TWI790069B - Learning system for appraising literature - Google Patents

Abstract

A learning system for appraising literature is adapted for solving the problem of limitations and ill effects on the learning for appraising in conventional way. The learning system for appraising literature comprises: a test management module and an evaluating module. The test management module is used for providing a test question to a user and receiving a responded answer made by the user. The test question includes a predetermined answer. The evaluating module includes a text extraction module and a semantic analysis module. The text extraction module is used for applying a text extracting process on the responded answer so as to obtain an extracted information. The semantic analysis module is used for applying a word embedding process on the extracted information and the predetermined answer so as to determine whether the extracted information and the predetermined answer are the same or not and hence to generate an analyzed result.

Description

Literature Review Learning System

The present invention relates to a learning system, in particular to a document review learning system for assisting users to independently learn how to review documents.

In order to train medical professionals to have the core competencies of evidence-based medicine and keep pace with the latest and correct knowledge at any time, medical education will provide many training courses related to evidence-based medicine. In addition to knowledge teaching, the most needed practice includes literature search and training in research and evaluation skills. Traditional teaching methods include classroom teaching, practical workshops, case-oriented discussion learning or online courses, etc. Among them, the most effective teaching method is practical workshops. After class and practice, there will be feedback from the teacher, who can know the standard answer, especially the training of evaluation ability requires such clear feedback. Therefore, schools or teaching hospitals will spend a lot of time, manpower, and cost to set up such evaluation workshops to train professional medical personnel to evaluate empirical research.

However, this method prevents learners from continuing to participate in time-consuming training courses due to the time and space constraints of the learners and the pressure of clinical work. Although the courses can improve the ability of evaluation, the short-term study can only With limited effect, once back in clinical practice, these abilities are not internalized into the behaviors that can be naturally integrated into the daily care of patients. Therefore, how to overcome the limitations of learning, break through time, space and cost considerations, so that all medical professionals can train and research at any time The ability to evaluate and read and achieve the same results as physical workshops is an urgent need to develop. However, there is currently no such evaluation and learning system for users to improve their evaluation ability.

In view of this, in the absence of this type of evaluation and learning system, the present invention provides a document evaluation and learning system for users who need to improve their evaluation ability.

In order to solve the above problems, the object of the present invention is to provide a document evaluation and learning system, which can reduce the constraints of time, space and cost, and can be used and studied by users at any time, so as to improve the user's evaluation ability.

The second object of the present invention is to provide a document evaluation and learning system, which can provide an analysis result for the user's evaluation, so as to understand the user's understanding of the content read, as a reference for the user's evaluation ability and learning improvement. in accordance with.

Another object of the present invention is to provide a document review and learning system, which can provide a judgment result for the user's review, so that the user can understand the key points of the test questions, and learn how to grasp the context of the key points, so as to improve the use of reader's reading ability.

Another object of the present invention is to provide a document evaluation and learning system, which can assign corresponding test questions according to the user's current ability, so as to achieve the effect of teaching students according to their aptitude, and improve the learning interest and efficiency of the user.

The use of "a" or "an" in the elements, components and modules described throughout the present invention is only for convenience and to provide the usual meaning of the scope of the present invention; it should be interpreted as including one or at least one in the present invention , and the singular concept also includes the plural, unless it clearly means otherwise.

The term "coupling" mentioned throughout the present invention includes electrical and/or signal direct or indirect connection, which can be selected by those skilled in the art according to usage requirements.

The terms "keywords" and "synonyms" mentioned throughout the present invention include at least one of single characters, single words, compound words, and compound words, and include words or phrases formed by the aforementioned single words, single words, compound words, and compound words .

