Valerie Hyde, PhD

United States Contact Info
2K followers 500+ connections

Join to view profile

About

Specialties: Statistical Analysis, Data Mining, Predictive Modeling, Machine Learning…

Activity

Join now to see all activity

Experience & Education

  • Google

View Valerie’s full experience

See their title, tenure and more.

or

By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.

Licenses & Certifications

Volunteer Experience

  • Judge

    New York City Science and Engineering Fair

    - 8 years 1 month

    Education

Publications

  • A Family of Growth Models for Representing the Price Evolution in Online Auctions

    Wiley & Sons

    Other authors

  • Transformations for Semi-Continuous Data

    Computational Statistics and Data Analysis

    Other authors

  • Investigating Concurrency in Online Auctions Through Visualization

    The American Statistician

    Other authors

Projects

  • Predictive Model for Early Onset of Congestive Heart Failure

    -

    Using encounter, demographic and clinical data describing medical history for 300,000 patients, I and my team built several predictive models that successfully predicted which patients were at the greatest risk of developing CHF in the next year. To create these models, we used IBM SPSS Modeler. A key challenge was in identifying patients who had already developed CHF versus those who developed the condition in our analysis period. We used 7 different approaches to classify patients as to…

    Using encounter, demographic and clinical data describing medical history for 300,000 patients, I and my team built several predictive models that successfully predicted which patients were at the greatest risk of developing CHF in the next year. To create these models, we used IBM SPSS Modeler. A key challenge was in identifying patients who had already developed CHF versus those who developed the condition in our analysis period. We used 7 different approaches to classify patients as to their CHF status. One interesting result was the use of freeform text information to identify left ventricular ejection fraction as a criterion for determining CHF status. This textual data improved the classification significantly. CHF was identified in over 800 patients through ejection fraction measurement and no other method. Furthermore over 2000 patients were excluded using this data alone as they had already developed CHF during the look-back period. This was a captivating project for me and I am looking forward to my next project that begins January 2014 for a different IBM customer.

    Other creators

Languages

  • English

    Native or bilingual proficiency

  • Spanish

    Limited working proficiency

Recommendations received

4 people have recommended Valerie

Join now to view

More activity by Valerie

View Valerie’s full profile

  • See who you know in common
  • Get introduced
  • Contact Valerie directly
Join to view full profile

Other similar profiles

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses

See all courses