TWI453703B - Method and system for assessment of learning - Google Patents

Description

Learning diagnosis analysis method and learning diagnosis analysis system

The present invention relates to a learning diagnosis analysis method and system, and in particular to a method for diagnosing a learner's learning effect by comparing a learner's test result and a reading record with a multi-knowledge point, and recommending a learning suitable reading material. Diagnostic analysis methods and systems.

For learners and teachers, the effectiveness of learning must be learned by fair and accurate test methods. For example, the traditional practice is to let the learners answer the papers by recording the test papers of the test questions, and correct the test papers after the test. Get the score of the test, and the score is generally treated directly as the learner's learning effect.

The traditional test paper records the test questions on paper and answers them on paper. This type of test makes the review of test papers more complicated. In addition, today's environmental awareness is on the rise, and the papers that use paper in large quantities are not in line with the current environmental protection trend. The computer-reading answer card can improve the above problem. It selects the field on the answer card of the computer by pen circle, and judges whether the learner's answer is correct by the computer for the circled field. Since the computer-reading answer card is only needed to judge the learner's answer, the test paper can be reused, and because the computer-reading answer card is small, the paper waste can be effectively suppressed. However, the way the computer reads the answer card is limited to the multiple-choice questions or non-titles that can be circled, but it cannot be applied to other questions. On the other hand, even if the computer-reading answer card has a small paper size, paper is still needed to answer the question.

In recent years, computer technology has developed rapidly, and many areas have become paperless or electronic. In terms of teaching, there is also an electronic way to measure the learning outcomes of learners to replace traditional paper papers. In the prior art, the electronic test paper allows the learner to see the test content on the computer and answer it on the computer. The electronic test papers can be combined by the test questions in the electronic question bank. Similarly, the answers to the questions are also stored in the electronic question bank. Therefore, the processes such as generating electronic test papers, conducting tests, and correcting test papers can all be performed by computer. Automatically correct and avoid wasting paper.

Although the learning outcome can be roughly seen from the test scores, it is not possible to know in detail from the test scores which aspects of the learner's knowledge is weak. For example, in a test paper, if the A concept and the B concept are each Half of the proportion, but the learner takes 50 points in the A concept test, and the B concept takes 20 points. From the total score of 70, it is impossible to see that the learner needs to strengthen the B concept. The test questions in the above electronic question bank can be marked with a knowledge point according to the knowledge categories to be tested in each test question. The assessor can know which aspects of the learner are weak by using the correct rate of each knowledge point answer, and can recommend the learner. Relevant textbooks or books to strengthen.

However, a test question usually does not use only one main idea (knowledge point). For example, a test point is a trigonometric function test. The main idea is a trigonometric function, but other concepts such as algebra are still needed to solve the problem. Familiar with the trigonometric function but not familiar with algebra and answering this question, resulting in the erroneous evaluation of the learner as a trigonometric function is weak, and recommend a less appropriate textbook for learners.

In addition to the above-mentioned inappropriate textbooks due to analysis errors, even if the analysis of the test points is more accurate, it is still possible to recommend inappropriate textbooks because of the learner's reading status. For example, by analyzing the electronic test paper, the learner's textbooks are recommended to be weak in both the trigonometric function and the algebra. However, the learner may have read a sufficient number and breadth of textbooks in the trigonometric function, resulting in recommendations. The same textbook resulted in not receiving good analysis and recommendation functions.

Therefore, one aspect of the present invention is to provide a learning diagnostic analysis method that can effectively solve the problems of the prior art.

According to a specific embodiment, the learning diagnostic analysis method of the present invention comprises the following steps: first, each test question in the electronic test paper is marked with a plurality of knowledge points; then, the learner is tested by the electronic test paper, and at the end of the test After analyzing the test results, the correct percentage of answers to each knowledge point is obtained; according to the learner's reading record and the above knowledge points, the knowledge point reading record corresponding to each knowledge point is obtained; and, according to the correct percentage of the answer and the knowledge point reading Record, recommend reading materials to learners.

In the method of the specific embodiment, the learning content of the reading teaching material can be labeled with a plurality of knowledge points as well as the test questions, and the reading materials required by the learner can be recommended. By sorting the correct percentage of answers and reading points of knowledge points, it is possible to more effectively find out which part of the learner needs to enhance reading, while avoiding repeatedly recommending the same textbook.

