Article
Computer Science, Information Systems
Osman Doluca, Kaya Oguz
Summary: APAL is a novel algorithm that enhances the detection of overlapping communities by considering the importance of adjacent vertices. In comparisons with other algorithms, APAL performs significantly better in networks with an increasing number of overlapping vertex memberships.
INFORMATION SCIENCES
(2021)
Article
Computer Science, Artificial Intelligence
Sondos Bahadori, Hadi Zare, Parham Moradi
Summary: A probabilistic overlapping community detection method called PODCD is proposed in this work, which considers dense connections between communities and utilizes a probabilistic model to control the dynamics of community structure. Experimental results demonstrate that the proposed method outperforms earlier algorithms on evolving networks.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Artificial Intelligence
Yanhao Yang, Pan Shi, Yuyi Wang, Kun He
Summary: Community detection is a crucial task in network analysis. This study introduces QOCE, an overlapping community detection method that does not require prior knowledge. The method expands seed sets and uses quadratic optimization to determine community boundaries, with competitive performance shown in evaluations on synthetic and real-world networks.
KNOWLEDGE-BASED SYSTEMS
(2022)
Article
Multidisciplinary Sciences
Stefano Benati, Justo Puerto, Antonio Rodriguez-Chia, Francisco Temprano
Summary: In this paper, a new model based on cooperative games and mathematical programming is proposed to detect overlapping communities in a network. Communities are defined as stable coalitions of a weighted graph community game and are revealed as the optimal solution of a mixed-integer linear programming problem. Exact optimal solutions are obtained for small and medium sized instances, providing useful information about the network structure. A heuristic algorithm is also developed to solve larger instances and compare two variations of the objective function.
Article
Computer Science, Theory & Methods
Chaobo He, Hai Liu, Yong Tang, Shuangyin Liu, Xiang Fei, Qiwei Cheng, Hanchao Li
Summary: Community detection in signed networks is a challenging research problem, and overlapping community detection is a less explored direction. This paper proposes a similarity preserving overlapping community detection method (SPOCD) that fuses node similarity and geometric structure information to better preserve nodes with high similarity in the same community.
FUTURE GENERATION COMPUTER SYSTEMS-THE INTERNATIONAL JOURNAL OF ESCIENCE
(2021)
Article
Engineering, Electrical & Electronic
Xinguo Lu, Zhenghao Zhu, Xianghua Peng, Qiumai Miao, Yuansheng Luo, Xiangtao Chen
Summary: Community detection is crucial in network analysis, especially in biological networks. InFun utilizes a two-way clustering method to detect gene communities, offering a new approach for understanding and utilizing cancer mechanisms.
SIGNAL IMAGE AND VIDEO PROCESSING
(2021)
Article
Computer Science, Information Systems
Baohua Sun, Richard Al-Bayaty, Qiuyuan Huang, Dapeng Wu
Summary: This research evaluates the performance of clustering algorithms on large networks for the first time and proposes a new algorithm called Game Theoretical Approach for Clustering (GTAC), which offers scalability and tunability for big data applications.
TSINGHUA SCIENCE AND TECHNOLOGY
(2021)
Article
Physics, Multidisciplinary
Hanyang Lin, Yongzhao Zhan, Zizheng Zhao, Yuzhong Chen, Chen Dong
Summary: The paper introduces an algorithm for detecting overlapping communities in social networks using an augmented attribute graph and an improved weight adjustment strategy. The algorithm also automatically determines the number of communities, resulting in successful detection of overlapping communities in experiments.
Article
Computer Science, Information Systems
Chaobo He, Yulong Zheng, Junwei Cheng, Yong Tang, Guohua Chen, Hai Liu
Summary: This paper proposes a semi-supervised overlapping community detection method named SSGCAE, which is based on graph neural networks. It addresses the problems of link and attribute fusion, prior information integration, and overlapping community detection in attributed graphs.
INFORMATION SCIENCES
(2022)
Article
Computer Science, Artificial Intelligence
Vincenzo Moscato, Giancarlo Sperli
Summary: Detecting users' communities in Online Social Networks is crucial for enhancing the effectiveness of diffusion of new ideas, improving recommendation suggestions, and finding experts. Different community detection techniques based on game theory, artificial intelligence, and fuzzy strategies are compared for various OSN models, highlighting pros and cons. Challenges and open issues in the community detection problem are discussed for future research.
KNOWLEDGE-BASED SYSTEMS
(2021)
Article
Physics, Multidisciplinary
Stefano Benati, Justo Puerto, Antonio M. Rodriguez-Chia, Francisco Temprano
Summary: In this article, a new optimization model is proposed to detect overlapping communities in networks. The model not only addresses the biases of previous models but also reveals additional structural properties. Furthermore, two heuristic algorithms are introduced to handle larger instances, which show favorable performance compared to other methodologies.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2022)
Article
Multidisciplinary Sciences
Alexander Ponomarenko, Leonidas Pitsoulis, Marat Shamshetdinov
Summary: The LPAM method introduces a new approach for detecting overlapping communities in networks, considering different distance functions and evaluating its performance on real life instances and synthetic network benchmarks. It utilizes link partitioning and partitioning around medoids to detect overlapping communities in graphs.
