Rameen Mahmood
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Rameen Mahmood has a total of 14 co-authors. Here are the top co-authors by number of shared publications:
- Danny Yuxing Huang
New York University, USA
4 shared publications - Annabelle David
New York University, New York City, New York, USA
3 shared publications - Donghan Hu
Computer Science, New York University, New York City, New York, USA
3 shared publications - Andrew M. Kiselica
Institute of Gerontology, University of Georgia
2 shared publications - Chao-Yi Wu
Department of Neurology, Massachusetts General Hospital, Harvard Medical School , Charlestown, Massachusetts
2 shared publications - Chenxi Qiu
Department of Genetics, Harvard Medical School
2 shared publications
Latest Publications
- Open Access
- 20 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 551-563
Characterizing Students’ Everyday AI Tool Use Through Privacy-Preserving Network-Traffic Logs
- 20 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 551-563
Open AccessCharacterizing Students’ Everyday AI Tool Use Through Privacy-Preserving Network-Traffic Logs
- Open Access
- 1 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 4 (CoNEXT2) , 1-26
What's on My Network? Using Large Language Models to Identify Real-World IoT Devices at Scale
The growth of IoT devices in shared environments has outpaced our ability to identify them, posing urgent risks to privacy, safety, and accountability. This challenge is especially pronounced in open-world environments, where network traffic metadata is often sparse, noisy, or adversarial. To address this problem, we introduce a semantic inference pipeline that reframes device identification as a language modeling task over real-world network metadata. As this approach depends on reliable supervision, we first construct high-fidelity vendor labels for the IoT Inspector dataset—the largest real-world corpus of its kind—using an ensemble of large language models guided by mutual-information and entropy-based stability scores. We then instruction-tune a quantized LLaMA 3.1 8B model on this dataset using curriculum learning to support generalization under sparsity and long-tail vendor distributions. Our model achieves 98.69% top-1 and 90.73% macro accuracy across 2,015 vendors, while remaining robust to missing fields, protocol drift, and adversarial manipulation. We also evaluate the model on an independent IoT testbed dataset, assess explanation quality, and conduct adversarial tests to probe robustness under spoofed and obfuscated input. These results position instruction-tuned LLMs as a scalable, interpretable foundation for trustworthy device identification at scale.
- 1 June 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Networking
- Vol. 4 (CoNEXT2) , 1-26
Open AccessWhat's on My Network? Using Large Language Models to Identify Real-World IoT Devices at Scale
The growth of IoT devices in shared environments has outpaced our ability to identify them, posing urgent risks to privacy, safety, and accountability. This challenge is especially pronounced in open-world environments, where network traffic metadata is often sparse, noisy, or adversarial. To address this problem, we introduce a semantic inference pipeline that reframes device identification as a language modeling task over real-world network metadata. As this approach depends on reliable supervision, we first construct high-fidelity vendor labels for the IoT Inspector dataset—the largest real-world corpus of its kind—using an ensemble of large language models guided by mutual-information and entropy-based stability scores. We then instruction-tune a quantized LLaMA 3.1 8B model on this dataset using curriculum learning to support generalization under sparsity and long-tail vendor distributions. Our model achieves 98.69% top-1 and 90.73% macro accuracy across 2,015 vendors, while remaining robust to missing fields, protocol drift, and adversarial manipulation. We also evaluate the model on an independent IoT testbed dataset, assess explanation quality, and conduct adversarial tests to probe robustness under spoofed and obfuscated input. These results position instruction-tuned LLMs as a scalable, interpretable foundation for trustworthy device identification at scale.
