Fire Alarms Systems

Making sense of Twitter noise is going to be easy

For its part, Twitter is a corporate image. Very few are aware of this fact

When a shock occurs, it is too late: a company suddenly sees a facet of your business - usually a looming crisis and developing countries - broke into Twitter. As always when it comes to a society, there is money to be made by converting the problem into an opportunity: the social network intelligence is destined to become a large company

In theory, when it comes to assessing the presence of the media of a brand, Facebook is the place to go. But as brands go to the dominant network, social noise and the signal becomes overwhelming - what people are saying about the brand - it is difficult to extract

In comparison, Twitter more quickly reflects the mood of the users of a product or service.

All

in the field of marketing / communication becomes more and more anxious to know what Twitter is said about a product defect, the perception of a strike or an environmental crisis. Twitter is the echo chamber, the pulse of public opinion. Therefore, it takes enormous courage.

Datamining

Twitter is not trivial

comparison, immersed in libraries and newspaper archives, blog is easy. The sentences are (usually) well-built, the names are written in full slang, jargon, and a new kind are relatively rare. However, on Twitter, the 140 character limit requires a lot of creativity.

Twitter slang is constantly evolving, with new names and characterizations shines all the time, which excludes simple text analysis. 250 tweets a day is a moving target. Reliable quantitative analysis of the current mood is a great challenge.

Companies like DataSift (released last month) exploit Twitter flexible based on metadata from more than 40 included in a message. Because, in case you do not know, a Tweet like this innocent aa ...

... is a rich treasure of data. A year ago, Raffi Krikorian, a developer platform team Twitter API (thanks to this story on ReadWriteWeb seen) reveals what is behind the 140 characters. The image below ...

... is a dismantling of a large (in this case, the blog Krikorian) which shows the depth of the metadata associated with a tweet. Each comes with information such as biography of the author, the level of commitment, popularity, how often, location (which can be very precise in the case of an access point georeferenced), etc. WiredUK In this interview, founder Nick Halstead DataSift cites the example of Starbucks Twitter:

saved literally everywhere in recent months about people checking Starbucks. Needless to say they are in a Starbucks, which can not be in a place that is Starbucks, which can be people on Twitter that allows you to save your position is. Therefore, I can say that the average age of people entering the Starbucks in the UK. Businesses can come and say. "I'm a retailer, if I provide the geographical data of where all my stores, tell me what people say when they are near him, or" Some traders are not many check-ins , but overall, more than a month is very rare that you can not get a good sample.

Well, think about it the next time a tweet from Starbucks.

, a startup based in Paris, with a foothold in California, plans to focus different.

instead of trying

guess
the feeling of a multitude of Twitter, it will create a network of connections between people, the terms and concepts . In other words, it creates a "structured serendipity" in which the user, of course, to expand the scope of a search form beyond the original query. With a web application called voluntary Mesagraph is scheduled for a private beta this week and the public in January.

Find best price for : --Wikipedia----Mesagraph----Starbucks----tweet----DataSift----Facebook----Twitter--