Scofolio soft skills development and assessment using intelligent system, competency framework, and social media-like interaction methods

Abstract

A unique worldwide digital system, comprises of a competency framework, and methods for the automated assessment and development of human soft skills. The primary innovation of the system is that it contextualizes data from a user's social media and social media-like interactions; Data engines analyze user data to create a preliminary human soft skills portfolio. An important component of the integration is a human-in-the-loop calibration interface, in which an expert reviewer adjusts the AI's output to create a calibrated assessment. This human-calibrated data is also used to retrain the models and reduce bias. The system then generates a personalized, actionable mentorship roadmap. The invention utilizes a dynamic, scientifically-grounded competency framework that evolves through ongoing data analysis and R&D, producing a holistic innovation index for inddividuals and organizations.

Description

Scofolio Soft Skills Development and Assessment using Intelligent System, Competency Framework, and Social Media-like Interaction Methods.Name of the inventor: Sally Kawtharany

[0001] Title of Invention: Scofolio Soft Skills Development and Assessment using Intelligent System, Competency Framework, and Social Media-like Interaction Methods.

[0002] The current invention pertains to educational and technological approaches for assessing and enhancing human soft skills in digital settings. Specifically, it focuses on the user's interactions with social media-like posts, utilizing unique in-house data models, a hierarchical recurrent neural network to produce dynamic skill scores, and reinforcement learning to determine the best mentorship feedback for each user’s interaction.

[0003] Current platforms and apps offer soft-skills assessment, through surveys, peer reviews, abstract games, or static AI analysis of user self-evaluated data. These systems are constrained by: Incapacity to incorporate behavioral data from multiple sources; feedback that is delayed and less useful for developing skills; one-size-fits-all rather than culturally specific; static scoring methods that do not adjust to the user's changing skill level or ongoing potential; more abstract and detached from the real world and contextual situations; the division between development and assessment, which prevents the development process from being viewed as continuous; predetermined responses that are susceptible to fraud and AI-generated responses; stressful and uninteresting, with insufficient personalization for the next generation; failed to consider the specifics of soft skills as part of ongoing, social, immersive, and situational learning and development. There is still a need for a system that automatically assesses human soft skills in real time, uses future-oriented metrics, is simple to use on a daily basis as a living, ongoing immersive experience, integrates a large number of data points, focuses on human natural incidents, is tailored to each culture, involves real-world context situations, and integrates development and assessment.

[0004] The present invention fulfills the long-standing need for a system that moves beyond static, biased, and backward-looking soft skill assessments by providing a living, immersive, and culturally-aware platform for continuous development and assessment. This invention offers a system, methods, and novel competency framework for automatically assessing and mentoring soft skills through specialized AI-driven analysis of user interactions with real-world topics that resemble social media interactions.

[0005] The technical problem is the absence of a unified computational framework capable of fusing objective, data-driven analysis with validated human-centric criteria for human soft skills. Existing systems are bifurcated: they are either purely subjective (human-led, slow, and non-scalable) or purely algorithmic (AI-led, biased, and context-blind), with no architecture to integrate them for continuous, contextualized assessment based on day-to-day interactions.

[0006] The present invention solves this problem through an adaptive digital system that functions as a continuous, contextual human soft skill intelligence platform. At its core is a dynamically-evolving competency framework that structures human skills into a measurable indicators. The system gathers both structured and unstructured social data, enriches it with situational metadata through a contextual mapping process, and immerses the user in real-world situated interaction using a variety of techniques. This contextualized data is then analyzed by a hybrid AI-human calibration engine, where artificial intelligence generates preliminary metrics that are refined by human expert judgment to mitigate bias and add nuanced understanding. The output is not a static report, but a personalized, actionable developmental roadmap and a unique "innovation index," which are integrated directly into the user's workflow. This creates a living, immersive feedback loop where assessment and development are seamlessly unified, and the system's own models continuously improve, ensuring culturally-relevant and scientifically-grounded measurement of human potential.

[0007] Objective, Quantifiable Measurement of the Subjective: Transforms soft skills from abstract concepts into a set of measurable, data-driven indicators based on the defined taxonomy. This provides the first true, objective baseline for assessing skills like collaboration or critical thinking.

[0008] Dramatic Reduction in Algorithmic and Human Bias: The human-in-the-loop feedback mechanism continuously calibrates the AI, preventing it from reinforcing hidden biases in its training data. The contextual layer ensures behavior is judged relative to the situation, not an absolute standard. This creates a fairer and more equitable assessment system.