- Open Access
- 27 April 2026
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 10, e84618
Digital Phenotyping via Passive Network Traffic Monitoring: Prospective Observational Study in University Students
Background: Digital behaviors such as sleep, social interactions, and productivity reflect how individuals structure their daily lives. Among university students, online activity patterns mirror academic schedules, social rhythms, and lifestyle habits, with disruptions linked to sleep, stress, and well-being. Existing approaches—including wearables, apps, and surveys—depend on self-report or active participation, limiting long-term adherence. Passive sensing of network traffic offers a scalable alternative for the unobtrusive capture of smartphone usage patterns that preserves privacy. Objective: This study evaluated the degree to which encrypted smartphone network traffic, collected via a standard virtual private network (VPN), can capture patterns of digital behavior. We assessed feasibility (sustained data capture) and acceptability (usability, burden, and privacy perceptions) and examined how traffic-derived features reveal aspects of digital behavior—including timing, intensity, and regularity—relevant to health and daily functioning. Methods: We conducted a 2-week prospective observational study at New York University. Participants installed the WireGuard VPN client on personal smartphones, enabling passive capture of encrypted network traffic. Feasibility was assessed using a mixed methods approach combining quantitative measures of user retention and data coverage with qualitative analysis of semistructured exit interviews. Acceptability was evaluated using the System Usability Scale, NASA Task Load Index, and qualitative interview analysis. Exploratory analyses visualized traffic-derived features in relation to digital activity patterns. Results: Thirty-eight students consented, of whom 29 (76.3%) contributed valid network traffic data and formed the analytic cohort. Within this cohort, 93% of participants (27/29; Wilson 95% CI 78%‐98%) contributed at least 5 days of monitoring, corresponding to 71% retention relative to all consented participants (27/38; Wilson 95% CI 55%‐83%). The mean data coverage within the analytic cohort (n=24) was 74.1% (SD 19.3%; median 77.1%, IQR 63.6%-90.0%; bootstrap 95% CI 66.3%‐81.4%). These participants contributed an average of 311.6 (∼13 d, SD 3.5) hours of monitored traffic, ranging from 121 to 496 hours. Acceptability outcomes were evaluated among participants completing the exit survey and interview. Usability ratings were high (System Usability Scale score: mean 78, SD 14.96), and perceived workload was low (NASA Task Load Index scores were minimal). Participants described the system as easy to install, unobtrusive, and generally trustworthy, although some reported temporarily disabling the VPN during activities they considered private. No inferential statistical tests were conducted; analyses were descriptive. Exploratory analyses indicated that traffic-derived features reflected daily digital activity rhythms and revealed distinctive lifestyle patterns, including gaming and irregular late-night food delivery use. Conclusions: VPN-based monitoring of encrypted smartphone traffic was feasible and acceptable, enabling sustained passive data collection with minimal burden. This approach shows promise as a scalable, device-agnostic method for digital phenotyping that captures fine-grained behavioral rhythms while preserving privacy. With broader validation, this technique could expand the toolkit for studying health and well-being in everyday life.