Another aspect of the present invention is to provide a learning diagnostic analysis system that can effectively identify possible learning problems of learners and recommend appropriate reading materials to solve the problems of the prior art.

According to another specific embodiment, the learning diagnostic analysis system includes a first knowledge point annotator, a test analyzer, a read record analyzer, and a comparison processor. The first knowledge point annotator can be used to mark a plurality of knowledge points for each test question in the electronic test paper, and the electronic test paper marked with the knowledge point allows the learner to perform the test and obtain a test result. The test analyzer can be used to analyze the learner's test results and obtain the correct percentage of questions for each knowledge point. The reading record analyzer is connected to the test analyzer, which can analyze the learner's reading record according to the knowledge points to obtain a knowledge point reading record for each knowledge point. The comparison processor is connected to the test analyzer and the reading record analyzer for receiving the correct percentage of the answer and the knowledge point reading record, and according to the recommended reading material for the learner.

In this embodiment, the learning diagnostic analysis system may further include a second knowledge point identifier for marking knowledge points in the learning content in the reading textbook, so the comparison processor may recommend reading the teaching material. Therefore, the learning diagnosis analysis system can more accurately find out the part that the learner needs to strengthen reading according to the correct percentage of the correct answer and the reading point of the knowledge point, and avoid repeatedly recommending the same teaching material.

The advantages and spirit of the present invention will be further understood from the following detailed description of the invention.

Referring to FIG. 1, FIG. 1 is a schematic diagram of a learning diagnostic analysis system 1 according to an embodiment of the present invention. As shown in FIG. 1, the learning diagnostic analysis system 1 includes a first knowledge point identifier 10, a test analyzer 12, a read record analyzer 14 and a comparison processor 16, wherein the first knowledge point identifier 10 can be connected to a The electronic question bank 20 is marked with a plurality of knowledge points for the electronic test questions stored therein. In practice, these electronic test questions with multiple knowledge points can be taken out and combined into electronic test papers for the learner to test. In another embodiment, each test question in the pre-formed electronic test paper may also be labeled by the first knowledge point identifier 10 for multiple knowledge points.

In this embodiment, the labeled electronic test paper allows the learner to perform a test and determine the test result after the test. In practice, the test result, that is, the correctness of each test question and the total score of the test, can be evaluated by the assessor or evaluated by the processor included in the learning diagnostic analysis system 1, and the present invention does not limit. The test analyzer 12 can receive the test results described above, that is, whether the test questions are correct, and then perform analysis according to each knowledge point marked by each test question, and obtain the correct percentage of answers for each knowledge point.

Generally, the learner can perform a test of multiple electronic test papers, wherein each test question of each electronic test paper is marked by a multi-knowledge point by the first knowledge point identifier 10, and then all the electrons are analyzed by the test analyzer 12. The test results of the test paper to obtain the correct percentage of answers to each knowledge point. From the correct percentage of each knowledge point, you can learn in detail the learner's learning level of various knowledge categories. For example, learners are familiar with trigonometric functions and are not familiar with algebra, although they are labeled with trigonometric functions and algebras at the same time. The test of the knowledge point has a lower accuracy rate, but the test question that does not contain the algebraic knowledge point but contains the trigonometric knowledge point has a higher accuracy rate, so respectively, for the trigonometric function and the algebraic knowledge point, The calculated correct percentage of the answer is not the same (the correct percentage of the answer to the trigonometric knowledge point should be higher than the correct percentage of the answer to the algebraic knowledge point).

The reading record analyzer 14 can accept each knowledge point of the test question, and analyze the learner's reading record according to each knowledge point to obtain a knowledge point reading record of each knowledge point, and the knowledge point reading record of each knowledge point includes There are detailed information on the total number of reading materials and teaching materials that learners have read related to each knowledge point. In practice, the reading record can be automatically recorded by a data processing device or the learning diagnostic analysis system 1 or manually recorded by the learner, which is not limited by the present invention.

