KR20200115966A - A problem curation system and method reflecting the similarity of the user learning information and the similarity of the problem context in the question bank - Google Patents

Abstract

The present invention relates to a system or the like for providing a question to a learner in a question bank. More particularly, the present invention relates to a system and a method for recommending a question based on a question of a similar learner and the context of the question to a person to learn in view of the context of the question and the question solving learning information of the learner who has solved the question. The question curation system of the present invention includes: an analysis module (200) extracting a question list to be provided for a learner by calculating the context similarity of a learning question and the behavior-based similarity of the learner; and a learning information module (300) storing the correct/wrong answer distribution of the question learned by the learner as data and providing the learning question list to the learner. The analysis module (200) includes: a question context analysis module (210) analyzing a morpheme by question, calculating inter-question similarity based on morpheme inclusion/non-inclusion, and deriving a similar question list information (530) by question; a learner learning similarity analysis module (220) calculating inter-learner similarity based on the inter-learner correct/wrong answer distribution of a question solving information (600) by learner provided from the learning information module (300) and deriving similar learner list information (620) by learner; and a learner-specific question extraction module (230) selecting a similar learner by the similar learner list information (620) by learner for each learner, extracting a question list of the similar learner from the question solving information (600) by learner, and deriving learner-specific question curation information (710).

Description

A problem curation system and method reflecting the similarity of the user learning information and the similarity of the problem context in the question bank}

The present invention relates to a system for providing a problem to a learner in a problem bank, and in detail, in consideration of the problem solving learning information and the context of the problem, the learner's problem similar to the learner, and the It relates to a system and method for recommending problems based on the context of the problem.

The system's method of providing customized problems to learners on the system is a technology that can increase the learning effect by intelligently providing suitable problems to learners, unlike simply learning a fixed list of problems in printed workbooks.

Among the customized problem provision methods of the prior art, there is a method of recommending learning problems of other learners similar to learners by grasping similarity such as progress based on the learner's learning information as a problem recommendation technology that reflects the learner's learning information. In addition, when recommending a problem, not only does not reflect the learner's additional learning information, but also does not reflect the characteristics of the learning problem.

The technical task to be achieved by the present invention is to identify the learner's behavior-based similarity and the context similarity of the learning problem and reflect it in the recommendation of the problem, so that the learner can receive problems in a context similar to the problem to be learned, and It is intended to provide an apparatus and method for providing customized learning that can increase the effectiveness of learning by predicting the learning trend of learners based on similarity and providing problems.

The technical problem to be achieved by the present invention is not limited to the technical problems mentioned above, and other technical problems that are not mentioned can be clearly understood by those of ordinary skill in the technical field to which the present invention belongs from the following description. There will be.

In order to achieve the above technical problem, the apparatus and method of the present invention include a problem bank module for storing problems, a problem context analysis module for calculating the context of a document, and a learner learning similarity analysis that classifies similar learners based on the user's learning information. It includes a module, a learning information module, and a problem extraction module for each learner that finally curates the problem.

The learning problem curation system according to the present invention may be operated by including a system that connects companies, organizations, and individuals providing learning problems, and a server on which the specified modules are mounted.

The problem curation system apparatus and method according to the present invention can provide a tailored problem intensively to a learner, so that the effect of learning can be further improved.

The effects of the present invention are not limited to the above effects, and should be understood to include all effects that can be inferred from the configuration of the invention described in the detailed description or claims of the present invention.

Fig. 1. A schematic system configuration diagram for configuring the system of the present invention.
Fig. 2. An example of the procedure for calculating the similarity for each question in the question stored in the question bank.
Fig. 3. An example of the procedure for calculating similarity for each learner based on learning information for each learner.
Fig. 4. The storage structure of problems recommended by the server for problem curation.
Fig. 5. An example screen for solving a problem recommended by problem curation on a web or mobile device.
Fig. 6. An example of adding additional selectivity in the selection of similar learners.
Fig. 7. An example of adding additional selectivity in the selection of similar learning problems.
Fig. 8. A variation on the selection of similar learners.

