JPH05303597A - Record clasififying system of information processor - Google Patents

Abstract

PURPOSE:To easily specify a group that record of decision analytic processing belongs to by means of the information processor which classifies records by using the decision analytic processing. CONSTITUTION:When the decision analytic processing is performed, a group that each record being input data belongs to needs to be clear. For the designation of the group that the input record belongs to, a classification scale field prescribing the group is selected (201). Then a condition which is applied to the data in the selected classification scale field to determine which group the data belongs to is incorporated (202). Lastly, groups that respective records belong to are determined according to the incorporated condition (203 and 204). Then proceeding to the decision analytic processing is made right after the group are determined (207-210).

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a record classification system of an information processing apparatus, and is particularly suitable for application to a system utilizing discriminant analysis processing.

[0002]

2. Description of the Related Art As an information processing device, there is an information processing device that processes information in so-called record units, which is a set of a plurality of types of data (so-called fields). In this case, a plurality of records may be classified from a predetermined viewpoint, and records may be processed for each classification or for any of the classification records. For example, records that consist of fields such as branch name, branch location, sales floor area, sales, profit, and number of employees are classified based on sales, and there is a difference between groups classified as profits per employee. It is also performed to detect whether or not it is present.

Here, when a new record is input, it is determined which group it belongs to. In this case, if there is data in the classification scale field that defines the group to be classified, classification may be performed based on that value. However, if the data in the classification scale field is unknown, the group belonging to other fields may be used. Is determined (classified). A discriminant analysis process, which is a method of multivariate analysis, is often used to determine such belonging groups. The discriminant analysis process is also used to judge the validity of the selection of the classification scale, the validity of the classification conditions, and the like.

The concept of discriminant analysis processing (strictly speaking, educational discriminant analysis processing) will be briefly described below with reference to FIG. FIG. 2 shows an example in which the belonging groups G1 and G2 are discriminated from the two field data x1 and x2.

From the records R11, R12, ... And R21, R22, ... Which the belonging group G1 or G2 is known in advance, the discriminant function D which is a plane defining the boundary between the groups is determined.
As is apparent from FIG. 2, the discriminant function D is represented by a straight line in the two-dimensional coordinate system, and each field data x1
And x2 linear combination, which is the projection plane of the coordinates of each record that maximizes the difference between the groups and minimizes the variation within the group.

The records R11, R12, ... And R21,
The determination of the belonging group G1 or G2 of R22, ... Is based on the data (x3) of the classification scale field different from the above-mentioned field data x1 and x2.

It is assumed that a record Rx whose data value x3 in the classification scale field is unknown is input. At this time, the position in the coordinate system is obtained from the field data x1 and x2 of the record Rx, and based on the position DRx obtained by projecting this position on the surface defined by the discriminant function D, the belonging group (G1) of this record Rx is determined. decide.

Therefore, a conventional series of processes for determining the belonging group of records whose belongings are unknown is shown in FIG. First, the discriminant analysis process is started, and then, for example, some records are selected from the records already stored in the auxiliary storage device, and the designation of the group to which the selected record belongs is taken in (steps 101 to 103). .. After that, various calculations are executed to determine the discriminant function and the like, and the user is caused to display and recognize the obtained processing result of the discriminant function and the like (step 1).
04, 105). When such a process is completed, a record whose affiliation is unknown is taken in, the affiliation group is determined, the result is displayed in, for example, a table format, and the series of processes is terminated (10
6, 107).

Thus, in order to obtain the discriminant function,
It is necessary to select and specify multiple sets of records whose affiliation is already known. As a concrete method, a separate data file for each group is formed based on an existing data file, or group identification information is added to a record.

[0010]

However, in the past, when specifying a group to which a record to be input for calculating a discriminant function or the like is assigned, the user classifies the record regardless of which concrete method is adopted. The data value of the scale field was judged for each record, and the belonging group was designated one by one. Therefore, it takes a lot of time and labor.

In order to improve the classification accuracy of records, a large number of input records whose affiliations are designated are required, so that the above-described problem of performing the designation operation for each record is great.

The present invention has been made in consideration of the above points, and the information processing that can easily specify the belonging group for the input record of the discriminant analysis process and can shorten the series of processing time. It is intended to provide a record classification system for a device.

[0013]

In order to solve such a problem, in the present invention, a record whose belonging group is unknown is classified into one of the groups, and each record in which the belonging group is designated belongs to that group. In the record classification method of the information processing device that executes the discriminant analysis process for calculating the probability of belonging, the affiliation group of each record that is the input data of the discriminant analysis process is specified by the following process.

That is, a selection process for selecting a classification scale field that defines a belonging group, and a condition fetching process for incorporating a condition for applying to the data of the selected classification scale field to determine which group it belongs to, The determination process is to determine the group to which each record belongs based on the captured conditions.

