Research: Add new features easily

[oliviadargel]

User story

1. As a NLP component developer
2. I want to have a solution to add new features dynamically
3. So that I can add new features efficiently

Acceptance criteria

• document solution and communicate it to the rest of the NLP team

Definition of done

• Code was reviewed by another person
• GitHub CI runs successfully
• Feature is merged into main branch

[NebiMucaj]

sample query: Suche mir eine Route von Berlin nach Paris mit charging stations

_queryobject: { ... charging_stations:true ... }

1. Option:

   • extend our existing model by new entities
     Steps:
     1. extend the training data by adding new entities in chatette file
     2. retrain the model
     3. add post-process function

2. Option:

   • implement fuzzy matching with the levenshtein distance Steps:
     1. detect if a token in the user query matches the target feature(e.g, charging station) via the levenshtein distance
     2. check if the detected feature is negated (e.g., without charging station)

--> We discussed in our team meeting that option 2 is the best solution that enable us to add more features in an efficient way.