Hereinafter, the present invention will be described with reference to the accompanying drawings. However, the present invention may be implemented in various different forms, and therefore is not limited to the embodiments described herein.

Throughout the specification, when a part is said to be "connected (connected, contacted, bonded)" with another part, it is not only "directly connected", but also "indirectly connected" with another member in the middle. "Including the case. In addition, when a part "includes" a certain component, this means that other components may be further provided, not excluding other components, unless specifically stated to the contrary.

The terms used in the present specification are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprise" or "have" are intended to designate the presence of features, numbers, steps, actions, components, parts, or a combination thereof described in the specification, but one or more other features. It is to be understood that the presence or addition of elements or numbers, steps, actions, components, parts, or combinations thereof, does not preclude in advance.

The present invention relates to an apparatus and method for providing a learner learning problem that provides a customized learning problem to a learner, and calculates the similarity between the learner and the learning problem and calculates the behavior-based similarity based on the context similarity of the learning problem and the learner's learning information. Provides selective intensive learning to learners by providing problems.

Hereinafter, preferred embodiments of the present invention will be described with reference to the accompanying drawings so that those of ordinary skill in the art can easily implement them.

First, referring to each drawing, FIG. 1 is a schematic system configuration diagram for configuring the present invention as a system, wherein the question bank module 100 stores data transmitted from the problem provider terminal 10, etc., and the analysis module 200 ) Brings problem data from the problem bank module 100 to calculate the context similarity of the problem, and provides a problem set, which is data extracted by fetching the learner's learning information from the learning information module 300 and calculating the learner's behavior-based similarity. It represents providing to learners by the problem processing module 320. 2 is an exemplary procedure for calculating the similarity for each problem with respect to a problem stored in the problem bank module 100 by the analysis module 200, and the jacquard similarity of the context similarity for each problem based on the morpheme analysis information 500 for each problem It shows a procedure of deriving and listing similar problems for each problem by calculating it using (Jaccard Similarity) or Cosine Similarity method.

3 is an example procedure for calculating the similarity of each learner based on the learning information of each learner, and the behavior-based similarity between learners is determined based on the correct and incorrect answer distribution data stored for each problem learned by the learner, Jaccard Similarity or Cosine Similarity. Similarity) method, etc., to derive and list similar learners for each learner. 4 is a diagram illustrating a problem curation and a storage structure of a problem in a server, and shows that a problem is provided to a learner by extracting a learning problem based on data of a specific learner having a high behavior-based similarity at the request of a learner. 5 is an exemplary view showing a screen for solving a problem recommended by problem curation on a web or mobile device, where the learner selects a problem that is likely to be correct, a problem that is likely to be wrong, etc. It shows a structure that is effectively provided.

FIG. 6 is an example of adding additional selectivity in the selection of similar learners, and shows the results obtained by measuring the correct and incorrect rate of each learner and dividing similar learners into upward similar learners and non-going similar learners based on this. 7 is an example of adding additional selectivity in the selection of similar learning problems, by measuring the correct and incorrect rate for each problem, and derives similar learning problems by dividing them into difficulty-level upward-similar learning problems and difficulty-level non-oriented similar learning problems. Indicates one thing. FIG. 8 shows an example of selecting a similar learner based on the required learning time as a modified embodiment of the selection of similar learners.

