KR20050095667A - Method and apparatus for providing grade information - Google Patents

Abstract

The present invention relates to a method and apparatus for providing grade information, and more particularly, to a method and apparatus for providing grade information evaluated based on a nationwide distribution regardless of the difficulty of the test and the number of candidates. In accordance with an aspect of the present invention, a method of providing grade information includes: (a) identifying a group corresponding to a candidate and classifying the candidate by a group preset according to the group; (b) extracting test taker score information, calculating average and standard deviation of the score information for each cluster, and calculating gaze standardization scores for each cluster corresponding to the score information using the average and standard deviation of each score; (c) calculating the number of stares for each stage by classifying the standardization scores by clustering according to the preset stages for each cluster, and then calculating the expected number of stages for each cluster; (d) generating a virtual standardization score and inserting the virtual standardization score in each cluster stage according to the estimated number of persons in each cluster stage; (e) generating a reduction score corresponding to the score information using the group-specific mean and standard deviation of the standardized score including the gaze standardized score and the virtual standardized score; And (f) arranging the reduction scores in order of magnitude, and generating ranking information corresponding to the score information.

Description

Method and apparatus for providing grade information {Method and apparatus for providing grade information}

The present invention relates to a method and apparatus for providing grade information, and more particularly, to a method and apparatus for providing grade information evaluated based on a national distribution regardless of the difficulty and the number of test takers.

In addition to the importance of college entrance exams, the importance of mock exams is increasing, and based on the results of these mock exams, it is used as basic information for selecting one's own position and the school to enter.

However, according to the prior art, it is difficult for candidates of the national scale to take a single exam at a large number of mock exams conducted by educational institutions, schools, educational organizations, etc. It is difficult to provide.

In other words, the reliability of the mock test results depends on the identity of the test taker and the test taker who wants to take the mock exam, but it is practically impossible to make all mock test takers the same as the test taker who wants to take the test.

In addition, due to the difference in the level and number of test takers, there is a fundamental limitation in evaluating the level of the college entrance exam using the mock test results due to the difference in the difficulty of the mock test conducted every time or every year.

The present invention has been made to solve the above-described problems, and an object thereof is to provide a method and apparatus for providing grade information evaluated based on a national distribution regardless of the difficulty and the number of test takers. In particular, it is an object of the present invention to provide reliable national predicted seating and percentile calculation information regardless of the number of test takers and the difficulty of the test.

In addition, another object of the present invention to provide a nationwide distribution characteristic criterion, to provide reliable grade information by using a predetermined number of years of SAT information and the average grade for each school, school.

Further, another object of the present invention is to provide a cluster formation method and apparatus having the best characteristics for each cluster in providing grade information by generating clusters for each school.

According to an aspect of the present invention for achieving the above object can be provided a method that can provide grade information evaluated based on the national distribution regardless of the difficulty and number of test takers.

According to a preferred embodiment, the grade information providing method comprises the steps of: (a) identifying a group corresponding to the candidate and classifying the candidate by a predetermined group corresponding to the group; (b) extracting the test taker score information, calculating the average and standard deviation of the score information for each cluster, and then calculating the standardization score for each group corresponding to the score information using the average and standard deviation for each cluster. Making; (c) classifying the grouping standardization scores according to the preset stages for each cluster, calculating the number of stares for each cluster, and calculating the expected number of clusters for each cluster; (d) generating a virtual standardization score and inserting the virtual standardization score in each cluster by a number corresponding to a difference between the number of gazing persons for each cluster and the estimated number of persons for each cluster; (e) generating a reduction score corresponding to the score information using the group-specific mean and the standard deviation of the standardized score including the gaze standardized score and the virtual standardized score; And (f) arranging the reduction scores in the order of size, and generating order information corresponding to the score information and displaying the result on the display device so that the user can recognize the test scores. Full ranking information can be provided.

Here, the step (a) may include classifying the cluster into a predetermined number by performing cluster analysis by using grade information corresponding to a preset period for each group; And determining a cluster corresponding to the group. In addition, the method of providing grade information is performed by performing the cluster analysis using an average score corresponding to the group, and in order to verify the validity of the cluster analysis, the correlation between the grade information and a test gazed by the existing total rankings is examined. The method may further include analyzing.

And if there is no cluster information corresponding to the group in step (a), calculating an average score of the group and determining a cluster corresponding to an average score similar to the average score as a cluster of the group. It may further include. And in the step (b), the gaze standardization score is expressed as Can be calculated by

In addition, the step (c) comprises the step of calculating the number of people for each group by dividing the estimated number of people for each group by the number of gaze; Extracting the expected number of people per cluster; And calculating the estimated number of persons in each group by multiplying the number of persons in each group by the estimated number of persons in each group. And (d) generating the virtual normalized score using a random number function; Classifying the virtual standardized scores by clusters; The method may include inserting a virtual standardization score corresponding to the category for each cluster in the clustered stage region.

