Senior Data Scientist
TomTom
Sep 2018- Present
I joined TomTom as Sr. data scientist at 2018 to develop ML powered features for mobile applications and improve TomTom services using ML. Here are example of projects I have been working on:
- Create AI-powered features using smartphone sensor data
- Design a pipeline to collect, analyse sensor data, and create a machine learning model to classify different events in signals.
- Tools: Several ML models including SVM, Neural Networks, TensorFlow (Lite)
- Designed the initial architecture of app analytic platform using UML to deliver insights on products and user behavior.
- Lead a Computer Vision project to detect failed rendered images in navigation device displays.
- Tools: CNN, Keras, Tensorflow
- Cluster analysis for User profiling
- In collaboration with UX research, I worked on clustering analysis on users by mining survey data and driving behaviour. Insights used for improving user personas.
- Point of interest analysis
- Conduct a clustering analysis on POI categories around users’ destination. The outcome of this projects is used to improve user profiles.
- Tools: K-mean, HDBSCAN, UMAP
Data Science Intern
March 2018 – August 2018
Worked on a customer project to improve forecast for a new line of production using time-series analysis and Machine Learning models. The solution I developed was two-folded: (1) create similarity index (clusters) for products, (2) predict the index of the new product, and then adjust the forecast of the first three months based on the predicted product index.
- Tools: Several Machine Learning models including XGBoost, Random forest, K-mean, Time series techniques including Dynamic Time Wrapping.
Zeinab Bakhtiari 