Man-machine interaction system and method for developmental reading disorder language ability

Abstract

The invention discloses a man-machine interaction system and method for developmental reading disorder language ability. The man-machine interaction system comprises an evaluation unit which is used for evaluating multiple language abilities of a user; the calculation unit is connected with the evaluation unit and is used for calculating initial task difficulty corresponding to each language capability for the user; the interaction unit is connected with the calculation unit and is provided with a plurality of interaction modules, and each interaction module comprises interaction tasks of a plurality of difficulty levels; a personalized interaction scheme for the user is formed by matching with the computing unit and is used for man-machine interaction; the difficulty adjusting unit is connected with the interaction unit and used for dynamically adjusting the difficulty level of each interaction task; and the feedback unit is connected with the interaction unit and is used for feeding back the man-machine interaction result to the user. Starting from the basic composition of the language ability, multiple basic abilities of the user are improved, and the task difficulty adapting to the user level is matched through the preset algorithm.

Description

Technical Field

[0001] The invention relates to a human-computer interaction system for language ability of developmental dyslexia, and also relates to a corresponding human-computer interaction method, belonging to the technical field of human-computer interaction. Background Art

[0002] Developmental dyslexia is one of the most common problems among children with learning difficulties. According to the 10th edition of the International Classification of Diseases (ICD-10) issued by the World Health Organization, it is defined as an individual's reading performance being significantly lower than the expected level for the corresponding age, with no difference in intelligence, motivation, living environment and educational conditions from other children, and without obvious visual, hearing or neurological disorders. This situation often leads to poor academic performance in school, which is often mistaken for intellectual retardation, resulting in frustration, which is extremely detrimental to the overall development of children.

[0003] From a linguistic perspective, the defects of developmental dyslexia are mainly manifested in aspects such as phonemic awareness, morpheme awareness, orthographic awareness and rapid naming. In recent years, cognitive interventions for these defects have gradually increased. For example, Chinese patent application CN113499525A discloses a training system that aims to improve children's working memory level, reading, writing and other related cognitive skills through training of central executive system modules, visual-spatial working memory modules, buffer modules and verbal working memory modules. The system trains children's abilities in working memory, visual-spatial information processing, spatial-verbal integration and separation, and verbal working memory through various tasks.

[0004] However, the intervention products currently on the market are mostly influenced by research on phonetic reading disorders, focusing mainly on the intervention effects of phonetic awareness, while there are relatively few studies on morpheme awareness, orthographic awareness, and rapid naming. The existing intervention system lacks systematic language intervention and has a low degree of electronicization, which cannot meet the needs of continuous health management such as family intervention and follow-up in the Internet era. In addition, the existing language intervention training tasks for reading disorders are relatively few, mainly focusing on training a specific consciousness defect, and there is no personalized training plan matched through big data calculations, which makes it difficult to meet the needs of most users. Summary of the invention

[0005] The primary technical problem to be solved by the present invention is to provide a human-computer interaction system for improving the language ability of developmental dyslexia.

[0006] Another technical problem to be solved by the present invention is to provide a human-computer interaction method for language ability of developmental dyslexia.

[0007] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0008] According to a first aspect of an embodiment of the present invention, a human-computer interaction system for language ability of developmental dyslexia is provided, comprising:

[0009] An evaluation unit is pre-set with a plurality of evaluation tasks for evaluating a user's multiple language abilities and obtaining evaluation results of each language ability; wherein the multiple language abilities at least include: phonetic awareness, morpheme awareness, orthographic awareness and form-sound mapping;

[0010] A calculation unit, connected to the evaluation unit and having a preset algorithm built therein, for calculating the initial task difficulty corresponding to each language ability based on the evaluation results of each language ability of the user;

[0011] The interaction unit includes a plurality of interaction modules, wherein the plurality of interaction modules correspond to the plurality of language abilities one by one, each of the interaction modules includes a plurality of interaction tasks, and each of the interaction tasks is preset with a plurality of difficulty levels; the interaction unit is connected to the calculation unit to perform human-computer interaction with the user based on the interaction tasks of the corresponding difficulty levels under the corresponding interaction modules according to the initial task difficulty corresponding to each language ability of the user;

[0012] A difficulty adjustment unit connected to the interaction unit to dynamically adjust the difficulty of each interactive task during the human-computer interaction of the user;

[0013] The feedback unit is connected to the interaction unit to feed back the result of the human-computer interaction to the user based on the behavioral response data of the user during the human-computer interaction.