The document evaluation and learning system of the present invention is executed on a host computer and includes a test management module for providing a test question to a user and receiving a reply answer from the user. The test question includes a default Answer, the default answer and the reply answer contain at least one text; the quiz topic is an article or a short passage and has a corresponding question, the default answer includes at least one default answer keyword corresponding to the question, the The extracted information includes at least one extracted keyword; an evaluation module, including a text extraction module and a semantic analysis module; the text extraction module is used to perform a semantic extraction process on the reply answer to obtain an extracted information; The semantic analysis module is used to carry out a text vectorized comparison program between the extracted information and the preset answer, so as to judge whether the extracted information is the same as the preset answer, and generate an analysis result; and a qualified judgment module, The test topic may include a qualifying standard, and the qualifying determination module is used to compare whether the analysis result meets the qualifying standard, and produce a corresponding determining result; wherein, the preset answer keyword may have a specific number of answers keyword, the qualification standard sets the judgment logic that must be met or not to be met respectively for the answer keywords of the specific group number, so as to obtain the respective initial judgment results for the answer keywords of the specific group number, and in each initial judgment When all judgment results are pass, the judgment result is qualified.

Accordingly, the document evaluation and learning system of the present invention can be used and studied by the user at any time through the construction of the system, and through the analysis results, the user can understand the degree of understanding of the content read by the user as a The user's evaluation ability and the basis for learning improvement have the effect of improving the user's evaluation ability. In addition, through the default answer keyword and the extracted keyword, the analysis result can show whether the above keywords are the same or synonymous, so as to understand the user's understanding of the content read, as the user's evaluation ability It has the effect of improving the user's evaluation ability. In addition, through the qualification judgment module, the qualification standard and the judgment As a result, the user can understand the key points of the test questions, so as to learn how to grasp the context of the key points, which has the effect of improving the user's evaluation ability.

Wherein, the literature evaluation and learning system may further include a user end device coupled to the host, the user end device may include a display device and an input device, and the display device is used to display the test questions to the user , the input device is used for the user to input the answer. In this way, through the client device, the user can conveniently use the document review and learning system of the present invention for learning, which can achieve the effect of improving the user's review ability.

Wherein, in the character vectorization comparison program, it is compared whether the vector similarity between the extracted keyword and the preset answer keyword is higher than a standard, so as to generate the analysis result. In this way, through the text vectorized comparison program, the generation of the analysis results can be facilitated, so as to understand the user's understanding of the content read, which can be used as the basis for the user's evaluation ability and learning improvement, and can improve the use of The effect of readers' critical reading ability.

Wherein, the document evaluation learning system may further include an identity management module, the identity management module may have a user code and a qualification information, the user code is used to correspond to the user, the qualification information and the Associated with the user code, the test topic may include a question-given condition. When the qualification information meets the question-given condition of the test question, the test question can be tested by the user. In this way, through the use of the qualification information and the question-given conditions, it can be ensured that the questions assigned to the user are in line with the user's qualification, level or ability, so that the overall system can achieve the effect of teaching students according to their aptitude, and can improve the user's learning interest and efficiency.

Wherein, the quiz question may include a question difficulty information, and the qualification information is the user's answering history, and the answering history may include the question difficulty information and corresponding A cumulative number of passing times, the given condition is the minimum number of passing times required for the test item to require at least one item difficulty information. In this way, through the use of the difficulty information of the topic and the cumulative number of passes, it can be ensured that the topic assigned to the user is in line with According to the user's level, the overall system can achieve the effect of teaching students according to their aptitude, and can improve the user's learning interest and efficiency.

Wherein, the topic difficulty information has a first difficulty level, a second difficulty level and a third difficulty level, and the given condition corresponding to the third difficulty level is: the topic difficulty information of the first difficulty level must be Accumulate no less than one first cumulative number of passes, and require that the difficulty information of the topic of the second difficulty level must accumulate no less than one second cumulative number of passes; the first cumulative number of passes and the second cumulative number of passes are not equal to zero. In this way, through the application of the difficulty information of the topic and the cumulative number of passing times, it can be ensured that the topics assigned to the user are in line with the user's level, so that the overall system can achieve the effect of teaching students according to their aptitude, and can improve the user's learning interest and efficiency .

Wherein, the test topic may include a topic code, and each of the topic codes is different from each other, so that each of the topic codes can correspond to different test topics. The identity management module may include a test record module, and the test record module uses The user code of the user is associated with the historical test results of the user, and the historical test results include the subject codes and analysis results of the test subjects that the user has taken. In this way, the user or an instructor can have a comprehensive understanding of the user's overall learning situation, so as to facilitate the user's self-adjustment and improvement, or serve as a reference for the instructor to evaluate or guide the user's learning, thereby having The effect of improving the learning efficiency of the user.

Wherein, in the case that the preset answer is modified, the historical test result will be correspondingly modified, so that the analysis result of the historical test result is generated corresponding to the modified preset answer. In this way, the user or an instructor can understand the most suitable insights at present, can ensure correct or updated information, and can understand the latest learning status of the user.

1: Document review learning system

1': Host

10: User device

11: Display device

12: Input device

2: Test Management Module

20: Question bank database

3: Evaluation module

31: Text extraction module

32: Semantic analysis module

4: Qualification judgment module

5: Identity Management Module

51: Quiz record module

[Fig. 1] A system block diagram of a preferred embodiment of the present invention.

In order to make the above-mentioned and other objects, features and advantages of the present invention more obvious and easy to understand, the preferred embodiments of the present invention are specifically cited below, together with the attached drawings, and are described in detail as follows: Please refer to Fig. 1, It is a block diagram of a document review and learning system 1 according to a preferred embodiment of the present invention. The document review and learning system 1 is executed on a host 1' and includes a test management module 2 and an assessment module 3 .

The document review learning system 1 may include a client device 10 coupled to the host 1' for mutual data transmission, reception, reading or storage. The user-end device 10 may include a display device 11 and an input device 12. The user-end device 10 may be a computer or other remote computer electrically connected to the host 1', or may be a tablet computer or A mobile phone and so on. The display device 11 can be a computer screen, a tablet screen or a mobile phone screen, etc.; the input device 12 can be a keyboard, a mouse, a touch panel, or a microphone.

The test management module 2 is used to provide a test question to a user and receive a reply answer from the user. The test question includes a default answer, and the default answer and the reply answer include at least one text. Specifically, the quiz topic is an article or a short passage, and has a question corresponding to the article or short passage, and the preset answer includes at least one preset answer keyword corresponding to the question. Preferably, the test management module 2 can include a question bank database 20, and the question bank database 20 can include at least one test question and at least one preset answer; in detail, in terms of data structure, each test question has Contains the corresponding preset answer associated with it. The question bank database 20 is coupled with the test management module 2 for transmitting the test questions to the test management module 2 for the user to answer; the display device 11 can at least display the test questions to the user , the input device 12 can be used by the user to input the answer.

The assessment module 3 includes a text extraction module 31 and a semantic analysis module 32; The text extraction module 31 is used to perform a context extraction procedure on the reply to obtain an extraction information; wherein, the extraction information includes at least one extraction keyword. The semantic analysis module 32 is used to perform a character vectorized comparison program between the extracted information and the preset answer, so as to determine whether the extracted information and the preset answer are identical or synonymous, and generate an analysis result. Based on the analysis results, the user or others can understand the user's understanding of the material after reading it, as a basis for the user's self-improvement study, or as the user's instructor (such as a teacher) , professors, or professionals in the corresponding field, etc.) are used to assist the user to strengthen the basis of learning. In addition, through the construction of this system, the user can use and learn at any time, especially when the user is in a good learning state, so that the learning efficiency is improved.

