arXiv subjects

Dhruv Dixit

Publications and source records attributed to Dhruv Dixit.

2 recordsLinked to original sources

arXiv2026

Mapping the Climate-Health Evidence Base (2007-2023): A Bibliometric, Statistical, and NLP Multi-Label Text Analysis of 22,695 Records

We analyzed a curated climate--health bibliographic corpus of 22,695 multi-labeled records from 2007--2023 to characterize growth, thematic concentration, and evolving methods. Annual publication counts rose sharply, with multiple change-points indicating phase-structured expansion; Negative Binomial models estimated roughly 10--11% year-over-year growth. Exposure--health co-occurrence departed strongly from independence, with canonical hazard--outcome dyads (e.g., extreme heat with heat-related impacts; floods/hurricanes with mental health) occurring far more often than expected even after accounting for marginal term popularity. A hierarchical logistic model for asthma-tagged records showed strong alignment with air-pollution-related exposures (including ozone and particulate matter) and relative under-representation of generic heat/temperature terms. Methodologically, modeling timescales shifted toward longer horizons over time, while at least one legacy method tag declined. Finally, we detected time- and geography-dependent annotation completeness, including decreased exposure-term coding in recent years, underscoring the need to model missingness when interpreting temporal trends.

cs.DL↗

arXiv2026

Detecting Speculative Language in Biomedical Texts using Recurrent Neural Tensor Networks

In this investigation, we delve into the automated detection of speculative language within biomedical articles by utilizing distributed sentence representations and advanced deep learning techniques. The implications of such identification extend to information retrieval, multi-document summarization, and the exploration of new knowledge. Our exploration encompasses two distinct approaches for acquiring distributed sentence representations: the Paragraph Vector model and the Recursive Neural Tensor Network. These methodologies are then rigorously compared against three foundational baseline algorithms: Support Vector Machines, Naive Bayes, and pattern matching. Our findings reveal that the Recursive Neural Tensor Network (RNTN) demonstrates a slight performance edge (F1 = 0.885) over the top-performing baseline, the linear bigram SVM (F1 = 0.881). Meanwhile, the Paragraph Vector model proves less effective (F1 = 0.368), even after extensive training using an expansive, unlabeled dataset. We engage in a comprehensive discourse on the factors influencing these performance disparities and provide insightful recommendations for future research directions.

cs.CL↗