[0014] Preferably, the preset algorithm includes the following steps:

[0015] The evaluation results of all current users in the human-computer interaction system are converted into standard scores according to the norm, so as to obtain a mapping relationship between the evaluation scores of all users and the language ability;

[0016] Based on the mapping relationship, the language ability corresponding to the evaluation scores of all users is used as the dependent variable, and the personal information of all users is used as the independent variable to establish a generalized linear mixed model; wherein the personal information of the user includes at least: age, education level and evaluation score;

[0017] The Monte Carlo maximum likelihood algorithm is used to fit the parameters and analyze the proportion of regression prediction parameters of the user's personal information to the user's language ability;

[0018] Inputting the personal information of the current user into the generalized linear mixed model to obtain prediction results of various language abilities of the user;

[0019] Taking a weighted average of the evaluation results of the various language abilities of the current user obtained through the evaluation and the predicted results of the various language abilities of the current user, so as to obtain the correction results of the various language abilities of the current user;

[0020] According to the correction results of the current user's various language abilities, the position of the correction results in the overall interactive task model is retrieved, so as to obtain the initial task difficulty and interaction time corresponding to the various language abilities of the current user;

[0021] Among them, in the overall interactive task model, for each language ability, different initial task difficulties and interaction durations are pre-set according to different language abilities.

[0022] Preferably, the interaction unit includes:

[0023] A speech interaction module presents a first interaction task based on Chinese character recognition and pronunciation and metalanguage training to the user, so as to train the user's speech awareness;

[0024] A morpheme interaction module presents a second interactive task of analyzing the structure and meaning of words in a language to the user, so as to train the user's morpheme awareness;

[0025] An orthographic interaction module presents a third interactive task of constructing and combining Chinese character structures, radicals, and radical positions to the user, so as to train the user's orthographic awareness;

[0026] The shape-sound mapping interaction module presents a fourth interaction task linking pronunciation with character shape to the user, so as to train the shape-sound mapping ability of the user.

[0027] Preferably, the first interactive task includes:

[0028] A syllable recognition task, which is used to play the pronunciation of letters to the user, and then a plurality of letter balls fall from the top of the screen, requiring the user to "catch" the ball with the same pronunciation as the played letter;

[0029] Pinyin recognition task, for playing the pronunciation of Chinese characters to the user, and then presenting a plurality of stimuli printed with pinyin on the screen, so as to require the user to click on the stimulus on the screen that has the same pronunciation as the played Chinese characters;

[0030] The pinyin spelling task is used to play the pronunciation of Chinese characters to the user, and then require the user to spell the pinyin of the played Chinese characters one by one in order on the screen.

[0031] Preferably, the second interactive task includes:

[0032] A word matching task is used to present a plurality of word bubbles to the user, each of which contains a same Chinese character, and the meaning of the Chinese character in at least one of the word bubbles is different from the meaning of the Chinese character in other word bubbles, so as to require the user to pop the word bubbles with different meanings;

[0033] A task of distinguishing similar characters is used to present a word composed of different similar characters to the user, so as to require the user to judge whether the word is composed of correct characters after seeing the word displayed on the screen;

[0034] The homophone identification task is used to present a word with pinyin to the user, and one word in the word is blank, so as to require the user to select a target option from multiple options of similar-sounding words and a target option.

[0035] Preferably, the third interactive task includes:

[0036] A four-sided character formation task is used to present Chinese character components to the user through a conveyor belt, so as to require the user to place the Chinese character components on the conveyor belt into four boxes above, below, left and right of the target component;

[0037] A Chinese character splitting task is used to present a given radical to the user, and require the user to combine the given radical with the components provided by the interface into a correct character; or present scrambled radicals and components to the user, and require the user to freely combine them into a correct character;

[0038] The Chinese character maze task is used to present a Chinese character maze to the user, requiring the user to find a suitable Chinese character to move the virtual image located at the starting point of the Chinese character maze until the virtual image successfully walks out of the Chinese character maze; wherein the suitable Chinese character represents a Chinese character with the same component.

[0039] Preferably, the fourth interactive task includes:

[0040] A literacy game task, for presenting a table of a*b to the user, with each grid having a corresponding Chinese character, and playing the pronunciation of a Chinese character to the user, so as to require the user to select a Chinese character with the same pronunciation as heard;

[0041] A Chinese character reading task is used to present a Chinese character table to the user and randomly light up a Chinese character, so as to require the user to read out the correct pronunciation according to the Chinese character lit up on the screen;

[0042] The task of finding homophones is used to present a paragraph of text to the user and play the pronunciation of a Chinese character, so as to require the user to find all Chinese characters with the same pronunciation as the played Chinese character from the paragraph of text.

