Semantic Tagging

This app attempts to assigns a best-guess tag to the meaning of a text. The single best tag is given for each paragraph, and a few tags are given for the whole text.

Note that the output per paragraph could be fed into SemSim for further analysis, e.g., using a valence filter.

Similarity scores are dot products.

Input

Text to analyze:

Tags (comma-separated list; if left empty, a default set of emotion-tags will be used):

Submit

Example

An example to check whether a few single-line paragraphs return the expected tags:

Text to analyze:
I am angry.
I own a fridge.
I am happy.
We had a party for my birthday.
I eat a lot of fish.
I went to the funeral.

Tags: Can be kept empty to use the emotional defaults, or set to: celebration, angry, happy, fish, grieving, fridge

Using the specified tags, you should see the correct tags being applied per paragraph, e.g., "grieving" should tag the sentence with "funeral" in it.