[0015]

In the information processing apparatus, a discriminant analysis is performed in a process of classifying a record whose belonging group is unknown into one of the groups or calculating a probability that each record in which the belonging group is designated belongs to the belonging group. Processing is used. When performing the discriminant analysis process, it is necessary that the belonging group is clear for each record that is the input data.

The present invention relates to designation of a group to which such an input record belongs. In the present invention, first, a classification scale field that defines the belonging group is selected, and then a condition that is applied to the data of the selected classification scale field to determine which group it belongs to is taken in. Finally, The group to which each record belongs is determined based on the embedded conditions. Then, after the affiliation group is determined, the process immediately proceeds to the discriminant analysis process.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. Like the general computer system, the information processing apparatus according to this embodiment is also configured by a CPU, a memory, a keyboard, a CRT display and the like in terms of hardware, although not shown. The information processing apparatus of this embodiment is configured as shown in FIG. 4 from the function of record classification. The functional configuration shown in FIG. 4 is actually configured by software.

As shown in FIG. 4, the functional configuration (program structure concept) is divided into a control module 11, an input processing module 12, an output processing module 13, a table management processing module 14, an emphasis processing module 15, and a statistical processing module 16. Be done.

The control module 11 controls the overall control, and the various modules 12 to 16 are under the control of the control module 11 and execute predetermined processing.

The input processing module 12 takes in records and commands input from an input device such as a keyboard. For example, a record having data of a classification scale field that determines a classification or a record of which the data of the classification scale field is unknown and the belonging group is unknown is taken. It also accepts the designation of the classification scale field and incorporates the conditions that determine the group of input records.

The output processing module 13 controls output to a CRT display, a printer and the like. For example, an image used for designating the affiliation group of an input record is output, the obtained discriminant function is output, or affiliation group information obtained for a record whose affiliation is unknown is output.

The table management processing module 14 manages table data including a plurality of records stored in the auxiliary recording device, the main memory or the like. In this embodiment, the tabular data managed by this module 14 may be provided with emphasis information indicating that a part of a plurality of records is emphasized over other records.

The emphasis processing module 15 emphasizes a part of the records in the table data managed by the table management processing module 14 that satisfy a predetermined condition, in distinction from other records. That is, the group to which the input record belongs, which is necessary for determining the discriminant function and the like, is designated by dividing into the emphasized record group and the unemphasized record group. The emphasis processing module 15
More specifically, the condition definition unit 15a, the determination unit 15b, and the emphasis processing unit 15c are included. The condition definition unit 15a activates the input processing module 12 and takes in the designation of the classification scale field and the condition for determining whether or not the record is an emphasis record based on the data value of the classification scale field. The determination unit 15b determines whether or not each record should be emphasized according to the condition. The emphasis processing unit 15c
Emphasized information is added to the table under the control of the table management processing module 14 based on the determination result.

The statistical processing module 16 calculates a discriminant function or the like and determines the affiliation destination of an unknown affiliation record. The statistical processing module 16 includes a discriminant analysis unit 16a and a probability calculation unit 16b. The discriminant analysis unit 16a calculates information such as a discriminant function using both the emphasized record group and the unemphasized record group.
The probability calculation unit 16b determines the group to which the record belongs by using the discriminant function calculated by the discriminant analysis unit 16a and other information when a record whose affiliation is unknown is given. Is represented by the probability of belonging to one belonging group. The probability calculator 16b also calculates the belonging probability of the input record for which the belonging group is designated.

Next, a series of record classification processing for determining the belonging group of records whose belongings are unknown (classifying the records) will be described with reference to FIGS. 1 and 5 to 7. 1. FIG. 1 is a flowchart showing the flow of processing. 5 and 6 are explanatory views showing table data which are input records used for determining a discriminant function, etc., FIG. 5 shows before emphasis processing, and FIG. 6 shows after emphasis processing. .. FIG. 7 is an explanatory diagram showing the determination result of the affiliation group of the affiliation unknown record. The example shown in FIG. 5 to FIG.
One record is formed by four fields A, B, C and D.

First, the classification scale field designated by the user is fetched (step 201). In this case, a plurality of fields may be designated as the classification scale fields. Next, the condition about the classification scale field for distributing each input record into two groups is taken in (step 202). When a plurality of fields are specified as the classification scale field, the condition for each field and the total condition for the result are taken in. And
A determination is made based on the fetched conditions, and information indicating whether the record is an emphasized record or a non-emphasized record is added to the record (step 203).

FIG. 6 shows an emphasis result when the field A is designated as the classification scale field and "the value of the field A>600" is specified as the condition of the emphasis record.