[0009] Contextually Relevant and Actionable Insights: By analyzing user interactions data, the assessments are no longer based on abstract test questions but on an individual's actual real-world context. The resulting mentorship is therefore highly specific, relevant, and immediately applicable, increasing its effectiveness and user adoption.

[0010] Scalable and Efficient Personalization: The Scofolio system makes personalized soft skill development feasible for organizations of thousands, not just for a select few executives. It is "more quick" and scalable.

[0011] Creation of a Proprietary and Defensible Data Asset (The "Unique Data Moat"): The system generates a unique, high-value dataset consisting of contextual behavioral data paired with expert-human-calibrated assessments. This dataset is incredibly difficult to replicate and becomes a sustainable competitive advantage, as it is used to continually improve the system's accuracy and reliability.

[0012] Synergistic Human-AI Collaboration: Solves the "in-between dilemma" by creating a unified system where AI and human expertise augment each other. The AI handles scale and pattern recognition; the human provides nuanced judgment and ethical oversight. The whole is greater than the sum of its parts.

[0013] shows the social media landscape patterns that correlate to the expression of human soft skills and what could be advantageous instead of their current conventional use.

[0014] depicts the first method: Re-generate social data into structured "folios" that fit the competency framework, reconnecting users with different data models and an architectured environment of interactions and human soft skills.

[0015] The direct path from real user social data interactions to direct human soft skills assessment processing.

[0016] Linking skills to real-world examples, case studies, and scenarios. Discover how people define human soft skills by connecting them to real-world examples and expressing them through social data.

[0017] The co-motion: learning human soft skills through observation and imitation of skilled interactions.

[0018] With reference to, The current social media landscape is only used to transfer information between communities. While the Scofolio system was designed to be reused for the purpose of empowering human soft skills.

[0019] With reference to, 3, 4, 5 Scofolio System for adaptive competency intelligence is shown. The system is typically implemented as a software-as-a-service (SaaS) platform hosted on cloud computing infrastructure, comprising one or more physical servers, databases, and client interfaces.

[0020] The ingestion of social data from various sources and for various purposes is the initial stage of the system's operation. Initially, as shown inthe first method of data ingestion and contextualization is set up to receive an ongoing stream of unstructured social data from multiple digital social platforms. Publicly available data, without identifying individuals, is used to analyze top trends and topics in specific subjects, reflecting the real-world context. With strict adherence to moral, ethical, and privacy standards. These data will then be converted into "folios"— cleaned, synthetic, elaborated, and reformulated small posts.

[0021] Method one's process includes context-awareness folios that connect first to the "Architectured Skills Data Model," which filters the generated folios by emerging skills and their interaction relevance, and then involves users to interact with as they interact with social media-like content. The user's interaction will be aligned with the scofolio competency framework, allowing the system to provide continuous assessment and personalized mentoring.

[0022] The intrinsic link between the Architectured Skills Data Model and the Calibrated Intelligence Processor forms the foundational core of this invention, ensuring scientific validity and operational efficacy. The Data Model serves as the definitive ontology, rigorously structuring human competencies into a hierarchical framework of pillars, skills, and measurable behavioral indicators—it defines what to measure. The Calibrated Intelligence Processor is the execution engine that operationalizes this model; it employs AI to analyze real-world behavioral data against the model's criteria and integrates human expert judgment to calibrate outputs, thereby determining how to measure. This symbiotic relationship is crucial: the Model provides the essential taxonomy and rules that ground the Processor's analysis, preventing algorithmic bias and ensuring assessments are contextually relevant and psychologically grounded. Simultaneously, the Processor's continuous feedback loop, enriched by human calibration, enables the Data Model to dynamically evolve and improve.

[0023] As shown in, The second method relied on previous interactions that user had on real social media channels; user can reuse this method for human soft skill assessment and development purposes by uploading / integrating the interactions data from their account to the Scofolio system. This would not be feasible without the user's permission to use their data for that purpose while maintaining the highest standards of security and privacy protection.

[0024] In the first method, a mode of operation for structured input data, the system maps the data directly to the Architectured Skills Data Model before refinement by the Calibrated Intelligence Processor. In the second method, a mode of operation for unstructured input data, the Calibrated Intelligence Processor first analyzes the data to extract behavioral indicators, which are subsequently classified using the Architectured Skills Data Model.

[0025] The second method will result in continuous assessment and mentoring after matching with the Scofolio competency framework.

