Description:FIELD OF THE INVENTION
This invention relates to An Edge-to-Cloud AI Framework for Real-Time Multimodal Assessment and Enhancement of Critical Thinking Skills in Education
BACKGROUND OF THE INVENTION
Educational system gives huge emphasise on cultivating critical thinking among students for better academic performance and career growth. But traditional evaluation methods such as written assessment or individual teacher assessment fail to capture the actual behavioural and cognitive indicators that reflect learner’s thinking capabilities. So, there is a need of an smart intelligent system that can analyse multimodal data collected from learner interaction behaviour, speech pattern, facial expression and provides real time feedback. However, the current AI solutions are based on text-based inputs or do not use edge computing with dynamic data. This gap results in delay feedback, no personalisation and superficial knowledge about learners critical thinking skills. Therefore, there is a need of AI driven sensor based scalable system that can automatically collect the behavioural and verbal data, analyse and assess the critical thinking of learner in an ethical manner.
Critical thinking is competent skill of 21st century required for problem solving, decision making and innovation. Still the current education system does not provide an effective system to evaluate and cultivate critical thinking skill in learners. They are dependent on traditional methods of written examination, essays, verbal discussions and results are solely depends on course and teacher discretion, which has numerous limitations. These assessments suffer from inconsistencies, hard to scale and biasness. Timely feedback and personalized learning environment are the limitation of traditional system.
A new possibility of personalized and automated learning has been opened with integration of Artificial intelligence (AI) in education. Most AI assisted technology in education system are based on Information recall or content-based suggestion. But there are very few AI-assisted system to evaluate critical thinking in learner by understanding and analyzing the complex cognitive skills. Available AI models are mostly depended on NLP based written text analysis. though they are useful, but they overlook the behavioral and emotional cues that are of utmost important to assess the critical thinking skill.
There is lack of research in assessment of critical thinking using AI integrated multimodal data like facial expression, eye tracking, speech pattern recognition and learner’s communication. Current AI tools do not use sensor collected data from real time environment which limits a deeper insight in assessing cognitive abilities.
Most AI learning models works only on cloud leading to inflexibility, data privacy concerns, latency issues in real time or remote classrooms. Edge computing collects the data from local devices and processes the raw data before sending to cloud for more deeper analysis, is still not explored in context to critical thinking assessment. The edge computing integrated AI system promises many benefits like higher speed, privacy, and situation awareness, but is not yet adopted or tested in critical thinking assessment.
Ethical and transparent use of AI model assessing cognitive skill is another concern. Also, whether the framework can provide understandable, fair and educational aligned interpretation and assessment is another concern. Educators and learners if not trained well, may mistrust or misinterpret the AI generated feedback.
In conclusion, Significant research gap exists in sensor based multimodal, edge to cloud AI technology for real time assessment of critical thinking. Filling this gap can provide a transformative, powerful, scalable educational solution for cultivating critical thinking among learners.
US11551570B2 A skills learning method for a student gathers objective data relating to the student in response to various stimuli, and produces a predicted feedback units as a function of the objective data using a machine learning-base classifier. The method can include training a neural network using objective data of student interactions and associated subjective assessments of a skill of each objective data. The method includes receiving a new dataset with objective data of a new student and an associated subjective assessment of a skill of the first student represented by the new objective data. A predicted assessment of the skill of the new objective data is calculated by inputting the new objective data into the neural network. The method can include updating the neural network by combining the initial dataset and the new dataset and recompiling the neural network to fit the model dataset based on a learning algorithm.
RESEARCH GAP: The proposed system differs by utilizing real-time multimodal data (audio and video) collected through Raspberry Pi-based IoT devices to assess critical thinking skills, whereas the other patent relies on static datasets and subjective feedback for general skill prediction.
US20200302296A1 The present invention is directed, in one particular implementation, to a cloud computing-based categorization system that comprises at least one electronic database having one or more performance assessment data associated with a plurality of entities matriculated at one or more educational institutions. The system further includes a processor, communicatively coupled to the at least one database, and configured to execute an electronic process that analyzes and converts said performance assessment data. Through one or more modules, the processor is configured to select performance assessment data corresponding to at least one structured assessment data value; and at least one unstructured assessment data set for an individual and evaluate the structured and un-structed data of the individual using an assessment model configured to classify the entity into one of a plurality of assessment categories. The processor is further configured by one or more modules to generate a graphical representation, for display and output to one or more remote users, of the likelihood that the individual is assigned to one of the plurality of assessment categories.
