Protik Bose Pranto, Isaac Buo, Ariane Middel
CAP Poster Symposium 2026 Poster
This research proposes a comprehensive methodology using LiDAR data to create a detailed tree inventory for the CAP-LTER region. The process involves acquiring and preprocessing data, followed by advanced analysis to measure tree height, canopy width, crown shape, and species classification. The study will use Machine learning models to classify tree species based on ground truth data from Tempe’s existing tree inventory. The study will explore the distribution of native versus non-native species and examine structural characteristics across different geographic areas.
Protik Bose Pranto, Isaac Buo, Ariane Middel
CAP Poster Symposium 2026 Poster
This research proposes a comprehensive methodology using LiDAR data to create a detailed tree inventory for the CAP-LTER region. The process involves acquiring and preprocessing data, followed by advanced analysis to measure tree height, canopy width, crown shape, and species classification. The study will use Machine learning models to classify tree species based on ground truth data from Tempe’s existing tree inventory. The study will explore the distribution of native versus non-native species and examine structural characteristics across different geographic areas.
Protik Bose Pranto, Isaac Buo, Ariane Middel
Journal of Transport Geography (Under review) 2026 Paper
Protik Bose Pranto, Isaac Buo, Ariane Middel
Journal of Transport Geography (Under review) 2026 Paper
Protik Bose Pranto, Minhazul Islam, Ripon Kumar Saha, Abimelec Mercado Rivera, Namig Abbasov
Urban AI (ACM SIGSPATIAL 2025) 2025 Paper
Cultural infrastructures, such as libraries, museums, theaters, and galleries, support learning, civic life, health, and local economies, yet access is uneven across cities. We present a novel, scalable, and open-data framework to measure spatial equity in cultural access. We map cultural infrastructures and compute a metric called Cultural Infrastructure Accessibility Score (CIAS) using exponential distance decay at fine spatial resolution, then aggregate the score per capita and integrate socio-demographic indicators. Interpretable tree-ensemble models with SHapley Additive exPlanation (SHAP) are used to explain associations between accessibility, income, density, and tract-level racial/ethnic composition.
Protik Bose Pranto, Minhazul Islam, Ripon Kumar Saha, Abimelec Mercado Rivera, Namig Abbasov
Urban AI (ACM SIGSPATIAL 2025) 2025 Paper
Cultural infrastructures, such as libraries, museums, theaters, and galleries, support learning, civic life, health, and local economies, yet access is uneven across cities. We present a novel, scalable, and open-data framework to measure spatial equity in cultural access. We map cultural infrastructures and compute a metric called Cultural Infrastructure Accessibility Score (CIAS) using exponential distance decay at fine spatial resolution, then aggregate the score per capita and integrate socio-demographic indicators. Interpretable tree-ensemble models with SHapley Additive exPlanation (SHAP) are used to explain associations between accessibility, income, density, and tract-level racial/ethnic composition.
Protik Bose Pranto, Isaac Buo, Ariane Middel
AGU Annual Meeting 2025 2025 Abstract
Pedestrian infrastructure is inadequately or inconsistently mapped in most cities, limiting efforts to assess walkability, accessibility, and equitable transportation systems. This study introduces WalkNet, a city-wide scalable framework that generates detailed, georeferenced pedestrian networks by integrating street-level imagery with open geospatial data. The framework applies a semantic segmentation model on panoramic images to extract sidewalks and roads, then computes a sidewalk-to-road height ratio as a geometric proxy for road type. It classifies roads as arterial, collector, or residential using the ratio, aligning with U.S. transportation codes.
Protik Bose Pranto, Isaac Buo, Ariane Middel
AGU Annual Meeting 2025 2025 Abstract
Pedestrian infrastructure is inadequately or inconsistently mapped in most cities, limiting efforts to assess walkability, accessibility, and equitable transportation systems. This study introduces WalkNet, a city-wide scalable framework that generates detailed, georeferenced pedestrian networks by integrating street-level imagery with open geospatial data. The framework applies a semantic segmentation model on panoramic images to extract sidewalks and roads, then computes a sidewalk-to-road height ratio as a geometric proxy for road type. It classifies roads as arterial, collector, or residential using the ratio, aligning with U.S. transportation codes.
