arXiv subjects

Jess Kropczynski

Publications and source records attributed to Jess Kropczynski.

15 recordsLinked to original sources

arXiv2026

Geographic Patterns in I2P Peer Selection: An Empirical Network Topology Analysis

The Invisible Internet Project (I2P) routes data via encrypted, decentralized tunnels. Peer selection can significantly affect security and performance. This empirical study examines whether geographic location systematically influences I2P's routing topology. Consistent with I2P's design principles, which include avoiding multiple peers from the same /16 IP subnet to maximize anonymity, we conducted assortativity analysis, community detection, and permutation testing on data from 327 routers and 254 connections (SWARM-I2P). We found a network-level absence of significant geographic homophily. The assortativity coefficient was r = 0.017 (p = 0.222). Same-country connections (11.1%) are statistically near random expectation (10.91%). Community detection found 110 highly modular groups (Q = 0.972) only moderately aligned geographically (NMI = 0.521). We conclude that aggregate peer selection in I2P leads to a highly heterogeneous, random geographical mixing, providing a foundation for understanding the performance-anonymity tradeoff.

cs.NI↗

arXiv2026

Persona-Based Requirements Engineering for Explainable Multi-Agent Educational Systems: A Scenario Simulator for Clinical Reasoning Training

As Artificial Intelligence (AI) and Agentic AI become increasingly integrated across sectors such as education and healthcare, it is critical to ensure that Multi-Agent Education System (MAES) is explainable from the early stages of requirements engineering (RE) within the AI software development lifecycle. Explainability is essential to build trust, promote transparency, and enable effective human-AI collaboration. Although personas are well-established in human-computer interaction to represent users and capture their needs and behaviors, their role in RE for explainable MAES remains underexplored. This paper proposes a human-first, persona-driven, explainable MAES RE framework and demonstrates the framework through a MAES for clinical reasoning training. The framework integrates personas and user stories throughout the RE process to capture the needs, goals, and interactions of various stakeholders, including medical educators, medical students, AI patient agent, and clinical agents (physical exam agent, diagnostic agent, clinical intervention agent, supervisor agent, evaluation agent). The goals, underlying models, and knowledge base shape agent interactions and inform explainability requirements that guided the clinical reasoning training of medical students. A post-usage survey found that more than 78\% of medical students reported that MAES improved their clinical reasoning skills. These findings demonstrate that RE based on persona effectively connects technical requirements with non-technical medical students from a human-centered approach, ensuring that explainable MAES are trustworthy, interpretable, and aligned with authentic clinical scenarios from the early stages of the AI system engineering. The partial MAES for the clinical scenario simulator is~\href{https://github.com/2sigmaEdTech/MAS/}{open sourced here}.

cs.SE↗

arXiv2025

Examining I2P Resilience: Effect of Centrality-based Attack

This study examines the robustness of I2P, a well-regarded anonymous and decentralized peer-to-peer network designed to ensure anonymity, confidentiality, and circumvention of censorship. Unlike its more widely researched counterpart, TOR, I2P's resilience has received less scholarly attention. Employing network analysis, this research evaluates I2P's susceptibility to adversarial percolation. By utilizing the degree centrality as a measure of nodes' influence in the network, the finding suggests the network is vulnerable to targeted disruptions. Before percolation, the network exhibited a density of 0.01065443 and an average path length of 6.842194. At the end of the percolation process, the density decreased by approximately 10%, and the average path length increased by 33%, indicating a decline in efficiency and connectivity. These results highlight that even decentralized networks, such as I2P, exhibit structural fragility under targeted attacks, emphasizing the need for improved design strategies to enhance resilience against adversarial disruptions.