In addition, the reduction score in the step (e) is the following equation Can be calculated by

In the grade information providing method, when the reduction score corresponds to a preset deletion range, deleting the reduction score may precede step (e), and the preset deletion range may include the reduction score. Exceeding perfect score; The reduction score is less than zero; The reduction score is higher than the highest scorer; It may include at least one of.

According to another aspect of the present invention for achieving the above object can be provided an apparatus capable of providing the grade information evaluated based on the national distribution regardless of the difficulty and number of test takers.

According to a preferred embodiment, an apparatus for providing grade information includes a cluster designation unit for identifying a group corresponding to the candidate and classifying the candidate by a preset cluster corresponding to the group; Standardized score for extracting the test taker score information, calculating the average and standard deviation of each of the score information, and then using the average and standard deviation for each group to calculate the standardized score for each group corresponding to the score information A calculator; An estimated number calculation unit for classifying the grouping standardized scores, calculating the number of stares for each group of clusters, and calculating an estimated number of grouped clusters for each cluster in accordance with a preset step for each cluster; A virtual standardization score calculator for generating a virtual standardization score and inserting the virtual standardization score in each cluster stage by a number corresponding to a difference between the estimated number of persons in each cluster and the estimated number of persons in each cluster; A reduction score generator for generating a reduction score corresponding to the score information by using a standardized score including the gaze standardized score and a virtual standardized score using the group-specific mean and standard deviation; And after sorting the reduction scores in the order of size, including a full seating information generating unit for generating a seating information corresponding to the score information, it is possible to provide the full seating information including the examinee who did not take the test.

Here, the grade information providing apparatus performs a cluster analysis by using grade information corresponding to a preset period for each group, classifies the cluster into a predetermined number, and then averages the cluster analysis with the average score corresponding to the group. Performed using, but may further include a cluster analysis unit for analyzing the correlation between the grade information and the test gaze the existing total order to verify the validity of the cluster analysis.

The grade information providing apparatus calculates an average score of the group when there is no cluster information corresponding to the group, and determines a cluster corresponding to an average score similar to the average score as a cluster of the group. It may further include a new group designator.

In addition, the expected number of people calculation unit may calculate the number of people for each group by step, and after extracting the expected number of people by group, multiply the number of people by each group by the estimated number of people by each group to calculate the estimated number of people by each group.

The virtual standardization score calculator generates the virtual standardization score by using a random number function, classifies the virtual standardization score by cluster stage, and then calculates the virtual standardization score corresponding to the category by stage for each cluster. Can be inserted into

Hereinafter, the configuration of the sexual information providing method and apparatus according to the present invention will be described in detail with reference to the accompanying drawings, the same or corresponding components regardless of reference numerals in the description with reference to the accompanying drawings the same reference numerals. And duplicate description thereof will be omitted.

Configuration and Overall Flowchart

The present invention utilizes the cumulative data of the simulated test to estimate the national estimated distribution, and is a method and apparatus for providing the expected score for the SAT using the simulated test scores. According to the present invention, since the difficulty of the SAT and the examinee change every year, after the standardization of scores by preset groups, the average scores of the groups are used. Hereinafter, the configuration of the sexual information providing apparatus will be described in FIG. 1, and the configuration of the sexual information providing apparatus will be described in FIG. 2.

1 is a diagram showing the configuration of a sexual information providing apparatus according to a preferred embodiment of the present invention.

The sexual information providing apparatus 300 according to the present invention is coupled to the terminal 100 through the network 200, and includes a cluster information generating unit 310 and a seating information providing unit 350.

The cluster information generation unit 310 may include a cluster analyzer 315, a cluster designator 320, and a new group designator 325.

The cluster analysis unit 315 classifies the cluster into a predetermined number by performing cluster analysis by using the grade information corresponding to the preset period for each group, and then calculates the cluster score by an average score corresponding to the group. In order to verify the validity of the cluster analysis, the correlation between the grade information and the test taken by the existing full scale can be analyzed. According to the embodiment of the present invention, as a result of analyzing the average score for each group, the group including all educational institutions in the country is classified into five clusters, and using the characteristics of each cluster for the estimation of the national expected distribution, reliable results Information can be provided.

The cluster designation unit 320 may identify a group corresponding to the test taker who has taken the test, and classify the test taker by a preset cluster corresponding to the group. If there is no cluster information corresponding to the group, the new group designation unit 325 may calculate an average score of the group, and determine a cluster corresponding to an average score similar to the average score as the cluster of the group. have.

The seat size information providing unit 350 includes a standardized score calculator 355, an estimated number of persons calculator 360, a virtual standardized score calculator 365, a reduction score generator 370, and an entire seat information generator 375. It may include.

The standardized score calculator 355 extracts score information of the test taker, calculates a mean and standard deviation for each cluster of the score information, and then for each cluster corresponding to the score information using the mean and standard deviation for each cluster. The standardization score for the test can be calculated. Here, the staring standardized score refers to the standardized score corresponding to the test taker who actually tested.