Protik Bose Pranto, Waqar Hassan Khan, Ariane Middel
Urban Climate Research Center Poster Event 2025 Poster
In a newly evolving research domain called “Urban Climate Informatics” (UCI), computer science methods have recently been increasingly used in urban climate research to process and analyze large-scale data from sensors, satellites, and IoT devices. Techniques such as machine learning, artificial intelligence, GIS, and cloud computing help assess climate impacts on urban systems at various scales. We conducted a comprehensive systematic literature review to explore the intersection of urban climate and computer science and examine how this cross-disciplinary approach addresses various urban climate challenges.
Protik Bose Pranto, Waqar Hassan Khan, Ariane Middel
Urban Climate Research Center Poster Event 2025 Poster
In a newly evolving research domain called “Urban Climate Informatics” (UCI), computer science methods have recently been increasingly used in urban climate research to process and analyze large-scale data from sensors, satellites, and IoT devices. Techniques such as machine learning, artificial intelligence, GIS, and cloud computing help assess climate impacts on urban systems at various scales. We conducted a comprehensive systematic literature review to explore the intersection of urban climate and computer science and examine how this cross-disciplinary approach addresses various urban climate challenges.
Protik Bose Pranto
International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) 2024 Paper
The rise of the internet and social media platforms has led to a significant increase in the spread of fake news in recent years. This rapid dissemination, especially in sensitive areas such as politics and finance, has caused considerable harm, prompting researchers to explore ways of automatically identifying misinformation through language structure. However, distinguishing between fake news and satirical content remains a challenge due to the nuanced nature of satire. While much research has focused on detecting fake news, relatively little attention has been given to the specific task of differentiating it from satirical news. To bridge this gap, this study investigates the performance of nine traditional AI models and three trans-former models.
Protik Bose Pranto
International Conference on Electrical, Computer, Communications and Mechatronics Engineering (ICECCME) 2024 Paper
The rise of the internet and social media platforms has led to a significant increase in the spread of fake news in recent years. This rapid dissemination, especially in sensitive areas such as politics and finance, has caused considerable harm, prompting researchers to explore ways of automatically identifying misinformation through language structure. However, distinguishing between fake news and satirical content remains a challenge due to the nuanced nature of satire. While much research has focused on detecting fake news, relatively little attention has been given to the specific task of differentiating it from satirical news. To bridge this gap, this study investigates the performance of nine traditional AI models and three trans-former models.
Protik Bose Pranto, Waqar Hassan Khan, Sahar Abdelnabi, Rebecca Weil, Mario Fritz, Rakibul Hasan
Design X Policy (CHI Workshop) 2023 Paper Abstract
We outline a planned experiment to investigate if personal data (eg, demographics and behavioral patterns) can be used to selectively expose individuals to disinformation such that an adversary can spread disinformation more efficiently compared to broadcasting the same information to everyone. This mechanism, if effective, will have devastating consequences as modern technologies collect and infer a plethora of private data that can be abused to target with disinformation. We believe this research will inform designing policies and regulations for online platforms.
Protik Bose Pranto, Waqar Hassan Khan, Sahar Abdelnabi, Rebecca Weil, Mario Fritz, Rakibul Hasan
Design X Policy (CHI Workshop) 2023 Paper Abstract
We outline a planned experiment to investigate if personal data (eg, demographics and behavioral patterns) can be used to selectively expose individuals to disinformation such that an adversary can spread disinformation more efficiently compared to broadcasting the same information to everyone. This mechanism, if effective, will have devastating consequences as modern technologies collect and infer a plethora of private data that can be abused to target with disinformation. We believe this research will inform designing policies and regulations for online platforms.
Bishal Basak Papan, Protik Bose Pranto, Md Saidur Rahman
The Computer Journal 2023 Paper
A graph $G = (V, E)$ is called a pairwise compatibility graph (PCG) if it admits a tuple $(T, d_{min}, d_{max})$ of an edge-weighted tree $T$ of non-negative edge weights with leaf set $L$, two non-negative real numbers $d_{min} \leq d_{max}$ such that each vertex $u' \in V$ represents a leaf $u \in L$ and $G$ has an edge $(u', v') \in E$ if and only if the distance between the two leaves $u$ and $v$ in the tree $T$ lies within interval $[d_{min}, d_{max}]$. It has been proven that not all graphs are PCGs. A graph $G$ is called a $k$-interval PCG if there exists an edge-weighted tree $T$ and $k$ mutually exclusive intervals of non-negative real numbers such that there is an edge between two vertices in $G$ if and only if the distance between their corresponding leaves in $T$ lies within any of the $k$ intervals. It is known that every graph $G$ is a $k$-interval PCG for $k = |E|$, where $E$ is the set of edges of $G$. It is thus interesting to know the smallest value of $k$ for which $G$ is a $k$-interval PCG. In this paper, we show that grid graphs and a subclass of 3D grid graphs are 2-interval PCGs.