cs.CR↗

arXiv2025

CryptoGuard: An AI-Based Cryptojacking Detection Dashboard Prototype

With the widespread adoption of cryptocurrencies, cryptojacking has become a significant security threat to crypto wallet users. This paper presents a front-end prototype of an AI-powered security dashboard, namely, CryptoGuard. Developed through a user-centered design process, the prototype was constructed as a high-fidelity, click-through model from Figma mockups to simulate key user interactions. It is designed to assist users in monitoring their login and transaction activity, identifying any suspicious behavior, and enabling them to take action directly within the wallet interface. The dashboard is designed for a general audience, prioritizing an intuitive user experience for non-technical individuals. Although its AI functionality is conceptual, the prototype demonstrates features like visual alerts and reporting. This work is positioned explicitly as a design concept, bridging cryptojacking detection research with human-centered interface design. This paper also demonstrates how usability heuristics can directly inform a tool's ability to support rapid and confident decision-making under real-world threats. This paper argues that practical security tools require not only robust backend functionality but also a user-centric design that communicates risk and empowers users to take meaningful action.

cs.CR↗

arXiv2025

A Fuzzy Supervisor Agent Design for Clinical Reasoning Assistance in a Multi-Agent Educational Clinical Scenario Simulation

Assisting medical students with clinical reasoning (CR) during clinical scenario training remains a persistent challenge in medical education. This paper presents the design and architecture of the Fuzzy Supervisor Agent (FSA), a novel component for the Multi-Agent Educational Clinical Scenario Simulation (MAECSS) platform. The FSA leverages a Fuzzy Inference System (FIS) to continuously interpret student interactions with specialized clinical agents (e.g., patient, physical exam, diagnostic, intervention) using pre-defined fuzzy rule bases for professionalism, medical relevance, ethical behavior, and contextual distraction. By analyzing student decision-making processes in real-time, the FSA is designed to deliver adaptive, context-aware feedback and provides assistance precisely when students encounter difficulties. This work focuses on the technical framework and rationale of the FSA, highlighting its potential to provide scalable, flexible, and human-like supervision in simulation-based medical education. Future work will include empirical evaluation and integration into broader educational settings. More detailed design and implementation is~\href{https://github.com/2sigmaEdTech/MAS/}{open sourced here}.

cs.CY↗

arXiv2025

LLM-as-a-Fuzzy-Judge: Fine-Tuning Large Language Models as a Clinical Evaluation Judge with Fuzzy Logic

Clinical communication skills are critical in medical education, and practicing and assessing clinical communication skills on a scale is challenging. Although LLM-powered clinical scenario simulations have shown promise in enhancing medical students' clinical practice, providing automated and scalable clinical evaluation that follows nuanced physician judgment is difficult. This paper combines fuzzy logic and Large Language Model (LLM) and proposes LLM-as-a-Fuzzy-Judge to address the challenge of aligning the automated evaluation of medical students' clinical skills with subjective physicians' preferences. LLM-as-a-Fuzzy-Judge is an approach that LLM is fine-tuned to evaluate medical students' utterances within student-AI patient conversation scripts based on human annotations from four fuzzy sets, including Professionalism, Medical Relevance, Ethical Behavior, and Contextual Distraction. The methodology of this paper started from data collection from the LLM-powered medical education system, data annotation based on multidimensional fuzzy sets, followed by prompt engineering and the supervised fine-tuning (SFT) of the pre-trained LLMs using these human annotations. The results show that the LLM-as-a-Fuzzy-Judge achieves over 80\% accuracy, with major criteria items over 90\%, effectively leveraging fuzzy logic and LLM as a solution to deliver interpretable, human-aligned assessment. This work suggests the viability of leveraging fuzzy logic and LLM to align with human preferences, advances automated evaluation in medical education, and supports more robust assessment and judgment practices. The GitHub repository of this work is available at https://github.com/2sigmaEdTech/LLMAsAJudge

cs.AI↗

arXiv2024

Examining Caregiving Roles to Differentiate the Effects of Using a Mobile App for Community Oversight for Privacy and Security

We conducted a 4-week field study with 101 smartphone users who self-organized into 22 small groups of family, friends, and neighbors to use ``CO-oPS,'' a mobile app for co-managing mobile privacy and security. We differentiated between those who provided oversight (i.e., caregivers) and those who did not (i.e., caregivees) to examine differential effects on their experiences and behaviors while using CO-oPS. Caregivers reported higher power use, community trust, belonging, collective efficacy, and self-efficacy than caregivees. Both groups' self-efficacy and collective efficacy for mobile privacy and security increased after using CO-oPS. However, this increase was significantly stronger for caregivees. Our research demonstrates how community-based approaches can benefit people who need additional help managing their digital privacy and security. We provide recommendations to support community-based oversight for managing privacy and security within communities of different roles and skills.