The estimated number of people calculation unit 360 classifies the staring standardization scores according to the preset stages for each of the clusters, calculates the number of stares for each stage of the cluster, and then calculates the expected number of stages for each cluster. According to a preferred embodiment, after calculating the number of people per group, extract the estimated number of people by cluster, multiply the number of people by each group by the estimated number of people by each cluster to calculate the estimated number of people by each cluster.

The virtual standardization score calculator 365 may generate a virtual standardization score, and insert the virtual standardization score in each cluster by the number corresponding to the difference between the number of gazing persons for each cluster and the estimated number of persons for each cluster. The virtual standardization score calculator 365 generates the virtual standardization score by using a random number function, classifies the virtual standardization score by stage for each cluster, and then calculates the virtual standardization score corresponding to the category for each cluster by stage. Can be inserted in the step-by-step area.

The reduction score generator 370 may generate a reduction score corresponding to the score information by using the standardized score including the gaze standardization score and the virtual standardization score as the group average and standard deviation. The reduction score generation unit may be configured to delete the reduction score when the reduction score corresponds to a preset deletion range.

The full order information generator 375 may generate the order information corresponding to the score information after sorting the reduction scores in order of size.

2 is a flow chart showing the overall operating principle of the sexual information providing procedure according to a preferred embodiment of the present invention.

The grade information providing procedure according to the present invention can be largely divided into a cluster information generating step for grasping a nationwide distribution and a step of generating total seating information using the generated cluster information.

Hereinafter, referring to FIG. 2, cluster information is generated by analyzing test information (for example, simulation test information) for a predetermined period (for example, three years) in the cluster information generation step of step S210 ( 3).

In the cluster identification step of step S220, a group corresponding to the candidate may be identified, and the candidate may be classified according to a preset cluster corresponding to the group (see FIG. 4).

In the clustering standardization step of step S230, the score information of the test taker is extracted, the average and the standard deviation of the score information are calculated for each cluster, and the clusters are corresponding to the score information using the cluster average and standard deviation. A standardization score of gaze is calculated (see FIG. 5).

In the step of calculating the expected number of people in step S240, corresponding to the pre-set step for each of the clusters, after classifying the gaze standardization score for each of the clusters, after calculating the number of stares for each of the clusters, the estimated number of people for each cluster is calculated ( 6).

In operation S250, the virtual standardization score may be generated, and the virtual standardization score may be inserted in consideration of the estimated number of persons in each cluster (see FIG. 7). According to a preferred embodiment, the virtual standardization score may be inserted in each cluster by the number corresponding to the difference between the number of gaze per stage for each group and the expected number of stages for each cluster.

In the step of generating a reduction score in step S260, a standardization score including the gaze standardization score and a virtual standardization score is used to generate a reduction score corresponding to the score information using the average and standard deviation for each group (see FIG. 8).

After arranging the reduction scores in the order of magnitude in the step S270 generation of the step S270, generating the seating order information corresponding to the score information, the terminal 100 or the grade information providing apparatus 300 so as to be recognized by the user. It can be displayed on the combined display device.

Specific operating procedure

As described above, the procedure for providing ranking information according to the present invention may include a cluster information generation process, a cluster identification process, a standardization process for each cluster, a process of calculating an expected number of persons, a virtual standardization score generation and insertion process, and a reduction score generation process. have.

Hereinafter, a cluster information generation process using cumulative data will be described with reference to the accompanying drawings, and a cluster identification process will be described with reference to FIG. 4. In FIG. 5, the standardization process for each cluster will be described, and in FIG. 6, the process for calculating the expected number of people will be described. In FIG. 7, a process of generating and inserting a virtual normalized score will be described, and a process of generating a reduction score will be described in detail with reference to FIG. 8.

3 is a flowchart illustrating a cluster information generation process according to a preferred embodiment of the present invention.

The sexual information providing apparatus 300 according to the present invention may determine the format of the analysis data according to a preset format, and construct analysis data (eg, sexual information) for extracting cluster information according to the format. . According to the embodiment, the analytical data according to the present invention is configured to generate based on a group of the examinees divided into preset groups, not the grade data corresponding to each examinee. Since the difficulty and examinees of the College Scholastic Ability Test differ every year, the distribution may be configured to estimate the national distribution by generating basic data based on educational institutions such as schools or institutes that maintain the same in order to accumulate a plurality of examination information. .

Hereinafter, the cluster information generation procedure according to the present invention will be described in detail with reference to FIG. 3.

First, in step S310, the grade information including the simulation test data, etc. is analyzed, scores are calculated according to preset criteria such as series, area, total score, and standardized. The standardization according to the present invention may use a variety of standardization methods, which will be developed conventionally or in the future, and will be described based on standardization using a Z normal distribution for convenience of description below.