Wherein, the evaluation module 3 may include a language analysis module, and the language analysis module includes a pre-established synonym database and a pre-trained language model. The pre-established synonym database refers to the creation of corresponding keywords at least for knowledge in a specific field, and the formation of associations between the keywords and their synonyms, such as forming a specific group/set of synonyms, so that the The pre-established synonym database includes multiple sets of synonym groups. Therefore, in the context extraction procedure (which may include, for example, a text cleaning step and a text normalization step), an extraction information can be obtained, and the extraction information preferably includes the above-mentioned at least one extraction keyword; more preferably, the extraction keyword is for Keywords or synonyms in the synonym database should be pre-built. The pre-trained language model can be an application of natural language processing (NLP, Natural Language Processing) technology, and uses technologies such as big data and deep learning to perform the above-mentioned text on target information (such as the aforementioned extracted information) Vectorized alignment program. Preferably, the pre-trained language model may be the Word2Vec model proposed by Google ^TM , but is not limited thereto. The text vectorization comparison program uses the pre-trained language model to encode the text in a vectorized manner, forming a computer that can use cosine similarity (Cosine Similarity) to calculate the similarity between two text information, and according to a certain similarity The degree standard judges whether two words are the same/similar. In the text vectorization comparison program, based on the pre-trained language model, the extracted keywords and the preset answer keywords are converted into vectors; and the vector similarity comparison technology is used to determine whether the two words are the same/similar, and Generate the results of this analysis. If after comparison, the similarity between the extracted keyword and the vector of the preset answer exceeds a certain standard, it means that the extracted keyword and the preset answer keyword are the same or synonymous, and the analysis result is the same; otherwise, the analysis The result is not the same. Wherein, the keywords or synonyms defined as the same group in the above-mentioned multiple groups of synonyms means that the analysis results are the same after they are subjected to the text vectorization comparison program. It should be noted that the above-mentioned pre-trained language model and pre-established synonym database are examples of an implementation mode used in the present invention, and the technology used in the present invention is not limited thereto.

According to the description of the aforementioned system architecture, function and technology, an implementation of the present invention can be illustrated as follows: the test topic: this article is about the effect of disease A improved after performing a specific fasting method (the test topic The content is only an exemplary summary of the key points of the corresponding article or essay to help illustrate this example).

Corresponding question: What method is used in this research to improve symptom A?

Corresponding preset answer: fasting method.

Corresponding answer: In this study, the subjects used the intermittent fasting process... (This answer is only an example to present part of the content to assist in explaining this example).

Corresponding extraction information: intermittent fasting.

Corresponding analysis result: match (for example, through the semantic analysis module 32, it is determined that "intermittent fasting" and "fasting method" are the same/synonymous).

In another embodiment of the present invention, when the default answer of the question of the quiz topic includes multiple sets of preset answer keywords, the text extraction module 31 obtains the extracted text from the reply answer of the user. information, and at least one keyword in the extracted information is analyzed and judged through the semantic analysis module 32; wherein, the following test questions, questions and answers are taken as examples, and the An example of the corresponding analysis and judgment logic is shown in Table 1 below.

The title of the test: This article is about the improvement of symptom A after implementing a specific fasting method (the content of this test title is only an exemplary summary of the key points of the corresponding article or short article to help illustrate this example).

Corresponding question: What method is used in this research and what kind of diet is recommended to improve symptom A?

Corresponding answer: In this study, the subjects used intermittent fasting for at least 30 days to reduce their starch intake... example).

Preferably, the test topic may include a passing standard, and the document evaluation and learning system 1 may further include a passing judgment module 4, which is used to compare whether the analysis result meets the passing standard, and Corresponding to the production-one judgment result. Specifically, the at least one preset answer keyword has a specific number of answer keywords (that is, includes one set of answer keywords or multiple sets of answer keywords), and the qualification standard corresponds to the at least one preset answer keyword Set the judgment conditions of "must meet" or "must not meet" for the answer keywords of the specific group number (that is, the judgment conditions of "must meet" or "must not meet" can be set for each of the default answer keywords) , to obtain the respective preliminary judgment results with the answer keywords corresponding to the specific group number. When the preliminary analysis result is completely passed, that is, when each of the preliminary judgment results is passed, the judgment result is qualified; When the analysis result is not completely passed, that is, when at least one of the preliminary judgment results is not passed, the judgment result is unqualified. Wherein when the eligibility criteria When the qualification standard is "must be met", and the analysis result is in compliance, the preliminary judgment result is passed; All passed. Preferably, the test management module 2 or the question bank database 20 may include the eligibility criteria, and each quiz question includes corresponding eligibility criteria associated with each of the preset answer keywords. Corresponding to the example in Table 1, the analysis and judgment logic of adding the eligibility criteria can be exemplified in the following Tables 2 and 3.