Article
Computer Science, Information Systems
Dong Liu, Guoliang Yang, Yanwei Wang, Hu Jin, Enhong Chen
Summary: Overlapping community detection algorithms have gained more attention in recent years, revealing real social relations and communication channels between communities. However, some individuals may not want to be found in overlapping areas, prompting the question of how to avoid community detection algorithms.
IEEE TRANSACTIONS ON BIG DATA
(2022)
Article
Computer Science, Artificial Intelligence
Yang Gao, Xiangzhan Yu, Hongli Zhang
Summary: Local community detection methods aim at finding communities around initial nodes in a network, addressing efficiency problems faced by global clustering methods. Techniques like personalized PageRank and heat kernel diffusion are used to rank proximity scores of vertices nearby.
EXPERT SYSTEMS WITH APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Zhigao Zheng, Xuanhua Shi, Hai Jin
Summary: In this paper, the authors propose Lugger, a GPU-based overlapping community detection algorithm that reduces branch divergence and improves performance in graph mining applications. Through a cache-aware parallel searching policy and a positive node splitting scheme, Lugger outperforms existing works on scalability and detection quality.
IEEE TRANSACTIONS ON BIG DATA
(2023)
Article
Biology
Saumya Yashmohini Sahai, Saket Gurukar, Wasiur R. KhudaBukhsh, Srinivasan Parthasarathy, Grzegorz A. Rempala
Summary: Due to reporting delays, the accuracy of daily national and statewide COVID-19 incidence counts is often questionable, necessitating estimation from recent data. This paper presents a simple random forest statistical model for nowcasting the daily new infection counts based on historical data and simple covariates such as current reported infection counts, day of the week, and time since first reporting. Application of the model in adjusting daily infection counts in Ohio demonstrates that this data-driven method outperforms a complex hierarchical Bayesian model in terms of prediction quality and computational burden. The interactive notebook for nowcasting can be accessed online at https://tinyurl.com/simpleMLnowcasting.
MATHEMATICAL BIOSCIENCES
(2022)
Article
Psychology, Experimental
Goonmeet Bajaj, Sean Current, Daniel Schmidt, Bortik Bandyopadhyay, Christopher W. Myers, Srinivasan Parthasarathy
Summary: The study of human cognition and artificial intelligence have a symbiotic relationship, with human cognition possessing abilities that modern AI systems cannot compete with, such as the detection, identification, and resolution of knowledge gaps. Researchers aim to incorporate these capabilities into artificial agents to explore the understanding of knowledge gaps in visual-linguistic communication.
TOPICS IN COGNITIVE SCIENCE
(2022)
Review
Biochemical Research Methods
Debomita Chakraborty, Raghunathan Rengaswamy, Karthik Raman
Summary: This paper systematically organizes key works in the field of genetic circuit design using the framework of generalized morphological analysis. It maps literature based on design methodologies, modeling techniques, circuit functionalities, design characteristics, and strategies for robust design. The paper concludes with an outlook on future research areas based on the assessment of research gaps.
ACS SYNTHETIC BIOLOGY
(2022)
Article
Genetics & Heredity
Malvika Sudhakar, Raghunathan Rengaswamy, Karthik Raman
Summary: The study develops a multi-omic approach called PIVOT, which uses a machine learning model to classify genes as tumor suppressor genes, oncogenes, or neutral genes based on their functional impact in patients. The models trained on multi-omic data improve predictions and identify both common and rare driver genes.
FRONTIERS IN GENETICS
(2022)
Article
Physics, Multidisciplinary
Sai Saranga Das, Karthik Raman
Summary: This study proposes a strategy to enhance the robustness of networks by allocating spare capacity for vulnerable nodes, resulting in significant improvements in both scale-free and real-world networks.
PHYSICA A-STATISTICAL MECHANICS AND ITS APPLICATIONS
(2022)
Article
Computer Science, Artificial Intelligence
Bo Peng, Zhiyun Ren, Srinivasan Parthasarathy, Xia Ning
Summary: Sequential recommendation is a useful tool for helping users select preferred items based on their purchase/rating history. This paper proposes a hybrid associations model (HAM) that considers users' long-term preferences, association patterns in recent purchases/ratings, and item synergies to generate sequential recommendations. Experimental results show that HAM models outperform the state-of-the-art methods in terms of performance and runtime efficiency.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Multidisciplinary Sciences
Sankalpa Venkatraghavan, Sathvik Anantakrishnan, Karthik Raman
Summary: Microbial consortia exhibit spatiotemporal organization through bacterial communication. Simulations show that secretion rates play a key role in controlling the behavior of the coupled consortia. These models provide a simplified and controllable approach to pattern formation in synthetic biology.
