All items about sentiment



I have a rather awkward subject to discuss. The last time I brought it up in mixed company, someone slapped me. But I’m going to do it anyway, because it’s worth discussing.

Natural language processing and semantic analysis allows us to extract sentiment from documents. Marketing organizations and community managers rely on tools from Scoutlabs, Radian6, and others that try to understand how online communities feel about their brands and products.

As we share more of our lives online, there’s more to analyze. Researchers from Northeastern University and Harvard University analyzed Twitter’s mood over the day. This kind of sentiment analysis can look at someone’s online messages and decide whether they’re angry or content, happy or sad. Given data over time, it can likely recognize patterns of mood, even cycles.

Such as those that occur every twenty-eight days.

(It’s at this point that my dinner companion launched a well-aimed palm at my somewhat scruffy chin.)
Read more »


Hashtag searches (note that this may be a technical constraint of how Twitter parses emoticon characters)
Hashtags are the standard way of adding meaning and context to online content, providing explicit context and making it easier for computers to understand what’s being said. And emoticons are a de facto standard for expressing sentiment that work across cultures and languages. Why haven’t we combined the two?

Read more »

Powered by WordPress, based on Mina theme.