RESEARCH GAP: The proposed system differs by actively capturing real-time behavioral and verbal data using Raspberry Pi-based audio-visual sensors to assess critical thinking skills, whereas the other invention focuses on analyzing pre-existing structured and unstructured academic performance data for category classification.
None of the prior art indicate above either alone or in combination with one another disclose what the present invention has disclosed. This invention relates to An Edge-to-Cloud AI Framework for Real-Time Multimodal Assessment and Enhancement of Critical Thinking Skills in Education
SUMMARY OF THE INVENTION
This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the invention.
This summary is neither intended to identify key or essential inventive concepts of the invention and nor is it intended for determining the scope of the invention.
To further clarify advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which is illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail with the accompanying drawings.
The proposed solution is a sensor integrated edge to cloud computing-based AI solution designed to measure and improve critical thinking skills of learner in educational context. This system uses variety of hardware and software components to collect the audio, video and behavioural data from the learner, analyse it to measure the critical thinking index and proposes AI generated feedback and suggestion to enhance it. The core component of this system are HD cameras, microphones and optional touch panels. These devices collect the real time data such as facial expression, eye movements, voice patterns and learner interaction, which provides a deep insight into learner cognitive engagement, attention, reasoning and analytical skills. The cameras are used to capture the nonverbal cues like eye movement, eye brow movements and rest of facial expression associated with various moods of person like curiosity, confusion etc. the microphone collect the verbal data of learner like speech pattern, pauses, tonality, exclamatory remarks to assess the state of mind like thought clarity or logical analysis. An optional touch panel captures the way learner interact with the environment. It captures hesitation while writing, typing patterns and response time. The preprocessing of initial data is done through embedded computing unit such as raspberry pi. This initial preprocessing includes facial expression, voice filtering, and timestamping of event to reduce latency and ensure the identity protection of learner by anonymizing information which is to be transmitted to cloud via Wi-Fi.
BRIEF DESCRIPTION OF THE DRAWINGS
The illustrated embodiments of the subject matter will be understood by reference to the drawings, wherein like parts are designated by like numerals throughout. The following description is intended only by way of example, and simply illustrates certain selected embodiments of devices, systems, and methods that are consistent with the subject matter as claimed herein, wherein:
FIGURE 1: OVERALL ARCHITECTURE
FIGURE 2: EDGE VISION DATA COLLECTOR DEVICE
The figures depict embodiments of the present subject matter for the purposes of illustration only. A person skilled in the art will easily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the disclosure described herein.
DETAILED DESCRIPTION OF THE INVENTION
The detailed description of various exemplary embodiments of the disclosure is described herein with reference to the accompanying drawings. It should be noted that the embodiments are described herein in such details as to clearly communicate the disclosure. However, the amount of details provided herein is not intended to limit the anticipated variations of embodiments; on the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the present disclosure as defined by the appended claims.
It is also to be understood that various arrangements may be devised that, although not explicitly described or shown herein, embody the principles of the present disclosure. Moreover, all statements herein reciting principles, aspects, and embodiments of the present disclosure, as well as specific examples, are intended to encompass equivalents thereof.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments. As used herein, the singular forms “a",” “an” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises,” “comprising,” “includes” and/or “including,” when used herein, specify the presence of stated features, integers, steps, operations, elements and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groups thereof.
It should also be noted that in some alternative implementations, the functions/acts noted may occur out of the order noted in the figures. For example, two figures shown in succession may, in fact, be executed concurrently or may sometimes be executed in the reverse order, depending upon the functionality/acts involved.
In addition, the descriptions of "first", "second", “third”, and the like in the present invention are used for the purpose of description only, and are not to be construed as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Thus, features defining "first" and "second" may include at least one of the features, either explicitly or implicitly.
Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which example embodiments belong. It will be further understood that terms, e.g., those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.
