Lucia Chen (Co-founder of Psyfy)

Lucia Chen

About Me

I am a researcher and founder developer dedicated to understanding human behavior using machine learning and natural language processing. As a former postdoctoral scholar at Stanford University, my research focuses on fostering fairness in algorithmic performances. I am placing this principle at the core of Psyfy, a mental wellness chatbot that I proudly founded. I'm also a machine learning consultant, always eager to bring my expertise to new projects.

Projects

Mental Wellness Chabot

Psyfy is a sophisticated mental wellness chatbot that utilizes a large language model. It offers users a comprehensive profile, summarizing their interactions with Psyfy and tracking their mood fluctuations, while also detecting elevated levels of anxiety or depression symptoms. Our utmost concerns are user safety and the trustworthiness and fairness of the product. With a research team composed of experts in the fields of clinical psychology, machine learning fairness, and large language model, Psyfy is also designed to be equitable for users from various cultural backgrounds .

Tech Advice for NGO

Every week, I allocate two hours to offering free tech advice to budding NGOs anywhere in the world that tackle issues like poverty, education, healthcare, environmental degradation, women's empowerment, and disaster relief. This includes guidance on web development and AI solutions to help them tackle their challenges more effectively. Feel free to email me at luciasalar AT gmail if you're associated with an NGO that could use my assistance.

Pandemic Anxiety

COVID-19 poses disproportionate mental health consequences to the public during different phases of the pandemic. We use a computational approach to capture the specific aspects that trigger the public's anxiety about the pandemic and investigate how these aspects change over time.

StreetLink App

We partnership with Homeless Link (HL), a UK-based charity, in developing a data-driven approach to better connect people sleeping rough on the streets with outreach service providers.

Opiod Overdose Risk Algorithms

Predictive models-driven Clinical Decision Support (CDS) systems hold immense potential in refining the precision and speed of clinical decision-making. However, without adequate validation, they could also misguide healthcare professionals and endanger patients. This risk is significantly high when these systems are utilized by healthcare practitioners prescribing and dispensing opioids, as inaccurate predictions can lead to direct patient harm.

June 21, 2021

How to retrieve all the tweets of a Twitter account?

The first version of Twitter API only allows us to retrieve 3200 tweets from each account. Now the good news is academic API2 allows us to retrieve all the tweets from an account.

July 8, 2020

Feed Forward Neural Network

I created a Youtube Channel to explain my work and some of the machine learning techniques I use. Here's the first episode of the deep learning series. Enjoy!

April 4, 2020

Social Media Data Tells Mental Status

Mood disorders such as anxiety and depression can be invisible but take a heavy toll on one's life daily. This article introduces several of my works that explain how we infer social media user's mental wellness from there social media posts

Sept 24, 2018

Machine Learning Career Path of Chinese Political Leaders

Recently JunYan Jiang published a Chinese Political Elite Database (CPED), which contains demographic and career information of Chinese political leaders from multiple levels. In this article, I tried to use maching learning to predict the promotion path of a Chinese political leader.

Oct 13, 2017

Workshop: Introduction to Qualitative Analysis II

I'm always interested in qualitative analysis out of curiosity, although I might or might not include it in my studies in the future. After years of doing quantitative studies, it might be a good idea to learn something from the other side. Therefore, I attended this workshop: 'Introduction to qualitative analysis' in the Welcome Trust Research Centre.

Aug 31, 2017

Book review: Gathering Data for Your PhD

This book gives a brief introduction the advantages and disadvantages of different data collection approaches. For instance, questionnaire, interview(structured, semi-structured, unstructured, mobile interview, face-to-fact), focus group, participant observation, etc.

Aug 30, 2017

Book Review: The Research Companion

Resources for better understanding how to plan your research