E
Eemil Lagerspetz
Researcher at University of Helsinki
Publications - 71
Citations - 1903
Eemil Lagerspetz is an academic researcher from University of Helsinki. The author has contributed to research in topics: Mobile computing & Mobile device. The author has an hindex of 18, co-authored 69 publications receiving 1468 citations. Previous affiliations of Eemil Lagerspetz include Helsinki Institute for Information Technology.
Papers
More filters
Journal ArticleDOI
Predicting Depression From Smartphone Behavioral Markers Using Machine Learning Methods, Hyperparameter Optimization, and Feature Importance Analysis: Exploratory Study
Kennedy Opoku Asare,Yannik Terhorst,Julio Vega,Ella Peltonen,Eemil Lagerspetz,Denzil Ferreira +5 more
TL;DR: In this paper, the authors explored the relationship between the behavioral features and depression using correlation and bivariate linear mixed models (LMMs) and leveraged 5 supervised machine learning (ML) algorithms with hyperparameter optimization, nested cross-validation, and imbalanced data handling to predict depression.
Proceedings ArticleDOI
Product retrieval for grocery stores
TL;DR: A grocery retrieval system that maps shopping lists written in natural language into actual products in a grocery store using nine months of shopping basket data from a large Finnish supermarket is introduced.
Posted Content
Differentially Private Bayesian Learning on Distributed Data
TL;DR: In this article, the authors consider DP Bayesian learning in a distributed setting, where each party only holds a single sample or a few samples of the data and propose a learning strategy based on a secure multi-party sum function for aggregating summaries from data holders and the Gaussian mechanism for DP.
Proceedings Article
Differentially private Bayesian learning on distributed data
TL;DR: This work builds on an asymptotically optimal and practically efficient DP Bayesian inference with rapidly diminishing extra cost and proposes a learning strategy based on a secure multi-party sum function for aggregating summaries from data holders and the Gaussian mechanism for DP.
Proceedings ArticleDOI
MegaSense: Feasibility of Low-Cost Sensors for Pollution Hot-spot Detection
Eemil Lagerspetz,Sasu Tarkoma,Tareq Hussein,Naser Hossein Motlagh,Martha A. Zaidan,Pak Lun Fung,Julien Mineraud,Samu Varjonen,Matti Siekkinen,Petteri Nurmi,Yutaka Matsumi +10 more
TL;DR: A 44-day measurement campaign is conducted to assess performance of low-cost air quality monitors under different environmental conditions and shows that the accuracy is sufficient for applications relying on variations in air quality index values, such as hot spot detection.