[0026] As shown in, the third method involves enhancing the Scofolio system with pertinent real-world context topics and subjects that have already been explicitly or implicitly labeled with soft skills. Exploring situations, scenarios, and subject examples through skill filtration is the foundation of experiential learning. The mapper will align this data with the entire competency framework and another model reformulates them in a structured Learning Record.

[0027] As shown in, The concept of co-motion is to foster the experience by involving the audience to interact with the same folio of the user that they interact with, as a social immersive experience to observe and then learn from skilled people by imitating words, thoughts and expressions.

[0028] In order to enhance the entire process, particularly the Architectured Skills Data Model and the Scofolio competency framework, the experience of the Skills Assessment Data Model and Mentorship Roadmaps will pass with analysis engines.

[0029] To give the user a live score, the system uses several data points from various sources and interactions.Examples

[0030] Example: A user interacted with a generated-folio about a common issue that occurs for those who use a stock trading platform. The user suggested a step-by-step advice to avoid this, but the Skills Assessment Data Model, which is based on the Architectured Skills Data Model, detects less proficiency of this as an indicator in Creativity, so it triggers the mentoring roadmaps to assist the user with motivational questions on elaborating more on the step details, and the user finally interacts with second polished interactions that match that indicator.

[0031] Another example of a user had a lot of comments and interactions on their Linkedin profile that advising, supporting others with solving issues and encouraging trying different approaches, this new source of data elevates the scoring system accuracy and level for their Scofolio (The Portfolio in Scofolio).

[0032] Another example of the third method: rather than writing theories and theoterical descriptions about creativity, scraping social data revealed, for instance, that the most popular topics labeled with creativity were: creative expression, digital and technology, educational development, entertainment, and interactive media. Gathering examples from public posts indicating situations, scenarios, and examples of creativity and connecting them to the Scofolio competency framework will engage users in real-world context learning and inspiration.

[0033] 100: Networked knowledge: is the conventional method of using social media platforms today, primarily for community knowledge transfer.

[0034] 101: Social Media Landscape: The social media landscape is an environment of user-generated data with three primary characteristics that are completely compatible with human soft skills.

[0035] 102: User-Generated Content (UGC) of social data is any form of content—such as text, photos, videos, reviews, or audio—created and shared by individual users (people) on digital platforms and social networks. This content, in its aggregate, forms a vast body of data that reveals human behavior, opinions, trends, and social interactions.

[0036] 103: Real World Data & Context: Unfiltered, organic record of human behavior, opinions, and interactions that occur outside of controlled environments like labs or surveys.

[0037] 104: Natural Interaction: Means communication and behavior that occurs spontaneously and without the influence of a structured research setting, reflecting a user's genuine intent and state of mind.

[0038] 105: Digital Identity: Social data represents a user's digital identity because it is a curated, yet dynamic, aggregate of their self-expression, connections, and interactions that collectively forms their online persona.

[0039] 106: The first Scofolio method is ethical sourcing, which is an engine for contextual data that is extracted from social data and converted into structured data while maintaining all security and privacy.

[0040] 107: Import user-authorized historical interactions: Leveraging a user's past, explicitly permitted interactions and data from one platform or many to enhance their experience on a the Scofolio system.

[0041] 108: Filtering Soft Skills-related Data: capture scenarios and situations that are clearly labeled by any human soft skills to correspond with how these skills are expressed in real-world discussions and subjects.

[0042] 109: Purposeful Adaptation: The data Adaptor is a system component comprising software modules and configuration rules that perform data transformation and context mapping to ensure relevant input data and interpreted by the Architectured Skills Data Model and processed by the Calibrated Intelligence Processor.

[0043] 110: Customized Folios generation Model: It's social media-like generated content, and the component that translates the system's complex assessments into a practical, actionable, and personalized asset for the user.

[0044] 111: The Architectured Skills Data Model is the core ontology and rulebook that defines, structures, and gives scientific meaning to the concept of "soft skills." It is the foundational layer that makes your system more than just another opinion-based assessment.

[0045] 112: User’s interaction: Spontenous interaction with the topic without specific questions or requests, similar to interactions on social media.

[0046] 113: The Calibrated Intelligence Processor is the core computational engine that executes a two-stage analysis of behavioral data. First, it uses artificial intelligence (AI) to analyze input data against the Architectured Skills Data Model to generate a preliminary, data-driven assessment. Second, and most critically, it integrates human expert judgment to review, adjust, and validate this assessment, creating a "calibrated" output.

