Hierarchical Modeling for User Personality Prediction: The Role of Message-Level Attention
- Veronica E. Lynn, Niranjan Balasubramanian, H. A. Schwartz
- Computer Science, PsychologyAnnual Meeting of the Association for…
- 1 July 2020
A novel model is presented that uses message-level attention to learn the relative weight of users’ social media posts for assessing their five factor personality traits and yields state-of-the-art accuracies for all five traits.
Human Centered NLP with User-Factor Adaptation
- Veronica E. Lynn, Youngseo Son, Vivek Kulkarni, Niranjan Balasubramanian, H. A. Schwartz
- Computer Science, PsychologyConference on Empirical Methods in Natural…
- 1 September 2017
A continuous adaptation technique is introduced, suited for real-valued user factors that are common in social science and bringing us closer to personalized NLP, adapting to each user uniquely.
CLPsych 2018 Shared Task: Predicting Current and Future Psychological Health from Childhood Essays
- Veronica E. Lynn, A. Goodman, Kate Niederhoffer, Kate Loveys, P. Resnik, H. A. Schwartz
- Psychology, MedicineCLPsych@NAACL-HTL
- 1 June 2018
This shared task represents one of the first attempts to evaluate the use of early language to predict future health and has the potential to support a wide variety of clinical health care tasks, from early assessment of lifetime risk for mental health problems, to optimal timing for targeted interventions aimed at both prevention and treatment.
Tweet Classification without the Tweet: An Empirical Examination of User versus Document Attributes
- Veronica E. Lynn, Salvatore Giorgi, Niranjan Balasubramanian, H. A. Schwartz
- Computer ScienceProceedings of the Third Workshop on Natural…
- 1 June 2019
The predictive power of user-level features alone versus document- level features for document-level tasks is investigated, showing the performance of strong document-only models can often be improved with user attributes, particularly benefiting tasks with stable “trait-like” outcomes.
Correcting Sociodemographic Selection Biases for Accurate Population Prediction from Social Media
- Salvatore Giorgi, Veronica E. Lynn, S. Matz, Lyle Ungar, H. A. Schwartz
- Computer ScienceArXiv
- 10 November 2019
Three methods to address predictive redistribution to account for shrinking, as well as adaptive binning and informed smoothing to handle sparse socio-demographic estimates are evaluated and it is shown each of these methods can significantly improve over the standard restratification approaches.
Correcting Sociodemographic Selection Biases for Population Prediction from Social Media
- Salvatore Giorgi, Veronica E. Lynn, H. A. Schwartz
- Environmental ScienceInternational Conference on Web and Social Media
- 10 November 2019
Three methods are developed and evaluated to address shrunken estimates (reduced variance of model predicted values) and sparse estimates of each population’s socio-demographic estimates: estimator redistribution to account for shrinking, adaptive binning and informed smoothing to handle sparse socio- Demographic estimates and combined approaches find substantial improvements over non-restratified models.
Residualized Factor Adaptation for Community Social Media Prediction Tasks
- Mohammadzaman Zamani, H. A. Schwartz, Veronica E. Lynn, Salvatore Giorgi, Niranjan Balasubramanian
- Computer ScienceConference on Empirical Methods in Natural…
- 28 August 2018
This paper presents residualized factor adaptation, a novel approach to community prediction tasks which both (a) effectively integrates community attributes, as well as (b) adapts linguistic features to community attributes (factors).
Semantic Role Labeling for Process Recognition Questions
- Samuel Louvan, Chetan Naik, Veronica E. Lynn, A. Arun, Niranjan Balasubramanian, Peter Clark
- Computer Science
Empirical evaluation shows that manually generated roles provide a 12% relative improvement in accuracy over a simpler bag-of-words representation, but automatic role identification is noisy and doesn’t provide gains even with distant supervision and domain adaptation modifications to account for the limited training data.
POE: A Pathology Extraction Tool for Finding Attribute-Value Pairs in Glioma Pathology Reports
- Veronica E. Lynn, Niranjan Balasubramanian, T. Kurç, J. Saltz, Rebecca S. Jacobson
- MedicineAmerican Medical Informatics Association Annual…
New integrative approaches that combine information from histologic, imaging and genomic features have the potential to advance methodologies for cancer classification and will likely necessitate development of new classification schemes.