- 27 April 2026
- Published by JMIR Publications Inc. in JMIR Formative Research
- Vol. 10, e84618
Open AccessDigital Phenotyping via Passive Network Traffic Monitoring: Prospective Observational Study in University Students
Background: Digital behaviors such as sleep, social interactions, and productivity reflect how individuals structure their daily lives. Among university students, online activity patterns mirror academic schedules, social rhythms, and lifestyle habits, with disruptions linked to sleep, stress, and well-being. Existing approaches—including wearables, apps, and surveys—depend on self-report or active participation, limiting long-term adherence. Passive sensing of network traffic offers a scalable alternative for the unobtrusive capture of smartphone usage patterns that preserves privacy. Objective: This study evaluated the degree to which encrypted smartphone network traffic, collected via a standard virtual private network (VPN), can capture patterns of digital behavior. We assessed feasibility (sustained data capture) and acceptability (usability, burden, and privacy perceptions) and examined how traffic-derived features reveal aspects of digital behavior—including timing, intensity, and regularity—relevant to health and daily functioning. Methods: We conducted a 2-week prospective observational study at New York University. Participants installed the WireGuard VPN client on personal smartphones, enabling passive capture of encrypted network traffic. Feasibility was assessed using a mixed methods approach combining quantitative measures of user retention and data coverage with qualitative analysis of semistructured exit interviews. Acceptability was evaluated using the System Usability Scale, NASA Task Load Index, and qualitative interview analysis. Exploratory analyses visualized traffic-derived features in relation to digital activity patterns. Results: Thirty-eight students consented, of whom 29 (76.3%) contributed valid network traffic data and formed the analytic cohort. Within this cohort, 93% of participants (27/29; Wilson 95% CI 78%‐98%) contributed at least 5 days of monitoring, corresponding to 71% retention relative to all consented participants (27/38; Wilson 95% CI 55%‐83%). The mean data coverage within the analytic cohort (n=24) was 74.1% (SD 19.3%; median 77.1%, IQR 63.6%-90.0%; bootstrap 95% CI 66.3%‐81.4%). These participants contributed an average of 311.6 (∼13 d, SD 3.5) hours of monitored traffic, ranging from 121 to 496 hours. Acceptability outcomes were evaluated among participants completing the exit survey and interview. Usability ratings were high (System Usability Scale score: mean 78, SD 14.96), and perceived workload was low (NASA Task Load Index scores were minimal). Participants described the system as easy to install, unobtrusive, and generally trustworthy, although some reported temporarily disabling the VPN during activities they considered private. No inferential statistical tests were conducted; analyses were descriptive. Exploratory analyses indicated that traffic-derived features reflected daily digital activity rhythms and revealed distinctive lifestyle patterns, including gaming and irregular late-night food delivery use. Conclusions: VPN-based monitoring of encrypted smartphone traffic was feasible and acceptable, enabling sustained passive data collection with minimal burden. This approach shows promise as a scalable, device-agnostic method for digital phenotyping that captures fine-grained behavioral rhythms while preserving privacy. With broader validation, this technique could expand the toolkit for studying health and well-being in everyday life.
- Open Access
- 24 September 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Digital Phenotyping via Passive Network Traffic Monitoring: Feasibility and Acceptability in University Students (Preprint)
BACKGROUND Digital behaviors such as sleep, social interaction, and productivity reflect how individuals structure daily life. Among university students, online activity patterns mirror academic schedules, social rhythms, and lifestyle habits, with disruptions linked to sleep, stress, and well-being. Existing approaches—including wearables, apps, and surveys—yield useful insights but depend on self-report or active participation, limiting adherence in real-world use. Passive sensing of network traffic provides a scalable and less burdensome alternative, enabling unobtrusive capture of smartphone usage patterns while preserving privacy. OBJECTIVE This study evaluated whether encrypted smartphone network traffic, collected via a standard virtual private network (VPN), can be used to capture patterns of digital behavior. We assessed feasibility (sustained data capture) and acceptability (usability, burden, and privacy perceptions), and examined whether traffic-derived features reveal aspects of digital behavior—including timing, intensity, and regularity—relevant to health and daily functioning. METHODS We conducted a two-week prospective observational study at New York University. Thirty-eight students enrolled; 29 provided valid network data, 27 remained active for more than five days, and 25 completed the exit interview. Participants installed the WireGuard VPN client on personal smartphones, which enabled passive capture of encrypted network traffic. Feasibility was assessed across two domains: user retention and data coverage. Acceptability was evaluated using the System Usability Scale (SUS), NASA Task Load Index (NASA-TLX), and semi-structured exit interviews. Beyond evaluating feasibility and acceptability, we conducted exploratory analyses that visualized traffic-derived features in relation to digital activity patterns. RESULTS Of the 29 participants who contributed valid data, 27 (93%) remained active for more than five days. Mean data coverage was 74.1% (median 77.1%). Participants contributed an average of 311.6 hours of monitored traffic (~13 days, SD 3.5), with totals ranging from 121 to 496 hours. Usability ratings were high (mean SUS score = 78) and perceived workload low (NASA-TLX scores minimal). Participants described the system as easy to install, unobtrusive, and generally trustworthy, though some reported temporarily disabling the VPN during activities they considered private. No inferential statistical tests were conducted; analyses were descriptive. Exploratory analyses indicated that traffic-derived features reflected daily digital activity rhythms and revealed distinctive lifestyle patterns, including gaming and irregular late-night food delivery use. CONCLUSIONS VPN-based monitoring of encrypted smartphone traffic was feasible and acceptable, enabling sustained passive data collection with minimal burden. The findings demonstrate the potential of this approach as a scalable and device-agnostic method for digital phenotyping—capable of capturing fine-grained behavioral rhythms while preserving privacy. With broader validation and deployment, the technique could expand the toolkit for studying health, well-being, and cognitive function in everyday life. CLINICALTRIAL Not applicable. This study was not registered as a clinical trial because it did not involve randomization.