[0043] Preferably, the feedback unit comprises:

[0044] A score feedback module, connected to the interaction unit, to provide the user with feedback on the score of each interaction task based on the behavioral response data of the user during human-computer interaction;

[0045] The wrong question feedback module is connected to the interaction unit to provide feedback to the user on the wrongly answered questions based on the behavioral response data of the user during human-computer interaction, and to review and repeat learning based on the wrongly answered questions.

[0046] According to a second aspect of an embodiment of the present invention, a human-computer interaction method for language ability of developmental dyslexia is provided, comprising the following steps:

[0047] Obtain the evaluation results of users' multiple language abilities;

[0048] Based on the evaluation results of the user's multiple language abilities, calculating the initial task difficulty corresponding to each language ability of the user;

[0049] Based on the initial task difficulty corresponding to each language ability of the user, performing human-computer interaction with the user using an interactive task of a corresponding difficulty level under a corresponding interactive module;

[0050] Behavior response data of the user during human-computer interaction is obtained, and the result of the human-computer interaction is fed back to the user based on the behavior response data.

[0051] Preferably, the human-computer interaction method further comprises:

[0052] If the user is a new user, then based on the evaluation results of each language ability of the new user, the initial task difficulty corresponding to each language ability is calculated respectively;

[0053] If the user is a historical user, based on the result of the previous human-computer interaction of the historical user, the language ability levels at the end of the previous human-computer interaction are obtained, and then the position of each language ability level in the overall model of the interaction task is retrieved, so as to obtain the initial task difficulty and interaction time of each language ability of the historical user in this human-computer interaction.

[0054] Compared with the prior art, the present invention has the following technical effects:

[0055] (1) Language training tasks and framework system adapted to children with dyslexia: Different from the existing single language training such as literacy, reading and writing on the market, the present invention starts from the basic structure of language ability, improves multiple basic abilities such as morpheme awareness, phonetic awareness, orthographic awareness and shape-sound mapping, and adopts a standardized and normalized evaluation paradigm, so that the intervention effect can be accurately evaluated.

[0056] (2) Gamification adaptation to enhance user experience: All interactive tasks are professionally designed to be gamified, which ensures their original effectiveness while increasing their fun, thus helping to improve children’s compliance during the training process.

[0057] (3) A difficulty system that adapts to user levels through preset algorithms: Different difficulty levels are set for each interactive task during the development process to widely adapt to users with different functional levels. Through preset algorithms, each user can receive personalized language cognition training that suits them to achieve the best training effect.

[0058] (4) The product is easily available and has no restrictions on usage scenarios: The language task unit provided by the present invention realizes the visualization of the training process and the freedom of training time and location. The scientific electronicization of the product enables the implementation of training tasks to save a lot of manpower and material resources, making it easier for parents and doctors to understand the children's situation in a timely manner, improving training efficiency, and making clinical intervention for dyslexia more convenient. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 A schematic diagram of the structure of a human-computer interaction system for developmental dyslexia language ability provided by the first embodiment of the present invention;

[0060] Figure 2 This is a schematic diagram of the structure of an interaction unit in the first embodiment of the present invention;

[0061] Figure 3 This is a flow chart of a human-computer interaction method for language ability of developmental dyslexia provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0062] The technical content of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0063] The embodiment of the present invention aims to develop a linguistic cognitive intervention system that uses language interaction tasks to improve the cognitive abilities of users (preferably children with dyslexia, but not limited to this, and may also include elderly people with cognitive impairments, etc.), including intervention contents such as morpheme awareness, phonetic awareness, orthographic awareness, and form-sound mapping. In the embodiment of the present invention, a systematic evaluation tool is formulated according to the characteristics of Chinese, the pertinence of interactive tasks is improved, and the impact of linguistic cognition on reading is fundamentally improved. In addition, electronic and gamified interactive tasks are used to improve user compliance and meet the needs of parents and doctors for remote intervention. In addition, through big data algorithms, the most effective personalized interaction plan is matched according to the actual situation of each user to achieve personalized intervention for each user.

[0064] First embodiment

[0065] like Figure 1 As shown, a human-computer interaction system for language ability of developmental dyslexia provided by the first embodiment of the present invention includes an evaluation unit 1, a calculation unit 2, an interaction unit 3, a difficulty adjustment unit 4 and a feedback unit 5. Among them, the evaluation unit 1 is used to evaluate multiple language abilities of the user. The calculation unit 2 is connected to the evaluation unit 1 to calculate the initial task difficulty corresponding to each language ability for the user according to the evaluation results of the user. The interaction unit 3 is connected to the calculation unit 2 and has multiple interaction modules, and each interaction module includes multiple interactive tasks of difficulty levels; to cooperate with the calculation unit 2 to form a personalized interaction plan for the user, so as to be used for human-computer interaction with the user. The difficulty adjustment unit 4 is connected to the interaction unit 3 to dynamically adjust the difficulty level of each interactive task. The feedback unit 5 is connected to the interaction unit 3 to feedback the results of this human-computer interaction to the user.

[0066] Below, each functional unit of the human-computer interaction system is described in detail:

[0067] 1. Evaluation Unit

[0068] In this embodiment, the evaluation unit 1 is preset with multiple evaluation tasks for evaluating the user's multiple language abilities, wherein the multiple language abilities at least include: phonetic awareness, morpheme awareness, orthographic awareness and form-sound mapping.

[0069] When a user logs into the system, personal information, including age, gender, education level, etc., needs to be collected and stored in the user database to generate a unique user portrait. In addition, different language abilities are evaluated based on the various evaluation tasks in the evaluation unit 1, thereby obtaining the evaluation results of the user's various language abilities.

[0070] 2. Computing Unit

[0071] In this embodiment, the calculation unit 2 is connected to the evaluation unit 1 and has a preset algorithm built in, so as to calculate the initial task difficulty corresponding to each language ability based on the evaluation results of each language ability of the user.

[0072] Specifically, the preset algorithm includes the following steps:

[0073] (1) The evaluation results of all current users in the human-computer interaction system are converted into standard scores according to the norms to obtain the mapping relationship between the evaluation scores of all users and their language abilities.

[0074] (2) Based on this mapping relationship, the language ability corresponding to the evaluation scores of all users is used as the dependent variable, and the personal information of all users is used as the independent variable to establish a generalized linear mixed model. The personal information of users includes at least: age, education level and evaluation score, etc.

[0075] (3) Using the Monte Carlo maximum likelihood algorithm to fit parameters, analyze and obtain the regression prediction parameter ratio of the user's personal information to the user's language ability. For example: obtain the regression prediction parameter ratio of the user's age, education level, and current assessment score to the user's cognitive ability.

[0076] (4) Input the current user's personal information into the generalized linear mixed model to obtain the prediction results of the user's various language abilities.

[0077] (5) The evaluation results of the various language abilities of the current user obtained through the evaluation are weighted averaged with the prediction results of the various language abilities of the current user to obtain the correction results of the various language abilities of the current user. In this embodiment, the weighted weight of the prediction result is n, where n represents the ratio of a single user to all users.

[0078] (6) According to the correction results of the current user's various language abilities, the position of the correction results in the overall interactive task model is retrieved, thereby obtaining the initial task difficulty and interaction time corresponding to each language ability of the current user. In the overall interactive task model, different initial task difficulties and interaction times are pre-set for each language ability according to different language abilities.

[0079] In addition, it should be understood that the above calculation process is only for new users. If the current user is a historical user, the above calculation process is omitted, and the user's language ability levels obtained at the end of the last human-computer interaction are inherited; then, based on the position of each language ability level in the overall model of the interactive task, the corresponding task difficulty level and training duration are obtained as the initial task difficulty and training duration of the user's current training.

[0080] It is understandable that, in this embodiment, the calculation unit 2 performs separate calculations for each language ability, thereby obtaining the initial task difficulty (which may be equal or unequal) corresponding to each language ability according to different language ability levels, so as to adapt to each language ability of the user. Thus, in subsequent human-computer interaction, a personalized human-computer interaction plan can be formed to improve the human-computer interaction effect.

[0081] 3. Interaction Unit

[0082] In this embodiment, the interaction unit 3 includes multiple interaction modules, and the multiple interaction modules correspond to multiple language abilities one by one. Each interaction module includes multiple interaction tasks, and each interaction task is preset with multiple difficulty levels. In addition, the interaction unit 3 is connected to the calculation unit 2 to perform human-computer interaction with the user based on the interaction tasks of the corresponding difficulty level under the corresponding interaction module (i.e., personalized human-computer interaction scheme) according to the initial task difficulty corresponding to each language ability of the user.

[0083] Specifically, Figure 2 As shown, the interaction unit 3 specifically includes the following four interaction modules:

[0084] (1) Voice interaction module

[0085] Phonological awareness can be broadly understood as sound recognition, that is, recognizing the structure of sound combinations in a language, including being aware of words, syllables, initial sounds and final sounds in a language. The voice interaction module 31 presents the user with a first interactive task based on Chinese character recognition and pronunciation and metalanguage training to train the user's phonetic awareness, which can effectively improve the sound recognition ability of children with dyslexia, thereby improving reading difficulties.

[0086] Specifically, the first interactive task includes the following three task types:

[0087] ① Syllable recognition task, which is used to play the pronunciation of letters to the user, and then multiple letter balls fall from the top of the screen, requiring the user to "catch" the ball with the same pronunciation as the played letter.

[0088] ② Pinyin recognition task, which is used to play the pronunciation of Chinese characters to the user, and then present multiple stimuli printed with pinyin on the screen (for example, potion bottles), requiring the user to click on the stimulus on the screen that has the same pronunciation as the played Chinese characters (used to make potions).

[0089] ③ Pinyin spelling task, which is used to play the pronunciation of Chinese characters to the user, and then require the user to spell the pinyin of the played Chinese characters one by one in order on the screen, for example: select the initial consonant, medial vowel, final vowel, etc. in order.

[0090] (2) Morpheme Interaction Module

[0091] Morpheme awareness is the ability to analyze the structure and meaning of words in a language. During reading, morpheme awareness can convey semantic, phonetic and syntactic information and is an important way to understand meaning. The morpheme interaction module 32 presents the user with a second interactive task of analyzing the structure and meaning of words in a language to train the user's morpheme awareness.

[0092] Specifically, the second interactive task includes the following three task types:

[0093] ① The word matching task is used to present multiple word bubbles to the user. Each word bubble contains an identical Chinese character, and the meaning of the Chinese character in at least one word bubble is different from that in other word bubbles, so that the user is required to break the word bubbles with different meanings so that the remaining word bubbles can be merged together.

[0094] ② The task of distinguishing similar characters is used to present a word composed of different similar characters to the user, such as "pingguo". The user is required to judge whether the word is composed of correct characters after seeing the word displayed on the screen. This training can help users further accurately master the learned similar morphemes, and gradually become familiar with the unlearned similar morphemes during the game process until they master them, so as to achieve the effect of "learning by practicing".

[0095] ③ Homophone identification task, which is used to present a word with pinyin to the user (for example: a basketball hoop with pinyin), and one of the words in the word is blank, so that the user is required to select a target option from multiple options of similar-sounding words and a target option (for example: throw a basketball with the correct word printed on it into the hoop).

[0096] (3) Orthographic Interaction Module,

[0097] Chinese orthography refers to children's understanding of the rules for constructing and combining Chinese characters, including Chinese character structure, radicals, and radical positions. The orthography interaction module 33 presents the user with a third interactive task of constructing and combining Chinese character structure, radicals, and radical positions, so as to train the user's orthography awareness, thereby improving the user's Chinese character writing ability, text comprehension, and reading comprehension ability.

[0098] Specifically, the third interactive task includes the following three task types:

[0099] ① The four-sided character combination task is used to present Chinese character components to users through a conveyor belt, requiring users to place the Chinese character components on the conveyor belt into the four boxes above, below, left, and right of the target component. This task enhances users' familiarity and manipulation ability of components by combining Chinese character components, and improves the writing ability and orthographic awareness of children with dyslexia.

[0100] ② Chinese character splitting task, this training task is aimed at the structure of Chinese character components, and aims to enhance the user's ability to recognize and spell Chinese character components. The primary version of the training task will give the radical, and the user needs to combine it with the components provided in the interface to form a correct word. The advanced version of the training task will shuffle the radical and components and present them together in the interface, and the user can freely combine them into the correct word.

[0101] ③ Chinese character maze task, this task aims to improve the user's overall perception of the character shape and the ability to recognize the detailed information of the character shape by distinguishing the character shape. After the training task begins, the user needs to find the appropriate Chinese character to move the little man according to the requirements. The appropriate Chinese character is the word with the same component, such as file, tree, and power. If the choice is correct, the little man will move to the next square, and if it is wrong, it will be blocked. Continuous operation will allow the little man to successfully walk out of the Chinese character maze.

[0102] (4) Shape-sound mapping interaction module

[0103] The training purpose of this module is to enhance the connection between form and sound through the practice of activating the form of the characters through voice, improve the speed of children using Chinese characters, and improve the problem of dictation difficulties for children with reading disabilities. The form-sound mapping interaction module 34 presents the fourth interactive task of connecting voice with the form of the characters to the user, so as to train the form-sound mapping ability of the user.

[0104] Specifically, the fourth interactive task includes the following three task types:

[0105] ① The word recognition game task is used to present a table of a*b to the user, with each grid having a corresponding Chinese character, and play the pronunciation of a Chinese character to the user, requiring the user to select the Chinese character with the same pronunciation as heard;

[0106] ② Chinese character reading task, which is used to present a Chinese character table to the user and randomly light up a Chinese character, requiring the user to read the correct pronunciation according to the Chinese character lit up on the screen. This task requires children to quickly and accurately say the pronunciation of a series of Chinese characters to improve children's basic language processing ability.