The proposed solution is a sensor integrated edge to cloud computing-based AI solution designed to measure and improve critical thinking skills of learner in educational context. This system uses variety of hardware and software components to collect the audio, video and behavioural data from the learner, analyse it to measure the critical thinking index and proposes AI generated feedback and suggestion to enhance it. The core component of this system are HD cameras, microphones and optional touch panels. These devices collect the real time data such as facial expression, eye movements, voice patterns and learner interaction, which provides a deep insight into learner cognitive engagement, attention, reasoning and analytical skills. The cameras are used to capture the nonverbal cues like eye movement, eye brow movements and rest of facial expression associated with various moods of person like curiosity, confusion etc. the microphone collect the verbal data of learner like speech pattern, pauses, tonality, exclamatory remarks to assess the state of mind like thought clarity or logical analysis. An optional touch panel captures the way learner interact with the environment. It captures hesitation while writing, typing patterns and response time. The preprocessing of initial data is done through embedded computing unit such as raspberry pi. This initial preprocessing includes facial expression, voice filtering, and timestamping of event to reduce latency and ensure the identity protection of learner by anonymizing information which is to be transmitted to cloud via Wi-Fi.
Once the data reaches the cloud server, it is processed through a worldly multi-layered architecture designed for high-speed, accurate, and context-aware inference. The first method is the data preprocessing module, which cleans, formats, and categorizes the incoming multimodal data. Deep learning models analyse video frames to detect emotions, eye movements, and how engaged a person is. Speech is converted into text and the audio is analysed for tone, pause, fluency and clarity. This version keeps the meaning clear and easy to understand. Textual responses are tokenized, cleaned, and examined for linguistic and semantic structure using advanced natural language processing (NLP) techniques, such as dependency parsing, part-of-speech tagging, and sentiment analysis. These data system combines video, audio, and text data in real time. It merges these inputs into single set of features that captures the learner’s behaviour and mental state during an activity.
The newly integrated feature vectors are incorporated into specially designed pre-trained AI models, including deep learning neural networks such as convolutional neural networks (CNN) for image data, high-quality transformer-based models such as BERT for language understanding, and recurrent neural networks (RNN) or signal-based models for temporal and sequence analysis. The AI tool is then trained on an annotated learnt database using authentic classroom interactions, debates, essays, and problem-solving sessions, aligned with an authentic framework of deep-thinking criteria including Bloom's Taxonomy model. The model assesses various dimensions of critical thinking, including analysis, evaluation, inference and simplification, and self-regulation.
The AI interface uses natural language generation (NLG) models for generating feedback. This feedback suggests the strengths and weak areas where learner need to work. The feedback may suggest the learner to go through different modules, or reflect upon some questions, for his improvement. This feedback is delivered through output screen which could be a mobile or web application where the learner can see it in form of charts, tables, dashboards, progress bars, heat maps to help him understand his critical thinking index or progress.
The instructors and educators’ access to dashboard may give him deeper insight about individual taking the course. They can keep a tab about the class, their individual growth, individual struggles and make alteration in course as per AI recommendation. This dashboard also equipped with alert system whenever it finds gaps in learner critical thinking and can suggest the alternatives. To comply with the ethical concerns, this system have robust data privacy and security system. User data is anonymized and encrypted at the local preprocessing in edge before transmitting to cloud. Not ensure the privacy right and data integrity, this system fulfils the security compliances of General Data Protection Regulation (GDPR) and the Family Educational Rights and Privacy Act (FERPA).
This system promises the novel and scalable solution for measuring and enhancing critical thinking combining input from multiple real time devices, advanced machine learning, edge to cloud computing architecture and generating personalised feedback.
A multimodal AI algorithm is used in this system which combine all the audio, video, and textual interactive data as an input to quantify critical thinking skill. It uses various models for extracting features like facial and eye movements are analyzed through convolutional neural networks (CNNs), vocal cues are analyzed through speech processing models, and transformer-based language model for analyzing textual or written responses. These are combined and form attention based fusion layer that puts weight to each input based on their importance. Combined data is processed through deep neural network trained on datasets based on critical thinking frameworks like bloom taxonomy or Damien Halpern model. The output is then generated along with different subsets like analysis, inference, and evaluation abilities. This algorithm also uses natural language generation (NLG) to provide personalised suggestions on predictive outcomes which makes system analytic as well as responsive.
