I particularly like the logic that the responsibility for creating engagement should be shared – it’s as much for the commentator to make themselves heard, as for the subject of comment to listen.

I’ve often felt uncomfortable that – for all of the shift in language in SMM over time from “monitoring” to “listening” – “engagement” is framed as something that a company does to its (passive) customers: a sham substitute for real dialogue. The idea of moving the concept as well as the tools forward really appeals.

Thanks for a great analysis in your recent posts of the current state of affairs in sentiment analysis. One of the most balanced and constructive critiques imho.

Widespread, accurate and cheap sentiment analysis would be a wonderful thing, particularly if it was so cheap as to be free or near-free to consumers too. Anything that brings organisations and their stakeholders closer together, openly and transparently, can only do good work.

But that future is a long, long way off, and it would be great to see if we could derive some of the value earlier than that, to everyone’s advantage.

It’s for this reason we’ve started looking at something I’ve called The Ontology for Feelings About Things. This Semantic Web approach aims to help social Web participants meet the analytics service half-way, or perhaps only quarter-way, but sufficiently so that critical momentum can be established behind widespread, accurate and cheap automated sentiment analysis.

Just as social Web participants understand the value of metadata today, whether they refer to it as metadata or not, such as tags and location, The Ontology for Feelings About Things will manifest in software that permits the social Web participant to quickly and painlessly mark-up the sentiment of his contribution. (Although in some instances, the metadata will be the data… it will suffice of itself to communicate the individual’s feelings if they don’t feel like writing a lengthy post for example.) They have incentive to do this because their contributions will be seen to resonate more loudly than equivalent contributions that do not have this metadata.

http://influencescorecard.wikispaces.com/The+ontology+for+feelings+about+things

I think there is a role for automated sentiment within social media monitoring given the amount of data/conversations that need to be tracked and reviewed. It’s important to remember that it’s still early days for sentiment technology so its reliability and accurateness will improve over time.

At the same time, social media monitoring services should still offer the ability for human editing. Within Sysomos’ MAP service, for example, users can manually adjust sentiment settings.

Mark

(I’m the director of communications with Sysomos.com)

Actually, my point in using the mood ring metaphor is to say that sentiment is more interested as a filter than as a standalone metric. Watching the sentiment line go up and down is like watching a mood ring change color–interesting, but lacking in depth.

http://net-savvy.com/executive/measurement/sentiment-analysis-is-not-a-mood-ring.html

I mostly wanted to address what you say in #1 re: vendors not disclosing if their analysis is at a topic or article level. I can’t speak for other vendors, but at Biz360, we do both, and our accuracy is slightly higher at a topic (aka entity) level analysis. I also think that if you are a brand, topic-level analysis is a lot more useful, because there are more articles that deal with several brands at the same time. Our topic level accuracy is 65%, while 62% accuracy at article level (we have 4 sentiments BTW vs. some other vendors’ 3). We trained our data via Mechanical Turk until incremental increase in accuracy was <1%. Users have a chance to override sentiment, as well as toss an article or a source altogether from your results.

Automated sentiment is useful when you have 50,000 articles – you will never be able to read them all! If you want sentiment of an individual article – well, you should probably read it

Cheers!

@themaria @biz360

It’s an interesting metaphor (although I had to resort to Wikipedia to find out what a mood ring is ).

I frame the issue slightly differently. Human sentiment analysis is a very established part of traditional media content analysis (around since the 70s / 80s). This is what many customers – particularly the ones with a PR / comms background – think they are getting: an automated; cheap; effective substitute for expensive human content analysis.

But as you guys know, it’s not the same thing at all – automated sentiment analysis was developed to do something different, follows a different logic path and produces very different results.

Best,
Michelle @Synthesio

This post was mentioned on Twitter by christhomasuk: Just posted: Why automated sentiment analysis shouldn’t feature in social media monitoring tools (http://bit.ly/aNYDdv)…