[0107] ③ Homophone search task, which is used to present a paragraph of text to the user and play the pronunciation of a Chinese character, and ask the user to find all the Chinese characters with the same pronunciation as the played Chinese character from the paragraph of text. This task can enhance children's ability to correspond to shape and sound and improve reading speed by training them to find homophones.

[0108] 4. Difficulty Adjustment Unit

[0109] In this embodiment, the difficulty adjustment unit 4 is connected to the interaction unit 3 to dynamically adjust the difficulty of each interactive task during the process of human-computer interaction performed by the user.

[0110] Specifically, during the human-computer interaction process, for each interactive task, if the user answers question a correctly continuously, the task difficulty increases by one level; if the user answers question b incorrectly, the task difficulty decreases by one level. Wherein, a and b are both positive integers, and the specific values ​​can be adaptively set according to needs, for example: a=2, b=1.

[0111] It can be understood that the difficulty adjustment unit 4 has the same adjustment rules for each interactive task, but based on the different behavioral response data of the user in each interactive task, the final adjustment result for each interactive task is also different, so that the task difficulty of each interactive task matches the user's language ability in all aspects.

[0112] 5. Feedback Unit

[0113] In this embodiment, the feedback unit 5 is connected to the interaction unit 3 to feed back the result of the human-computer interaction to the user based on the behavioral response data of the user during the human-computer interaction.

[0114] Specifically, the feedback unit 5 includes a score feedback module 51 and a wrong question feedback module 52. The score feedback module 51 is connected to the interaction unit 3 to provide feedback to the user on the score of each interactive task based on the user's behavioral response data during human-computer interaction. In addition, the wrong question feedback module 52 is connected to the interaction unit 3 to provide feedback to the user on the wrongly answered questions based on the user's behavioral response data during human-computer interaction, and review and repeat learning based on the wrongly answered questions, for example: the user can click on the wrongly answered questions and follow the voice prompts.

[0115] Second embodiment

[0116] like Figure 3 As shown, based on the above first embodiment, the second embodiment of the present invention provides a human-computer interaction method for language ability of developmental dyslexia, which specifically includes steps S1 to S4:

[0117] S1: Obtain the evaluation results of the user's multiple language abilities.

[0118] Specifically, when a user logs into the system, personal information, including age, gender, education level, etc., needs to be collected and stored in the user database to generate a unique user portrait. In addition, different language abilities are evaluated based on the various evaluation tasks in the evaluation unit 1, thereby obtaining the evaluation results of the user's various language abilities.

[0119] S2: Based on the evaluation results of the user's multiple language abilities, calculate the initial task difficulty corresponding to each language ability of the user.

[0120] Specifically, if the user is a new user, the calculation unit 2 performs calculations using a preset algorithm to obtain the initial task difficulty corresponding to each language ability of the new user. The specific calculation process refers to the content in the first embodiment and will not be repeated here.

[0121] If the user is a historical user, the calculation process is omitted and the user's language ability levels obtained at the end of the last human-computer interaction are inherited. Then, based on the position of each language ability level in the overall model of the interactive task, the corresponding task difficulty level and training duration are obtained as the initial task difficulty and training duration of the user's current training.

[0122] S3: Based on the initial task difficulty corresponding to the user's language abilities, human-computer interaction is performed with the user using interactive tasks of corresponding difficulty levels under the corresponding interactive modules.

[0123] Specifically, after obtaining the initial task difficulty corresponding to each language ability of the user based on step S2, the interactive tasks of corresponding difficulty are matched from each interactive module of the interactive unit 3, thereby forming a personalized interactive plan for the user. And, human-computer interaction is performed with the user based on the personalized interactive plan. Among them, the training process of each interactive task refers to the content in the first embodiment, and will not be repeated here.

[0124] In addition, in this embodiment, during the human-computer interaction of the user, the difficulty of each interactive task needs to be dynamically adjusted by the difficulty adjustment unit 4 to adapt to the language ability level of the user in all aspects. When adjusting specifically, the difficulty is adjusted according to a increase and b decrease, that is, if a question is answered correctly continuously, the task difficulty increases by one level; if b question is answered incorrectly, the task difficulty decreases by one level.

[0125] S4: Obtaining behavioral response data of the user during human-computer interaction, and feeding back the result of this human-computer interaction to the user based on the behavioral response data.

[0126] In this embodiment, step S4 is divided into two parts. One part is the feedback on the score of each task; the other part is the review and repeated learning of the language knowledge such as letters, pinyin, Chinese characters, etc. that the user answered incorrectly during the task.

