Statistically Solving Sneezes and Sniffles Step by Step

on YouTube ↗39m

" summary

Project records a year of one-person, timestamped symptom logs with GPS and backs them up to Segment IO, plus external London Air, weather, MyFitnessPal and Oyster-scraped commute data. An open-source iOS app and data prep tools get built. Models (logistic regression, random forests, SVMs, recursive feature elimination; scikit-learn, statsmodels) train on ~330 rows and 84 engineered features and expose overfitting. Key findings: loratadine shows ~2h onset and ~9h effective window and low humidity correlates with more sneezes; pollution shows no clear signal yet.

AI-written summary · about

" chapters (7)

1. 0:00Introduction
2. 2:53Sneezing patterns
3. 6:45iOS app for data
4. 17:15Data relationships
5. 22:19Predictive models
6. 29:35Next steps
7. 31:19Q&A

" related talks

• 27mHARSHA URLAM & SWETA SINGH at Cypher 2017Harsha Urlam, Sweta Singh · Cypher 2017
• 06mPredicting Subjective Measures of Social Anxiety from Sparsely Collected Mobile Sensor DataBethany A. Teachman, Laura E. Barnes, Haroon Rashid, Sanjana Mendu, Katharine E Daniel, Miranda L. Beltzer, Mehdi Boukhechba · UbiComp / ISWC 2020
• 26mMachine Learning for Personalised HealthDanielle Belgrave · Strange Loop 2018
• 01h 00mTake Control of the Data of YouNigel Parker · NDC Oslo 2017
• 01h 04mTake Control of the Data of YouNigel Parker · NDC London 2018
• 22mCOVID-19: Data Science & ExpertiseRajiv Shah · GOTO Chicago 2020
• 44mQuantified Self: Analyzing the Big Data of our Daily LifeAndreas Schreiber · PyData Berlin 2014
• 42mMachine learning with ventilator dataIan Ozsvald, Giles Weaver, Gusztav Belteki · PyData London 2017