As shown in Fig. 1, the problem curation system and method in a problem bank for providing customized learning problems to learners according to an embodiment of the present invention is a problem provider providing data such as learning problems and related information. The terminal 10, the question bank module 100 in which the data is stored, the analysis module 200 for calculating the context similarity of the learning problem and the learner's behavior-based similarity and extracting the problem to be provided to the learner, And a learning information module 300 that stores data such as distribution of correct and incorrect answers, and stores problems to be provided to learners, and a learner terminal 400 that transmits and receives data related to learning of the learner. The analysis module 200 calculates the similarity between problems based on the morpheme analysis information 500 for each problem to derive the similar problem list information 530 for each problem, and the similarity for each learner based on the problem solving information 600 for each learner. Learner list information 620 is derived, and problem curation information 710 for each learner is derived in a list form using the similar problem list information 530 for each problem and the similar learner list information 620 for each learner. Based on the listed problem curation information for each learner, the problem that the learner has already learned is selectively processed, and based on the listed learner data, based on the learner's behavior-based similarity, it is expected to be provided or correct or incorrect. Provides an optional problem.

In an embodiment of the present invention, the problem bank module 100 receives problem data from the problem provider terminal 10. The problem data includes learning problems and related information for the learner's learning, and may include, but is not limited to, specific types, grades, units, difficulty levels, subjects and fields, multiple choices/subjects, and multiple choices. In addition, the number of problems is not limited to a specific number, but a method of performing each group or category by dividing into a predetermined group or category in consideration of calculation speed, storage space, etc. may be employed.

In an embodiment of the present invention, the analysis module 200 analyzes and processes a learning problem and a learner, and also performs a function of curating a problem by calculating the context similarity of the learning problem and the behavior-based similarity of the learner.

First, when a procedure for calculating the context similarity for each problem is described, the problem context analysis module 210 analyzes the morphemes for each problem and derives the morpheme analysis information 500 for each problem that determines whether each morpheme is included. Based on the morpheme analysis information for each problem, common (True/True), unilateral (True/False), by counting the common, one-sided, the other, and non-appearance of morphemes in both learning problems targeting specific problems and other problems. The other (False/True) and non-appearance (False/False) are arranged to prepare the True/False calculation information 510 between problems.

Next, the procedure for calculating the similarity based on learners' behavior is performed by calculating the similarity between learners by the learning information storage module 300 recording the learner's learning information in a log format. As an example, the distribution of correct and incorrect answers of a problem that has been learned equally between a specific learner and other learners is counted, and the count of the distribution of incorrect answers is prepared by writing a True/False table.

Fig. 2(a) shows the learning problem and the morphemes (Terms) appearing for each problem in a matrix form to derive the morpheme analysis information 500 for each problem. For example, learning based on four morphemes (Term) The information on the appearance of morphemes of Problem 1 to Problem 5 is displayed. Whether or not the specific morpheme is included is recorded as 0 and 1, when the specific morpheme is provided, it is recorded as 1, and when the specific morpheme is not provided, it is recorded as 0. The morpheme (Term) is a unit of language and may include words, idioms, sentences, etc. in addition to morphemes, and may include mathematical expressions or symbols, and the number of morphemes is not limited to a specific range.

2(b) shows the information 510 for calculating True/False between problems, and as an example, the appearance of morphemes between problems between learning problem 1 (Problem1) and learning problem 2 (Problem2) is shown in a True/False table. It was recorded. The True/True item records the number of morphemes commonly included in both learning questions, and True/False or False/True records the number of morphemes included in only one or the other of both learning questions, respectively, and False/False means The number of morphemes that did not appear anywhere in both learning problems was recorded. In the example, in the False/False item, a value of zero (0) is recorded by calculating based on four morphemes, but if more morphemes are used as the reference, it can be counted as an arbitrary value. Specifically, the value aggregated in the False/False item matches the total number of target morphemes minus the values of the preceding True/True items, True/False items, and False/True items.

2(c) and 2(d) are illustrative representations of similarity calculation information 520 for each problem, and one implementation of calculating similarity by using Jaccard similarity or Cosine similarity, respectively. It shows an example. In this case, when the jacquard similarity is calculated, the similarity value appears as a value between 0 and 1, and when the similarity value is 1, both learning problems are the same problem.