As described above, the basic data for estimating the national distribution is not based on individual data but on average scores of groups based on educational institutions such as schools or institutes. According to an embodiment of the present invention, in order to estimate the national distribution, the test results for the pre-set year (for example, three years, 2001-2003) of all educational institutions in the country is extracted, and the extracted results After standardizing by the predetermined method, calculate the average score for each institution. In the analysis data according to the present invention, new simulation test data is generated, and when data is accumulated for a certain period of time, classification analysis may be newly performed to determine the change in cluster for each educational institution.

In step S320, an average score is calculated based on a school, a school, or other educational institution (hereinafter, referred to as a group). That is, the average score of the standardized score can be obtained based on the group. Referring to Table 1 below, the average score table for each group according to an embodiment of the present invention is shown.

Group code Language average Mathematical average Foreign language average Social search average Common Social Average … Overtaking Choice 2 Average Overtaking Choice 3 Average Overtaking Choice 4 Average Raw score total score average 11101 11102 11103 :

In step S330, cluster analysis is performed on the group. According to a preferred embodiment of the present invention, cluster analysis using K-means can be performed. Unlike the hierarchical cluster analysis, the K-means clustering method is a mutually exclusive clustering method in which an individual belongs to only one cluster. According to the present invention, the number of clusters is designated as five, and the variable may designate all Z-score averages for each institution of all subjects, domains, and total score combinations.

In step S340, clusters are designated for each group according to the cluster analysis. According to the embodiment of the present invention, when the analysis is completed, cluster codes of 1 to 5 for each group are assigned. The group is classified into clusters classified into preset numbers based on the average score. According to the embodiment of the present invention, when the average score for each school is analyzed, it can be seen that the characteristics of each cluster are most clearly shown when all schools in the country are classified into five clusters. In this classification, clusters can clearly differentiate characteristics of clusters, such as clusters that show excellent grades in all subjects and clusters that show relatively poor grades in all subjects. That is, cluster analysis may be performed by giving k = 5 in k mean cluster analysis.

After classifying the clusters and assigning the clusters to the respective groups, a procedure for verifying the suitability of the simulated test data for estimating the national distribution in step S350 is performed. According to the present invention, it is possible to prove the validity of estimating the national distribution using the simulated test data by verifying whether the simulated test results reflect the actual national distribution characteristics using the reference data.

The verification method according to the present invention may be performed using the correlation method according to Equation 1 below. For example, the results of analyzing the relationship between the average scores of 2002 SAT and the average scores of the simulated exams for three years 2001-2003 are as follows.

: Average scores of SAT in 2002

: Average scores of mock exams by educational institution from 2001 to 2003

The correction coefficient of the explanatory power of the model of Equation 1 is 0.95, and the three-year simulation test results have a correlation of about 95% with the 2002 Mathematics test. Here, students who took the mock test in 2001 would have taken the actual SAT test in 2002, The coefficient of can be set to a high slope, such as 0.68. Where Of course, the coefficient of can be changed annually to improve the correlation with reference to the simulated test data.

Table 2 below is a table showing the correlation between the SAT and the mock exam based on the school according to the preferred embodiment of the present invention.

M ₀₁ M ₀₂ M ₀₃ S ₀₂ M ₀₁ 1.00000 0.85888 0.85733 0.96027 M ₀₂ - 1.00000 0.90199 0.92732 M ₀₃ - - 1.00000 0.91132 S ₀₂ - - - 1.00000

Referring to Table 2 above, it can be seen that there is a very close positive correlation between the mock test scores and the actual SAT scores when judging based on the average grades of the schools. Therefore, the mock test data can be used to estimate the actual national distribution. It can be seen that.

In step S360, it is configured to calculate the estimated number of people per cluster. The preliminary candidate distribution information of the students is generated according to the cluster. Table 3 below is a table for calculating the 2005 preliminary candidate distribution estimation for the national distribution estimation according to a preferred embodiment of the present invention.

association School (current student) School (restart) Sum Number percent Number percent One 183,260 (38.0%) 20,322 (12.0%) 203,582 2 60,549 (12.6%) 2,031 (1.2%) 63,580 3 108,505 (22.5%) 106,691 (63.0%) 215,196 4 99,286 (20.6%) 5,081 (3.0%) 104,367 5 30,400 (6.3%) 35,225 (20.8%) 65,625 Sum 482,000 (74.1%) 169,350 (25.9%) 651,350

4 is a flowchart illustrating a cluster identification process according to a preferred embodiment of the present invention.

Hereinafter, referring to FIG. 4, a cluster identification process will be described. First, in step S410, a group is identified and a cluster corresponding to the group is determined. If there is no cluster corresponding to the group, that is, if the group is a new educational institution, the process proceeds to step S420.