Algorithm
Algorithm AssessCriticalThinking
Input: VideoStream, AudioStream, TextResponse
Output: CriticalThinkingIndex (CTI), FeedbackReport
BEGIN
// Step 1: Preprocessing
VisualData ← ExtractFrames(VideoStream)
AudioData ← ExtractFeatures(AudioStream)
TranscribedText ← SpeechToText(AudioStream)
CleanedText ← PreprocessText(TextResponse + TranscribedText)
// Step 2: Feature Extraction
VisualFeatures ← CNN_Model(VisualData)
AudioFeatures ← AudioAnalysisModel(AudioData)
TextFeatures ← TransformerModel(CleanedText)
// Step 3: Multimodal Fusion
CombinedFeatures ← AttentionFusion(VisualFeatures, AudioFeatures, TextFeatures)
// Step 4: Prediction
CTI_Score, Subscores ← CriticalThinkingModel(CombinedFeatures)
// Step 5: Feedback Generation
Feedback ← GenerateFeedback(CTI_Score, Subscores)
// Step 6: Output Results
Display(CTI_Score, Subscores, Feedback)
END
ADVANTAGES OF THE INVENTION:
• The proposed system provides real time data capturing and fair evaluation of critical thinking skills. In contrast to traditional system, it captures the real time data from multiple sources such as facial expression, eye tracking, speech patterns and textual interactions continuously and measures critical thinking without biasness leading to a transparent and fair assessment.
• Secondly it provides personalized learning space and flexible feedback system. Each learner gets a feedback based on individual performance. AI generated suggestion helps individual to improve, support their learning and guide them for deeper reflection. This approach provides more meaningful learning and helps learner to proceed with their own speed.
• Thirdly system is scalable and accessible to all educational environments. It can be applied to traditional classrooms as well as remote learning system. low weight and affordability of edge devices make it a low investment solution.
• The architecture of system is secure and efficient. Videos and audios are pre-processed to protect the identity of the learner before sending them to cloud. This approach provides data security as well as reduces latency allowing prompt feedback and better user experience.
• The system generates valuable insights based on data which can aid an instructor in better decision making. Educators can have more detailed reports, identify learner’s weak areas and plan accordingly. This is more logical and fact-based approach for developing learner’s critical thinking.
 
, Claims:1. A system for measuring critical thinking comprising a sensor-integrated arrangement with edge devices including an HD camera and microphone for capturing real-time audio-video data, an embedded device for local preprocessing of raw data, a wireless communication module for transmitting pre-processed data to a cloud server, an artificial intelligence engine in the cloud configured to analyse multimodal input using a trained model, and an output screen to display a critical thinking index with AI-generated personalised feedback to improve critical thinking.
2. The system as claimed in claim 1, wherein the embedded device preprocesses audio and video data by filtering, facial expression analysis, voice segmentation, and anonymisation before transmission to the cloud server.
3. The system as claimed in claim 1, wherein the cloud-based artificial intelligence engine applies multimodal fusion to integrate video, audio, and textual data into combined features representing learner behaviour and cognitive state.
4. The system as claimed in claim 1, wherein convolutional neural networks are employed for analysing visual data including facial expressions and eye movements to determine engagement and emotional state.
5. The system as claimed in claim 1, wherein transformer-based language models are applied for analysing textual and transcribed speech responses to extract linguistic and semantic features.
6. The system as claimed in claim 1, wherein audio analysis models process voice features including tone, pauses, fluency, and clarity for assessment of cognitive reasoning.
7. The system as claimed in claim 1, wherein the artificial intelligence engine generates a Critical Thinking Index score and subscores corresponding to dimensions of analysis, evaluation, inference, and regulation.
8. The system as claimed in claim 1, wherein personalised feedback is generated through natural language generation models that suggest learning improvements, reflection tasks, and module recommendations.
9. The system as claimed in claim 1, wherein the output screen presents the Critical Thinking Index, subscores, and feedback in the form of dashboards, charts, heat maps, or progress bars for the learner and instructor.
10. The system as claimed in claim 1, wherein anonymisation and encryption of learner data are performed at the edge device level to ensure compliance with data privacy and educational security regulations.