[0047] 114: Contextual Mapping is the computational process of enriching raw behavioral data with metadata that defines the circumstances, environment, and relationships of the interaction. It transforms a generic action into a situated behavior, allowing the system to accurately interpret the demonstration of a soft skill.

[0048] 115: The Learning Record Store (LRS) Model is a structured, semantic repository that ingests, tags, and reorganizes learning content (articles, case studies, videos, exercises) into a curated collection of "topics." These topics are dynamically reformulated and indexed against the skills and sub-skills defined in the Scofolio Competency Framework.

[0049] 116: Experiential learning: is based on connecting skills to real-world examples, case studies, and scenarios. Instead of defining skills, learn how people define them by connecting them to real-world examples and expressing them through social data.

[0050] 117: Scofolio Competency Framework: It consists of five pillars, and their relevant skills are as follows: Pillar 1: Conceptual and thinking: Abstract Thinking, Creativity, Critical thinking. Pillar 2: Social Intelligence: Communication, Interaction Ability, Openness. Pillar 3: Leadership: Decision Making, Problem Solving, Self Discipline. Pillar 4: Data Wise: Analysis Ability, Knowledge Management, Curiosity. Pillar 5: Personal Effectiveness: Empathy, Self-Reflective, Positive Minded.

[0051] 118: Skills Assessment Data Model: It is the data schema that transforms raw interactive data into quantifiable metrics for evaluating human soft skills.

[0052] 119: Mentorship Roadmaps are dynamic, personalized guides that use motivational questioning and targeted interventions to help a user consciously develop and master the specific soft skills they have targeted for growth.

[0053] 120: social immersive co-motion method: pick up human soft skills by observing others and imitating their thoughts and expressions.

[0054] 121: Human soft skills outcomes and other people's participation as a simulation experience; the user will observe and learn by matching interactions and skill labels and taking it as part of learning input.

[0055] The invention is susceptible of industrial application as a software platform and service in multiple fields, including but not limited to:

[0056] The Corporate Sector: For use in human resources departments for objective hiring, promotion, employee development, and team building, thereby improving workforce productivity and reducing turnover.

[0057] The Government Sector: For use in public administration for civil service evaluation, training, and assembling project teams with complementary skill sets, leading to more effective public service delivery.

[0058] The Higher Education Sector: For use by universities and vocational schools to provide students with certified soft skill assessments, personalized learning pathways, and career readiness portfolios, enhancing their employability.

[0059] The invention is implemented as a tangible software-as-a-service (SaaS) platform, operable on standard computing devices and networks, making it capable of being produced and used on an industrial scale by everyone, from students in educational institutions to employees across corporate and governmental sectors.Patent Literature

[0060] PI 2023005492; Entitled: “System and methods for sustainable career success prediction using psychometrics and job performance”. Whereas the prior art predicts career success based on static psychometric and performance data to recommend corrective measures, our invention generates a dynamic, contextualized soft skill assessment and development roadmap by fusing AI analysis of real-world interactive data with continuous human expert calibration.

[0061] US10796217B2, Xianchao Wu. Entitled: “Systems and methods for performing automated interviews”. Whereas this performs automated interviews by having an AI bot conduct a scripted, Q&A-style interaction to evaluate a candidate, our invention assesses soft skills contextually by analyzing a user's real-world interaction and generates a personalized mentorship roadmap, without the need for an artificial interview process.Non-Patent Literature

[0062] Soft skills (e.g., communication, creative thinking, and problem-solving) are often vaguely defined, resulting in inconsistent interpretation among educators and students. This ambiguity complicates the development of standardized assessment tools. Furthermore, these skills are inherently subjective, shaped by personal values, cultural norms, and situational factors. Consequently, traditional assessments—effective for technical knowledge—struggle to capture such nuanced and dynamic attributes. Moreover, many existing assessment tools are developed within specific cultural contexts and may not align with the values or expectations of other societies. This limits their validity and fairness across diverse populations. Additionally, the reliance on subjective judgments by instructors can also introduce bias and reduce reliability. Therefore, effective assessment requires nuanced, culturally aware, and multi-method approaches that extend beyond conventional testing.

[0063] Conversely, temporary or short-term training programs for soft skills are widely used in education and the workplace, yet their epistemological foundations and long-term impact remain debated. Because soft skills are defined by behaviors, attitudes, and interpersonal dynamics, they are less tangible and harder to standardize than technical skills. This inherent subjectivity complicates both development and assessment, particularly within brief interventions. Ultimately, the sustainable development of soft skills requires ongoing, contextually embedded practice and robust, longitudinal assessment frameworks.