The comparison processor 16 is connected to the test analyzer 12 and the reading record analyzer 14 to receive the correct percentage of the answers of the knowledge points and the knowledge point reading record, and then, according to the received correct points of each knowledge point, the correct percentage and knowledge points Reading records, you can recommend reading materials to learners, so that learners can strengthen the knowledge points that are less familiar.

Referring to FIG. 2, FIG. 2 is a flow chart showing the steps of the method for learning diagnostic analysis according to another embodiment of the present invention. As shown in FIG. 2, the learning diagnosis analysis method of the specific embodiment includes the following steps: in step S30, each test question in the electronic test paper is respectively marked with a plurality of knowledge points; in step S32, the learner performs the electronic test paper. After the test, the test result is analyzed to obtain the correct percentage of the answer corresponding to each knowledge point; in step S34, according to the marked knowledge points and the learner's reading record, the knowledge point reading record corresponding to each knowledge point is obtained; In step S36, the reading material is recommended to the learner according to the correct percentage of the answer to each knowledge point and the knowledge point reading record.

The learning diagnosis analysis method of the specific embodiment can be performed by the learning diagnosis analysis system 1 of the previous specific embodiment. In step S30, the first knowledge point identifier 10 of the learning diagnosis analysis system 1 can be used for the test questions in the electronic test paper. Label knowledge points. Step S32 can analyze the test result by the test analyzer 12 and obtain the correct percentage of the answer corresponding to each knowledge point. Step S34 can analyze the learner's reading record by reading the record analyzer 14 to obtain a knowledge point reading record corresponding to each knowledge point. Step S36 can then use the comparison processor 16 to recommend the reading material to the learner according to the correct percentage of the answer and the knowledge point reading record.

In the above specific embodiments, the learning content in various reading materials may also be marked with a plurality of knowledge points. Referring to FIG. 1 again, the learning diagnostic analysis system 1 of the specific embodiment further includes a second knowledge point identifier 18 connected to a reading teaching material database 22, and can mark the learning content of each reading teaching material stored therein. Knowledge point. Please note that the knowledge points of the learning content and the test questions are one of the system points, and the problem of the knowledge points of different names is not generated for the learning content and the test questions of the same knowledge category.

Referring to FIG. 3, FIG. 3 is a flow chart showing steps of a method for learning diagnostic analysis according to another embodiment of the present invention. As shown in FIG. 3, step S400 and step S402 respectively perform unified knowledge point labeling on the test questions and the learning content in the reading textbook. Similarly, step S402 can also mark the learning content in the reading textbook through the second knowledge point identifier 18 of the previous embodiment. Please note that the other steps of the learning diagnostic analysis method of the present embodiment are substantially the same as the corresponding steps of the learning diagnostic analysis method of FIG. 2, and thus will not be described again.

In the above specific embodiment, the learning content in the reading textbook may be marked with at least one knowledge point. For example, in a chapter on trigonometric functions in a reading textbook, you can mark the knowledge points of the trigonometric function, but if the chapter paragraphs also mention algebra or apply to the equation to help explain, you can add algebra and equations. Knowledge point. Reading the knowledge points marked in the teaching materials can be used as a basis for obtaining the reading records of the knowledge points and recommending reading materials. For example, the reading record analyzer 14 finds the reading materials read according to the reading records of the learners, and reads according to each reading. The knowledge points marked in the textbook are classified, and the degree to which the learner reads or learns the knowledge points, that is, the knowledge point reading record, is known. In addition, the comparison processor 16 may also recommend reading materials marked with these knowledge points based on knowledge points that the learner is less familiar with.

In each of the above specific embodiments, steps S36 and S46 may recommend reading the teaching material according to the correct percentage of the answer and the knowledge point reading record by the comparison processor 16. In detail, the comparison processor 16 may perform the test result and the reading record according to the learner. Compare the learner's familiarity with each knowledge point, and then generate a priority recommendation criterion and recommend reading materials that meet the learner's needs according to the criteria.

Please refer to FIG. 4A and FIG. 4B together. FIG. 4A is a detailed schematic diagram of the comparison processor 16 of the learning diagnostic analysis system 1 of FIG. 1 , and FIG. 4B is a schematic diagram of the learning diagnosis analysis method of FIG. 2 . Step flow chart for reading the textbook. Please note that the steps of the recommended reading material of FIG. 4B can be performed according to the ratio of FIG. 4A to the processor 16, so the following steps and devices are described together.

