Intelligent recommendation method, system and storage medium for individual education plan of autism user

Abstract

This application discloses an intelligent recommendation method, system, and storage medium for individualized education plans for autistic users. The method includes: collecting multimodal data of autistic users to determine their autism type; constructing a rule-based first recommendation model, which stores predefined rules for recommending individualized education plans and recommends plans based on these rules; constructing a case-based second recommendation model, which filters and recommends individualized education plans for similar users from a case resource library based on the autism type of the autistic user; if historical users with similar autism types exist in the case resource library, the second recommendation model is invoked to recommend individualized education plans; otherwise, the first recommendation model is invoked. This invention can intelligently and automatically generate individualized education plans that meet the needs of different autistic users.

Description

Technical Field

[0001] This application relates to the field of digital learning and recommendation technology, and more specifically, to a method, system, and storage medium for intelligent recommendation of individualized education plans for autistic users. Background Technology

[0002] Individualized Education Programs (IEPs) are the cornerstone of education for children with disabilities, ensuring that they receive education tailored to their specific needs within a mainstream educational environment. A systematic and scientifically sound IEP is the foundation and guarantee for personalized educational interventions, serving as a crucial basis and guide for their implementation. For children with autism, IEPs are a key link between developmental assessments and the implementation of personalized educational interventions, acting as a catalyst for effectively improving the quality and effectiveness of personalized education and interventions.

[0003] Existing research indicates that individualized education programs are crucial for the education of children with special needs and are an important guarantee for achieving personalized educational intervention for children with autism. However, due to the complex and time-consuming professional assessments required for the development and implementation of individualized education programs, there are problems such as strong reliance on experts, lack of learning resources and activities, and insufficient application of intelligent technologies. As a result, the development and implementation of individualized education programs for children with autism are plagued by irregularities, heavy workloads, difficulties in implementation, low effectiveness, and insufficient levels of intelligence. Furthermore, the heterogeneity of individual characteristics and developmental trajectories of autistic individuals urgently requires diverse individualized education programs, but traditional methods of developing individualized education programs are difficult to match the developmental levels of children with autism in various functional areas with intervention resources, and cannot meet the needs of personalized intervention. Summary of the Invention

[0004] To address at least one deficiency or improvement need in the prior art, the present invention provides a method, system, and storage medium for intelligent recommendation of individualized education plans for autistic users, which can intelligently and automatically generate individualized education plans that meet the needs of different autistic users.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a method for intelligent recommendation of individualized education plans for autistic users is provided, comprising: Collect multimodal data of autistic users, and determine the autism type of the autistic user based on the multimodal data; A rule-based first recommendation model is constructed. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based. Individual education plans are recommended according to the recommendation rules. A second recommendation model based on cases is constructed. In the second recommendation model, individual education plans are selected from the case ontology resource library for recommendation according to the autism type to which the autism user belongs. The case ontology resource library stores individual education plans corresponding to multiple historical users. If the case ontology resource library contains historical users whose similarity to the autism type is higher than a preset threshold, the second recommendation model is invoked to recommend individual education plans; otherwise, the first recommendation model is invoked to recommend individual education plans. Furthermore, the multimodal data includes prior data of the autistic user before participating in the intervention project, and real-time learning status data of the autistic user during the intervention project. The prior data includes basic information of the autistic user, developmental level data of each functional domain, and cognitive style type data. The real-time learning status data includes behavioral data, cognitive data, and physiological data of the autistic user during the intervention project.

[0006] Furthermore, determining the autism type of the autism user based on the multimodal data includes: Behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors are extracted from the behavioral data, cognitive data, and physiological data, respectively. The behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors are then fused and fed into a first classifier to obtain a first decision vector. The developmental level feature vector is extracted from the basic information of the autistic user and the developmental level data of each functional domain. The cognitive style feature vector is extracted from the cognitive style type data. The developmental level feature vector and the cognitive style feature vector are fused and then fed into the second classifier to obtain the second decision vector. The first decision vector and the second decision vector are weighted and fused to obtain the autism type of the autism user.

