An interactive framework to visualize and analyze your AutoML process in real-time.
Official Code for "Non-Probability Sampling Network for Stochastic Human Trajectory Prediction (CVPR 2022)"
Faster, better, smarter ecological niche modeling and species distribution modeling
Portfolio-grade audit of a student mental health & academic pressure survey. Measures coverage and sample imbalance, runs validity checks, highlights measurement and selection bias risks, and converts messy open-text “stress causes” into a transparent taxonomy. Ships a Markdown report, figures, and a Streamlit dashboard.
Longform data analysis article arguing every “dataset” is actually three: Observed (captured rows), Missing (what should exist but doesn’t), and Excluded (what filters/joins/dropna removed). Includes dataset accounting, join-loss and missingness audits, segmentation checks, and practical templates to prevent biased KPIs and wrong conclusions.
R code used for the analyses of the paper: Spatial conservation prioritisation in data-poor countries: a quantitative sensitivity analysis using different taxa
Sensitivity analysis suite for phenogenetic study designs and estimation under assortment and selection
Pipelines to evaluate Breast Cancer Purity Score and to correct sampling bias
Efficient Multistream Classification using Direct DensIty Ratio Estimation
This project presents an exploratory and statistical analysis of amphibian occurrence records in Brazil using data from the Global Biodiversity Information Facility (GBIF).
A global, taxon-stratified, high-resolution sampling-effort dataset from GBIF for bias-aware ecological modelling — reproducible R-based workflow for generating global sampling-effort rasters from GBIF occurrence records across spatial, temporal, and taxonomic scales. For more details, see El-Gabbas (2026), Diversity and distributions (accepted).
Market selection on the 2017 freeCodeCamp survey. Bootstrapping shows the conventional answer is noise: India vs Canada is a coin flip (P=0.514), three different second-place countries appear depending on the outlier cutoff, and the median monthly spend is $0 in three of four markets. Dataquest guided project.
🚀📐 Representación gráfica de distribución de muestreos aleatorios.
Portfolio artifact 1 — applied statistical analysis of public UN SDG data for East Africa; Python.
Analysis of the accuracy of U.S. presidential election polls (2008–2024) based on prediction error data. Compares ideal vs. practical polling methodology, presents three contrasting visualizations of the same dataset to illustrate framing effects, and identifies selection/measurement biases with recommendations to improve accuracy.
Historical sampling bias, observation processes, and their implications for contemporary biodiversity modelling.
How much of a reported advance in first flowering dates comes from observer effort alone, and what to use instead: mean dates, model-based low quantiles, or first dates rarefied to a common sample size.
Simulate the impact of sampling bias in the derivation of effect sizes
Overcoming Selection Bias in Statistical Studies With Amortized Bayesian Inference
📊 Evaluate survey quality and bias through coverage, representativeness, and measurement risk audits for reliable insights and data validity checks.
To associate your repository with the sampling-bias topic, visit your repo's landing page and select "manage topics."