Some Terminologies:

• Corpus → All the data
• Documents → One row (sentence)
• Vocabulary → All unique words in data
• Word

NLP Task:

1. Dataset
2. Text Preprocessing
   1. Tokenization, Lowering the words
   2. Stemming, Lemmatization, Stop Words
3. Words to Vectors (BOW | TFIDF | Word2Vec)

One Hot Encoding:

• Advantage
  • Simple to implement
  • Intuitive
• Disadvantage
  • Create a sparse matrix (All values is zero except one value is one ) → يخلي التدريب بطيء كبيره RAM ويحتاج .
  • OOV → Out Of Vocabulary → خلي الموديل مش قادر يتعامل مع بيانات جديدة بشكل صحيح يعنى لو شاف كلمه جديده متدربش عليها بيعملها كلها اصفار
  • Not fixed sentences (size)
  • Semantic meaning between words is not captured

BOW (Bag Of Words):

D1 → He is a good boy

D2 → She is a good girl

D3 → Boy and girl are good

• First apply stop words & Lower the characters
  • D1 → good boy
  • D2 → good girl
  • D3 → Boy girl good