Jacquard similarity between problems is obtained as the ratio of the number of morphemes commonly included in both problems to the number of morphemes included in any of the problems, and in the example, there are 2 morphemes commonly included in both problems, and Since there are 4 morphemes included everywhere, the similarity value is 0.5. In addition, since the jacquard similarity between the same problems is inevitably calculated as 1, the calculation of the similarity may be omitted, and when there is no common morpheme, the jacquard similarity between the two problems is zero (0).

,

와, 내적값 1X1+0X1+1X1+1X0 = 2를 이용하여 계산하면 코사인 유사도 값은 0.666 이다. 또한, 동일 문제간의 코사인 유사도는 1로 산출될 수밖에 없으므로 해당 유사도 계산은 생략될 수 있으며, 공통되는 형태소가 하나도 없는 경우에 양 문제간의 코사인 유사도는 영(0)이다.The cosine similarity between problems can be obtained by calculating the cosine value of the inner angles of the vectors of both problems by considering the list representing the appearance of the morpheme in each problem as coordinate value 1 as a vector. It can be calculated by treating the dot product of both vectors as a ratio. In the example, problems 1 and 2 are regarded as vectors (1,0,1,1) and (1,1,1,0), respectively, and their respective sizes

,

And, if calculated using the dot product value 1X1+0X1+1X1+1X0 = 2, the cosine similarity value is 0.666. In addition, since the cosine similarity between the same problems is inevitably calculated as 1, the calculation of the similarity may be omitted, and when there is no common morpheme, the cosine similarity between the two problems is zero (0).

Meanwhile, in calculating the true/false calculation information 510 between problems or the similarity calculation information 520 between problems, the similarity calculation may be supplemented by reflecting the relationship between morphemes. That is, in the case of having a relationship such as synonyms and/or antonyms, foreign words among the morphemes, the morphemes are grouped and the appearance of the morphemes is calculated, and the true/false calculation information 510 between problems or the similarity calculation information 520 between problems Is to produce. To this end, the analysis module 200 may additionally include a relational word dictionary module 240. By providing such a relational word dictionary module 240, it is possible to reduce the burden on the amount of calculation and storage space.

The relational word dictionary module 240 provides the problem context analysis module 210 with the determination of the relationship between the morphemes according to the system setting or, in addition, provides grouping information of the morphemes based on the relationship determination. At this time, the relationship may be limited to a simple synonym (including a foreign word) relationship according to the system setting, or may be extended to not only a synonym relationship but also an antonym, a related word, etc., and the related word dictionary module 240 is basically provided inside. It may be configured to determine the synonym or antonym from a language dictionary or a language dictionary connected through a network, but is not limited thereto.

For example, when deriving the True/False calculation information 510 between problems as a preparation for calculating the cosine similarity between problems, when considering the synonym relationship between the morphemes provided from the relational word dictionary module 240 Each item value of the inter-true/false calculation information 510 appears as a result of being individually changed, and thus, the similarity calculation information 520 between problems and the similarity problem list information 530 for each problem are also changed.

Here, in the case of reflecting the relationship between the morphemes, it is preferable not to excessively expand the relationship between the morphemes because noise may be included in the similarity calculation if the determination accuracy of whether or not synonymous is not secured.

FIG. 2(e) shows an example of the similar problem list information 530 for each problem, and shows recorded information listing the most similar problem for each problem. In the example, learning problem 3 (Problem3) is a case in which the similarity of the problem is high between learning problem 5 (Problem5), and in consideration of the similarity of the problem, the learner is provided with problems with high learning effect and intensive learning. You will be able to. In addition, the learner can further improve the learning effect by receiving problems of a desired level of difficulty among problems with high similarity to a specific problem. In addition, in the case of a problem in which a learner feels difficult or gives an incorrect answer to a specific problem, problems similar to the corresponding problem are provided so that they can intensively learn the insufficient type and the learning ability to cultivate.

