Text mining for meaning

The more content we have to scan daily, the harder it becomes to sift out the most important things. Calais is a simple text mining tool that analyses an input text and tells us the topic it is about, the entities it contains (cities, companies, countries, currencies, organizations, people, political events, positions, religion), the events and facts. It shows us a combination of the most important tags, so we can determine if the text is of further interest to us before we read it. Entity keywords are underlined in specific colors, so we can more easily find the place at which they occur. This is a useful tool that can help us spend less time on less relevant texts or ones that would otherwise take too long to read.

Although it helps us see useful details, we'll still need to understand what they mean by seeking out the larger context, because we won't know in what sense the many tags are connected. This makes it very likely that we'll need to go through the text anyway, and maybe not only at the place where the keywords are. (Having the ore doesn't mean we have the metal.) Nevertheless, its a useful tool that can complement our search for meaning, not replace it. There are other text mining tools as well, but so far I haven't seen one with such a simple interface. Maybe you know of better ones that are equally easy to use.

Such initiatives are great because the speed with which we create content now has left behind our capability to deal with obsolete one and in general to understand things.

bit.ly/195goXd