SSSomya Sharma

Somya Sharma

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Somya Sharma has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Classifying Violent Anti-government Conflicts in Mexico: A Machine Learning Framework

    by Vishal Subedi, Harsh Hemant Shah, Benjamin Bagozzi, Somya Sharma, Snigdhansu Chatterjee

    Domestic crime, conflict, and instability pose a significant threat to many contemporary governments. These challenges have proven to be particularly acute within modern-day Mexico. While there have been significant developments in predicting intrastate armed and electoral conflict in various contemporary settings, such efforts have thus far been limited in their use of spatial as well as temporal correlations, as well as in the features they have consid- ered. Machine learning, especially deep learning, has been proven to be highly effective in predicting future conflicts using word embeddings in Convolutional Neural Networks (CNN) but lacks the spatial structure and, due to the black box nature, cannot explain the importance of predictors. We develop a novel methodology using machine learning that can accurately classify future anti-government violence in Mexico. We further demonstrate that our approach can identify important leading predictors of such violence. This can help policymakers make informed decisions and can also help governments and NGOs better allocate security and humanitarian resources, which could prove beneficial in tackling this problem. Using a variety of political event aggregations from the ICEWS database alongside other textual and demographic features, we trained various classical machine learning algorithms, including but not limited to Logistic Regression, Random Forest, XGBoost, and a Voting classifier. We considered a mix of CNN + Long Short Term Memory (LSTM) networks to decode the spatial and temporal relationship in the data. The performance of all the black box deep learning models was not at par with the classical machine learning models. We also evaluated the temporal relationships in the data to identify important predictors that fuel anti-government violence along with their directions using Shapley additive values. The voting classifier, utilizing a subset of features derived from LASSO across 100 simulations, consistently surpasses alternative models in performance and demonstrates efficacy in accu- rately classifying future anti-government conflicts. Notably, Random Forest feature importance indicates that some features, including but not limited to homicides, accidents, material conflicts, and positive worded citizen information sentiments emerge as pivotal predictors in the classification of anti-government conflicts. Finally, we conclude the research by analysing the spatial structure of the data using Moran’s I index extended version for spatiotemporal data to identify the global spatial dependency and local clusters followed by modelling the data spatially and evaluating the same using Gaussian Process Boost(GPBoost). The global spatial autocorrelation is minimal, characterized by local- ized conflicts cluster within the region. Furthermore, in Phase 3, the Voting Classifier method demonstrates superior performance over GPBoost, leading to the inference that no substantial spatial dependency exists among the various locations.

  • Open Access

    Entity Aware Modelling: A Survey

    by Rahul Ghosh, Haoyu Yang, Ankush Khandelwal, Erhu He, Arvind Renganathan, Somya Sharma, Xiaowei Jia, Vipin Kumar

    Personalized prediction of responses for individual entities caused by external drivers is vital across many disciplines. Recent machine learning (ML) advances have led to new state-of-the-art response prediction models. Models built at a population level often lead to sub-optimal performance in many personalized prediction settings due to heterogeneity in data across entities (tasks). In personalized prediction, the goal is to incorporate inherent characteristics of different entities to improve prediction performance. In this survey, we focus on the recent developments in the ML community for such entity-aware modeling approaches. ML algorithms often modulate the network using these entity characteristics when they are readily available. However, these entity characteristics are not readily available in many real-world scenarios, and different ML methods have been proposed to infer these characteristics from the data. In this survey, we have organized the current literature on entity-aware modeling based on the availability of these characteristics as well as the amount of training data. We highlight how recent innovations in other disciplines, such as uncertainty quantification, fairness, and knowledge-guided machine learning, can improve entity-aware modeling.

  • Open Access

    Machine Learning Methods for Multiscale Physics and Urban Engineering Problems

    by Somya Sharma, Marten Thompson, Debra Laefer, Michael Lawler, Kevin McIlhany, Olivier Pauluis, Dallas R. Trinkle, Snigdhansu Chatterjee

    We present an overview of four challenging research areas in multiscale physics and engineering as well as four data science topics that may be developed for addressing these challenges. We focus on multiscale spatiotemporal problems in light of the importance of understanding the accompanying scientific processes and engineering ideas, where “multiscale” refers to concurrent, non-trivial and coupled models over scales separated by orders of magnitude in either space, time, energy, momenta, or any other relevant parameter. Specifically, we consider problems where the data may be obtained at various resolutions; analyzing such data and constructing coupled models led to open research questions in various applications of data science. Numeric studies are reported for one of the data science techniques discussed here for illustration, namely, on approximate Bayesian computations.

  • Open Access

    Winsorization for Robust Bayesian Neural Networks

    by Somya Sharma, Snigdhansu Chatterjee

    With the advent of big data and the popularity of black-box deep learning methods, it is imperative to address the robustness of neural networks to noise and outliers. We propose the use of Winsorization to recover model performances when the data may have outliers and other aberrant observations. We provide a comparative analysis of several probabilistic artificial intelligence and machine learning techniques for supervised learning case studies. Broadly, Winsorization is a versatile technique for accounting for outliers in data. However, different probabilistic machine learning techniques have different levels of efficiency when used on outlier-prone data, with or without Winsorization. We notice that Gaussian processes are extremely vulnerable to outliers, while deep learning techniques in general are more robust.


  • AgriAcT: Agricultural Activity Training using multimedia and wearable sensing

    by Somya Sharma, Bhushan Jagyasi, Jabal Raval, Prashant Patil

    There has been immense work in past on the human activities detection and context recognition using the wearable sensing technologies. However, a more challenging problem of providing training on the activities to the users with the help of wearable sensors has not been adequately attempted. Specially, in the agriculture applications, an appropriate training to the farmers on performing the agricultural activities would result in the sustainable agriculture practices for achieving higher and better quality yield. In this paper, a novel first-of-a-kind, multimedia and wearable sensors based Agricultural Activity Training (AgriAcT) system has been proposed for the dissemination of agricultural technologies to the remotely located farmers. In the proposed system, a training video of an expert farmer performing an activity is captured along with the gesture data obtained from the wearable motion sensors from the expert's body while the activity is being performed. A trainee farmer, can learn a selected activity by watching the multimedia content of the expert performing that activity on the mobile phone and subsequently perform the activity by wearing the required motion sensors. We present a novel K-Nearest Neighbor based Agriculture Activity Performance Score (KAAPS) engine to generate an Activity performance score (AcT-Score) which suggest how efficiently the activity had been performed by the trainee as compared to the expert's performance. The exhaustive experimental results by collecting data from eight experts and ten trainees for two different activities are used to present the inferences on the impact made by the Act-Score on the performance of the trainee farmers.