Bishal Basak Papan, Protik Bose Pranto, Md Saidur Rahman
The Computer Journal 2023 Paper
A graph $G = (V, E)$ is called a pairwise compatibility graph (PCG) if it admits a tuple $(T, d_{min}, d_{max})$ of an edge-weighted tree $T$ of non-negative edge weights with leaf set $L$, two non-negative real numbers $d_{min} \leq d_{max}$ such that each vertex $u' \in V$ represents a leaf $u \in L$ and $G$ has an edge $(u', v') \in E$ if and only if the distance between the two leaves $u$ and $v$ in the tree $T$ lies within interval $[d_{min}, d_{max}]$. It has been proven that not all graphs are PCGs. A graph $G$ is called a $k$-interval PCG if there exists an edge-weighted tree $T$ and $k$ mutually exclusive intervals of non-negative real numbers such that there is an edge between two vertices in $G$ if and only if the distance between their corresponding leaves in $T$ lies within any of the $k$ intervals. It is known that every graph $G$ is a $k$-interval PCG for $k = |E|$, where $E$ is the set of edges of $G$. It is thus interesting to know the smallest value of $k$ for which $G$ is a $k$-interval PCG. In this paper, we show that grid graphs and a subclass of 3D grid graphs are 2-interval PCGs.
Protik Bose Pranto, Waqar Hassan Khan, Sahar Abdelnabi, Rebecca Weil, Mario Fritz, Rakibul Hasan
USENIX Symposium on Usable Privacy and Security (SOUPS) 2023 Paper Abstract
Two of the major issues society is currently facing are Privacy threats from data collection by digital platforms, and the quick and large-scale propagation of disinformation on the same platforms. What if the first problem further fuels the second? Specifically, we hypothesize that private data, such as demographics, interests, and psychological and physiological states could be used to expose people to certain disinformation, resulting in higher engagement, and ultimately enabling an adversary to propagate disinformation more efficiently and effectively. This abstract details the experiment design to test this hypothesis and initial findings.
Protik Bose Pranto, Waqar Hassan Khan, Sahar Abdelnabi, Rebecca Weil, Mario Fritz, Rakibul Hasan
USENIX Symposium on Usable Privacy and Security (SOUPS) 2023 Paper Abstract
Two of the major issues society is currently facing are Privacy threats from data collection by digital platforms, and the quick and large-scale propagation of disinformation on the same platforms. What if the first problem further fuels the second? Specifically, we hypothesize that private data, such as demographics, interests, and psychological and physiological states could be used to expose people to certain disinformation, resulting in higher engagement, and ultimately enabling an adversary to propagate disinformation more efficiently and effectively. This abstract details the experiment design to test this hypothesis and initial findings.
Waqar Hassan Khan, Protik Bose Pranto, Tianyi Yang, Rakibul Hasan
USENIX Symposium on Usable Privacy and Security (SOUPS) 2023 Paper Abstract
The rapid growth of technology’s use in educational institutes, accompanied by numerous incidents of data breaches as well as data abuse for profit, has raised concerns regarding users’ privacy, security, and safety. Different from other contexts (eg, social media), institutionalized use of technologies rarely offers any option to opt out and involves multiple user groups (eg, students and instructors) with power asymmetries, further complicating the situation. To discover perceptions and concerns from different user groups, we manually analyzed 3,300 online reviews of 33 education technologies. We conducted a thematic analysis of the 163 reviews that expressed concerns about privacy/security harms from the applications and identified five themes. Additionally, we identified 77 reviews (through keyword search and then manual annotations) where users anticipated harm from other users and found one additional theme, totaling six themes.
