CN111985282A - A training and evaluation system for learning ability - Google Patents

Abstract

The present invention provides a training and evaluation system for learning ability, which includes the following program modules: a training course content unit; a data processing center; a training model content matching unit; a key information database unit; an ability training model unit for generating a personalized model Training content; it includes a reading training model, a word dictation training model, an application question training model, a voice training model, a picture training model, a "find a friend" training model, and a graphic association training model; model training content units; answering result judgment units; and an ability evaluation center, which is used to provide corresponding evaluation results according to the evaluation criteria pre-stored in the system and combined with the judgment results given by the answering result judgment unit; wherein, the learning terminal is equipped with an auxiliary device for assisting students' concentration. assessment of ability. It can be organically combined with the individualized learning of students, and realize the ability training and cultivation of Jiuli in a comprehensive and seamless manner.

Description

A training and evaluation system for learning ability

technical field

The present invention relates to the cultivation of learning ability, in particular to the training and evaluation of learning ability.

Background technique

Education pays more and more attention to the ability training and cultivation of students' observation, concentration, memory, imagination, thinking, reading comprehension, extension, analytical thinking, problem solving, etc., but most of the learning ability evaluation The system is implemented manually or in combination with computers. Students manually complete the evaluation items or complete the evaluation items on the computer display interface. The test results of the evaluation items are collected manually or by software, and the judges manually determine the test level of the students according to these test results. . Because students of different ages have great differences in cognitive ability, and the evaluation and training items are not comprehensive enough, the fixed and standardized evaluation content, items and methods cannot intelligently adapt to different types of students.

The existing ability training and training for the above-mentioned nine powers mainly tests and trains students through additional training systems and items that are not related to students' learning content. Students need to spend additional time for training and evaluation after completing their homework. It is difficult for most students to take the initiative to complete a set of training and evaluation items without supervision, and most of the evaluation content is not only separated from the content of each student's study, but also the same. For students of the same age and gender, their The content of training and evaluation is the same, and relevant ability training and evaluation cannot be carried out according to the individualized learning of each student.

SUMMARY OF THE INVENTION

The technical problem to be solved by the present invention is that, aiming at the above-mentioned defects of the prior art, a learning ability training and evaluation system is proposed, which can be organically combined with the individualized learning of students, so as to realize the comprehensive and seamless realization of the nine powers. Ability training and development.

The technical scheme adopted by the present invention to solve the technical problem is: a training and evaluation system of learning ability is provided, including the following program modules:

The content unit of the training course is used to provide the electronic version of the learning content for students to learn independently on the learning terminal;

A data processing center for computing and processing related information;

A training model content matching unit, used for format and parameterized identification and input of the content in the input text provided by the training course content unit;

The key information base unit stores the text and picture information extracted from the content matching unit of the training model and that can be recognized by all the files in the system;

The ability training model unit is used to generate personalized model training content; it includes a reading training model, a word dictation training model, an application question training model, a voice training model, a picture training model, a "find a friend" training model, and graphic association training. Model;

The model training content unit is used to provide the ability training model unit with personalized model training content, which is displayed on the interface of the learning terminal and interacts with students;

an answering result judgment unit for judging the student's question-answer interaction in the model training content unit; and

The ability evaluation center is used to provide the corresponding evaluation results according to the evaluation criteria pre-stored in the system and in combination with the judgment results given by the answering result judgment unit;

Wherein, the learning terminal is equipped with an auxiliary device for assisting the evaluation of the students' ability to concentrate.

In some embodiments, the auxiliary device includes: an eye tracker, for detecting the movement of the students' eyeballs; and a camera, for recognizing the facial features of the students.

In some embodiments, the electronic version of the learning content includes preview and practice exercises and courseware uploaded by the teacher, and learning content pre-stored or downloaded to the learning terminal by the parent.

In some embodiments, the training model content matching unit performs normalized format processing on the electronic version of the learning content, and extracts all convertible file information.

In some embodiments, the key information library unit is composed of a key word library unit and a key picture library unit.

In some embodiments, the keyword library unit includes a textbook word and idiom library, math application problems, and text field chapters; the key image library unit includes pictures of people, pictures of animals, and pictures of objects.

In some embodiments, the system uses game models, language models, convolutional neural network models, and hidden Markov models to establish the input text of the training course content unit and all models of the competency training model unit.

In some embodiments, the capability training model unit fills and combines the elements in the key information base unit in the game, picture frame, interactive interface, and learning tutorial model, and generates personalized content through program operations. Model training content.