Thereafter, the records are automatically sorted into two groups based on the emphasis result without waiting for the designation by the user (step 204). In the case of the example of FIG. 6, it is divided into a record group having record numbers 1, 2, 4, 5 and 6 and a record group having record numbers 3, 7, 8, 9 and 10.

In this state, the discriminant analysis process is activated, and firstly, each sorted record group is automatically designated as an input record group to be used for discriminant analysis (step 2).
05, 206). After that, a discriminant analysis process is executed to determine a discriminant function and the like, and the user is caused to display and recognize the processing result of the discriminant function obtained (steps 207, 20).
8). When the preprocessing for such classification is completed, the records whose affiliation is unknown are taken in, the affiliation group is determined based on the discriminant function, etc. (for example, the affiliation probability is calculated), and the result is displayed in, for example, a table format. Display and end the series of processing (20
9, 210).

FIG. 7 shows that there are two records of which the value of the classification scale field A is unknown and whose membership is unknown (record numbers 11 and 1).
2) shows the result when given, and in the case of this example, these records are both the group on the emphasis side (field A).
Probability of belonging to a group whose value of is greater than 600) (emphasis probability)
Is greater than the probability of belonging to the non-emphasized group.

Note that FIG. 7 also shows the probabilities for the input records for which the affiliation group is designated. For example, although the record with the record number 6 is designated as belonging to the non-emphasized group, the probability of belonging to the emphasized group is considerably high from the result of the discriminant analysis. Such results are also used for selecting the classification scale field and analyzing the validity of the classification conditions.

Therefore, according to the above-mentioned embodiment, in order to execute the discriminant analysis process of classifying records whose affiliation is unknown and calculating the probability that the record whose affiliation is designated belongs to the affiliation group, Since the process of allocating the input record of 2 to the group of 2 or more is automatically performed by the designation of the classification scale field and the condition setting, the processing time and the labor of the user can be significantly reduced. In other words, there is no need to perform input operation after sorting records into multiple groups, it is sufficient to input without being aware of the group, and then specify the classification scale field and set the conditions, which improves the usability of the user. To do. Also, already
Even when using the record stored in the storage device,
The affiliation group can be easily specified.

In practice, in an information processing device such as a database device or a spreadsheet device, a large number of records are processed,
Since it is necessary to classify the records as needed and to judge the appropriateness of the classification, which is often used as in the above-mentioned embodiment, the above-described effect is great.

Incidentally, some conventional information processing devices such as database devices and spreadsheet devices have a process of emphasizing some records so that they are different from other records. It is a process that is provided separately from the analysis process, so it cannot be used directly in the discriminant analysis process.
It is necessary for the user to carry out the emphasizing process and then process it so that it can be used in the discriminant analysis process, which is time-consuming and time-consuming as compared with the embodiment.

In the above-mentioned embodiment, the case of discriminating into two groups is shown, but the invention can be applied to the case of discriminating (determining) into three or more groups. Further, although one record is shown to be composed of four field data, it is not limited to this. Furthermore, the present invention can be applied to the discriminant analysis process in which the classification process is not executed for records whose belonging group is unknown.
That is, the record classification referred to in the claims includes calculating the belonging probability for the record in which the belonging group is designated.

[0036]

As described above, according to the present invention, the specification of a plurality of groups of input records, the specification of the classification scale field, and the conditions for the field data thereof, which are necessary in the discriminant analysis process for classifying records, are performed. Since it is automatically performed according to the determination of, the operation by the user can be significantly reduced, and the processing time including the designation of the input record can be significantly shortened as compared with the conventional case.

[Brief description of drawings]

FIG. 1 is a processing flowchart of an embodiment method.

FIG. 2 is a schematic explanatory diagram of a discriminant analysis process.

FIG. 3 is a processing flowchart of a conventional method.

FIG. 4 is a functional block diagram of an embodiment system.

FIG. 5 is an explanatory diagram of table data before emphasis processing.

FIG. 6 is an explanatory diagram of table data after emphasis processing.

FIG. 7 is an explanatory diagram of a final processing result.

[Explanation of symbols]

11 ... Control module, 12 ... Input processing module, 1
3 ... Output processing module, 14 ... Table management processing module, 15 ... Emphasis processing module, 16 ... Statistical processing module.

Claims (1)

[Claims] 1. An information processing apparatus for executing a discriminant analysis process of classifying a record whose belonging group is unknown into one of the groups, and calculating a probability that each record in which the belonging group is designated belongs to the belonging group. In the record classification method of, the selection process that selects the classification scale field that defines the belonging group, and the condition capture process that captures the condition that determines which group it belongs to by applying to the data of the selected classification scale field. The determination process for determining the affiliation group of each record based on the fetched conditions specifies the affiliation group of each record that is the input data of the discriminant analysis process, and then shifts to the discriminant analysis process. Record classification method of information processing device.