Publication Types:

Sort by year:

Automated ontology-based annotation of scientific literature using deep learning.

Somya D. Mohanty, Prashanti Manda, and Saed Sayedahmed
ACM Special Interest Group onManagement of Data (SIGMOD) 2020
Publication year: 2021

Automated ontology-based annotation of scientific literature using deep learning.

Prashanti Manda, Saed Sayedahmed, Somya D. Mohanty
Proceedings of the International Workshop on Semantic Big Data, SBD - In conjunction with the 2020 ACM SIGMOD/PODS Conference
Publication year: 2020

Representing scientific knowledge using ontologies enables data integration, consistent machine-readable data representation, and allows for large-scale computational analyses. Text mining approaches that can automatically process and annotate scientific literature with ontology concepts are necessary to keep up with the rapid pace of scientific publishing. Here, we present deep learning models (Gated Recurrent Units (GRU) and Long Short Term Memory (LSTM)) combined with different input encoding formats for automated Named Entity Recognition (NER) of ontology concepts from text. The Colorado Richly Annotated Full Text (CRAFT) gold standard corpus was used to train and test our models. Precision, Recall, F-1, and Jaccard semantic similarity were used to evaluate the performance of the models. We found that GRU-based models outperform LSTM models across all evaluation metrics. Surprisingly, considering the top two probabilistic predictions of the model for each instance instead of the top one resulted in a substantial increase in accuracy. Inclusion of ontology semantics via subsumption reasoning yielded modest performance improvement.

An Active Learning Pipeline to Detect Hurricane Washover in Post-Storm Aerial Images.

Evan B. Goldstein, Somya D. Mohanty, Shah N. Rafique, Jamison Valentine
Workshop on AI in Earth Science, 2020 Conference on Neural Information Processing Systems, 2020.
Publication year: 2020

We present an active learning pipeline to identify hurricane impacts on coastal landscapes. Previously unlabeled post-storm images are used in a three component workflow — first an online interface is used to crowd-source labels for imagery; second, a convolutional neural network is trained using the labeled images; third, model predictions are displayed on an interactive map. Both the labeler and interactive map allow coastal scientists to provide additional labels that will be used to develop a large labeled dataset, a refined model, and improved hurricane impact assessments.

A Trustworthy Assurance-as-a-Service Architecture

Book ChapterWorkshop
Mahalingam Ramkumar and Somya D Mohanty
Frontiers in Artificial Intelligence and Applications, 831 - 840
Publication year: 2014

Increasing complexity and inter-dependency of information systems (IS), and the lack of transparency regarding system components and policies, have rendered traditional security mechanisms (applied at different OSI levels) inadequate to provide convincing confidentiality-integrity-availability (CIA) assurances regarding any IS. We present an architecture for a generic, trustworthy assurance-as-a-service IS, which can actively monitor the integrity of any IS, and provide convincing system-specific CIA assurances to users of the IS. More importantly no component of the monitored IS itself is trusted in order to provide assurances regarding the monitored IS.