[0007] Furthermore, the construction of the case ontology resource library includes: Create three classes and class attributes: Autism Domain, Person Domain, and Intervention Activity Domain. The Autism Domain and Person Domain are used to describe the autism type of the historical user from two dimensions: the autism domain of the historical user and the basic information of the historical user, respectively. The Intervention Activity Domain is used to describe the corresponding individual education plan of the historical user. Based on historical users and their corresponding individual education plans, instances of the three classes—Autism Domain, Human Domain, and Intervention Activity Domain—and the index relationships between the instances are generated.

[0008] Furthermore, the autism domain includes attributes: autism definition, autism etiology, and autism symptoms, used to describe the autism domain of historical users from the three aspects of autism definition, autism etiology, and autism symptoms, respectively; The human domain includes attributes: human and population, which are used to describe the basic information of historical users based on their age, gender, and population affiliation, respectively. The intervention activity areas include attributes: cognition, language, application, imitation, emotional expression and social activities, and self-care ability, which are used to describe the specific content of the corresponding individualized education plan for historical users from the aspects of cognition, language, application, imitation, emotional expression and social activities, and self-care ability, respectively.

[0009] Furthermore, the step of recommending individual educational plans from the case ontology resource library based on the autism type of the autism user includes: Calculate the similarity between the autism type of the autism user and the autism type of each historical user in the case ontology resource library, and select the individual education plan corresponding to the historical user with the highest similarity for recommendation.

[0010] Furthermore, the intelligent recommendation method for individualized education plans for autistic users also includes: presenting the recommended individualized education plans to the autistic user, and supporting the autistic user to view, save, and print them at any time.

[0011] According to a second aspect of the present invention, an intelligent recommendation system for individualized education plans for autistic users is also provided, comprising: The assessment module is used to collect multimodal data of autistic users and determine the autism type of the autistic user based on the multimodal data. The first recommendation module is used to construct a rule-based first recommendation model. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based, and recommend individual education plans according to the recommendation rules. The second recommendation module is used to construct a case-based second recommendation model. In the second recommendation model, individual educational plans are selected from the case ontology resource library for recommendation based on the autism type of the autism user. The case ontology resource library stores individual educational plans corresponding to multiple historical users. The calling module is used to call the second recommendation model to recommend an individual education plan if there are historical users in the case ontology resource library whose similarity to the autism type is higher than a preset threshold; otherwise, it calls the first recommendation model to recommend an individual education plan.

[0012] Furthermore, the multimodal data includes prior data of the autistic user before participating in the intervention project, and real-time learning status data of the autistic user during the intervention project. The prior data includes basic information of the autistic user, developmental level data of each functional domain, and cognitive style type data. The real-time learning status data includes behavioral data, cognitive data, and physiological data of the autistic user during the intervention project.

[0013] According to a third aspect of the invention, a storage medium is also provided that stores a computer program executable by a processor, which, when run on the processor, causes the processor to perform the steps of any of the methods described above.

[0014] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects: (1) This invention overcomes the problems of traditional individual education plan formulation being heavy, difficult to implement, low in effectiveness, and lacking in intelligence. Based on in-depth research on the fit between the developmental scope of autistic children and intervention resources, it effectively organizes personalized intervention resources, integrates intelligent recommendation algorithms, and realizes intelligent recommendation of individual education plans, providing important basis and reference for personalized intervention and filling the gap in the field of intelligent customization of individual education plans for autism.

[0015] (2) This invention innovatively constructs an ontology resource library for the autism domain, which includes an autism domain knowledge base, a human domain, and an intervention activity domain. This provides rich digital intervention resources for the development of individualized education plans, overcoming the problem of insufficient resources faced by manually developing individualized education plans. Currently, there is no ontology knowledge base specifically for the autism domain, which is also a technical bottleneck restricting the application of intelligent recommendation technology in the field of autism intervention. The autism ontology knowledge base constructed by this method can provide necessary support for intelligent intervention.

[0016] (3) This invention proposes a hybrid recommendation method based on rules and cases, which intelligently recommends individualized education plans based on the category to which autistic children belong. This solves the problems of cold start and limited coverage in traditional recommendation systems, while also meeting the personalized and differentiated needs of autistic children. Specifically, the rule-based recommendation method can provide new users with individualized education plans that match their abilities and developmental levels based on rules formulated by experts; the case-based recommendation method can quickly recommend the same individualized education plans to similar users, thus improving the efficiency of the system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the intelligent recommendation process for individualized education plans for children with autism according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the intelligent recommendation system for individualized education plans for children with autism according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the autism feature fusion and clustering method according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the conceptual structure of the autism domain ontology library according to an embodiment of the present invention.

