07.11.2025 aktualisiert
100 % verfügbar
Simulation & Data Science Experte
Göttingen, Niedersachsen, Deutschland
Deutschland
M.Sc. Mathematics (Specialization in Data Science)Skills
3D ModellierungForschungIndesignKünstliche IntelligenzData AnalysisBig DataEntscheidungsbaum LernenDatenvisualisierungEntscheidungsanalyseForecastingSkalierbarkeitPythonPostgresqlMachine LearningMathematikMobilitätsmanagementMysqlNumerische AnalyseNumpyOnline-UmfragenStatistische AuswertungenSimulationenStatistikenWorkflowsAblaufplanungUmfragenData ScienceMulti-Agent SystemsDeep LearningPandasMatplotlibScikit-learn
English:
I am a freelance Data Scientist and Mathematician specializing in data-driven analysis, statistical modeling, and simulation.
What I offer:
– Data analysis, forecasting, and machine learning using Python (pandas, numpy, scikit-learn, matplotlib, seaborn)
– Statistical modeling and multivariate evaluation of complex datasets
– Simulation and optimization of systems using agent-based or data-driven approaches
– Analytical data visualization and reporting for research and technical applications
– Reproducible data pipelines and automated analytical workflows
– Support for scientific projects in design, analysis, and evaluation
I combine a strong mathematical background with hands-on expertise in Data Science, Machine Learning, Statistics, and Simulation, delivering transparent, reproducible, and scalable solutions.
_________________________________________________
Deutsch:
Ich bin freiberufliche Data Scientist und Mathematikerin mit Fokus auf datengetriebene Analysen, Simulationen und statistische Modellierung.
Leistungsangebot:
– Datenanalyse, Prognosemodelle und maschinelles Lernen mit Python (pandas, numpy, scikit-learn, matplotlib, seaborn)
– Statistische Modellierung und multivariate Auswertungen für komplexe Datensätze
– Simulation und Optimierung von Systemen mittels agentenbasierter oder datengesteuerter Ansätze
– Datenvisualisierung und Reporting für wissenschaftliche und technische Projekte
– Erstellung reproduzierbarer Analysepipelines und automatisierter Auswertungen
– Unterstützung bei der Konzeption, Planung und Evaluation von Forschungsprojekten
Ich kombiniere fundierte mathematische Kenntnisse mit praktischer Erfahrung in Data Science, Machine Learning, Statistik und Simulation, für präzise, nachvollziehbare und skalierbare Ergebnisse.
Sprachen
DeutschgutEnglischverhandlungssicher
Projekthistorie
- Designed KPI-based quality metrics for on-demand transport systems (comfort, reliability, digital maturity).
- Processed and analyzed operational trip data using Python (pandas, GeoPandas, seaborn).
- Built reproducible scripts for data cleaning, visualization, and model evaluation.
- Delivered interpretable results for mobility and public-transport applications.
- Generated synthetic populations from household travel surveys using Python (pandas, numpy).
- Applied Iterative Proportional Fitting (IPF) and Hot Deck Imputation to fill missing socio-demographic variables.
- Validated imputed data distributions with variance analysis and visualization (matplotlib, seaborn).
- Prepared MATSim-compatible input data for urban transport simulation.
Portfolio
Demand-Responsive Transport Systems
Developed a KPI-based framework for auditing service quality in on-demand transport systems.
Demand-Responsive Transport Systems
Analyzed trip logs using Python (pandas, GeoPandas) to derive comfort, reliability, and digital maturity indicators.
Implemented in Python (pandas, GeoPandas, seaborn) using real operational trip logs.
Accessibility Analysis and Coverage Radius
Visualized cumulative coverage of user trip origins relative to stop locations.
Calculated accessibility radius (rα) showing that 95% of trips were within 36 meters.
Demonstrated spatial efficiency and equity in DRT system accessibility.
Emission Reduction Benchmark (CO₂ per P_Km)
Compared DRT system emissions against benchmark vehicles.
Achieved 7.95% CO₂ reduction per passenger-kilometer and 72.6% efficiency per vehicle-kilometer.
Provided evidence for sustainable mobility optimization.
Population Synthesis, Imputation and Simulation
Generated 25 000+ synthetic individuals using Iterative Proportional Fitting (IPF) and Hot Deck Imputation.
Created complete socio-demographic profiles for agent-based simulations and validated behavioral distributions.
Validation of Data Distributions (MATSim Pipeline)
Python function developed for validating imputed socio-demographic data in population synthesis.
Compares original vs. imputed variable distributions using matplotlib and seaborn for visual and statistical evaluation.
Validation of Imputed vs. Original Distributions
Shaded areas represent ±1σ variance, indicating stability and overlap between the two datasets.
This analysis confirms that the imputation process preserves overall distributional characteristics and variability.