3 shows a process of deriving a similar learner list by the learner learning similarity analysis module 220 of the analysis module 200, provided from the learning information storage module 310 that records the learner's learning information in a log format. A process of deriving a list of similar learners by calculating the similarity between learners based on the distribution of correct and incorrect answers among learners using the problem solving information 600 for each learner is illustrated as an example.

3(a) shows the problem solving information 600 for each learner, and shows an example of recording correct and incorrect answers or unlearned states for each question for each learner. 'T' is the record that the learner has submitted a correct answer (True) to the question,'F' is the record that the learner has submitted an incorrect answer (False) to the question, and if the learner has not yet learned, T and F It maintains the state of an arbitrary initial value, not a value. For example, if the learner did not learn, it may optionally include cases where the learning problem was not provided, but also stopped learning due to lack of time, or disconnected due to an error in the device or communication network. can do. In addition, the problem solving information 600 for each learner may include correct and incorrect answer information and additional information such as a learning date, and may be used in a problem curation process.

3(b) shows the similarity value between learners by using the problem solving information 600 for each learner, and between learners storing the similarity value between learners by calculating the similarity between one learner and another learner. The similarity calculation information 610 is shown. Here, for example, the similarity value of learner 1 (Student1) is derived as 1, so the display of the similarity value is omitted, and the similarity value of learner 2 (Student2) to learner 1 (Student1) is learner 2 (Student2). ) Is the same as the similarity value of learner 1 (Student1), so the similarity value display is omitted. The similarity between learners can be calculated using the Jaccard similarity or the cosine similarity method.

3(c) is a list of similar learners for each learner selected based on the previously obtained similarity calculation information 610, and a list of similar learners for each learner organized by listing learners with high similarity to a specific learner. The information 620 is illustrated by way of example. Using the similar learner list information 620 for each learner, problem curation may be performed based on a problem solved by a learner having a high similarity to the learner. In the example, through the distribution of correct and incorrect answers of the same learned problem, it was found that learner 1 (Student1) was most similar to learner 23 (Student23), and using this, learner 23 (Student23) most similar to learner 1 (Student1). It is possible to provide problem curation based on this learned problem. The detailed process of problem curation is described below.

4 is a process of deriving problem curation information 710 for each learner by performing problem curation by reflecting the similarity of learning problems and similarity between learners by the problem extraction module 230 for each learner in the analysis module 200 Is shown.

In order to explain the detailed process of problem curation, if problem curation is performed for learner 1 (Student1) as a specific learner using the example of FIG. 4, learner 1 using the previously prepared similar learner list information 620 for each learner. Is identified as similar to learner 23, etc., and since learner 23's learning information can be obtained from learner-specific learning problem list information 700, basic problem curation can be provided for learner 1 based on learner 23's learning information. I can. That is, in an example, basic problem curation may be performed by providing learning problem 36, learning problem 7, learning problem 12, etc. to learner 1 with reference to learning information of learner 23. Here, the learning problem list information 700 for each learner may use the problem solving information 600 for each learner as it is or may be used by organizing it.

Through problem curation using the learning information of similar learners, the learner can receive learning problems reflecting the learning difficulty and progress of similar learners embedded in the learning information of similar learners, and thus the learning effect can be significantly improved. . In addition, by performing problem curation at predetermined periods, the learner's similar learner may change according to the change of the learner's own learning history and/or the change of the learning history of other learners. Will continue to provide. That is, for example, as the learner's learning achievement improves, the similar learner is changed accordingly to provide a new list of learning problems.

On the other hand, in the case of using basic problem curation, the learner can repeatedly provide the problem he has already learned, and the learner submits the memorized response as it is, so the increase in the learning effect may be weakened, so rather than simply removing the problem. Further complementary problem curation is needed for learners. In the example of FIG. 4, in the process of presenting the learning problem of the similar learner 23 to learner 1, it is an exemplary diagram that a learner identifies and supplements a problem that has already been learned from a problem list. That is, since learner 1 has already learned the learning problem 36 (Problem36) in the problem list, the problem extraction module 230 for each learner may derive an alternative learning problem that replaces the learning problem and present it to the learner 1. On the other hand, when additional information such as a learning date and the like are included in the problem solving information 600 for each learner, it may be presented again among learning problems already learned in consideration of the progress of the learning history.

