Selpi Selpi

Researcher at Mechanics and Maritime Sciences, Division of Vehicle Safety

+46317721369 http://orcid.org/0000-0003-2800-4479

Selpi is employed at the Division of Vehicle Safety at the Department of Mechanics and Maritime Sciences. She is also an affiliated teacher at the Division of Data Science and AI at the Department of Computer Science and Engineering.

Beside having experience in conducting research, leading research projects, supervising students, and teaching, she also has a long experience as a Director of Doctoral Studies at Chalmers.

Outside academic work, Selpi has several years of experience in software and IT industry.

She has a PhD degree in Computing. Her PhD thesis is on application of a machine learning technique, called Inductive Logic Programming, for a bioinformatics problem related to gene regulation.

Her current interests include applications of machine learning and data mining for transport-related domain (e.g., und erstanding driving styles/driver behaviour from naturalistic driving data, travel time and traffic volume predictions, text-mining for text data in transport) and studying the impact of mixed traffic (with different automation levels) on both traffic safety and traffic efficiency.
Recent Master's thesis projects that I supervised:
  • Resource optimal Neural Networks for safety critical real time systems
  • Detecting and predicting automation expectation mismatch
  • Safety evaluation of heterogeneous traffic: using modified models in SUMO
  • Using pre-trained language models for extractive text summarisation of academic papers
  • AI-based road friction estimation using road weather information
  • Predicting pedestrian movement for street segments in urban environments
  • Exploring semi-supervised learning for deciding the order of vehicles in rear-end crashes
  • Lane-level map matching using hidden Markov models
  • Influence area of speed cameras
  • Highway tollgates traffic prediction using a stacked autoencoder neural network
  • Highway tollgates travel time & volume predictions using Support Vector Regression with scaling methods
  • Topic modelling and clustering for analysis of road traffic accidents
  • Automating text categorization with Machine Learning: Error responsibility routing in a multi-layer hierarchy

Page manager Published: Mon 14 Dec 2020.