicating variation in the difference for each of the plurality of groups before and after movement of the boundary.
The determination processing performs the determination to move the boundary when the index value indicating the variation of the difference after the movement of the boundary is smaller than the index value indicating the variation of the difference before the movement of the boundary. ,
The grouping processing method according to claim 3, characterized in that:   The grouping processing method according to claim 6, wherein the index value indicating the variation of the difference is a standard deviation of the difference for each of the plurality of groups. The classification processing further includes comparing the difference and the standard deviation for each of the plurality of groups, and determining the difference to be compared when the difference is smaller than the standard deviation. Classify elements into the same subgroup,
The grouping processing method according to claim 7, characterized in that: When classifying the plurality of elements into a plurality of groups based on the attribute value of each of a plurality of elements, an index value indicating a variation of the attribute value of an element classified into at least one group among the plurality of groups Is controlled to be smaller than the index value indicating the variation of the attribute value for the elements other than the element having the attribute value of the maximum and / or the minimum among the elements classified into the group,
A grouping processing program that causes a computer to execute the processing. When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, an index value indicating a variation in attribute values for the element candidates included in the first group is an attribute value from the element candidate The element candidate with the smallest attribute value is classified into the second group if it is smaller than the index value indicating the dispersion of the attribute value of the element candidate excluding the smallest element candidate.
A grouping processing program that causes a computer to execute the processing. The plurality of elements are arranged in order from the element having the largest attribute value of each of the plurality of elements and classified into a plurality of groups, and the difference between the attribute values of the respective elements is calculated,
It is determined whether to move the boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference.
The elements are classified into the respective groups based on the result of the determination.
A grouping processing program that causes a computer to execute the processing. When classifying the plurality of elements into a plurality of groups based on the attribute value of each of a plurality of elements, an index value indicating a variation of the attribute value of an element classified into at least one group among the plurality of groups A control unit configured to control to be smaller than an index value indicating a variation in attribute value of an element other than the element having the attribute value of maximum and / or minimum among the elements classified into the group,
An information processing apparatus comprising: When classifying the plurality of elements into a plurality of groups based on the attribute values of each of a plurality of elements, an index value indicating a variation in attribute values for the element candidates included in the first group is an attribute value from the element candidate A classification unit that classifies the element candidate with the smallest attribute value into a second group if the index value is smaller than the index value indicating the dispersion of the attribute value of the element candidate excluding the smallest element candidate;
An information processing apparatus comprising: A calculation unit which arranges a plurality of elements in order from the element having the largest attribute value of each of the plurality of elements and classifies them into a plurality of groups, and calculates differences between the attribute values of the respective elements;
A determination unit that determines whether to move a boundary between a first group and a second group adjacent to the first group among the plurality of groups based on the difference;
A classification unit that classifies each of the elements into the respective groups based on the result of the determination;
An information processing apparatus comprising:

Images