In step S420, in the case of a new educational institution that does not exist in the cumulative data for a predetermined period, the average of each cluster (three-year cumulative data analysis result) is compared with the average of the new educational institution, and configured to give a cluster corresponding to the average. Can be. For example, if the average score for each cluster is calculated with the results of the current month's mock test (for example, March 2004), if there is a new educational institution that does not have a cluster code, the average score for each existing institution's cluster The appropriate cluster can be set based on the criteria.

In step S430, a cluster is determined according to a preset cluster code, and a cluster code is assigned to an educational institution on the simulation test data of the current month. For example, if a new educational institution enters the practice test for the current month, the average score of the new educational institution can be calculated, and a cluster code having the average score most similar to the average score of the new educational institution can be assigned.

In step S440, it is configured to update the cluster information including the new group.

5 is a flowchart illustrating a clustering standardization process according to a preferred embodiment of the present invention.

In operation S510, the average and the standard deviation are calculated based on a preset reference. Here, the preset reference may include a series, an area, and the like. If the predetermined criterion is a criterion for each series, in extracting the cluster information, not only the average information according to the cluster, but also the national estimated distribution for each series may be separately calculated for the standardization of the test results. In this case, the standardization by series is limited to the case in which the national expected distribution is separately generated for each series, and in order to generate a nationwide distribution that does not require series classification, the standardization may be performed without separating the series.

In step S520, the gaze standardization score is calculated using the calculated average and standard deviation. According to an embodiment, when normalizing using a Z normal distribution, the gaze normalization score may be calculated by Equation 2 below.

Here, the reason for calculating the standardized score is to reduce the raw score after using the standardized score. Referring to Table 4 below, a standardized score table calculated for each cluster on a region-by-region basis is shown.

Number of students Average Standard Deviation One 111398 -0.02782 0.89019 2 24053 -0.87957 0.97775 3 81256 0.38307 0.86893 4 49982 -0.45782 0.87782 5 14435 1.10922 0.87830

6 is a flowchart illustrating a process of calculating an expected number of people according to a preferred embodiment of the present invention.

In order to grasp the expected number of people by group and step, first, in step S610, each cluster may be subdivided into a plurality of stages using an average score. That is, by using the gaze standardized score for each cluster, the distribution status of each stage may be classified into preset criteria after being divided into preset stages. The detailed grouping step may be set to be subdivided (eg, 12 steps, 24 steps, 60 steps, etc.) according to the grade information. For example, in the case of subdividing into 60 stages, in order to grasp the distribution status of each stage by cluster based on the actual candidates of the current month, it may be divided into 60 stages by using the standardized Z score for each cluster.

In step S620, based on the actual number of the current month's stare, the number of people in each group and stage is determined, and then the number of people in each group is calculated. Here, the percentage of the number of people per stage (%) refers to the ratio of actual test takers of the current month's practice test.

In step S630, the estimated number of people information is extracted to calculate the expected number of people according to the actual number of people in each group. Here, the expected number of people can be extracted for each cluster.

Here, the estimated number of persons may be extracted for each series including the humanities, natural systems, and arts and physical systems, as shown in Table 5 below, wherein the extraction information is extracted from a preset ratio information table, and the ratio information for each series is estimated from the estimated number table. Can be configured to extract.

association Humanities nature Physical fitness Humanities + Arts One 53.6 35.8 10.6 64.2 2 62 20 18 80 3 50 42.3 7.7 57.7 4 54.9 31.3 13.8 68.7 5 53.5 43.3 3.2 56.7

In operation S630, the estimated number of persons per group may be multiplied by the ratio of the number of persons in each group, and the estimated number of persons in each group may be calculated.

That is, the total estimated number of persons can be extracted and the estimated number of persons can be calculated according to the preset invention. Referring to Table 6 below, it is shown in the group-specific expected personnel discharge table according to the preferred embodiment of the present invention.

One 2 3 … … . 60 Sum One real 8 469 1748 45 34,304 % 0.02 1.37 5.10 0.13 100 (%) Estimated Person 40 201,155 2 real 3 346 741 7 3,778 % 0.08 9.16 19.61 0.19 100 (%) Estimated Person 66,382 3 real One 161 600 80 25,045 % 0.00 0.64 2.40 0.32 100 (%) Estimated Person 207,820 4 real 7 395 1276 4 10,680 % 0.07 3.70 11.95 0.04 100 (%) Estimated Person 105,320 5 real 11 27 90 3,060 % 0.36 0.88 2.94 100 (%) Estimated Person 61,906

7 is a flowchart illustrating a process of calculating a virtual standardization score generation and insertion process according to a preferred embodiment of the present invention.

The present invention is configured to generate a virtual standardized score corresponding to the expected number of people for each stage and cluster by using a random number function according to a preset algorithm.

Hereinafter, a process of generating and inserting a virtual normalization score will be described with reference to FIG. 7. In operation S710, a virtual standardization score is generated using a random number function for each cluster. According to a preferred embodiment, it may be configured to calculate as many virtual standardized scores as possible from the normal distribution, using the cluster-specific mean and standard deviation as variables.

