Information and Emotion in Online Health-Related Discussions: Visualizing Connections and Causal Chains
|
If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL). |
Abstract
Regardless of one’s condition, coping with a health condition can be a difficult experience. Aside from the flood of emotions one might experience concerning their illness, patients are likely to have numerous questions – what their illness is, how to deal with it, and how to deal with others, and so on. In recent years, people have increasingly sought out online support groups for the answers to these questions.
Without question, affect is inextricably tied to the discussion content in online support groups. For example, anxiety is likely to be present in threads about biopsy results, anger might be present in threads that discuss the reaction of others to fibromyalgia, etc. Thus, a system that would support affect-based information retrieval could be helpful for users to identify content of interest.
However, health-related discussion forums today do not readily support affect-based search. Thus, an interactive interface was designed in order to facilitate a view of the interrelatedness of topic and emotion in health-related discussion forums. The current version uses a corpus of discussion content for three different conditions: breast cancer, type 1 diabetes, and fibromyalgia.
One of the views of the interface employs a timeline to facilitate a granular view of how multiple emotions and topics travel throughout one thread. Emotions are automatically identified using a lexicon generated based on Wordnet-Affect, and topics are automatically identified using Latent Dirichlet Allocation, using the MALLET toolkit. The interface also facilitates a view juxtaposing multiple threads that can be used to examine similarities and differences in emotional flows or topics.
This interface might be useful to many different types of users, including patients, forum administrators and researchers. Patients could use the interface to identify posts that resonate with their own thoughts and feelings, or provide alternative views to a topic that they are struggling with. The interface may also be a useful supplement for forum administrators to identify areas in the forum that with particular strong affective content and implement features to support this. As a research tool, the interface may be used to analyze the role of affect in online information exchange. For example, by juxtaposing multiple threads, one could investigate questions such as: "In relation to what topics do anger or fear emerge?" and "To what extent do certain emotions, e.g. anger or sadness, appear in the forums?" The presentation will provide examples of these possible uses, as well as discuss research that is currently in progress.
Without question, affect is inextricably tied to the discussion content in online support groups. For example, anxiety is likely to be present in threads about biopsy results, anger might be present in threads that discuss the reaction of others to fibromyalgia, etc. Thus, a system that would support affect-based information retrieval could be helpful for users to identify content of interest.
However, health-related discussion forums today do not readily support affect-based search. Thus, an interactive interface was designed in order to facilitate a view of the interrelatedness of topic and emotion in health-related discussion forums. The current version uses a corpus of discussion content for three different conditions: breast cancer, type 1 diabetes, and fibromyalgia.
One of the views of the interface employs a timeline to facilitate a granular view of how multiple emotions and topics travel throughout one thread. Emotions are automatically identified using a lexicon generated based on Wordnet-Affect, and topics are automatically identified using Latent Dirichlet Allocation, using the MALLET toolkit. The interface also facilitates a view juxtaposing multiple threads that can be used to examine similarities and differences in emotional flows or topics.
This interface might be useful to many different types of users, including patients, forum administrators and researchers. Patients could use the interface to identify posts that resonate with their own thoughts and feelings, or provide alternative views to a topic that they are struggling with. The interface may also be a useful supplement for forum administrators to identify areas in the forum that with particular strong affective content and implement features to support this. As a research tool, the interface may be used to analyze the role of affect in online information exchange. For example, by juxtaposing multiple threads, one could investigate questions such as: "In relation to what topics do anger or fear emerge?" and "To what extent do certain emotions, e.g. anger or sadness, appear in the forums?" The presentation will provide examples of these possible uses, as well as discuss research that is currently in progress.
Medicine 2.0® is happy to support and promote other conferences and workshops in this area. Contact us to produce, disseminate and promote your conference or workshop under this label and in this event series. In addition, we are always looking for hosts of future World Congresses. Medicine 2.0® is a registered trademark of JMIR Publications Inc., the leading academic ehealth publisher.

This work is licensed under a Creative Commons Attribution 3.0 License.