Parsing of noise values as entity

hi team,

I have trained a model on some (10-12 ) different entities with sufficient examples of these entities. but model is picking some noise values [values on which model is not being trained on] also as an entity.

for example -> ‘dell’ is an entity value of entity ‘brand’ , and i am getting ‘dell laptop’ also as brand value by entity extractor which is not needed.

@Tobias_Wochinger , @akelad

@shubham1140,

I think you just need to add more training data with more annotations.

For the example you provide, one idea that comes to mind is to do this:

  • Add a new entity, called ‘product’
  • Annotate additional training sentences, like: My dell laptop is great.