In step S720, the virtual standardization scores are classified and arranged in stages corresponding to the classification criteria for each cluster. That is, when divided into 60 stages, the generated virtual standardization scores may be selected for each cluster stage. Here, it may be configured to include cluster information and step information when placing the virtual standardization score.

For example, if one million virtual standardized scores are generated according to the mean and standard deviation of one cluster standardized score, they may be arranged step by step. Referring to Table 7, below, a virtual standardized score placement table in accordance with a preferred embodiment of the present invention is shown.

Virtual standardization score association step 2.44268 One 2 2.20268 One 2 2.32162 One 2 2.03689 One 2 2.11968 One 2 2.39267 One 2 2.13940 One 2 2.11367 One 2 2.34537 One 2 2.22762 One 2

In operation S720, the virtual standardization score may be inserted in each cluster by the number corresponding to the difference between the estimated number of persons in each cluster and the estimated number of persons in each cluster.

8 is a diagram illustrating a reduction score generation and ranking information generation process according to a preferred embodiment of the present invention.

The present invention may be configured to reduce the standardized score including the gaze standardized score and the virtual standardized score in the form of raw scores, and to calculate the estimated national ranking using the reduced score.

In operation S810, a reduction score is generated by Equation 3 below. Here, the reduced score is generated corresponding to the gaze standardization score and the virtual standardization score.

When the standardized scores are calculated for each series, the standardized scores, standard deviations, and averages calculated for each series can be used.

In operation S820, when the reduction score deviates from a preset criterion, the reduction score or the virtual normalization score corresponding to the reduction score is configured to be removed. Here, the preset criterion may include a case in which the standardized score is reduced to the original raw score, a perfect score, a negative score, or a larger score than the highest scorer.

In this manner, after deleting the standardized score that is out of the standard, a ranking information table for generating an expected ranking table in step S830 may be generated, and ranking information may be generated. That is, the reduced scores may be sorted in order of magnitude, and then a ranking may be given and percentile information may be calculated. Referring to Table 8 below, a predicted ranking table according to a preferred embodiment of the present invention is shown.

Standard score association line Reduction score Seat 2.48251 5 One 476.298 976 2.48249 5 One 476.296 977 2.81536 3 2 476.294 978 2.48244 5 One 476.292 979 2.48240 5 One 476.289 980 2.48233 5 One 476.282 981 2.48218 5 One 476.270 982 2.48214 5 One 476.266 983 2.48206 5 One 476.259 984 2.48188 One One 476.244 985 2.48173 One One 476.231 986

User interface screen

Hereinafter, a detailed method of providing grade information and ranking information calculated according to the above-described method will be described with reference to the user interface screen shown in the accompanying drawings. 9 illustrates a comprehensive analysis screen, FIG. 10 illustrates a score providing screen, and FIG. 11 illustrates a percentile graph providing screen according to a region combination.

9 is a diagram illustrating a comprehensive analysis screen according to a preferred embodiment of the present invention.

The comprehensive analysis screen may include an area grade field 910, an inquiry area grade field 920, a total score field 930, and a grade comparison graph 940.

The grade-specific grade field 910 may provide grades for the regions actually selected by students according to the SAT system through which individual students may select and stare at the zone. According to a preferred embodiment, in the case of the mathematics area, the type (A, B) that the student took is displayed, and in the case of the inquiry area, the grades are presented together with the name of the elective course selected by the student.

The inquiry area grade field 920 may provide adjusted scores for nationwide comparisons because the number of elective courses and subjects differs for each inquiry area. The adjusted score is calculated based on the grades of the subjects actually taken by the student. The adjusted score is calculated based on the scores of the subjects. According to an embodiment of the present invention, the method for calculating the adjusted score of the raw score is obtained by obtaining a standard score of a subject taken by the student, and the student does not select the raw score corresponding to the average of the standard scores in the subject with the largest number of candidates. Not the fourth grade. The method of calculating the adjusted score of the standard score is to calculate the average of the percentiles of the subjects (eg, three subjects) taken by the student, and then the standard score corresponding to the percentile of the subjects with the highest number of subjects among the subjects who did not take the subjects. It is configured to be the score of the fourth course that is not selected.

The total score field 930 analyzes the SAT utilization area of each university in the college admission process, and provides information that categorizes the most useful area combinations. Alternatively, it may be provided differently as a criterion for selecting a scientific research area.

The grade comparison graph (940) shows the percentiles of last month along with the percentiles for the current month's mock test, and provides information on how much the individual's grades have improved / decreased compared to the previous mock test by area / subject or total score type. can do. In addition, the percentiles of each area / subject and total score are shown as broken line graphs to identify relatively weak areas and subjects at a glance. The graphs are provided based on the national percentile, and in the case of inquiry areas, elective subjects can be presented together.