With the addition of the eligibility criteria, when setting the at least one default answer for each of the test questions, the focus of the at least one default answer can be further set, and allow the user to answer the questions given by the eligibility standard The result of the judgment is to understand the key points of the corresponding test questions, and then grasp the context of how to grasp the key points after reading a material.

Among them, in the above-mentioned examples of Tables 1 to 3, when the text extraction module 31 cannot extract any keywords from the reply submitted by the user, especially in the reply that cannot correspond to the predetermined In the case of creating any information in the synonym database, the extracted information or the extracted key can be an empty set or other default information to represent the situation (such as table 1 to "No Corresponding Keyword" shown in Table 3), and the corresponding analysis results will not match.

Preferably, the document evaluation and learning system 1 may include an identity management module 5, and the identity management module 5 has a user code and a qualification information. The user code is used to correspond to each user, that is, each user has its unique code, so that the user code can correspond to a correct single user. The qualification information is associated with the user code, and the test topic may also include a question-giving condition. When the qualification information meets the question-giving condition of the test question, the test question can be tested by the user, for example, in In one implementation example, the user can only select the test questions whose qualification information matches the test conditions on the display device 11, thereby ensuring that the test questions assigned to the user by the test module 2 are in line with the user qualifications , level or ability, so that the overall system can achieve the effect of teaching students according to their aptitude, allowing the user to take tests and learning step by step according to his ability, and allowing the user to balance the challenge and sense of accomplishment in learning, and improve the user's ability Learning interest and efficiency.

In an example, the test topic may include a topic difficulty information, and the topic difficulty information is used to display the difficulty of the test topic; preferably, the test topic may include a topic code, and each of the topic codes is different from each other, so that each The question code can correspond to different test questions, and the question difficulty information is associated with the question code. The qualification information may be the user's answering history, which may include, for example, the difficulty information of all the test questions that the user has completed and the results of the judges are qualified, the corresponding cumulative number of passing times, and the corresponding question-giving conditions It may be the minimum number of passing times required for the test item to require at least one item difficulty information. In terms of data structure, the answering process can be presented as a set or matrix data form such as [question difficulty information, cumulative number of passing times], and the condition of the question can be correspondingly presented as [question difficulty information, at least the cumulative number of passing times] ] in the data form of a set or matrix. Among them, for the same question, the accumulated passing times are only counted once. According to the technical characteristics such as the difficulty information of the above questions and the cumulative number of passing times, the corresponding analysis and judgment logic can be illustrated in the following Tables 4 and 5.

Table 4: An example of displaying information about user qualifications in Identity Management Module 5

The question number T4 in the above list 4 and table 5 is taken as an example, because the user (user number P1) has passed the test question of the difficulty information Lv1 more than 4 times (the cumulative number of passing times of the user is 5 times), and passed Question Difficulty Information Lv2 test questions more than 1 time (accumulative number of passing times for this user is 1 time), and meet the test conditions of question number T4, so the question number is T4 The test questions will be associated with the user, that is, the system can assign the test questions with the question number T4 to the user to take the test, or the user can select or perform the test questions with the question number T4 on the corresponding display device 11 . Taking question number T5 as an example, although the user (user number P1) has passed the test question of difficulty information Lv1 more than 2 times (the cumulative number of passing times of the user is 5 times), but failed the test of question difficulty information Lv2 The question has been given more than 2 times (the accumulative pass times of the user is only 1 time), but does not meet the test conditions of the question number T5, so the test question of the question number T5 will not be associated with the user, that is, the system will not assign The test question with the question number T5 is for the user to take the test, or the user cannot select or perform the test question with the question number T5 on the corresponding display device 11 .