SCIENTIFIC REPORTS
(2022)
Article
Biology
Priyan Bhattacharya, Karthik Raman, Arun K. Tangirala
Summary: In this work, a generic methodology inspired by systems theory is presented to discover the design principles for robust adaptation in two different contexts: deterministic and stochastic. Contrary to the existing approaches, the proposed methodologies provide admissible network structures without resorting to computationally burdensome techniques. The proposed frameworks do not assume prior knowledge about the particular rate kinetics, thereby validating the conclusions for a large class of biological networks.
MATHEMATICAL BIOSCIENCES
(2023)
Article
Multidisciplinary Sciences
Prem Jagadeesan, Karthik Raman, Arun K. Tangirala
Summary: Computational modelling of biological processes faces multiple challenges, such as identifiability, parameter estimation from limited data, informative experiments, and anisotropic sensitivity. Sloppiness, the property where model predictions are nearly identical over large regions in the parameter space, is a crucial but inconspicuous challenge. This study addresses critical unanswered questions about sloppiness, including its quantification, practical implications, and its impact on system identification.
Review
Multidisciplinary Sciences
Pratyay Sengupta, Shobhan Karthick Muthamilselvi Sivabalan, Amrita Mahesh, Indumathi Palanikumar, Dinesh Kumar Kuppa Baskaran, Karthik Raman
Summary: Microorganisms are widely distributed in nature and form complex networks to survive in different environments. The structure of these communities is influenced by factors like nutrient availability, temperature, pH, and microbial composition. Categorizing accessible biomes according to their habitats helps in understanding the complexity of environment-specific communities.
JOURNAL OF THE INDIAN INSTITUTE OF SCIENCE
(2023)
Article
Multidisciplinary Sciences
Melpakkam Pradeep, Karthik Raman
Summary: The COVID-19 pandemic has posed a significant challenge to global healthcare systems. Multiple waves of the disease have strained healthcare resources worldwide, leading to diligent data collection. This manuscript collates COVID-19 case data from the World Health Organization website to create a labelled dataset for building supervised learning classifiers. The dataset, along with a simple XGBoost model, demonstrates its utility for predicting future waves and will be valuable for epidemiologists and others interested in early prediction.
Article
Ecology
Dinesh Kumar Kuppa Baskaran, Shreyansh Umale, Zhichao Zhou, Karthik Raman, Karthik Anantharaman
Summary: Deep-sea hydrothermal vents are abundant and have important roles in ocean biogeochemistry. By studying microbial communities in the Guaymas Basin hydrothermal system, we identified key species and their interactions. Metabolic models were used to infer metabolic exchanges and horizontal gene transfer events within the community. Our findings highlight the importance of microbial interactions in driving community structure and organization in hydrothermal plume microbiomes.
ISME COMMUNICATIONS
(2023)
Proceedings Paper
Automation & Control Systems
Prem Jagadeesan, Karthik Raman, Arun K. Tangirala
Summary: In complex dynamical systems, the precise estimation of parameters and prediction quality rely on the information contained in experimental data. Optimal Experimental Design (OEL) refers to the selection of experimental schemes that maximize the information in the data. OED utilizes Fisher Information Matrix and variance-covariance matrix as central concepts. However, in sloppy models, applying OED leads to decreased predictive ability despite precise parameter estimation. This study introduces a new information gain index as an experiment design criterion in the Bayesian framework, demonstrating its effectiveness in minimizing prediction and parameter uncertainty in sloppy models through simulations.
Review
Biology
Priyan Bhattacharya, Karthik Raman, Arun K. Tangirala
Summary: Network architecture plays a crucial role in governing the dynamics of biological networks, and the mapping between network structures and output functionality aids in understanding biological systems and has applications in synthetic biology and therapeutics. This review provides a qualitative and quantitative study of computational efforts, rule-based methods, and systems-theoretic approaches, based on the well-researched biological phenotypes of oscillation, toggle switching, and adaptation.
JOURNAL OF BIOSCIENCES
(2022)
Article
Genetics & Heredity
Anjana Anilkumar Sithara, Devi Priyanka Maripuri, Keerthika Moorthy, Sai Sruthi Amirtha Ganesh, Philge Philip, Shayantan Banerjee, Malvika Sudhakar, Karthik Raman
Summary: This article introduces a user-friendly pipeline tool for analyzing cancer genomic data, capable of analyzing whole-genome and transcriptome data, predicting pathogenicity of mutations, and distinguishing tumor suppressor genes from oncogenes.
NAR GENOMICS AND BIOINFORMATICS
(2022)