[0019] Figure 5 This is a schematic diagram of a hybrid recommendation model for individualized education plans for children with autism according to an embodiment of the present invention; Figure 6 This is a schematic diagram of a rule-based individualized education plan recommendation model according to an embodiment of the present invention; Figure 7 This is a schematic diagram of a case-based individualized education plan recommendation model according to an embodiment of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0021] The terms "first," "second," "third," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or modules is not limited to the listed steps or modules, but may optionally include steps or modules not listed, or may optionally include other steps or modules inherent to such processes, methods, products, or apparatus.

[0022] The flowchart of an intelligent recommendation method for individualized education plans for autistic users according to an embodiment of the present invention is as follows: Figure 1 As shown, the principle is as follows Figure 2As shown, the steps include: S101, Collect multimodal data of autistic users and determine the autism type of the autistic user based on the multimodal data.

[0023] Specifically, multimodal data from users can be collected through human-computer interaction to determine the user's autism category, providing effective input for the individualized education plan recommendation module. Autism users can be children with autism.

[0024] The human-computer interaction module can include registration and login, and human-computer interaction assessment tasks. After completing registration and login, users can further complete human-computer interaction assessment tasks, which include psychological test tasks such as joint attention assessment, facial expression recognition, facial expression identification, and facial expression-situation matching.

[0025] Furthermore, the multimodal data includes: prior data of autistic users before participating in the intervention program, and real-time learning status data of autistic users during the intervention program.

[0026] Furthermore, the prior data included: basic information about autistic users, developmental level data in various functional areas, and cognitive style type data. Specifically, the developmental levels of autistic children's abilities in cognition, language expression, language comprehension, fine motor skills, gross motor skills, and imitation were obtained using the PEP psychoeducational assessment scale; children's cognitive styles were obtained using a learning style questionnaire.

[0027] Furthermore, real-time learning status data includes: behavioral, cognitive, and physiological data of autistic users participating in intervention programs. Eye-tracking devices acquire fixation pattern data, including areas of interest, fixation points, and fixation duration; electroencephalogram (EEG) devices or skin conductance monitoring bracelets acquire physiological data, including alpha, beta, theta, and gamma brainwave data or skin conductance data, to determine the psychological state of autistic children during intervention; other basic information about the children, including age and gender, can be entered by the assessor. Simultaneously, the system's backend database updates the developmental status information of autistic children.

[0028] Furthermore, the aforementioned multimodal data serves as input information for autistic users, establishing a mapping between it and the backend intervention resource library.

[0029] Furthermore, such as Figure 3 As shown, machine learning algorithms are used to fuse the above data at the feature layer and decision layer to determine the category of autism children to which the user belongs.

[0030] Specifically, the process involves: extracting behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors from behavioral data, cognitive data, and physiological data, respectively; fusing these feature vectors and feeding them into classifier 1 to obtain the first decision vector; extracting developmental level feature vectors from the basic information of autistic users and developmental level data in each functional domain, and extracting cognitive style feature vectors from cognitive style type data; fusing these feature vectors and feeding them into classifier 2 to obtain the second decision vector; and performing weighted decision fusion on the first and second decision vectors to obtain the autism type to which the autistic user belongs.

[0031] In this case, the decision result of classifier 1 is D1 with weight w1, the decision result of classifier 2 is D2 with weight w2, and the final weighted decision fusion result of classifier 3 is D = D1*w1⊙D2*w2.

[0032] Specifically, one of the following deep learning algorithms can be used: recurrent neural networks (RNN), convolutional neural networks (CNN), or K-means algorithm, to extract behavioral data features such as the proportion of eye-tracking fixation points in interest areas and the proportion of fixation time; one of the following algorithms can be used: Bayes classification algorithm, support vector machine (SVM), or convolutional neural network (CNN) to collect facial expression features; E-prime can be used to collect cognitive data features such as reaction time and accuracy in facial expression recognition tasks; convolutional neural networks and long short-term memory models (CNN-LSTM) can be used to extract features from electroencephalogram (EEG) signals or electrodermal data; and SPSS analysis can be used to extract cognitive style data with statistically significant differences and features representing the development level of abilities in various domains.

[0033] The feature layer is fused with time-synchronized source data, and the sub-decision obtained from feature fusion is fused at the decision layer. The final identification result is obtained through decision fusion, which determines the category to which the autistic child user belongs.

[0034] S102, Construct a rule-based first recommendation model. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based. Individual education plans are recommended according to the recommendation rules.

[0035] Individualized education plans may include: intervention scope, sub-scope, learning focus, intervention goals, success criteria, intervention methods, number of sessions, start date, and completion date.

[0036] By analyzing individualized education plans for children with autism provided by experts, and based on the following questions, we construct recommendation rules and models for individualized education plans: a) Why should we conduct personalized interventions? This means answering the question of the meaning and value of interventions.

[0037] b) In which areas is intervention necessary? Determine the basic information of individuals with autism, their developmental level in each functional area, and their cognitive style type.

[0038] c) How to implement the intervention? This requires answering questions about the intervention method, intervention goals, intervention activities, intervention intensity, and intervention duration.

[0039] Through the analysis above, the necessary facts and recommendation rules for reasoning were obtained, and relevant information on autistic children who contributed to the reasoning was collected to establish an individualized education program recommendation model. The specific recommendation rules were formulated by experts based on general child development trajectories and autism intervention guidelines, taking into account the child's age and developmental level. In other words, the rules clearly define the expected ability levels for children of different ages and developmental stages, and provide candidate intervention programs for each domain.

[0040] S103. Construct a case-based second recommendation model. In the second recommendation model, individual educational plans are selected from the case ontology resource library for recommendation based on the autism type of the autism user. The case ontology resource library stores individual educational plans corresponding to multiple historical users.

[0041] The second recommendation model is primarily based on a case ontology resource library. This library records basic information about existing historical users and individual educational plans for each user.

[0042] The construction of the case ontology resource library includes: (1) creating three classes and class attributes: autism domain, human domain and intervention activity domain. The autism domain and human domain are used to describe the autism type of the historical user from the two dimensions of the historical user's autism domain and the historical user's basic information, respectively. The intervention activity domain is used to describe the corresponding individual education plan of the historical user; (2) generating instances of the three classes: autism domain, human domain and intervention activity domain and the index relationship between instances based on the historical user and his / her corresponding individual education plan.

[0043] like Figure 3 As shown, the case ontology resource library includes three categories: autism domain, human domain, and intervention activity domain. Each category also includes sub-level classes and class attributes.

[0044] In one embodiment, the autism domain includes attributes such as autism definition, autism etiology, and autism symptoms, used to describe the autism domain of the historical user from these three aspects, respectively. The person domain includes attributes such as person and population, used to describe the historical user's basic information from the perspectives of age, gender, and population affiliation. The intervention activity domain includes attributes such as cognition, language, application, imitation, emotional expression and social activities, and self-care ability, used to describe the corresponding individualized education plan for the historical user from the aspects of cognition, language, application, imitation, emotional expression and social activities, and self-care ability, respectively.

[0045] In one embodiment, the construction principle of the case ontology resource library is as follows. Due to the specialized and complex nature of knowledge in the autism domain, this embodiment of the invention adopts professional knowledge standards as the basic specifications for ontology construction, standardizing ontology concepts. The construction of the autism ontology resource library fills a gap in the field of intelligent recommendation for autism.

[0046] The first step is to define the professional fields and scope of the autism ontology. The professional fields included in the ontology knowledge base mainly involve the autism domain, the personal information domain, and the intervention activity domain.

[0047] The second step involves reusing existing ontologies. This invention builds an ontology library for the autism domain based on Freebase. Based on standard specifications or ontology knowledge from other domains, some ambiguous concepts in the Freebase ontology are revised and corrected to meet the requirements of clarity and objectivity in autism domain knowledge concepts.

[0048] The third step is to list the key terms in the autism ontology, including the definition, etiology, and core symptoms of autism.

[0049] The fourth step is to define the classes and their hierarchical relationships. A top-down approach is used, starting with general concepts in the autism domain and then refining those concepts.