- 24 September 2025
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessDigital Phenotyping via Passive Network Traffic Monitoring: Feasibility and Acceptability in University Students (Preprint)
BACKGROUND Digital behaviors such as sleep, social interaction, and productivity reflect how individuals structure daily life. Among university students, online activity patterns mirror academic schedules, social rhythms, and lifestyle habits, with disruptions linked to sleep, stress, and well-being. Existing approaches—including wearables, apps, and surveys—yield useful insights but depend on self-report or active participation, limiting adherence in real-world use. Passive sensing of network traffic provides a scalable and less burdensome alternative, enabling unobtrusive capture of smartphone usage patterns while preserving privacy. OBJECTIVE This study evaluated whether encrypted smartphone network traffic, collected via a standard virtual private network (VPN), can be used to capture patterns of digital behavior. We assessed feasibility (sustained data capture) and acceptability (usability, burden, and privacy perceptions), and examined whether traffic-derived features reveal aspects of digital behavior—including timing, intensity, and regularity—relevant to health and daily functioning. METHODS We conducted a two-week prospective observational study at New York University. Thirty-eight students enrolled; 29 provided valid network data, 27 remained active for more than five days, and 25 completed the exit interview. Participants installed the WireGuard VPN client on personal smartphones, which enabled passive capture of encrypted network traffic. Feasibility was assessed across two domains: user retention and data coverage. Acceptability was evaluated using the System Usability Scale (SUS), NASA Task Load Index (NASA-TLX), and semi-structured exit interviews. Beyond evaluating feasibility and acceptability, we conducted exploratory analyses that visualized traffic-derived features in relation to digital activity patterns. RESULTS Of the 29 participants who contributed valid data, 27 (93%) remained active for more than five days. Mean data coverage was 74.1% (median 77.1%). Participants contributed an average of 311.6 hours of monitored traffic (~13 days, SD 3.5), with totals ranging from 121 to 496 hours. Usability ratings were high (mean SUS score = 78) and perceived workload low (NASA-TLX scores minimal). Participants described the system as easy to install, unobtrusive, and generally trustworthy, though some reported temporarily disabling the VPN during activities they considered private. No inferential statistical tests were conducted; analyses were descriptive. Exploratory analyses indicated that traffic-derived features reflected daily digital activity rhythms and revealed distinctive lifestyle patterns, including gaming and irregular late-night food delivery use. CONCLUSIONS VPN-based monitoring of encrypted smartphone traffic was feasible and acceptable, enabling sustained passive data collection with minimal burden. The findings demonstrate the potential of this approach as a scalable and device-agnostic method for digital phenotyping—capable of capturing fine-grained behavioral rhythms while preserving privacy. With broader validation and deployment, the technique could expand the toolkit for studying health, well-being, and cognitive function in everyday life. CLINICALTRIAL Not applicable. This study was not registered as a clinical trial because it did not involve randomization.