cs.HC↗

arXiv2024

Co-designing Community-based Sharing of Smarthome Devices for the Purpose of Co-monitoring In-home Emergencies

We conducted 26 co-design interviews with 50 smarthome device owners to understand the perceived benefits, drawbacks, and design considerations for developing a smarthome system that facilitates co-monitoring with emergency contacts who live outside of one's home. Participants felt that such a system would help ensure their personal safety, safeguard from material loss, and give them peace of mind by ensuring quick response and verifying potential threats. However, they also expressed concerns regarding privacy, overburdening others, and other potential threats, such as unauthorized access and security breaches. To alleviate these concerns, participants designed for flexible and granular access control and fail-safe back-up features. Our study reveals why peer-based co-monitoring of smarthomes for emergencies may be beneficial but also difficult to implement. Based on the insights gained from our study, we provide recommendations for designing technologies that facilitate such co-monitoring while mitigating its risks.

cs.HC↗

arXiv2024

CO-oPS: A Mobile App for Community Oversight of Privacy and Security

Smartphone users install numerous mobile apps that require access to different information from their devices. Much of this information is very sensitive, and users often struggle to manage these accesses due to their lack of tech expertise and knowledge regarding mobile privacy. Thus, they often seek help from others to make decisions regarding their mobile privacy and security. We embedded these social processes in a mobile app titled "CO-oPS'' ("Community Oversight for Privacy and Security"). CO-oPS allows trusted community members to review one another's apps installed and permissions granted to those apps. Community members can provide feedback to one another regarding their privacy behaviors. Users are also allowed to hide some of their mobile apps that they do not like others to see, ensuring their personal privacy.

cs.HC↗

arXiv2024

From Parental Control to Joint Family Oversight: Can Parents and Teens Manage Mobile Online Safety and Privacy as Equals?

Our research aims to highlight and alleviate the complex tensions around online safety, privacy, and smartphone usage in families so that parents and teens can work together to better manage mobile privacy and security-related risks. We developed a mobile application ("app") for Community Oversight of Privacy and Security ("CO-oPS") and had parents and teens assess whether it would be applicable for use with their families. CO-oPS is an Android app that allows a group of users to co-monitor the apps installed on one another's devices and the privacy permissions granted to those apps. We conducted a study with 19 parent-teen (ages 13-17) pairs to understand how they currently managed mobile safety and app privacy within their family and then had them install, use, and evaluate the CO-oPS app. We found that both parents and teens gave little consideration to online safety and privacy before installing new apps or granting privacy permissions. When using CO-oPS, participants liked how the app increased transparency into one another's devices in a way that facilitated communication, but were less inclined to use features for in-app messaging or to hide apps from one another. Key themes related to power imbalances between parents and teens surfaced that made co-management challenging. Parents were more open to collaborative oversight than teens, who felt that it was not their place to monitor their parents, even though both often believed parents lacked the technological expertise to monitor themselves. Our study sheds light on why collaborative practices for managing online safety and privacy within families may be beneficial but also quite difficult to implement in practice. We provide recommendations for overcoming these challenges based on the insights gained from our study.

cs.HC↗

arXiv2024

It Takes a Village: A Case for Including Extended Family Members in the Joint Oversight of Family-based Privacy and Security for Mobile Smartphones

We conducted a user study with 19 parent-teen dyads to understand the perceived benefits and drawbacks of using a mobile app that allows them to co-manage mobile privacy, safety, and security within their families. While the primary goal of the study was to understand the use case as it pertained to parents and teens, an emerging finding from our study was that participants found value in extending app use to other family members (siblings, cousins, and grandparents). Participants felt that it would help bring the necessary expertise into their immediate family network and help protect the older adults and children of the family from privacy and security risks. However, participants expressed that co-monitoring by extended family members might cause tensions in their families, creating interpersonal conflicts. To alleviate these concerns, participants suggested more control over the privacy features to facilitate sharing their installed apps with only trusted family members.

cs.HC↗

arXiv2024

Agricultural 4.0 Leveraging on Technological Solutions: Study for Smart Farming Sector