According to the previously prepared similar problem list information 530 for each problem, since there is a learning problem 435 (Problem435) as a problem similar to the learning problem 36 (Problem36), one of them can be selected as an alternative learning problem and presented to learner 1. The list of learning problems curated to learner 1 through the supplemented problem curation is based on the list of learning problems of similar learners 23, but some may be replaced. The problem curation information 710 for each learner shown in FIG. 4 shows learning problems replaced by the supplemented problem curation in bold for convenience of explanation.

The problem processing module 320 provided by the learning information module 300 according to an embodiment of the present invention includes problem curation information for each learner curated by the problem extraction module 230 for each learner of the analysis module 200 (710) is provided to the learner terminal 400. 5 illustrates an example of a learner screen 800 displayed on the learner terminal 400. Here, the learner terminal 400 may transmit a learner's request for problem data, etc. to the server, and transmit the answer information input by the learner, a solution process, etc. to the server in response to the problem data received from the server, and the like. It is preferable to be configured to display information such as received problem data, correct answer information, and solving process. Therefore, the learner terminal 400 is capable of communication, such as a personal computer (PC), a laptop, a personal digital assistant (PDA), a smart phone, a tablet PC, and the like, and preferably includes a display means. In addition, it is desirable to be implemented as a terminal supporting a data transmission rate of several to hundreds of megabytes for communication.

In configuring the learner screen 800, the provided problem processing module 320 may present a problem type selection screen so that the learner can select the type of the learning problem before or after presentation of the learning problem. In the example of FIG. 5, the selection of'all' in which the entire learning problem to be presented as a problem curation is learned as it is, and the'fit-up' in which a similar learner presents the correct answer and intensively learns the learning problem that is likely to present the correct answer. Learner's choice of learning by presenting the'Expected Problem Intensive Learning' option, and the'Probably Incorrect Problem Focused Learning' option in which similar learners present incorrect answers and learn more intensively learning problems that are likely to present incorrect answers. The structure that further improved the figure is presented as an example. Through this selection function, the learner can learn simple but steadily progressing the learning progress without omission, selectively re-check through familiar learning problems, or intensively check the areas where learning ability is insufficient. The learning effect can be improved by making more useful use of bank information.

On the other hand, in the problem curation method and apparatus in a problem bank that provides customized learning problems to learners according to an embodiment of the present invention, a similar learner is changed to an upward or downward similar learner in the problem curation process according to a learner's selection, etc. Therefore, it can be configured to perform problem curation or substitute learning problems in consideration of the difficulty of the problem.

6(a) is an exemplary display of information 601 obtained by calculating the correct answer rate for each learner from the problem-solving information 600 for each learner, and FIG. 6(b) shows the similar learner list information 620 for each learner described above. Based on the correct answer rate for each learner, similar learner identification information 621 is displayed, which distinguishes between upstream similar learners showing higher correct answer rates than learners and non-disciplinary similar learners showing the same or lower correct answer rate. Depending on the learner's selection or basic settings such as the system, a similar learner for a learner may be selected from an upstream similar learner or a non-oriented similar learner. The similar learner identification information 621 may be performed in the problem extraction module 230 for each learner of the analysis module 200. It is possible to derive supplemented problem discipline information for each learner by performing problem curation by adjusting the similar learner to the learner according to the learner's choice or basic setting such as the system, and by adjusting to an upstream similar learner or a non-going similar learner. .