[0050] Fifth, define the class attributes. Define the attributes for the autism domain, the person domain, and the intervention activity domain.

[0051] The sixth step is to define the object's attributes. Object attributes always connect individuals within a domain to individuals within a certain range, describing the attributes between different individuals. The ontology's framework structure is composed of classes, but the description of its internal relationships is constituted by the object attributes of those classes.

[0052] Step 7: Creating Instances. This invention is based on some Freebase instances and combines them with the needs of intelligent and personalized intervention to construct an ontology of individualized education programs in the field of autism. Each case's URI is uniquely labeled using specific encodings in Freebase, and the RDFS label attribute is used to determine the specific Chinese and English terminology of the instance and to establish the index relationship. Ontology instances have their own unique ID encoding; to retain this method, newly added ontology instances are also encoded using a similar method.

[0053] S104. If there are historical users with the same autism type in the case ontology resource library, call the second recommendation model to recommend individual education plans; otherwise, call the first recommendation model to recommend individual education plans. This step employs a hybrid recommendation method to intelligently recommend personalized individualized education plans for autistic users. This hybrid method includes rule-based and case-based recommendation approaches. The rule-based reasoning model for individualized education plan recommendation is a precise inference matching, while the case-based reasoning model is an improved personalized recommendation model based on collaborative filtering. It uses fuzzy reasoning when the rule-based model cannot match, serving as a solution and necessary supplement to situations where rule matching fails. It recommends individualized education plans for new users to similar users by querying a case library.

[0054] The case-based recommendation method described above is applicable to newly registered users or existing users.

[0055] Furthermore, such as Figure 5 As shown, for newly registered users, the system first searches the backend database based on assessment data to see if there are any historical users with similar autism types. If such users exist, a case-based recommendation method is used to generate an individualized education plan; otherwise, a rule-based recommendation method is used. This hybrid recommendation method, employing both rule-based and case-based reasoning, overcomes the cold-start and limited coverage issues in the field of intelligent recommendation, thus meeting the personalized needs of children with autism.

[0056] Similarity calculations are used to determine if there are historical users with similar autism types. The similarity calculation is as follows: The similarity between the autism type of the current autism user and the autism type of each historical user in the case ontology resource library is calculated. The individual educational plan corresponding to the historical user with the highest similarity is then recommended.

[0057] First, the user ontology for autistic children is compared with the case ontology. A semantic similarity calculation method is used to calculate the similarity value between the user ontology and each case ontology. The individual educational plan for the case ontology with the highest similarity to the user ontology is then recommended to the new user (see appendix). Figure 7 The similarity calculation formula is as follows:

[0058] Here, A represents a set of m users to be recommended, with the backend database updated each time a new user is added. B represents a set of n historical users in the case ontology, existing users stored in the backend database. By comparing the two sets, the highest similarity value is retained, and the average of these values ​​is the similarity value between the user ontology and the case ontology. sim(B,A) represents recommending an individualized education plan from the user with the highest similarity in set B to any user in set A.

[0059] Using the methods recommended above, we can intelligently recommend and generate personalized individual education plans for children with autism.

[0060] Furthermore, the intelligent recommendation method for individualized education plans for autistic users also includes S105: presenting the recommended individualized education plans to autistic users and supporting them to view, save, and print them at any time.

[0061] This module presents individualized intervention plans to users, allowing parents or assessment trainers to view, save, and print these plans at any time, providing a reference and basis for conducting personalized interventions on a daily basis.

[0062] The application scenarios of the intelligent recommendation method and system for individualized education plans for children with autism according to embodiments of the present invention are as follows.

[0063] Children with autism register and log in to the main control computer. The system's assessment module interface collects information on their developmental status and cognitive style. Developmental status information can be obtained using the traditional PEP scale, requiring assessment by a professional assessor, with the data then input into the system. Alternatively, the system can use a gamified version of the PEP scale, where children complete assessment games under the guidance of an assessor. The system backend records and provides data on developmental levels in each functional area, as well as gender and age. Guardians or teachers familiar with the child's life complete the FILS cognitive style online questionnaire, and the system backend records the data. Eye-tracking and EEG data are obtained through eye trackers and EEG devices deployed on the touchscreen all-in-one computer. The system backend uses a deep learning-based algorithm to classify children with autism. Based on the classification information, the system searches the autism domain ontology. First, if no similar users are found in the autism user model library, the system intelligently recommends individualized education plans based on rules established by experts. If similar users are found, a case-based recommendation model is used to recommend existing individualized education plans from the case library. Furthermore, individual educational plans are presented to system users, who can view, save, and print them for daily personalized educational interventions.

