Quality and Relevancy Matter in Health

Unlike tagging a photo of a castle with the word “castle,” people want reliable and factual information about health concerns, diseases and mental health issues. They want relevant results, but they also wants results that haven’t been tainted by others with a hidden agenda. Health (medical, and mental health) concerns are more complex than, “X causes Y and Z is the usual treatment for it.” Anybody who has worked in this field for more than a few years understands that even as we learn more and more everyday from reams of research, the relationsihps and knowledge becomes more and more complex in medicine and mental health care. So concepts such as relevancy, consistency, trust and reliability are key components in this area of the online world.

One way that Web 2.0 services try to address issues of bias and usefulness of a registered user’s tags or votes is by employing a rating system of registered members. Unfortunately, most such rating systems are fairly simple things which will be fairly unreliable if the community is still relatively small. Even as the community grows, ratings can be connected more to seniority and frequency of behaviors within the community than as an actual indication of that person’s quality of tagging or responses. Often, there’s little differentiation in these rating systems between 10 bad people rating 1 bad person especially good and 10 good people rating 1 bad person especially bad. The former should carry little weight, while the latter should carry extraordinary weight. But few user social networking systems understand this kind of value relationship.

Examining Results from Flickr and Del.icio.us

I was curious as to how two of the more popular Web 2.0 applications, Flickr and Del.icio.us, answer people’s health or mental health questions. Flickr is, of course, a photo sharing application, so its relevancy to health concerns is going to be limited. But by using it as an example, we can learn about the strengths and weaknesses of a system like Flickr, since it epitomizes the very definition of Web 2.0 services.

So let’s see how these two popular, Web 2.0 applications stand a random test of relevancy and accuracy. Let’s start first with Flickr, a service I use regularly. Flickr is interesting for more than one reason, because it doesn’t even pretend to make an attempt at the idea of “relevancy,” since every keyword or tag is user-created (using judgment that spans the quality spectrum). Instead, it allows you to sort through photos by “most recent” or “most interesting.” Most interesting is Flickr’s way of getting at a concept similar to relevancy:

There are lots of things that make a photo ‘interesting’ (or not) in the Flickr. Where the clickthroughs are coming from; who comments on it and when; who marks it as a favorite; its tags and many more things which are constantly changing. Interestingness changes over time, as more and more fantastic photos and stories are added to Flickr.

So let’s see how Flickr stands up…