By 2050, it is predicted that there will be 9 billion people on the planet, which will call for more production, lower costs, and the preservation of natural resources. It is anticipated that atypical occurrences and climate change will pose severe risks to agricultural output. It follows that a 70% or more significant rise in food output is anticipated. Smart farming, often known as agriculture 4.0, is a tech-driven revolution in agriculture with the goal of raising industry production and efficiency. Four primary trends are responsible for it: food waste, climate change, population shifts, and resource scarcity. The agriculture industry is changing as a result of the adoption of emerging technologies. Using cutting-edge technology like IoT, AI, and other sensors, smart farming transforms traditional production methods and international agricultural policies. The objective is to establish a value chain that is optimized to facilitate enhanced monitoring and decreased labor expenses. The agricultural sector has seen tremendous transformation as a result of the fourth industrial revolution, which has combined traditional farming methods with cutting-edge technology to increase productivity, sustainability, and efficiency. To effectively utilize the potential of technology gadgets in the agriculture sector, collaboration between governments, private sector entities, and other stakeholders is necessary. This paper covers Agriculture 4.0, looks at its possible benefits and drawbacks of the implementation methodologies, compatibility, reliability, and investigates the several digital tools that are being utilized to change the agriculture industry and how to mitigate the challenges.

cs.HC↗

arXiv2023

Evaluating the Impact of Community Oversight for Managing Mobile Privacy and Security

Mobile privacy and security can be a collaborative process where individuals seek advice and help from their trusted communities. To support such collective privacy and security management, we developed a mobile app for Community Oversight of Privacy and Security ("CO-oPS") that allows community members to review one another's apps installed and permissions granted to provide feedback. We conducted a four-week-long field study with 22 communities (101 participants) of friends, families, or co-workers who installed the CO-oPS app on their phones. Measures of transparency, trust, and awareness of one another's mobile privacy and security behaviors, along with individual and community participation in mobile privacy and security co-management, increased from pre- to post-study. Interview findings confirmed that the app features supported collective considerations of apps and permissions. However, participants expressed a range of concerns regarding having community members with different levels of technical expertise and knowledge regarding mobile privacy and security that can impact motivation to participate and perform oversight. Our study demonstrates the potential and challenges of community oversight mechanisms to support communities to co-manage mobile privacy and security.

cs.HC↗

arXiv2023

Machine Learning Methods for Evaluating Public Crisis: Meta-Analysis

This study examines machine learning methods used in crisis management. Analyzing detected patterns from a crisis involves the collection and evaluation of historical or near-real-time datasets through automated means. This paper utilized the meta-review method to analyze scientific literature that utilized machine learning techniques to evaluate human actions during crises. Selected studies were condensed into themes and emerging trends using a systematic literature evaluation of published works accessed from three scholarly databases. Results show that data from social media was prominent in the evaluated articles with 27% usage, followed by disaster management, health (COVID) and crisis informatics, amongst many other themes. Additionally, the supervised machine learning method, with an application of 69% across the board, was predominant. The classification technique stood out among other machine learning tasks with 41% usage. The algorithms that played major roles were the Support Vector Machine, Neural Networks, Naive Bayes, and Random Forest, with 23%, 16%, 15%, and 12% contributions, respectively.

cs.LG↗

arXiv2023

A Semantic Approach to Negation Detection and Word Disambiguation with Natural Language Processing

This study aims to demonstrate the methods for detecting negations in a sentence by uniquely evaluating the lexical structure of the text via word-sense disambiguation. The proposed framework examines all the unique features in the various expressions within a text to resolve the contextual usage of all tokens and decipher the effect of negation on sentiment analysis. The application of popular expression detectors skips this important step, thereby neglecting the root words caught in the web of negation and making text classification difficult for machine learning and sentiment analysis. This study adopts the Natural Language Processing (NLP) approach to discover and antonimize words that were negated for better accuracy in text classification using a knowledge base provided by an NLP library called WordHoard. Early results show that our initial analysis improved on traditional sentiment analysis, which sometimes neglects negations or assigns an inverse polarity score. The SentiWordNet analyzer was improved by 35%, the Vader analyzer by 20% and the TextBlob by 6%.

cs.CL↗