Does it help to charge the electric cars at intersections?

Accepted paper in Computer-Aided Civil and Infrastructure Engineering 

Khan, Khan Chowdhury, Safro, Ushijima-Mwesigwa "Wireless Charging Utility Maximization and Intersection Control Delay Minimization Framework for Electric Vehicles"

This study presents the Wireless Charging Utility Maximization (WCUM) framework, which aims to maximize the utility of Wireless Charging Units (WCUs) for electric vehicle (EV) charging through the optimal WCU deployment at signalized intersections. Furthermore, the framework aims to minimize the control delay at all signalized intersections of the network. The framework consists of a two‐step optimization formulation, a dynamic traffic assignment model to calculate the user equilibrium, a traffic microsimulator to formulate the objective functions, and a global Mixed Integer Non‐Linear Programming (MINLP) optimization solver. An optimization problem is formulated for each intersection, and another for the entire network. The performance of the WCUM framework is tested using the Sioux Falls network. We perform a comparative study of 12 global MINLP solvers with a case study. Based on solution quality and computation time, we choose the Couenne solver for this framework.

https://onlinelibrary.wiley.com/doi/abs/10.1111/mice.12439

Predicting bariatric surgery outcomes

Accepted paper in Annals of Operations Research

Razzaghi, Safro, Ewing, Sadrfaridpour, Scott "Predictive models for bariatric surgery risks with imbalanced medical datasets"

Bariatric surgery (BAR) has become a popular treatment for type 2 diabetes mellitus which is among the most critical obesity-related comorbidities. Patients who have bariatric surgery, are exposed to complications after surgery. Furthermore, the mid- to long-term complications after bariatric surgery can be deadly and increase the complexity of managing safety of these operations and healthcare costs. Current studies on BAR complications have mainly used risk scoring for identifying patients who are more likely to have complications after surgery. Though, these studies do not take into consideration the imbalanced nature of the data where the size of the class of interest (patients who have complications after surgery) is relatively small. We propose the use of imbalanced classification techniques to tackle the imbalanced bariatric surgery data: synthetic minority oversampling technique (SMOTE), random undersampling, and ensemble learning classification methods including Random Forest, Bagging, and AdaBoost. Moreover, we improve classification performance through using Chi-squared, Information Gain, and Correlation-based feature selection techniques. We study the Premier Healthcare Database with focus on the most-frequent complications including Diabetes, Angina, Heart Failure, and Stroke. Our results show that the ensemble learning-based classification techniques using any feature selection method mentioned above are the best approach for handling the imbalanced nature of the bariatric surgical outcome data. In our evaluation, we find a slight preference toward using SMOTE method compared to the random undersampling method. These results demonstrate the potential of machine-learning tools as clinical decision support in identifying risks/outcomes associated with bariatric surgery and their effectiveness in reducing the surgery complications as well as improving patient care.

AI@Clemson

It was great discussing AI and text mining at Clemson University research  symposium on AI.

Welcome new students!

Two new students, Zirou Qiu (MSc) and Korey Palmer (senior undergrad) are joining our research group.

Relaxation-Based Coarsening for Multilevel Hypergraph Partitioning

Accepted paper at SIAM Multiscale Modeling and Simulations

Ruslan Shaydulin, Jie Chen, Ilya Safro "Relaxation-Based Coarsening for Multilevel Hypergraph Partitioning", 2019, preprint at arXiv:1710.06552

Multilevel partitioning methods that are inspired by principles of multiscaling are the most powerful practical hypergraph partitioning solvers. Hypergraph partitioning has many applications in disciplines ranging from scientific computing to data science. In this paper we introduce the concept of algebraic distance on hypergraphs and demonstrate its use as an algorithmic component in the coarsening stage of multilevel hypergraph partitioning solvers. The algebraic distance is a vertex distance measure that extends hyperedge weights for capturing the local connectivity of vertices which is critical for hypergraph coarsening schemes. The practical effectiveness of the proposed measure and corresponding coarsening scheme is demonstrated through extensive computational experiments on a diverse set of problems. Finally, we propose a benchmark of hypergraph partitioning problems to compare the quality of other solvers.

Thesis defense

Congratulations to Varsha Chauhan for successfully defending her MSc thesis on planar graph generation! 

Travel awards

Congratulations to Justin Sybrandt and Ruslan Shaydulin for receiving travel awards to present their papers at IEEE Big Data 2018 and APS 2018!

Thesis defense

Congratulations to Dr. Hayato Ushijima-Mwesigwa for successfully defending his Ph.D. thesis "Models for Networks with Consumable Resources"!

Community detection on NISQ devices

Accepted paper at at 3rd International Workshop on Post Moore's Era 2018

Supercomputing (PMES 2018)

Ruslan Shaydulin, Haayto Ushijima-Mwesigwa, Ilya Safro, Susan Mniszewski, Yuri Alexeev "Community Detection Across Emerging Quantum Architectures", preprint at arXiv:1810.07765, 2018

Can we predict crimes in Chicago?

Our paper is accepted at IEEE Big Data 2018

Saroj K. Dash, I. Safro, Ravisutha S. Srinivasamurthy "Spatio-temporal prediction of crimes using network analytic approach", preprint at arXiv:1808.06241, 2018

It is quite evident that majority of the population lives in urban area today than in any time of the human history. This trend seems to increase in coming years. Studies say that nearly 80.7% of total population in USA stays in urban area. By 2030 nearly 60% of the population in the world will live in or move to cities. With the increase in urban population, it is important to keep an eye on criminal activities. By doing so, governments can enforce intelligent policing systems and hence many government agencies and local authorities have made the crime data publicly available. In this paper, we analyze Chicago city crime data fused with other social information sources using network analytic techniques to predict criminal activity for the next year. We observe that as we add more layers of data which represent different aspects of the society, the quality of prediction is improved. Our prediction models not just predict total number of crimes for the whole Chicago city, rather they predict number of crimes for all types of crimes and for different regions in City of Chicago.

Two papers accepted at IEEE Big Data 2018

Sybrandt, Carrabba, Herzog, Safro "Are Abstracts Enough for Hypothesis Generation?", arXiv:1804.05942

Sybrandt, Shtutman, Safro "Large-Scale Validation of Hypothesis Generation Systems via Candidate Ranking", arXiv:1802.03793