RDoC Task

Task 1:

Submitted scores are used to rank the articles from most to least relevant to their respective RDoC construct. Each column reports the Average Precision (AP) [1] for the respective RDoC construct. The last column reports the Mean Average Precision [1] for the RDoC constructs reported in the respective row.

Team	Acute Threat Fear	Arousal	Circadian Rhythms	Frustrative Nonreward	Loss	Potential Threat Anxiety	Sleep Wakefulness	Sustained Threat	Mean Average Precision
Team 10	0.89	0.93	1.00	0.69	0.74	0.87	1.00	0.64	0.85
Team 21	0.83	0.91	1.00	0.67	0.71	0.89	1.00	0.64	0.83
Team 22	0.67	0.84	1.00	0.61	0.61	0.81	0.98	0.41	0.74
Team 30	0.85	0.92	1.00	0.73	0.78	0.94	1.00	0.63	0.86

Team 10: Javad Rafiei Asl et. al., Team 21: Ramya Tekumalla et. al., Team 22: Daniel Laden et. al., Team 30: Yatin Chaudhary^, Pankaj Gupta^‡ and Hinrich Schütze^‡; * Siemens AG ‡ CIS LMU Munich.

[1] C. D. Manning, P. Raghavan, and H. Schütze, "Evaluation in information retrieval", in Introduction to Information Retrieval, Cambridge: Cambridge University Press, 2008, pp. 139-161.

Task 2:

Accuracy is the percentage of abstracts with correctly predicted context sentence. The last column reports the macro-averaged accuracy over all the RDoC constructs reported in the respective row.

Team	Acute Threat Fear	Arousal	Circadian Rhythms	Frustrative Nonreward	Loss	Potential Threat Anxiety	Sleep Wakefulness	Sustained Threat	Macro Average Accuracy
Team 30	0.74	0.73	0.47	0.37	0.74	0.59	0.60	0.42	0.58
Team 10	0.68	0.62	0.47	0.34	0.56	0.32	0.50	0.36	0.48
Team 22	0.47	0.42	0.60	0.29	0.62	0.38	0.57	0.47	0.48
Team 7	0.58	0.46	0.70	0.43	0.26	0.41	0.33	0.47	0.46
Team 21	0.37	0.12	0.10	0.11	0.26	0.15	0.17	0.14	0.18

Team 7: Fei Li et. al., Team 10: Javad Rafiei Asl et. al., Team 21: Ramya Tekumalla et. al., Team 22: Shyaman Jayasundara et. al., Team 30: Yatin Chaudhary^, Pankaj Gupta^‡ and Hinrich Schütze^‡; * Siemens AG ‡ CIS LMU Munich.