In some embodiments, the model training content unit includes reading training content, word dictation training content, word problem training content, speech training content, picture training content, "find a friend" training content, and graphic association training content.

In some embodiments, the ability assessment center includes a reading comprehension assessment, a memory assessment, a problem-solving assessment, an analytical thinking assessment, a thinking assessment, an observation assessment, an imaginative assessment, an extension assessment, and a concentration assessment .

The beneficial effect of the present invention is that, through the training course content unit, the data processing center, the training model content matching unit, the key information database unit, the ability training model unit, the model training content unit; the answering result judgment unit, the ability evaluation center and the learning terminal configuration The ingenious cooperation of some auxiliary devices can be organically combined with the individualized learning of students, so as to realize the ability training and cultivation of Jiuli in a comprehensive and seamless manner.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings and embodiments, in which:

FIG. 1 illustrates the frame structure of the learning ability training and evaluation system of the present invention.

Figure 2 illustrates some parts of the more detailed frame structure of the system of the present invention.

Figure 3 illustrates other partial and more detailed frame structures of the system of the present invention.

FIG. 4 illustrates the process of obtaining key information content from the training course content by the system of the present invention.

The reference numerals are described as follows: 100, system 200, students 10, training course content unit 20, data processing center 30, training model content matching 40, key information library unit 41, keyword library unit 411, textbook words and idioms Library 412, Math Application Questions 413, Text Field Chapters 42, Key Image Library Unit 421, Character Pictures 422, Animal Pictures 423, Physical Pictures 50, Ability Training Model Unit 501, Content Filling and Combination 502, Game Model 503, Language Model 504, Convolutional Neural Network Model 505, Hidden Markov Model 51, Reading Training Model 52, Word Dictation Training Model 53, Application Question Model Training Model 54, Voice Training Model 55, Picture Training Model 56, "Find a Friend" training Model 57, graphic association training model 60, ability training content unit 61, reading training content 62, word dictation training content 63, application question model training content 64, voice training content 65, picture training content 56, "find a friend" training content 57. Graphic correlation training content 70, Ability evaluation center 71, Reading comprehension evaluation 72, Memory evaluation 73, Problem solving evaluation 74, Analysis and thinking evaluation 75, Thinking evaluation 76, Imagination evaluation 77, Observation evaluation 78, Extension force evaluation 79, concentration evaluation 80, answering result evaluation unit 90, learning terminal 91, auxiliary device 95, answering feedback information.

Detailed ways

The preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

Referring to FIG. 1 to FIG. 4 , FIG. 1 illustrates the frame structure of the learning ability training and evaluation system of the present invention. Figure 2 illustrates some parts of the more detailed frame structure of the system of the present invention. Figure 3 illustrates other partial and more detailed frame structures of the system of the present invention. FIG. 4 illustrates the process of obtaining key information content from the training course content by the system of the present invention. The learning ability training and evaluation system 100 of the present invention includes a learning terminal 90, and the learning terminal 90 includes several processors; the program modules running on these processors include: a training course content unit 10, a data processing center 20, a training model content matching unit The unit 30 , the key information base unit 40 , the ability training model unit 50 , the model training content unit 60 , the answer result judgment unit 70 and the ability evaluation center 80 . Among them, the learning terminal 90 is equipped with an auxiliary device 91 such as an eye tracker and a camera.

The training course content unit 10 is used for providing the electronic version of the learning content for the students to learn independently on the learning terminal 90 . Specifically, it includes: preview and practice exercises, courseware, etc. uploaded by the teacher; learning content pre-stored or downloaded to the learning terminal 90 by parents, which may include files such as pictures, text, video, and audio. The training course content unit 10 uses the above-mentioned learning contents from teachers and parents as the input text of the system 100 .

The data processing center 20 is used for computing and processing related information. It involves a processor, a storage unit, data uploading and downloading, etc., and performs parameter calculation, file storage, and transmission of calculation parameters for the above input text of the training course content unit 10 , and analyzes and calculates the data of the auxiliary device 91 . The data processing center 20 processes the feedback information formed by the following student users' ability training answers: the answer feedback information 95, and sends the processing result to the answer result judgment unit 70 described below. For example, the data processing center 20 is composed of hardware such as a processor and a memory on the learning terminal 90 and some software running on the processor.

The training model content matching unit 30 is used for format and parameterized identification and input of the content in the input text provided by the training course content unit 10 . Since the input text contains various forms of files, the system 100 needs to perform normalized format processing on it, for example: audio information is converted into text form for recognition; video information is converted into picture information for recognition; then, in the recognition processing After that, extract all the file information that can be converted.