FIG. 7(a) is an exemplary display of information 602 obtained by calculating the correct answer rate for each question from the problem solving information 600 for each learner, and FIG. 7(b) is a similar problem list information 530 for each problem described above. Based on the correct answer rate for each question, similar problem classification information 531 is displayed for classifying similar learning problems with the same or higher level of difficulty with similar learning problems with the same or higher correct answer rate than the corresponding problem. Depending on the learner's choice or basic settings such as the system, the curation of problems for learners is based on the learning problems of similar learners, but individual learning problems are replaced by similar learning problems of upward difficulty or similar learning problems of non-directed difficulty. In addition, it is possible to derive the curation information of problems for each learner that has been supplemented.

On the other hand, the problem curation system and method in a problem bank for providing a customized learning problem to a learner according to an embodiment of the present invention, by measuring the answer input time of the learner input to the learner terminal 400 and accumulating data Depending on the learner's learning required time, it can be configured to perform problem curation by changing to an upward or downward similar learner, or to substitute and present a learning problem in consideration of the difficulty of the problem.

FIG. 8(a) shows information 603 obtained by statistically processing the learner's answer input time for each question from the problem solving information 600 for each learner to calculate the average time required. Here, the input time may be selected for all learners' answer input time or only the answer input time of learners who entered the correct answer. Additional information can be derived by using the calculated average time required for each problem. For example, the information 603 calculating the learner's time-out rate based on the average time required for each problem is also shown. Considering that the learner's learning ability can be referenced based on this time-out rate, it can be used for problem curation, and FIG. 8(b) shows similar learners according to basic settings such as the learner's choice or system, as in the previous embodiments. In the selection, the similar learner classification information 622 is derived by classifying upward similar learners or non-going similar learners by comparing with the learner's time-out rate.

Meanwhile, since the average time required may be understood to reflect the difficulty of the learning problem, it may be used to select a similar learning problem as in the embodiment described above with reference to FIG. 7.

The system or device described above may be implemented as a hardware component, a software component, and/or a combination of a hardware component and a software component. For example, the devices and components described in the embodiments include, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable array (FPA), It can be implemented using one or more general purpose computers or special purpose computers, such as a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing device may execute an operating system (OS) and one or more software applications executed on the operating system. In addition, the processing device may access, store, manipulate, process, and generate data in response to the execution of software. For the convenience of understanding, although it is sometimes described that one processing device is used, one of ordinary skill in the art, the processing device is a plurality of processing elements and/or a plurality of types of processing elements. It can be seen that it may include. For example, the processing device may include a plurality of processors or one processor and one controller. In addition, other processing configurations are possible, such as a parallel processor.

In addition, the software may include a computer program, code, instructions, or a combination of one or more of these, and may configure the processing unit to operate as desired, or independently or collectively. ) You can command the processing unit. Software and/or data may be interpreted by a processing device or to provide instructions or data to a processing device, of any type of machine, component, physical device, virtual equipment, computer storage medium or device. , Or may be permanently or temporarily embodyed in a transmitted signal wave. The software may be distributed over networked computer systems and stored or executed in a distributed manner. Software and data may be stored on one or more computer-readable recording media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the embodiment, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of the program instructions include not only machine language codes such as those produced by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like. The hardware device described above may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa.

In addition to this, the above description of the present invention is for illustrative purposes only, and those skilled in the art to which the present invention pertains will understand that it is possible to easily transform it into other specific forms without changing the technical spirit or essential features of the present invention. I will be able to. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not limiting. For example, each component described as a single type may be implemented by being distributed or divided, and similarly, components described as being distributed or divided may also be implemented in a combined form within the scope of the understanding of a person skilled in the art. have. Further, the steps of the method may be performed alone or in a plurality of times in combination with at least one other step.

The scope of the present invention is indicated by the claims to be described later, and all changes or modified forms derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present invention.