[0127] In summary, the human-computer interaction system and method for improving language ability of developmental dyslexia provided by the embodiments of the present invention have the following beneficial effects:

[0128] (1) Language training tasks and framework system adapted to children with dyslexia: Different from the existing single language training such as literacy, reading and writing on the market, the present invention starts from the basic structure of language ability, improves multiple basic abilities such as morpheme awareness, phonetic awareness, orthographic awareness and shape-sound mapping, and adopts a standardized and normalized evaluation paradigm, so that the intervention effect can be accurately evaluated.

[0129] (2) Gamification adaptation to enhance user experience: All interactive tasks are professionally designed to be gamified, which ensures their original effectiveness while increasing their fun, thus helping to improve children’s compliance during the training process.

[0130] (3) A difficulty system that adapts to user levels through preset algorithms: Different difficulty levels are set for each interactive task during the development process to widely adapt to users with different functional levels. Through preset algorithms, each user can receive personalized language cognition training that suits them to achieve the best training effect.

[0131] (4) The product is easily available and has no restrictions on usage scenarios: The language task unit provided by the present invention realizes the visualization of the training process and the freedom of training time and location. The scientific electronicization of the product enables the implementation of training tasks to save a lot of manpower and material resources, making it easier for parents and doctors to understand the children's situation in a timely manner, improving training efficiency, and making clinical intervention for dyslexia more convenient.

[0132] It should be noted that the above embodiments are only examples, and the technical solutions of the various embodiments can be combined, all within the protection scope of the present invention.

[0133] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0134] The above is a detailed description of the human-computer interaction system and method for developmental dyslexia language ability provided by the present invention. For those skilled in the art, any obvious changes made to it without departing from the essence of the present invention will constitute an infringement of the patent right of the present invention and will bear corresponding legal responsibilities.

Claims

1. A human-computer interaction system for developmental dyslexia language ability, characterized in that include: An evaluation unit is pre-set with a plurality of evaluation tasks for evaluating a user's multiple language abilities and obtaining evaluation results of each language ability; wherein the multiple language abilities at least include: phonetic awareness, morpheme awareness, orthographic awareness and form-sound mapping; A calculation unit, connected to the evaluation unit and having a preset algorithm built therein, for calculating the initial task difficulty corresponding to each language ability based on the evaluation results of each language ability of the user; The interaction unit includes a plurality of interaction modules, wherein the plurality of interaction modules correspond to the plurality of language abilities one by one, each of the interaction modules includes a plurality of interaction tasks, and each of the interaction tasks is preset with a plurality of difficulty levels; the interaction unit is connected to the calculation unit to perform human-computer interaction with the user based on the interaction tasks of the corresponding difficulty levels under the corresponding interaction modules according to the initial task difficulty corresponding to each language ability of the user; A difficulty adjustment unit connected to the interaction unit to dynamically adjust the difficulty of each interactive task during the human-computer interaction of the user; The feedback unit is connected to the interaction unit to feed back the result of the human-computer interaction to the user based on the behavioral response data of the user during the human-computer interaction.

2. The human-computer interaction system according to claim 1, characterized in that The preset algorithm includes the following steps: The evaluation results of all current users in the human-computer interaction system are converted into standard scores according to the norm, so as to obtain a mapping relationship between the evaluation scores of all users and the language ability; Based on the mapping relationship, the language ability corresponding to the evaluation scores of all users is used as the dependent variable, and the personal information of all users is used as the independent variable to establish a generalized linear mixed model; wherein the personal information of the user includes at least: age, education level and evaluation score; The Monte Carlo maximum likelihood algorithm is used to fit the parameters and analyze the proportion of regression prediction parameters of the user's personal information to the user's language ability; Inputting the personal information of the current user into the generalized linear mixed model to obtain prediction results of various language abilities of the user; Taking a weighted average of the evaluation results of the various language abilities of the current user obtained through the evaluation and the predicted results of the various language abilities of the current user, so as to obtain the correction results of the various language abilities of the current user; According to the correction results of the current user's various language abilities, the position of the correction results in the overall interactive task model is retrieved, so as to obtain the initial task difficulty and interaction time corresponding to the various language abilities of the current user; Among them, in the overall interactive task model, for each language ability, different initial task difficulties and interaction durations are pre-set according to different language abilities.

