- Describe how NLP is at the inter-section of multiple disciplines.
- Operationally define of natural language processing.
- Discuss the purpose of NLP and what problems it tries to solve.
- List the five types of applications of NLP.
- Disucss how language is ambiguous and give an example.
- Utilize and implement NLP with Transformers.
- Discuss PyTorch.
- Briefly describe a tranformer.
- Discuss the results and interesting outcomes or surprises.
- Discuss the purpose and importance of text/data pre-processing and data cleaning.
- Describe tokenization.
- Discuss cleaning options and examples.
- Describe normalization with stemming and lemmatization and the purpose.
- Describe one-hot encoding and its purpose.
- Discuss NLTK.
- Implement tokenization, case normalization, and stop word removal.
- Implemeting data stemming.
- Implement data one-hot encoding.
- Implement the Word2Vec model demo.
- Describe word embeddings and the purpose.
- Discuss the Word2Vec model.
- Discuss how words can be described as vector mappings.
- Discuss a multidimensional embedding (embedder).
- Discuss the results and interesting outcomes or surprises.
- Describe the CBOW versus Skip-Gram architectures.
- Discuss the essence of word embedding.
- Implement your own word embedding training.
- Implement training word vectors on different data sets.
- Discuss the results and interesting outcomes or surprises.