Alert button
Picture for Artyom Sosedka

Artyom Sosedka

Alert button

Sberbank, National University of Science and Technology MISIS

Logistics, Graphs, and Transformers: Towards improving Travel Time Estimation

Jul 12, 2022
Natalia Semenova, Vadim Porvatov, Vladislav Tishin, Artyom Sosedka, Vladislav Zamkovoy

Figure 1 for Logistics, Graphs, and Transformers: Towards improving Travel Time Estimation
Figure 2 for Logistics, Graphs, and Transformers: Towards improving Travel Time Estimation

The problem of travel time estimation is widely considered as the fundamental challenge of modern logistics. The complex nature of interconnections between spatial aspects of roads and temporal dynamics of ground transport still preserves an area to experiment with. However, the total volume of currently accumulated data encourages the construction of the learning models which have the perspective to significantly outperform earlier solutions. In order to address the problems of travel time estimation, we propose a new method based on transformer architecture - TransTTE.

* 4 pages, 1 figure, 1 table. Accepted at PKDD'22 demonstration track 
Viaarxiv icon

Citation network applications in a scientific co-authorship recommender system

Nov 22, 2021
Vladislav Tishin, Artyom Sosedka, Peter Ibragimov, Vadim Porvatov

Figure 1 for Citation network applications in a scientific co-authorship recommender system
Figure 2 for Citation network applications in a scientific co-authorship recommender system

The problem of co-authors selection in the area of scientific collaborations might be a daunting one. In this paper, we propose a new pipeline that effectively utilizes citation data in the link prediction task on the co-authorship network. In particular, we explore the capabilities of a recommender system based on data aggregation strategies on different graphs. Since graph neural networks proved their efficiency on a wide range of tasks related to recommendation systems, we leverage them as a relevant method for the forecasting of potential collaborations in the scientific community.

* 7 pages 
Viaarxiv icon