Figure 1. Examples of e-cigarette discussions in social media From: Mining e-cigarette adverse events in social media using Bi-LSTM recurrent neural network with word embedding representation J Am Med Inform Assoc. Published online May 13, 2017. doi:10.1093/jamia/ocx045 J Am Med Inform Assoc | © The Author 2017. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com
Figure 3. The Bi-LSTM RNN architecture From: Mining e-cigarette adverse events in social media using Bi-LSTM recurrent neural network with word embedding representation J Am Med Inform Assoc. Published online May 13, 2017. doi:10.1093/jamia/ocx045 J Am Med Inform Assoc | © The Author 2017. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com
Figure 2. Annotation example From: Mining e-cigarette adverse events in social media using Bi-LSTM recurrent neural network with word embedding representation J Am Med Inform Assoc. Published online May 13, 2017. doi:10.1093/jamia/ocx045 J Am Med Inform Assoc | © The Author 2017. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com
Figure 4. Word embedding visualization for e-cigarette related entities From: Mining e-cigarette adverse events in social media using Bi-LSTM recurrent neural network with word embedding representation J Am Med Inform Assoc. Published online May 13, 2017. doi:10.1093/jamia/ocx045 J Am Med Inform Assoc | © The Author 2017. Published by Oxford University Press on behalf of the American Medical Informatics Association. All rights reserved. For Permissions, please email: journals.permissions@oup.com