As shown in FIG. 4A and FIG. 4B, the comparison processor 16 further includes a sorting unit 160, an extracting unit 162, and a recommending unit 164, wherein the extracting unit 162 is connected to the sorting unit 160, and the recommending unit 164 is connected to the extracting unit 162. In step S360, the sorting unit 160 sorts each knowledge point according to the correct percentage of the answer of each knowledge point received; in step S362, the borrowing unit 162 extracts the knowledge point reading record according to the sorted knowledge points; In step S364, the recommendation unit 164 reads the record based on the sorted knowledge points and the extracted knowledge points, generates a priority recommendation criterion, and recommends reading the teaching material to the learner accordingly.

In detail, in step S360, the sorting unit 160 outputs a priority order to each knowledge point according to the correct percentage of the answer. In step S362, the extracting unit 162 extracts the learner's knowledge points for each knowledge point according to the priority order of each knowledge point. Record. Since the correct percentage of the answer directly reflects the learner's familiarity with this knowledge point, the knowledge points are sorted according to the correct percentage of the answer from the lowest to the highest, and the knowledge points in the prior position can be judged as the knowledge points that the learner is less familiar with. Priority reading recommended reading materials for learners to strengthen learning. In addition, the factors that learners are not familiar with this knowledge point may be that they have less reading materials at this knowledge point, so they can read the recommended reading materials for the knowledge points that contain less reading materials. As described above, the priority recommendation criteria generated by the recommendation unit 164 in step S364 is preferably in the order of the correct percentage of the answer from low to high, and the number of reading materials in the knowledge point reading record is as small as possible.

The correct percentage of answers is more positive than the number of reading materials in the reading record of the knowledge point. The learner's learning effect at each knowledge point. In other words, the learner may have read a lot of reading materials for a knowledge point, but the test may not be able to There is a higher percentage of correct answers, and learners still need to reinforce this knowledge. Therefore, in the above specific embodiment, the sorting unit 160 first sorts the knowledge points according to the correct percentage of the answer. However, if the correct percentage of the answers to the two knowledge points is similar, but there is a large gap between the reading materials that are read in the reading points of the two knowledge points, the priority recommendation criteria can be adjusted. For example, the learners' correct percentages for trigonometric functions and algebras are 61% and 62%, respectively, while learners have read 10 books on trigonometric functions but only 2 books on algebra. It is recommended that reading materials on algebra may have a stronger effect on the overall learning of learners, so it is recommended to first read reading materials or books about algebra.

In another specific embodiment, the second knowledge point identifier 18 can mark the recommended reading time when the knowledge content of the learning content of each reading textbook is marked, so that the comparison processor 16 can display the teaching material simultaneously when recommending reading the teaching material. Suggested reading time for learners to refer to.

The knowledge point reading records of the above specific embodiments include the reading materials and the number of reading materials that the learner has read at this knowledge point. Therefore, when comparing the processor 16 for recommendation, the original reading may be recorded in the knowledge. Read the reading materials in the record, and avoid repeating the recommendation of the same reading materials to the learners.

In summary, the learning and diagnosis analysis method and system of the present invention can first mark the unified multi-knowledge points of the test questions in the electronic question bank and the reading materials in the reading material database. After the learner conducts the test with the electronic test paper composed of the test questions with too many knowledge points, the learning diagnostic analysis method and the system can analyze the test results to obtain the correct percentage of the answers of each knowledge point. In addition, the learning diagnostic analysis method and system can analyze the learner's reading record, and then learn the learner's detailed reading records of each knowledge point, such as the textbooks read and the number thereof. According to the correct percentage of the answers of the above knowledge points and the comparison of the reading records of the knowledge points, the effect of the learner's learning can be more effectively analyzed, and the reading materials that meet the learner's needs are recommended and the same reading materials are avoided.