Waqar Hassan Khan, Protik Bose Pranto, Tianyi Yang, Rakibul Hasan
USENIX Symposium on Usable Privacy and Security (SOUPS) 2023 Paper Abstract
The rapid growth of technology’s use in educational institutes, accompanied by numerous incidents of data breaches as well as data abuse for profit, has raised concerns regarding users’ privacy, security, and safety. Different from other contexts (eg, social media), institutionalized use of technologies rarely offers any option to opt out and involves multiple user groups (eg, students and instructors) with power asymmetries, further complicating the situation. To discover perceptions and concerns from different user groups, we manually analyzed 3,300 online reviews of 33 education technologies. We conducted a thematic analysis of the 163 reviews that expressed concerns about privacy/security harms from the applications and identified five themes. Additionally, we identified 77 reviews (through keyword search and then manual annotations) where users anticipated harm from other users and found one additional theme, totaling six themes.
Protik Bose Pranto, Syed Zami-Ul-Haque Navid, Protik Dey, Gias Uddin, Anindya Iqbal
arXiv preprint 2022 Paper
Our opinions and views of life can be shaped by how we perceive the opinions of others on social media like Facebook. This dependence has increased during COVID-19 periods when we have fewer means to connect with others. However, fake news related to COVID-19 has become a significant problem on Facebook. Bengali is the seventh most spoken language worldwide, yet we are aware of no previous research that studied the prevalence of COVID-19 related fake news in Bengali on Facebook. In this paper, we develop machine learning models to detect fake news in Bengali automatically. The best performing model is BERT, with an F1-score of 0.97. We apply BERT on all Facebook Bengali posts related to COVID-19. We find 10 topics in the COVID-19 Bengali fake news grouped into three categories: System (e.g., medical system), belief (e.g., religious rituals), and social (e.g., scientific awareness).
Protik Bose Pranto, Syed Zami-Ul-Haque Navid, Protik Dey, Gias Uddin, Anindya Iqbal
arXiv preprint 2022 Paper
Our opinions and views of life can be shaped by how we perceive the opinions of others on social media like Facebook. This dependence has increased during COVID-19 periods when we have fewer means to connect with others. However, fake news related to COVID-19 has become a significant problem on Facebook. Bengali is the seventh most spoken language worldwide, yet we are aware of no previous research that studied the prevalence of COVID-19 related fake news in Bengali on Facebook. In this paper, we develop machine learning models to detect fake news in Bengali automatically. The best performing model is BERT, with an F1-score of 0.97. We apply BERT on all Facebook Bengali posts related to COVID-19. We find 10 topics in the COVID-19 Bengali fake news grouped into three categories: System (e.g., medical system), belief (e.g., religious rituals), and social (e.g., scientific awareness).
Protik Bose Pranto, Bishal Basak Papan,, Md Saidur Rahman
IEEE International Conference on Telecommunications and Photonics (ICTP) 2021 Paper
In a k-safe labeling of a graph G, each vertex is labeled by a distinct positive integer such that the difference of the labels of two adjacent vertices is at least k. The span of a k-safe labeling of G is the range between the minimum and the maximum labels used in G. The k-safe labeling problem asks to label all the vertices of G using the minimum span. This problem has practical applications in assigning frequencies of transmitters in a network. k-safe labeling problem has been proven to be NP-hard and there is not an exact upper bound on the span of k-safe labeling of a graph. In this paper, we give an upper bound on k-safe labelings of all connected graphs based on the size of the maximum clique in the graph. Our proof leads to a polynomial-time algorithm for finding a k-safe labeling of any connected graph attaining the bound.
Protik Bose Pranto, Bishal Basak Papan,, Md Saidur Rahman
IEEE International Conference on Telecommunications and Photonics (ICTP) 2021 Paper
In a k-safe labeling of a graph G, each vertex is labeled by a distinct positive integer such that the difference of the labels of two adjacent vertices is at least k. The span of a k-safe labeling of G is the range between the minimum and the maximum labels used in G. The k-safe labeling problem asks to label all the vertices of G using the minimum span. This problem has practical applications in assigning frequencies of transmitters in a network. k-safe labeling problem has been proven to be NP-hard and there is not an exact upper bound on the span of k-safe labeling of a graph. In this paper, we give an upper bound on k-safe labelings of all connected graphs based on the size of the maximum clique in the graph. Our proof leads to a polynomial-time algorithm for finding a k-safe labeling of any connected graph attaining the bound.