[0064] An intelligent recommendation system for individualized education plans for autistic users according to an embodiment of the present invention includes: The assessment module is used to collect multimodal data of autistic users and determine the type of autism the user belongs to based on the multimodal data; The first recommendation module is used to build a rule-based first recommendation model. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based, and recommend individual education plans according to the recommendation rules. The second recommendation module is used to build a case-based second recommendation model. The second recommendation model is used to select individual education plans from the case ontology resource library for recommendation based on the autism type of the autism user. The case ontology resource library stores individual education plans corresponding to multiple historical users. The module is used to call the second recommendation model to recommend individual education plans if there are historical users in the case ontology resource library whose similarity to the autism type is higher than a preset threshold; otherwise, it calls the first recommendation model to recommend individual education plans.

[0065] Furthermore, the multimodal data package includes prior data of autistic users before they participate in the intervention project, and real-time learning status data of autistic users during the intervention project. The prior data includes: basic information of autistic users, developmental level data of each functional domain, and cognitive style type data. The real-time learning status data includes: behavioral data, cognitive data, and physiological data of autistic users during the intervention project.

[0066] Furthermore, determining the autism type of autism users based on multimodal data includes: Behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors are extracted from behavioral data, cognitive data, and physiological data, respectively. The behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors are then fused and fed into the first classifier to obtain the first decision vector. The developmental level feature vector is extracted from the basic information of autistic users and the developmental level data of each functional domain. The cognitive style feature vector is extracted from the cognitive style type data. The developmental level feature vector and the cognitive style feature vector are fused and then fed into the second classifier to obtain the second decision vector. The first decision vector and the second decision vector are weighted and fused to obtain the autism type of the autism user.

[0067] Furthermore, the construction of the case ontology resource repository includes: Create three classes and their attributes: Autism Domain, Person Domain, and Intervention Activity Domain. The Autism Domain and Person Domain are used to describe the autism type of a historical user from two dimensions: the autism domain of the historical user and the basic information of the historical user, respectively. The Intervention Activity Domain is used to describe the corresponding individual education plan for the historical user. Based on historical users and their corresponding individual education plans, instances of three classes—Autism Domain, Human Domain, and Intervention Activity Domain—are generated, along with the index relationships between these instances.

[0068] The intelligent recommendation system for individualized education plans for autistic users also includes a presentation module, which presents the recommended individualized education plans to autistic users and allows them to view, save, and print them at any time.

[0069] The implementation principle and technical effects of the intelligent recommendation system for individualized education plans for autistic users are the same as those of the intelligent recommendation method for individualized education plans for autistic users, and will not be repeated here.

[0070] This application also provides a storage medium storing a computer program executable by a processor. When the computer program runs on the processor, it causes the processor to perform any of the steps of the aforementioned intelligent recommendation method for individualized education plans for autistic users. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0071] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0072] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0073] In the embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual coupling or direct coupling or communication connection may be through some service interfaces; the indirect coupling or communication connection of the system or modules may be electrical or other forms.

[0074] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0075] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0076] If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0077] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, which may include: a flash drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0078] The foregoing description is merely an exemplary embodiment of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Those skilled in the art will readily conceive of embodiments of this disclosure upon considering the specification and practicing the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described herein. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