3. The human-computer interaction system according to claim 1, characterized in that The interaction unit comprises: A speech interaction module presents a first interaction task based on Chinese character recognition and pronunciation and metalanguage training to the user, so as to train the user's speech awareness; A morpheme interaction module, presenting a second interactive task of analyzing the structure and meaning of words in a language to the user, so as to train the user's morpheme awareness; An orthographic interaction module presents a third interactive task of constructing and combining Chinese character structures, radicals, and radical positions to the user, so as to train the user's orthographic awareness; The shape-sound mapping interaction module presents a fourth interaction task linking pronunciation with character shape to the user, so as to train the shape-sound mapping ability of the user.

4. The human-computer interaction system according to claim 3, characterized in that The first interactive task includes: A syllable recognition task, which is used to play the pronunciation of letters to the user, and then a plurality of letter balls fall from the top of the screen, requiring the user to "catch" the ball with the same pronunciation as the played letter; Pinyin recognition task, for playing the pronunciation of Chinese characters to the user, and then presenting a plurality of stimuli printed with pinyin on the screen, so as to require the user to click on the stimulus on the screen that has the same pronunciation as the played Chinese characters; The pinyin spelling task is used to play the pronunciation of Chinese characters to the user, and then require the user to spell the pinyin of the played Chinese characters one by one in order on the screen.

5. The human-computer interaction system according to claim 3, characterized in that The second interactive task includes: A word matching task is used to present a plurality of word bubbles to the user, each of which contains a same Chinese character, and the meaning of the Chinese character in at least one of the word bubbles is different from the meaning of the Chinese character in other word bubbles, so as to require the user to pop the word bubbles with different meanings; A task of distinguishing similar characters is used to present a word composed of different similar characters to the user, so as to require the user to judge whether the word is composed of correct characters after seeing the word displayed on the screen; The homophone identification task is used to present a word with pinyin to the user, and one word in the word is blank, so as to require the user to select a target option from multiple options of similar-sounding words and a target option.

6. The human-computer interaction system according to claim 3, characterized in that The third interactive task includes: A four-sided character formation task is used to present Chinese character components to the user through a conveyor belt, so as to require the user to place the Chinese character components on the conveyor belt into four boxes above, below, left and right of the target component; A Chinese character splitting task is used to present a given radical to the user, and require the user to combine the given radical with the components provided by the interface into a correct character; or present scrambled radicals and components to the user, and require the user to freely combine them into a correct character; The Chinese character maze task is used to present a Chinese character maze to the user, requiring the user to find a suitable Chinese character to move the virtual image located at the starting point of the Chinese character maze until the virtual image successfully walks out of the Chinese character maze; wherein the suitable Chinese character represents a Chinese character with the same component.

7. The human-computer interaction system according to claim 1, characterized in that The fourth interactive task includes: A literacy game task, for presenting a table of a*b to the user, with each grid having a corresponding Chinese character, and playing the pronunciation of a Chinese character to the user, so as to require the user to select a Chinese character with the same pronunciation as heard; A Chinese character reading task is used to present a Chinese character table to the user and randomly light up a Chinese character, so as to require the user to read out the correct pronunciation according to the Chinese character lit up on the screen; The task of finding homophones is used to present a paragraph of text to the user and play the pronunciation of a Chinese character, so as to require the user to find all Chinese characters with the same pronunciation as the played Chinese character from the paragraph of text.

8. The human-computer interaction system according to claim 1, characterized in that The feedback unit comprises: A score feedback module, connected to the interaction unit, to provide the user with feedback on the score of each interaction task based on the behavioral response data of the user during human-computer interaction; The wrong question feedback module is connected to the interaction unit to provide feedback to the user on the wrongly answered questions based on the behavioral response data of the user during human-computer interaction, and to review and repeat learning based on the wrongly answered questions.

9. A human-computer interaction method for improving language ability of developmental dyslexic patients, implemented based on the human-computer interaction system according to any one of claims 1 to 8, characterized in that The steps include: Obtain the evaluation results of users' multiple language abilities; Based on the evaluation results of the user's multiple language abilities, calculating the initial task difficulty corresponding to each language ability of the user; Based on the initial task difficulty corresponding to each language ability of the user, performing human-computer interaction with the user using an interactive task of a corresponding difficulty level under a corresponding interactive module; Behavior response data of the user during human-computer interaction is obtained, and the result of the human-computer interaction is fed back to the user based on the behavior response data.

10. The human-computer interaction method according to claim 9, characterized in that Also includes: If the user is a new user, then based on the evaluation results of each language ability of the new user, the initial task difficulty corresponding to each language ability is calculated respectively; If the user is a historical user, based on the result of the previous human-computer interaction of the historical user, the language ability levels at the end of the previous human-computer interaction are obtained, and then the position of each language ability level in the overall model of the interaction task is retrieved, so as to obtain the initial task difficulty and interaction time of each language ability of the historical user in this human-computer interaction.

Images