10 is a diagram illustrating a score providing screen according to a preferred embodiment of the present invention.

The score providing screen includes the raw score field (1010), the standard score field (1020), the standard score field (1030), the previous year's SAT index field (1040), the percentile field (1050), the national predicted grading field (1060), and the grading field ( 1060).

The raw score field 1010 may provide scoring and scoring of raw scores for each area / subject and total score type, and the standard scoring field 1020 may provide scoring and scoring of standard scores for each region / subject and total scoring type. According to a preferred embodiment, the language, numeracy, and foreign language areas are given as a total score of 200 out of 100 points, and a total score of 100 out of 100 for each elective course of inquiry area, and the total score is based on the sum of the standard scores of the area / subject. It may be provided as.

The standard score field 1030 uses a score converted by the university's own score utilization method, rather than using the standard score presented in the SAT score report at each university. Therefore, standard scores can be provided that were recalculated using the score utilization method (100 points per area) adopted by many universities in 2005.

The previous year's SAT index field 1040 may provide a score obtained by converting the individual total scores of the current month's mock exams based on the score distribution of last year's college scholastic ability test. It can be used as a reference for comparing the score of the current test and the results of last year's entrance examination. The previous year's CSAT can provide the raw score and the conversion standard score.

Percentile field 1050 may provide percentile information on the number of candidates staring at the central practice test and the expected number of national standards for the current month. The national predicted grading field 1060 is provided with a grading estimated nationally based on the national percentile, and the ranking table 1060 is provided with a class and campus seats and a national estimated grading.

11 is a diagram illustrating a percentile graph providing screen according to an area combination according to an exemplary embodiment of the present invention.

The present invention may include advantageous area combination fields 1110 and other fields 1120.

The advantageous region combination field 1110 may show a bar graph of the percentiles by total score type based on the national expected percentile to suggest what type of total score is most favorable to the student. The percentile, favorable ranking, and SAT index fields 1120 show the national expected percentiles and the previous year's SAT index for each total score type, and can be presented in order of favorable total score types based on the national expected percentiles.

As described above, the method and apparatus for providing grade information according to the present invention has an effect of providing grade information evaluated based on a national distribution regardless of the difficulty and the number of test takers.

In addition, the present invention has an effect that can provide reliable national predicted car difference and percentile calculation information regardless of the number of test takers and the difficulty of the test.

In addition, the present invention also provides a national distribution characteristic criteria, it is also possible to provide highly reliable grade information by using a predetermined number of years of SAT information and the average grade for each school, school.

In addition, the present invention has the effect of providing a cluster forming method and apparatus having the best characteristics of each cluster in providing grade information by generating clusters for each school.

Although described above with reference to a preferred embodiment of the present invention, those skilled in the art that various modifications of the present invention without departing from the spirit and scope of the invention described in the claims below And can be changed.

1 is a diagram showing the configuration of a sexual information providing apparatus according to a preferred embodiment of the present invention.

2 is a flow chart showing the overall operating principle of the sexual information providing procedure according to a preferred embodiment of the present invention.

3 is a flow chart illustrating a cluster information generation process according to a preferred embodiment of the present invention.

Figure 4 is a flow chart illustrating a cluster identification process according to a preferred embodiment of the present invention.

Figure 5 is a flow chart showing a cluster-specific standardization process according to a preferred embodiment of the present invention.

6 is a flowchart illustrating a process of calculating an expected number of people according to an exemplary embodiment of the present invention.

7 is a flowchart illustrating a process of calculating a virtual standardization score generation and insertion process according to a preferred embodiment of the present invention.

8 is a diagram illustrating a process of generating reduction scores and ranking information according to a preferred embodiment of the present invention.

9 is a diagram showing a comprehensive analysis screen according to a preferred embodiment of the present invention.

10 is a diagram showing a score providing screen according to a preferred embodiment of the present invention.

11 is a diagram illustrating a percentile graph providing screen according to a region combination according to a preferred embodiment of the present invention.

<Description of the symbols for the main parts of the drawings>

100: terminal

200: network

300: grade information providing device

315: cluster analysis unit

320: cluster designation

325: new group designator

355: standardized score calculation unit

360: estimated number calculation unit

365: virtual standardized score calculation unit

370: reduction score generator

375: full seat information generation unit

Claims (19)