[0079] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0080] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for intelligently recommending individualized education plans for autistic users, characterized in that, include: Multimodal data of autistic users is collected, including prior data of the autistic users before participating in the intervention program and real-time learning status data of the autistic users during the intervention program. The prior data includes basic information of the autistic users, developmental level data in each functional domain, and cognitive style type data. The real-time learning status data includes behavioral data, cognitive data, and physiological data of the autistic users during the intervention program. Determining the autism type of the autism user based on the multimodal data specifically includes: extracting behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors from the behavioral data, cognitive data, and physiological data, respectively; fusing the behavioral data feature vectors, cognitive data feature vectors, and physiological data feature vectors; and then feeding them into a first classifier to obtain a first decision vector. The developmental level feature vector is extracted from the basic information of the autistic user and the developmental level data of each functional domain. The cognitive style feature vector is extracted from the cognitive style type data. The developmental level feature vector and the cognitive style feature vector are fused and then fed into the second classifier to obtain the second decision vector. The decision result of the first classifier is D1 with weight w1, and the decision result of the second classifier is D2 with weight w2. The first decision vector D1 and the second decision vector D2 are weighted and fused to obtain the weighted decision fusion result D = D1*w1 ⊙ D2*w2 of the third classifier, thereby determining the autism type of the autism user. A rule-based first recommendation model is constructed. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based. Individual education plans are recommended according to the recommendation rules. A second recommendation model based on cases is constructed. In the second recommendation model, individual education plans are selected from the case ontology resource library for recommendation according to the autism type to which the autism user belongs. The case ontology resource library stores individual education plans corresponding to multiple historical users. If the case ontology resource library contains historical users whose similarity to the autism type is higher than a preset threshold, the second recommendation model is invoked to recommend individual education plans; otherwise, the first recommendation model is invoked to recommend individual education plans.

2. The intelligent recommendation method for individualized education plans for autistic users as described in claim 1, characterized in that, The construction of the case ontology resource library includes: Create three classes and class attributes: Autism Domain, Person Domain, and Intervention Activity Domain. The Autism Domain and Person Domain are used to describe the autism type of the historical user from two dimensions: the autism domain of the historical user and the basic information of the historical user, respectively. The Intervention Activity Domain is used to describe the corresponding individual education plan of the historical user. Based on historical users and their corresponding individual education plans, instances of the three classes—Autism Domain, Human Domain, and Intervention Activity Domain—and the index relationships between the instances are generated.

3. The intelligent recommendation method for individualized education plans for autistic users as described in claim 2, characterized in that, The autism domain includes attributes: autism definition, autism etiology, and autism symptoms, which are used to describe the autism domain of historical users from the three aspects of autism definition, autism etiology, and autism symptoms, respectively. The human domain includes attributes: human and population, which are used to describe the basic information of historical users from the perspectives of age, gender, and population affiliation, respectively. The intervention activity areas include attributes: cognition, language, application, imitation, emotional expression and social activities, and self-care ability, which are used to describe the specific content of the corresponding individualized education plan for historical users from the aspects of cognition, language, application, imitation, emotional expression and social activities, and self-care ability, respectively.

4. The intelligent recommendation method for individualized education plans for autistic users as described in claim 1, characterized in that, The step of selecting and recommending individual educational plans from the case ontology resource library based on the autism type of the autism user includes: Calculate the similarity between the autism type of the autism user and the autism type of each historical user in the case ontology resource library, and select the individual education plan corresponding to the historical user with the highest similarity for recommendation.

5. The intelligent recommendation method for individualized education plans for autistic users as described in claim 4, characterized in that, Also includes: The system will present the individualized education plan recommended for the autistic user to the autistic user, and allow the autistic user to view, save, and print it at any time.

6. An intelligent recommendation system for individualized education plans for autistic users, characterized in that: The steps for performing the intelligent recommendation method for individualized education plans for autistic users as described in any one of claims 1 to 5 include: The assessment module is used to collect multimodal data of autistic users and determine the autism type of the autistic user based on the multimodal data. The first recommendation module is used to construct a rule-based first recommendation model. The first recommendation model is used to store predefined individual education plan recommendation rules. The recommendation rules define the facts and recommendation judgment rules on which the recommendation is based, and recommend individual education plans according to the recommendation rules. The second recommendation module is used to construct a case-based second recommendation model. In the second recommendation model, individual educational plans are selected from the case ontology resource library for recommendation based on the autism type of the autism user. The case ontology resource library stores individual educational plans corresponding to multiple historical users. The calling module is used to call the second recommendation model to recommend an individual education plan if there are historical users in the case ontology resource library whose similarity to the autism type is higher than a preset threshold; otherwise, it calls the first recommendation model to recommend an individual education plan.

7. A storage medium, characterized in that, It stores a computer program that, when run on a processor, causes the processor to perform the steps of the method according to any one of claims 1 to 5.

Images