The features and spirit of the present invention will be more apparent from the detailed description of the preferred embodiments. On the contrary, the intention is to cover various modifications and equivalents within the scope of the invention as claimed. Therefore, the scope of the patented scope of the invention should be construed as broadly construed in the

1. . . Learning diagnostic analysis system

10. . . First knowledge point identifier

12. . . Test analyzer

14. . . Reading record analyzer

16. . . Align processor

18. . . Second knowledge point identifier

20. . . Electronic question bank

twenty two. . . Reading textbook database

160. . . Sorting unit

162. . . Extraction unit

164. . . Recommended unit

S30～S36, S360～S364, S400～S402, S42～S46. . . Process step

1 is a schematic diagram of a learning diagnostic analysis system in accordance with an embodiment of the present invention.

2 is a flow chart showing the steps of a method for learning diagnostic analysis according to another embodiment of the present invention.

FIG. 3 is a flow chart showing the steps of a method for learning diagnostic analysis according to another embodiment of the present invention.

FIG. 4A is a detailed schematic diagram of the comparison processor of the learning diagnostic analysis system 1 of FIG. 1.

FIG. 4B is a flow chart showing the steps of the recommended reading method of the learning diagnosis analysis method of FIG. 2 .

1. . . Learning diagnostic analysis system

10. . . First knowledge point identifier

12. . . Test analyzer

14. . . Reading record analyzer

16. . . Align processor

18. . . Second knowledge point identifier

20. . . Electronic question bank

twenty two. . . Reading textbook database

Claims (10)

A learning diagnostic analysis method for analyzing a learner's learning effect comprises the following steps: marking each test question in at least one electronic test paper with a plurality of knowledge points; the learner performs the test with the at least one electronic test paper, And analyzing the test result after the test is finished to obtain a correct percentage of the answer to each knowledge point; according to the reading record of the learner and the knowledge points, obtaining a knowledge point reading record corresponding to each knowledge point; According to the correct percentage of the questions and the reading records of the knowledge points, at least one reading material is recommended to the learner. The method for learning and diagnosing analysis according to item 1 of the patent application scope further includes the following steps: sorting the knowledge points according to the correct percentage of the questions; and extracting the knowledge points according to the sorted knowledge points Recording; and generating a priority recommendation criterion based on the sorted knowledge points and the extracted knowledge point reading records, and recommending the at least one reading material to the learner according to the priority recommendation criteria. The method for learning and diagnosing analysis as described in claim 2, wherein the priority recommendation criterion is that the correct percentage of the answers to the knowledge points is from low to high, and the number of reading materials in the reading records of the knowledge points is as small as at most . The learning diagnostic analysis method according to claim 1, further comprising the step of: marking at least one of the knowledge points of the at least one reading material. The method for learning diagnostic analysis according to item 4 of the patent application scope further includes the step of: marking a recommended reading time for the learning content. A learning diagnostic analysis system for analyzing a learner's learning effect, comprising: a first knowledge point identifier, configured to mark a plurality of knowledge points for each test question in at least one electronic test paper; a test analyzer, And a test result obtained by the learner performing the test by using the at least one electronic test paper to obtain a correct percentage of each of the knowledge points; a reading record analyzer for receiving the knowledge points and according to the knowledge points Analyzing one of the learners' reading records to obtain a knowledge point reading record for each knowledge point; and a comparison processor connecting the test analyzer and the reading record analyzer to receive the correct percentage of the questions and The knowledge points read the records and recommend at least one reading material to the learner based on the correct percentage of the questions and the reading records of the knowledge points. The learning diagnostic analysis system of claim 6, wherein the comparison processor further comprises: a sorting unit for sorting the knowledge points according to the correct percentage of the answers; an extracting unit connecting the a sorting unit for respectively extracting the knowledge point reading records according to the sorted knowledge points; a recommendation unit connecting the extracting units for the sorting of the knowledge points and the extracted knowledge points The reading record generates a priority recommendation criterion, and the at least one reading material is recommended to the learner based on the priority recommendation criterion. The learning diagnostic analysis system according to claim 7, wherein the priority recommendation criterion is that the correct percentage of the answers to the knowledge points is from low to high, and the number of reading materials in the reading records of the knowledge points is as small as possible. . The learning diagnostic analysis system of claim 6, further comprising: a second knowledge point identifier, configured to mark at least one of the knowledge points of the at least one reading material. The learning diagnostic analysis system of claim 9, wherein the second knowledge point identifier further marks the learning content with a suggested reading time.