[0064] Soft skills frameworks and indicators are widely recognized as essential for employability and education, but they face persistent problems of vague definitions, inconsistent measurement, and limited transferability across contexts. Soft skills are defined and grouped in highly variable ways across research, education, and national systems, leading to confusion and lack of standardization. Moreover, on one hand, many frameworks lack precise, measurable, and low-inference indicators, resulting in subjective or inconsistent evaluations. On the other hand, Many frameworks do not address the dynamic, context-dependent nature of soft skills development and assessment.

Claims

A digital ecosystem for assessing and developing human soft skills, including a competency framework, data models, tools, and techniques. By connecting with extensive human soft skills engines and converting the idea of social data interactions into human soft skills scoring and experience, this ecosystem shapes social data usage from sharing and transmitting to purposeful everyday real-world interactions.The system of claim 1, wherein the interactive data inputs are ingested from a plurality of source types, which include: Connect and analyze real social media interactions derived by users to measure and develop human soft skills. User-generated digital interactions within brief content, including social media-like content generally and “folios” as they are referred to in the scofolio system; Contextualized data, ingested from external third-party platforms consisting of: online forums, discussion boards, and video conferencing systems.The system of claim 1, wherein contextualizing the interactive data comprises generating metadata that classifies the data according to both content-based dimensions including topic classification, and interaction-based dimensions, communication channel type and participant relationship dynamics.The system of claim 1, wherein the AI analysis engine comprises Adaptive Intelligence models that have been trained on a dataset of human-calibrated behavioral data, and wherein the instructions further configure the system to use subsequent calibration input from the human reviewer interface as training data to dynamically retrain the machine learning model, thereby creating a continuous feedback loop that reduces algorithmic bias.The system of claim 1, wherein the personalized mentorship roadmap is integrated into the user's digital workflow environment. Using hierarchical recurrent neural network to produce dynamic skill scores, and reinforcement learning wherein the roadmap specifies a set of contextual challenges that dynamically appear within the user's ongoing digital interactions, the challenges designed to target specific skill gaps identified in the calibrated competency assessment.Four foundational methods that the system has: Social data methods for generating structured social media-like folios. Inputting user social media histrocial interactions. Creating skill-labeled customized content based on social data, and co-motion social immersive experienceimmersing the user in social media-like content to serve as the foundation for user-generated immersive interactions to measure human soft skills and create a personalized mentorship roadmap for the user based on their assessed soft skill developmentObtaining user interaction data at the processor via actual user-generated unstructured historical data on social media platforms.Obtaining social data that demonstrates and exemplifies soft skills at the contextual mapper. Learn by observing the connections between human soft skills and real-world contexts through scenarios, examples, and case studies.Co-motion Social immersive learning experience by observing others' interactions with social media-like content or "folios", what human soft skills are acquired, and how to improve mentality learning through their word expressions and actions.The method of claim 6, wherein receiving the human-calibrated assessment initiates a calibration process comprising: receiving input that adjusts a quantitative metric of the preliminary assessment and provides a qualitative rationale for the adjustment, thereby generating a fused data point comprising the adjusted quantitative metric, the qualitative rationale, and the supporting contextualized interactive data, which is stored in a calibrated training dataset to iteratively refine the AI engine.A dynamically-evolving, comprehensive digital-implemented competency framework for assessing human soft skills, the framework comprising: a hierarchical structure stored in a database, the hierarchical structure defining: a plurality of core competency pillars; for each of the defined skills, a set of quantitative measurable indicators;wherein the hierarchical structure is configured to be updated over time based on aggregated, anonymized data from a plurality of users and ongoing research and development to refine the interrelationships between the pillars and skills, thereby generating a holistic innovation index for a user that is distinct from a simple aggregation of individual skill scores.The system of claim 1, wherein the AI analysis engine is configured to analyze the contextualized data by mapping it against the hierarchical structure of the dynamically-evolving competency framework of claim 8 to produce the preliminary soft skill profile.The system of claim 1, wherein generating the personalized mentorship roadmap includes calculating and presenting the holistic innovation index, wherein the innovation index is a computed metric representing a user's synergistic proficiency across the five core competency pillars.The competency framework of claim 8, wherein the hierarchical structure is updated by a framework evolution module configured to: receive the aggregated, anonymized data and performance outcomes; perform statistical and causal analysis on the data to identify correlations and causal relationships between the defined skills and successful outcomes; and automatically adjust the weighting or definition of the interrelationships within the hierarchical structure based on the analysis.

Images