In other words, it can be seen from the above content in Table 5 that in the example with the title number T4 or T5, the title difficulty information has a first difficulty level (corresponding to the title difficulty information Lv1), a second difficulty level (corresponding to the title difficulty information Lv2 ) and a third level of difficulty (corresponding to the difficulty information Lv3 of the topic), the condition for the given question corresponding to the third level of difficulty is: the difficulty information of the topic of the first level of difficulty must be accumulated not less than the first accumulated passing times , and it is required that the difficulty information of the topic of the second difficulty level must accumulate no less than a second cumulative number of passing times; the first cumulative number of passing times and the second cumulative number of passing times are not zero. Among them, taking the topic number T4 as an example, the first cumulative number of passing times is 4 times, and the second cumulative number of passing times is 1 time; taking the topic number T5 as an example, the first cumulative number of passing times is 2 times, and the second cumulative number of passing times is 1 time. The qualifying times are 2 times.

It should be noted that the qualification information and the question-giving conditions can be modified accordingly according to requirements. More preferably, besides the history of answering questions, the qualification information can also include identity category. In the environment where the user is a student, the identity category can be distinguished according to the user's grade or seniority, or can be determined according to whether the user has completed certain previous tasks. Set credits, courses or courses as judgment conditions. Among them, it is also possible to add classifications such as department, department, or group to the identity category, or to add new data and conditions to make the qualification information and the given question conditions related, so that the data related to each other and the judgment logic The hierarchical structure (Hierarchical Structure) is more detailed; or, the hierarchical structure of the overall associated data and judgment logic can also be simplified, for example, in the association between the qualification information and the given condition, the judgment condition can only be based on the number of passes.

Preferably, the identity management module 5 may include a test record module 51 for associating the user code of the user with the user's historical test results, the historical test results include, for example, that the user has Item codes, analysis results and/or judgment results and/or item difficulty information of the conducted test items. In this way, the user or an instructor can fully understand the user's overall learning situation, so as to facilitate the user's self-adjustment and improvement, or serve as a reference for the instructor to evaluate or guide the user's learning.

Preferably, the default answer (especially the default answer keyword) or the eligibility criteria can be modified/updated, and when the default answer or the eligibility criteria are modified, the historical test results can correspond to To modify the historical test results so that the analysis results or judgment results of the historical test results are generated corresponding to the modified preset answers or the modified eligibility criteria, so that the user or an instructor can understand the current most appropriate insights , to ensure that correct or updated information can be learned, and the latest learning status of the user can be understood.

It should be noted that the host mentioned in the present invention can be a host/server with an integrated architecture or a distributed architecture, or a host/server in the cloud, and includes functions for data transmission, reception, reading and storage General software and hardware configuration required for functions such as Wherein, the host computer includes at least one processor, which is specially used for calculating, analyzing, and judging whether various conditions are met, for example, it is used for analysis results, judgment results, or given conditions of the present invention, etc. judge.

To sum up, the document evaluation and learning system of the present invention can generate an analysis result by comparing whether the keywords in the preset answer and the user's answer have the same meaning through the evaluation module, and understand The user's understanding of the content read is used as the basis for the user's reading ability and learning progress. In addition, through the application of the qualification judgment module and the qualification standard, a judgment result obtained from the analysis result allows the user to understand the key points of the test questions and learn Learn how to grasp the context of key points. In addition, by associating the user's qualification information with the question-giving conditions, it can be ensured that the questions assigned to the user are in line with the user's ability, and the effect of teaching students according to their aptitude can be achieved, improving the user's learning interest and efficiency.

