Open positions
Open research positions in SNAP group are available at undergraduate, graduate and postdoctoral levels.

Dynamic Face-to-Face Interaction Networks

Dataset information

The dynamic face-to-face interaction networks represent the interactions that happen during discussions between a group of participants playing the Resistance game. This dataset contains networks extracted from 62 games. Each game is played by 5-8 participants and lasts between 45--60 minutes. We extract dynamically evolving networks from the free-form discussions using our algorithm, ICAF.

The networks are weighted, directed and temporal. Each node represents a participant. At each 1/3 second, a directed edge from node u to v is weighted by the probability of participant u looking at participant v or the laptop. Additionally, we also provide a binary version where an edge from u to v indicates participant u looks at participant v (or the laptop).

Project website with demo: Please refer to the project website for details about the network extraction algorithm and to visualize the networks.


Dataset statistics
Number of networks 62
Number of nodes 451
Number of edges 3,126,993
Average number of edges per network 50,435
Total temporal length 142,005 seconds
Average temporal length per network 2,290 seconds

Source (citation)

The following BibTeX citation can be used:
@inproceedings{bai2019predicting,
  title={Predicting the Visual Focus of Attention in Multi-Person Discussion Videos},
  author={Bai, Chongyang and Kumar, Srijan and Leskovec, Jure and Metzger, Miriam and Nunamaker, Jay and Subrahmanian, VS},
  booktitle={IJCAI 2019},
  year={2019},
  organization={International Joint Conferences on Artificial Intelligence}
  }

Files (Download)

File Description
network_list.csv List of network ID and number of participants
network/network[ID].csv Dynamic face-to-face interaction networks. One file per network.
network/network[ID]_weighted.csv Weighted version of dynamic face-to-face interaction networks. One file per network.
network_loader.py Code to easily load the data.

Data format

network_list.csv: Each network ID represents one discussion.
NETWORK,NUMBER_OF_PARTICIPANTS
0,7
1,8

where


For files network/network[ID].csv: One file per game. Each file has the time series of the who-looks-at-whom networks. Each row is a binary adjacency matrix. For example, in network0.csv:
TIME,P1_TO_LAPTOP,P1_TO_P1,P1_TO_P2,P1_TO_P3,P1_TO_P4,P1_TO_P5,P1_TO_P6,P1_TO_P7,P2_TO_LAPTOP,P2_TO_P1,P2_TO_P2,P2_TO_P3,P2_TO_P4,P2_TO_P5,P2_TO_P6,P2_TO_P7,P3_TO_LAPTOP,P3_TO_P1,P3_TO_P2,P3_TO_P3,P3_TO_P4,P3_TO_P5,P3_TO_P6,P3_TO_P7,P4_TO_LAPTOP,P4_TO_P1,P4_TO_P2,P4_TO_P3,P4_TO_P4,P4_TO_P5,P4_TO_P6,P4_TO_P7,P5_TO_LAPTOP,P5_TO_P1,P5_TO_P2,P5_TO_P3,P5_TO_P4,P5_TO_P5,P5_TO_P6,P5_TO_P7,P6_TO_LAPTOP,P6_TO_P1,P6_TO_P2,P6_TO_P3,P6_TO_P4,P6_TO_P5,P6_TO_P6,P6_TO_P7,P7_TO_LAPTOP,P7_TO_P1,P7_TO_P2,P7_TO_P3,P7_TO_P4,P7_TO_P5,P7_TO_P6,P7_TO_P7
0.0,0,0,0,0,1,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,1,0,0,0,0,0,0,1,0,0,0,0,1,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0

where, in each row:


For files network/network[ID]_weighted.csv: One file per game. This is a weighted version of the binary adjacency matrices. Each file has the time series of the weighted who-looks-at-whom networks. Each row is a weighted adjacency matrix. For example, in network0_weighted.csv:
TIME,P1_TO_LAPTOP,P1_TO_P1,P1_TO_P2,P1_TO_P3,P1_TO_P4,P1_TO_P5,P1_TO_P6,P1_TO_P7,P2_TO_LAPTOP,P2_TO_P1,P2_TO_P2,P2_TO_P3,P2_TO_P4,P2_TO_P5,P2_TO_P6,P2_TO_P7,P3_TO_LAPTOP,P3_TO_P1,P3_TO_P2,P3_TO_P3,P3_TO_P4,P3_TO_P5,P3_TO_P6,P3_TO_P7,P4_TO_LAPTOP,P4_TO_P1,P4_TO_P2,P4_TO_P3,P4_TO_P4,P4_TO_P5,P4_TO_P6,P4_TO_P7,P5_TO_LAPTOP,P5_TO_P1,P5_TO_P2,P5_TO_P3,P5_TO_P4,P5_TO_P5,P5_TO_P6,P5_TO_P7,P6_TO_LAPTOP,P6_TO_P1,P6_TO_P2,P6_TO_P3,P6_TO_P4,P6_TO_P5,P6_TO_P6,P6_TO_P7,P7_TO_LAPTOP,P7_TO_P1,P7_TO_P2,P7_TO_P3,P7_TO_P4,P7_TO_P5,P7_TO_P6,P7_TO_P7
0.0,0.016,0.0,0.0,0.0,0.572,0.0,0.0,0.534,0.0,0.046,0.083,0.0,0.041,0.029,0.067,0.077,0.023,0.0,0.127,0.011,0.025,0.034,0.503,0.079,0.03,0.041,0.0,0.039,0.041,0.037,0.533,0.1,0.049,0.019,0.063,0.0,0.073,0.377,0.12,0.043,0.034,0.04,0.357,0.02,0.0,0.05,0.256,0.15,0.111,0.131,0.128,0.044,0.115,0.0,0.12,0.094,0.071,0.136,0.172,0.053,0.165,0.109,0.0

where in each row: