What is the key difference between the n$-$gram model and negative sampling? N$-$gram models are based on sliding windows, while negative sampling focuses on token frequencies

N$-$gram models capture local context without neural networks, whereas negative sampling captures local context using neural networks

N$-$gram models are unsupervised, while negative sampling is a supervised learning technique

N$-$gram models are primarily used for image processing, while negative sampling is specific to text data