The key information library unit 40 is used to store the text and picture information in all the files that can be recognized by the system, extracted from the above-mentioned training model content matching unit 30 . It consists of a key word library unit 41 and a key picture library unit 42 .

The keyword library unit 41 includes textbook word and idiom library 411 , mathematics application questions 412 , and text field chapters 413 ;

The ability training model unit 50 includes: a reading training model 51, a word dictation training model 52, an application question training model 53, a voice training model 54, a picture training model 55, a "find a friend" training model 56 and a graphic association training model 57 .

It is worth mentioning that, for all the above models 51 to 57, the system 100 mainly uses the game model 502, the language model 503, the convolutional neural network model 504 and the hidden Markov model 505 to input the text and the corresponding ability training model 51. to the establishment of 57.

Specifically, the elements 411 to 413, 421 to 423 in the above-mentioned key information library unit 40 are filled and combined 501 with the content of the program framework in the models of games, picture frames, interactive interfaces, learning tutorials, etc., and individual characters are generated through program operations. The transformed model training content is sent to the model training content unit 60 .

The model training content unit 60 is configured to display the personalized model training content provided by the ability training model unit 50 on the interface of the learning terminal 90 to interact with the students 200 . It includes: reading training content 61 , word dictation training content 62 , application question training content 63 , voice training content 64 , picture training content 65 , “find a friend” training content 66 , and graphic association training content 67 .

The reading training content 61 includes: in this module, students read, set the number of words on the page, record the stay time and the interval time between page turning.

The word dictation and voice training contents 62 and 64 include: fetching words from the above models 51 to 57, performing voice broadcast, students answering on the operation interface, the system making judgments, and judging the user's voice.

The application problem training content 63 includes: fetching words from the above models 51 to 57 , matching of mathematical application problems or thinking problem models.

The pictures and graphics association training contents 65 and 67 include: fetching pictures from the above models 51 to 57 for matching, allowing students to write the corresponding words according to the pictures, and judge the category of the graphics.

The training content 66 of "finding friends" includes: fetching a certain number of pictures from the above models 51 to 57, giving keywords, allowing students to find the corresponding pictures, and the system will judge and record the time.

The answering result judgment unit 70 is used for judging the question-answer interaction of the students in the model training content unit 60 . The student 200 answers the feedback information 95 of the training content in the learning terminal 90, and performs parameter matching of the big data on the ability training content (stored in the ability training content unit 60) generated by the above-mentioned ability training model unit 50. Search for relevant answers in 40 or link to the external network cloud database for answer traversal to match the results.

The ability evaluation center 80 is configured to provide corresponding evaluation results according to the evaluation criteria pre-stored in the system and in combination with the judgment results provided by the answering result judgment unit. Including: Reading Comprehension Test 81, Memory Test 82, Problem Solving Test 83, Analytical Thinking Test 84, Thinking Power Test 85, Observation Test 86, Imagination Test 86, Extension Test 88, and Concentration Test 89 .

It is worth mentioning that the learning terminal 90 includes a mobile phone terminal, a PC terminal, a tablet computer and other learning devices with smart display. On the interface of the learning terminal 90, when the students perform the above-mentioned ability training, with the aid of the auxiliary device 91, they can grasp the area of concentration of the students' attention when they are studying in time, and then through data analysis, they can evaluate the concentration of the students throughout the training process, This evaluation method can fully or partially achieve the aforementioned focus evaluation89.

Hereinafter, the system 100 of the present invention will be described in more detail with several examples.

Referring to FIG. 1 , the user (ie, the student 200 ) is studying the school coursework in the learning terminal 90 , and the teacher publishes the electronic homework in the system 100 via the server (eg: cloud server) through the APP and other means, and the student 200 is in the system 100 After registering and logging in, the learning content can be downloaded and imported (achieved by the training course content unit 10); then, the data processing center 20 uploads the above-mentioned teacher to the course, and performs corresponding download and data storage analysis processing.

Referring to FIG. 2 , taking the training model 56 of “finding friends” as an example, to perform corresponding ability training and evaluation on the observation power of the students 200, the system 100 can preset 30 pictures as a group, and randomly grab the part of one of the pictures. corresponding text for feedback. It can be understood that the ability training model unit 50 includes the program structure and logic structure of the above-mentioned “find a friend” training model 56 .

The model training content matching unit 30 performs capture, format recognition, and content recognition on the data captured by the above-mentioned data processing center 20 for word and picture features. For example, the above-mentioned student 200 includes 100 pictures in all the courseware, including mathematical graphics, illustrations of Chinese passages, illustrations of English passages, etc. The data processing center 20 performs format recognition on all 100 pictures. Extraction and size and format screening resulted in 80 identified pictures.

Referring to FIG. 3, the model training content matching unit 30 performs feature recognition on key features such as characters, numbers, animals, and things in these 80 pictures, and performs feature matching through the external network link server and cloud data, and judges each picture. graphic content.

Referring to FIG. 4 , a process of obtaining the content stored in the key information base unit 40 from the input course provided by the training course content unit 10 is illustrated. Step S401 is to input the course; Step S402 is to obtain the key information of the first layer, such as: key paragraph chapter feature or key picture feature; Step S403 is to obtain the second layer of key information, such as: keyword word feature or key Element picture feature; in step S404, the obtained output text or picture.

For example, the system 100 randomly selects one of the 80 pictures for identification. If it is recognized that it contains a person's head portrait, the ability training model unit 50 performs the high-level feature identification and matching of the neural convolutional network on the above 80 pictures. , and fill all the pictures containing the avatars with the content of the "find friends" training model 56 to generate a new set of 50 pictures (assuming that each picture contains avatars).

The system 100 provides the above-mentioned picture containing the person's avatar on the learning terminal 90, the student 200 clicks a button such as "Start" on the interface of the learning terminal 90, the system 100 starts to present 50 pictures and starts timing, and the student 200 selects one in the interface After that, the system 100 automatically submits and records the picking time.

The student 200 sends the response feedback information 95 (the selected picture at this time) to the data processing center 20 to calculate and identify the picture parameters, and count the time parameters of the time spent. The answering result judging unit 70 performs a parameter comparison operation between the recognition result and the picture containing the avatar provided by the above-mentioned system 100 on the interface of the learning terminal 90 .

The answering result judgment unit 70 provides the judgment result and the time parameter to the ability evaluation center 80 . It can be understood that the ability evaluation center 80 includes a segmented level of the length of time to be searched for 50 pictures preset by the system 100 . When the judgment of the answer result judging unit 70 and the given picture feature repetition threshold exceed the set value, it can be judged that the selection of the student 200 is correct, and the time parameter for completing the task is further compared with the preset segmentation level of the system. Achieve a student's 200 observational assessment, and give the corresponding assessment conclusion.

It is worth mentioning that the user can set the given picture by himself. For example, it can be a picture containing a certain formula, a certain word, a certain word and other elements. With the help of this "find a friend" training and assessment to improve students' observation skills.

For example, the learning terminal 90 is provided with an eye tracker and a camera (that is, the auxiliary device 91 ). The above-mentioned student 200 performs the training and answering process of the “find a friend” training model. The student 200 clicks “Start” on the interface of the learning terminal 90 When the button is pressed, the eye tracker detects the eye movement of the student 200, and recognizes the facial features of the student in combination with the camera.

The data processing center 20 calculates the parameters collected by the eye tracker and the camera, and determines the length of time T1 that the eyes of the students 200 stay on the operation interface of the learning terminal 90 during the above training and evaluation process.

Assumption: In the foregoing example, the time when the student 200 searches for a picture is T0, and the data processing center 20 can calculate the value of β=T1/T0 according to T1 and T0. The system 100 can preset different β ratios and their corresponding concentration level segments. The system 100 matches the β values in the ability evaluation center 80 to give the concentration evaluation results of the students 200 .

The beneficial effects of the learning ability training and evaluation system 100 of the present invention include:

1. Through the program architecture based on the game model, language model, convolutional neural network model and hidden Markov model, feature extraction is performed on the learning content carried by the students 200 in the system 100, and according to the training framework and The evaluation framework fills parameters and generates content, which can achieve fully personalized training and evaluation, and can more efficiently and accurately fit specific students 200 .

2, by setting the reading training model 51, the word dictation training model 52, the application question training model 53, the voice training model 54, the picture training model 55, the "find a friend" training model 56, the graphic association training model 57 and the auxiliary device 91, It can train nine abilities such as reading comprehension, memory, problem-solving, analytical thinking, thinking, observation, imagination, extension and concentration.

3. The training project parameters of the nine abilities are calculated and the results are matched and judged by the data processing center 20, the grade thresholds of various models and the division of the ability level are preset by the system, and the answer result judgment unit 70 and the ability evaluation center 80 are used. , which can give students 200 ability evaluation.

