Classification from the imbalanced data is a common yet unresolved problem. Our scientists, Hubert Jegierski and dr. Stanisław Saganowski, introduce an “outside the box” solution to this problem. They suggest enriching the imbalanced data set with observations from other external data sets. They propose three enrichment scenarios: random, greedy, and supervised.
The ENGINE leader, Prof. Przemysław Kazienko gave an invited talk at Emotion AI Conference on “Emotion Recognition Using Wearables: Advances and Challenges” on May 6, 2020. Initially, the conference was scheduled for March 30, 2020 in New York, however, it had to be moved online due to pandemic crisis. The
The ENGINE group member, Radosław Michalski, altogether with Michał Weskida published an article on maximizing the spread of influence in social networks. In this work the authors perform a study on the usability of the evolutionary algorithm in this task. The results indicate that this approach is able to outperform
Researchers from Wroclaw University fo Science of Technology, Poland together with Poznań University of Technology, Poland just published a paper in Scientific Reports on a new generic and comprehensive method to generate complex networks (incl. social networks). The authors claim that networks are commonly created according to the priority attachment
3000 kilometers. 48 hours of intense coding. 20 hours of travel on the train. 4 hours of sleep. The 3 largest cities in Europe: London, Paris & Frankfurt. And all of this over 1 weekend. This is how the last edition of the unusual international “hackathon on the train” called
Next step in our research on Sequential Seeding This time we consider information spread which propagates with certain probability from nodes just activated to their not yet activated neighbors. Diffusion cascades can be triggered by activation of even a small set of nodes. Such activation is commonly performed in a single
Piotr Bródka, the ENGINE staff member, presented his research done within @Renoir Project at the @Sunbelt 2018 – the largest annual conference on social networks. The studies have been carried out in cooperation Jingwen Zhang and Cuihua (Cindy) Shen from University of California, Davis, and Jarosław Jankowski from West Pomeranian
Martin Víta from Masaryk University, Brno, Czech Republic visited gave a talk on “Siamese Convolutional Neural Networks with Static Attention for Recognizing Faceted Entailment” at the ENGINE Centre on May 14, 2018. Martin was invited by Maciej Piasecki, the leader of G4.19 NLP group. Recognizing textual entailment (RTE), i.e., a
The ENGINE Centre organized a datathon for their data science master students in Szklarska Poreba (Karkonosze mountains) on May 11-13, 2018. Almost 40 students analysed real datasets about smog measurements in Europe as well as about NSF and ERC scientific grants. Over 10 ENGINE staff members (including Halina Kwasnicka, Przemyslaw
The ENGINE Centre and the Rebel Campus organized a datathon for master students from Wroclaw University of Science and Technology, Wroclaw University of Economics and University of Wroclaw on April 21, 2018. Over 40 students in mixed groups analysed datasets provided by three local companies related to car fleet management,