10: problem provider terminal
100: question bank module
200: analysis module
210: problem context analysis module
220: learner learning similarity analysis module
230: problem extraction module for each learner
240: relational dictionary module
300: learning information module
310: learning information storage module
320: Provide problem processing module
400: learner terminal
500: Morphological analysis information for each problem
510: True/False calculation information between problems
520: Information on calculating similarity between problems
530: Similar problem list information for each problem
531: Difficulty classification similar problem list information
600: Problem solving information for each learner
610: Information on calculating similarity among learners
620: Similar learner list information for each learner
621: Information on the list of similar learners by learning level
700: learning question list information for each learner
710: Problem curation information for each learner
800: learner screen
810: Problem type selection screen

Claims (6)

In the learning problem curation system that extracts a part from the learning problem of the problem bank and provides it to the learner using learner learning information and problem context,
A question bank module 100 in which data of at least the learning question is stored; And
An analysis module 200 for extracting a problem list to be provided to the learner by calculating the context similarity of the learning problem and the learner's behavior-based similarity;
And a learning information module 300 that stores the distribution of correct and incorrect answers of the problems learned by the learner as data and provides the learning problem list to the learner,
The analysis module 200,
A problem context analysis module 210 that analyzes morphemes for each problem and calculates similarity between problems based on whether or not each morpheme is included to derive similar problem list information 530 for each problem; And
A learner learning similarity analysis module 220 that calculates the similarity between learners based on the distribution of correct and incorrect answers between learners of the problem solving information 600 for each learner provided from the learning information module 300 to derive similar learner list information 620 for each learner. );
For each learner, similar learners are selected based on the similar learner list information 620 for each learner, and problem curation information 710 for each learner is extracted by extracting the problem list of the similar learners from the problem solving information 600 for each learner. And a problem extraction module 230 for each learner to derive;
The learning information module 300 extracts the learning problem list for a corresponding learner from the problem curation information 710 for each learner and provides it to the learner terminal 400.
The method of claim 1,
The problem extraction module 230 for each learner of the analysis module 200,
In the process of extracting the problem list of the similar learner from the problem solving information 600 for each learner and deriving the problem curation information 710 for each learner, the learning problem already learned by the learner is the similar problem list information 530 for each problem. ), a learning problem curation system, characterized in that replacing the selected similar learning problem using.
The method of claim 2,
The problem extraction module 230 for each learner of the analysis module 200,
In replacing an already learned academic problem with a similar learning problem, it is configured to select and replace a similar learning problem of an upward difficulty level or a similar learning problem of a non-reversing difficulty level according to a personal setting or a basic setting of the system, Learning problem curation system.
The method of claim 1,
The problem extraction module 230 for each learner of the analysis module 200,
In the process of selecting a similar learner for each learner based on the similar learner list information 620 for each learner, it is configured to select an upstream similar learner or a non-reversing similar learner according to an individual setting or a basic setting of the system. A curation system for learning problems.
The method of claim 1,
The learning information module 300,
When extracting the learning problem list and providing it to the learner's terminal 400, whether to receive the learning problem list as it is or only the problems expected to be matched among them, or only problems expected to be incorrect among them The learning problem curation system, characterized in that configured to reflect the learner's selection by presenting a problem type selection screen 810 that the learner can select.
In the learning problem curation method of extracting a part from the learning problem of the problem bank and providing it to the learner based on the learner's learning information and the problem context using the learning problem curation system of any one of claims 1 to 5,
Analyzing morphemes for each problem, calculating similarity between problems based on whether or not each morpheme is included, and deriving similar problem list information 530 for each problem; And
Calculating a similarity between learners based on the distribution of correct and incorrect answers between learners of the problem solving information 600 for each learner to derive similar learner list information 620 for each learner;
For each learner, similar learners are selected based on the similar learner list information 620 for each learner, and problem curation information 710 for each learner is extracted by extracting the problem list of the similar learners from the problem solving information 600 for each learner. Deriving;
And extracting the learning problem list for a corresponding learner from the problem curation information 710 for each learner and providing the learning problem list to the learner terminal 400; and
In the step of deriving the similar problem list information 530 for each problem, in calculating the similarity between problems, the similarity between problems is calculated by reflecting the relationship between morphemes.

Images