Although the present invention has been disclosed by using the above-mentioned preferred embodiments, it is not intended to limit the present invention. It is still within the scope of this invention for anyone skilled in the art to make various changes and modifications relative to the above-mentioned embodiments without departing from the spirit and scope of the present invention. The technical scope protected by the invention, therefore, the scope of protection of the present invention should be defined by the scope of the appended patent application.

1: Document review learning system

1': Host

10: User device

11: Display device

12: Input device

2: Test Management Module

20: Question bank database

3: Evaluation module

31: Text extraction module

32: Semantic analysis module

4: Qualification judgment module

5: Identity Management Module

51: Quiz record module

Claims (8)

A literature evaluation and learning system, executed on a host computer, comprising: a test management module, used to provide a test question to a user and receive a reply answer from the user, the test question includes a default answer , the default answer and the reply answer contain at least one text; the quiz topic is an article or a short passage and has a corresponding question, the default answer includes at least one default answer keyword corresponding to the question, and the extraction The information includes at least one extraction keyword; an evaluation module, including a text extraction module and a semantic analysis module; the text extraction module is used to perform a text extraction process on the reply answer to obtain an extraction information; the The semantic analysis module is used to carry out a text vectorized comparison program between the extracted information and the preset answer, so as to judge whether the extracted information is the same as the preset answer, and generate an analysis result; and a qualified judgment module, the The test topic includes a qualified standard, and the qualified judgment module is used to compare whether the analysis result meets the qualified standard, and produce a corresponding judgment result; wherein, the preset answer keyword has a specific number of answer keywords, The eligibility criteria set the judgment logic that must be met or not to be met respectively for the answer keywords of the specific group number, so as to obtain the respective preliminary judgment results corresponding to the answer keywords of the specific group number, and in each of the preliminary judgment results. When it is pass, the judgment result is pass. The document review and learning system of claim 1 further includes a user-end device coupled to the host, the user-end device includes a display device and an input device, and the display device is used to display the test questions for the user Or, the input device is used for the user to input the answer. The document review and learning system according to claim 1, wherein, in the text vectorized comparison program, compare whether the vector similarity between the extracted keyword and the preset answer keyword is higher than a standard, so as to generate the analysis result. The document evaluation and learning system of any one of the request items 1~3 further includes an identity management module, the identity management module has a user code and a qualification information, the user code The code is used to correspond to the user, the qualification information is associated with the user code, and the test topic includes a given condition. When the qualification information meets the given condition of the test question, the test question can be used by the user. take the test. For example, the document evaluation and learning system of claim 4, wherein the test question includes a question difficulty information, the qualification information is the user's answer history, and the answer history includes all the test questions that the user has completed. Information about the difficulty of the subject of the passer and a corresponding cumulative number of times of passing the test. The condition for giving the question is the minimum number of times of accumulative passing required for at least one item of difficulty information for the test question. For example, the document evaluation and learning system of claim item 5, wherein the difficulty information of the topic has a first difficulty level, a second difficulty level and a third difficulty level, and the given condition corresponding to the third difficulty level is: Requirements The difficulty information of the topic of the first difficulty level must accumulate no less than a first cumulative pass times, and require that the topic difficulty information of the second difficulty level must accumulate no less than a second cumulative pass times; the first cumulative number of passes The qualified times and the second accumulated qualified times are not zero. For example, the document evaluation and learning system of claim 4, wherein the test topic includes a topic code, and each of the topic codes is different from each other, so that each of the topic codes can correspond to different test topics, and the identity management module includes a test record module, the test record module is used to associate the user code of the user with the historical test results of the user, and the historical test results include the subject codes and analysis results of the test subjects that the user has taken. As in the document evaluation and learning system of claim 7, wherein, in the case where the preset answer is modified, the historical test result will be modified accordingly, so that the analysis result of the historical test result is determined by the modified preset answer Correspondingly generated.

Images