In the method of providing sexual information, (a) identifying a group corresponding to the candidate, and classifying the candidate by a preset cluster corresponding to the group; (b) extracting the test taker's score information, calculating the average and standard deviation of each of the score information, and calculating the standardized score corresponding to the score information using the average and the standard deviation of each group. ; (c) calculating the number of stares for each stage by grouping the gaze standardization scores according to the preset stages for each of the clusters, and then calculating the expected number of stages for each cluster; generating virtual standardization scores and classifying the virtual standardization scores by clusters according to the estimated number of persons in each cluster; (e) generating a reduction score corresponding to the score information using the group-specific mean and the standard deviation of the standardized score including the gaze standardized score and the virtual standardized score; And (f) arranging the reduction scores in order of size, and generating order information corresponding to the score information and displaying the result on the display device so that the user can recognize the scores; Including, including the test takers who did not take the test, the grade information providing method, characterized in that to provide full information. The method of claim 1, In step (a), Classifying the cluster into a predetermined number by performing cluster analysis by using grade information corresponding to a preset period for each group; Determining a cluster corresponding to the group Sexual information providing method comprising a. The method of claim 1, Performing the cluster analysis using an average score corresponding to the group, Analyzing a correlation between the grade information and a test gazed by the existing overall rankings in order to verify the validity of the cluster analysis; Sexual information providing method characterized in that it further comprises. The method of claim 1, If cluster information corresponding to the group does not exist in the step (a), Calculating an average score of the group and determining a cluster corresponding to an average score similar to the average score as a cluster of the group Sexual information providing method characterized in that it further comprises. The method of claim 1, In the step (b), the gaze normalization score is represented by the following equation Method for providing sexual information, characterized in that calculated by. The method of claim 1, In step (c), Calculating the number of people in each group by dividing the estimated number of people in each group by the number of stares; Extracting the expected number of people per cluster; Calculating an estimated number of persons for each group by multiplying the number of persons for each group by the estimated number of persons for each group; Sexual information providing method comprising a. The method of claim 1, In step (d), Generating the virtual normalization score using a random number function; Classifying the virtual standardized scores by clusters; Inserting a virtual standardization score corresponding to the clustered staged category into the clustered staged region; Sexual information providing method comprising a. The method of claim 1, The reduction score in the step (e) is the following equation Method for providing sexual information, characterized in that calculated by. The method of claim 1, Deleting the reduction score when the reduction score corresponds to a preset deletion range; The sexual information providing method characterized in that the preceding step (e). The method of claim 9, The preset deletion range The reduction score exceeds a perfect score; The reduction score is less than zero; The reduction score is higher than the highest scorer; medium Sexual information providing method comprising at least one. In the device for providing sexual information, A cluster designation unit for identifying a group corresponding to the candidates and classifying the candidates by a preset cluster corresponding to the group; After the score information of the test taker is extracted, the average and standard deviation of the score information are calculated for each cluster, and then the standardization for calculating a standardized gaze score for each cluster corresponding to the score information is calculated using the average and standard deviation for each cluster. A score calculator; An estimated number of people calculation unit for classifying the standardization scores of the gaze, calculating the number of gazing persons for each of the clusters, and calculating an estimated number of people for each of the clusters according to a preset step for each cluster; A virtual standardization score calculator for generating a virtual standardization score and inserting the virtual standardization score in each cluster stage according to the estimated number of persons in each cluster; A reduction score generator for generating a reduction score corresponding to the score information by using a standardized score including the gaze standardized score and a virtual standardized score using the group-specific mean and standard deviation; And An overall order information generation unit for generating order information corresponding to the score information after sorting the reduction scores in order of magnitude; Including, the grade information providing device, characterized in that to provide full seating information including the examinee did not take the test. The method of claim 11, The grade information providing device Using the grade information corresponding to a predetermined period for each group, performing cluster analysis to classify the cluster into a predetermined number, and then performing the cluster analysis using an average score corresponding to the group, but performing the cluster analysis. Cluster analysis unit for analyzing the correlation between the above grade information and the test taken by the existing total order to validate the analysis Sexual information providing device further comprising a. The method of claim 11, The grade information providing device If there is no cluster information corresponding to the group, a new group designation unit for calculating an average score of the group and determining a cluster corresponding to an average score similar to the average score as a cluster of the group. Sexual information providing device further comprising a. The method of claim 11, The standardized score calculating unit is the following equation And calculating the gaze standardization score by using the score information providing device. The method of claim 11, The expected number of people calculation unit After dividing the estimated number of people in each group by the number of stares, calculating the number of people in each group, and extracting the number of people in each group, multiplying the number of people in each group by the number of people in each group to calculate the estimated number of people in each group. Sexual information providing device, characterized in that. The method of claim 11, Virtual standardized score calculation unit After generating the virtual standardization score by using a random number function, classify the virtual standardization score by stage for each cluster, and insert a virtual standardization score corresponding to the category for each cluster by stage into the clustered stage region. Sexual information providing device. The apparatus of claim 11, wherein the grade information providing device is The reduction score in the step (e) is the following equation Sex information providing apparatus, characterized in that calculated by. The method of claim 11, The reduction score generator And when the reduction score corresponds to a preset deletion range, deleting the reduction score. The method of claim 18, The preset deletion range The reduction score exceeds a perfect score; The reduction score is less than zero; The reduction score is higher than the highest scorer; medium Apparatus for providing sexual information comprising at least one.