Natalie Parde

Get In Touch

Prefer using email? Say hi at



My primary research interests lie in several areas:

 

Below, I provide a brief overview of the ongoing projects in which I’m involved; for additional details, refer to the relevant papers listed or feel free to send me a message. My recent work has been conducted in UIC’s Natural Language Processing Laboratory, which I co-direct. On my publications page, you can also find links to papers about earlier lines of research that, while not ongoing, still represent substantial interests of mine.

 

 

Healthcare Applications

My work at the intersection of natural language processing and healthcare focuses primarily on aspects of cognitive and mental health, caregiver support, and online behavior. My current collaborators in this area include researchers from UIC, UI Health, and the University of California, San Diego.  
 

In my group’s work related to cognitive and mental health, we have developed approaches for spoken language detection of cognitive and mental health conditions. In the area of cognitive health, we’ve worked on Alzheimer’s disease detection and prediction of fine-grained cognitive health scores; more recently, we’ve also developed a dataset to facilitate pioneering studies of language patterns in preclinical Alzheimer’s disease. In the area of mental health, we’ve worked on recognizing language characteristics in individuals with bipolar disorder and schizophrenia when participating in a spoken dialogue social skills performance assessment. Portions of this work were funded by the National Institutes of Health. Some of our recent publications in this area include:  

 
 

In my group’s project collaborating with researchers in UIC’s CPERL group, we are investigating ways to support caregivers of children with pediatric rehabilitation needs through the development of smart and connected tools for equitable early intervention service design. This project is funded by the National Science Foundation and involves the development of novel techniques for dialogue systems and caregiver strategy classification. Some papers describing more about aspects of the project background include:  

 
 

In my group’s work pertaining to health-related online behavior we have examined medical self-disclosure and (more recently) empathy in online health forums. This has resulted in two innovative datasets: one pertaining to medical self-disclosure, and AcnEmpathize, a novel dataset of domain-specific health-related online empathy. More information about this work, including details for accessing these datasets, can be found here:  

 

 

Multimodality

My work in multimodal natural language processing has primarily centered on language and vision. My group investigates numerous topics falling under this umbrella, recently including visual storytelling and the detection of misogynistic memes. Papers addressing these topics can be found here:  

 

 

Creative Language

My work on creative language has primarily focused on figurative language. Recently, my group has examined metaphor, idiom, and hyperbole in empathetic online communication. We also conducted a focused study on euphemism. Some publications regarding this, along with my earlier work on metaphor novelty and sarcasm in more generalized domains, include the following:  

 

 

Other Ongoing Research

A complementary focus of my ongoing research has been on the reproducibility and robust evaluation of natural language processing experiments. My research group has conducted a number of high-level and probing studies on this topic, in an effort to promote higher quality and more open science. A selection of our publications regarding this topic include: