Skip to content

Latest commit

 

History

History
43 lines (32 loc) · 3.12 KB

README.md

File metadata and controls

43 lines (32 loc) · 3.12 KB
Raw

interview-code-examples

Notebooks with example code explained for interviews.

I have often found when I am doing a tech interview as a canidate the following problems:

  • The interviewer asks very specific questions about the tasks I did for companies I worked for. These questions can often times be too specific and answering their questions could breach an NDA.
  • I honestly cannot remember exactly how I did something AI related from several years ago in great detail

And as someone who has also been a hiring manager, I am tasked to interview canidates, including tech interviews.

I thought it might be great to come up with some small challenges we could go through to aid canidates and interviewers alike.

NLU Intent classifiers

Strangely I couldn't find an example of building a basic intent matcher used in an NLU engine for voice assistants, where utterances (ie user commands/questions) are classified by their intent category. So I decided to make a simple one for everyone out there. NLU_Intent_matching.ipynb explores the idea of intent matching on utterances and entity extraction. Using the NLU Evaluation Dataset as a bases of data, both a word embedding approach (using Word2Vec) and a TFIDF approach are explored using the following algorithms to train intent classifiers:

More may be added in the future.

Reasons these classifiers were choosen

  • Using algorithms that are easy to find in other langauges (not just Python)
  • Computationally inexpensive (they could run in real time on a phone, for example)
  • Not just inference could be run on low-power device, training could also be performed (TinyML), therefore a user could always add/edit the data to create models that work better for them

Good questions for NLU intent classifiers

  • Why does Word2Vec perform much worse than TFIDF?
  • Is there any other word embedding based approach that might work better? (how costly is it to run inference on?)
  • Would TFIDF out-perform simple bag of words?
  • Is f1 a good measure of performance?
  • How would one fine tune the models and what kind of expectations are there for performance gains?
  • How could adding in a entity tagger boost intent classification?
  • Are there any other performance hacks?

Entity extraction (Named Entity Recognition)

This notebook uses conditional random fields (CRFs) for entity extraction.

And there is more to come, whenever I find time. ;)