Explorer takes your data and ‘explores’ it to analyze it, this process involves:

  • Collating Topics: Explorer finds common topics across the texts and ranks these topics according to their frequency.

  • Aggregating Sentiment: Explorer analyzes each text and each topic for sentiment expressed by the author (positive, negative and skeptical) and then sorts your texts according to each sentiment type.

  • Merging Synonyms: Explorer analyses how words are used in your data and groups these possible synonyms so that you can merge them to get better coverage. The reason for this is that people can express the same thing in many different ways and Explorer understands this out of the box.

  • Computing Related Topics: Explorer also finds related topics for each topic. Related topics are those topics that often occur together with other topics. By computing these Explorer can find more fine grained combination which serves to explain the contents of the data even better.

  • Finding Keywords: For each and every text Explorer determines which words are most significant for describing the text.

  • Establishing N-Grams: N-grams, or multi-word expressions, are sequences of words that often occur together and that represent special meaning in this form compared to when they are apart. Some examples are 'San Francisco' and 'academy awards'. Explorer knows n-grams out of the box; there is no need to manually enter them.

If you want to know more on how Explorer saves you time, read more at How does Explorer save me time?