Deep Learning Internship – Transport Mode

  • Internship
  • Belgium - Antwerp


You are currently studying in a field that is related to computer science, and data science in particular, and are looking for a fast-paced environment where you can work with the latest machine learning technologies and data science tools. You will be part of our Data Science Team and will be working on developing a robust transport mode classifier using smartphone sensor data (GPS, accelerometer, gyroscope, etc). The classifier is intended to be a deep-learning model, potentially a hybrid of CNN and LSTM architectures. Besides the raw sensor data, relevant external or high-level information could be used additionally to distinguish between similar transport modes (‘bus’ and ‘car’ for example). Challenges include the noise in sensor data (bad GPS fixes, for example) as well as the closeness between the fine-grained transport modes.

Job description

Your main objectives will be the design and implementation of a deep-learning CNN model that classifies transport modes based on smartphone sensor (as well as external contextual) data.  On top of the CNN classifier, add an LSTM based architecture to enforce temporal consistency on the timeline of transport modes classified for a user. For example. if someone commutes to work by car, then she will more likely commute back by car as well. Sentiance transforms mobile sensor data into human-relatable intelligence, and hence this project sits at the heart of its technology.

Desired skills and expertise

  • You are a student (21 years or older) currently pursuing a degree in computer science or related field.
  • You have worked on and are passionate about deep learning.
  • You have basic knowledge of Python and Tensorflow.
  • You have a strong mathematical background and analytical mindset.
  • You have a keen interest in machine learning and data analysis.
  • You are fluent in English. Dutch is a plus.
  • You can work independently and take matters into your own hands.
  • The ability to quickly learn new technologies and successfully implement them is essential.


Experience with any of the following is considered a plus:

  • Advanced Python & Tensorflow knowledge and experience
    • Scikit-learn, Pandas, Numpy, Matplotlib
  • Experience with Spark or the Hadoop eco-system
  • Machine learning, data mining, data visualization

Status: Internship
Location: Antwerp, Belgium

Please send us your resume and a strong motivation with reference sentiance/2017/DLITM.