To sum up, the learning ability training and evaluation system 100 of the present invention can fully realize the nine abilities of observation, concentration, memory, imagination, thinking, reading comprehension, extension, analytical thinking, and problem solving. Ability training and training; and, the content of Jiuli’s ability training and training is not separated from the content of students’ learning, so that students can realize ability training and training without trace in the process of personalized learning (complete homework, review homework alone). evaluation. It can be understood that both the training and evaluation implemented by the system 100 can improve the students' abilities; in other words, the training and evaluation are organically combined in the system 100 to form the students' ability-improving and mutually supportive legs.

It can be understood that, according to actual application requirements, the above-mentioned program modules 10 , 20 , 30 , 40 , 50 , 60 , 70 and 80 can be flexibly configured on the learning terminal 90 and the server (if any). For example, in one situation, the above-mentioned program modules 10 , 20 , 30 , 40 , 50 , 60 , 70 and 80 are all configured on the learning terminal 90 , at this time, the system 100 can be regarded as being completely controlled by the learning terminal 90 . And the software implementation on it, and has nothing to do with the server. In another case, the above-mentioned program modules 10, 20, 30, 40, 50, 60, 70 and 80 are distributed on the learning terminal 90 and the server (not shown in the figure), at this time the system 100 It can be regarded as being realized by the learning terminal 90 and the server and the software on it.

It should be understood that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them. For those skilled in the art, the technical solutions recorded in the above embodiments can be modified, or some of the technical features can be modified. Equivalent replacements are made; and these modifications and replacements shall fall within the protection scope of the appended claims of the present invention.

Claims (10)

1. A learning ability training and evaluation system, comprising the following program modules:

the training course content unit is used for providing electronic version learning content for students to independently learn on the learning terminal;

the data processing center is used for calculating and processing the related information;

the training model content matching unit is used for identifying and inputting the format and the parameterization of the content in the input text provided by the training course content unit;

the key information base unit stores the character and picture information in the file which is extracted from the training model content matching unit and can be identified by all systems;

the capability training model unit is used for generating personalized model training content; the method comprises a reading training model, a word dictation training model, an application question training model, a voice training model, a picture training model, a 'friend finding' training model and a graph association training model;

The model training content unit is used for presenting the capability training model unit with personalized model training content on the interface of the learning terminal and forming interaction with students;

the answer result judging unit is used for judging the question-answer interaction of the students in the model training content unit; and

the capability evaluation center is used for giving a corresponding evaluation result according to the evaluation standard prestored in the system and by combining the judgment result given by the answer result judgment unit;

wherein, this study terminal disposes auxiliary device for the evaluation of the ability of supplementary student's special attention.

2. A learning ability training and evaluation system according to claim 1, wherein: the auxiliary device includes: the eye tracker is used for detecting the eyeball rotation condition of the student; and the camera is used for identifying the facial features of the students.

3. A learning ability training and evaluation system according to claim 1, wherein: the electronic version learning content comprises pre-learning and exercise exercises and courseware uploaded by teachers and learning content prestored or downloaded to the learning terminal by parents.

4. A learning ability training and evaluation system according to claim 1, wherein: the training model content matching unit performs normalized format processing on the electronic version learning content and extracts all file information capable of being converted.

5. A learning ability training and evaluation system according to claim 1, wherein: the key information library unit is composed of a key character library unit and a key picture library unit.

6. A learning ability training and evaluation system according to claim 5, wherein: the keyword library unit comprises a textbook word and phrase library, a math application question and a text paragraph chapter; the key picture library unit comprises a figure picture, an animal picture and a real object picture.

7. A learning ability training and evaluation system according to claim 1, wherein: the system establishes the input text of the training course content unit and all models of the ability training model unit through a game model, a language model, a convolutional neural network model and a hidden Markov model.

8. A learning ability training and evaluation system according to claim 7, wherein: the ability training model unit fills and combines the elements in the key information base unit in the game, the drawing frame, the interactive interface and the learning course model, and generates personalized model training content through program operation.

9. A learning ability training and evaluation system according to claim 1, wherein: the model training content unit comprises reading training content, word dictation training content, application question training content, voice training content, picture training content, friend finding training content and graph association training content.

10. A learning ability training and evaluation system according to any one of claims 1 to 9, characterized in that: the ability evaluation center comprises reading comprehension evaluation, memory evaluation, problem solving evaluation, analysis thinking ability evaluation, observation ability evaluation, imagination evaluation, extendibility evaluation and concentration evaluation.

Images