This repository is dedicated to utilizing advanced models and data processing techniques for Named Entity Recognition (NER). It showcases three models, including a KNN model (model 1), a Feed-Forward Network (model 2), and a competitive LSTM model (competitive model). A distinctive aspect of our approach is the use of word embeddings, which were performed using a combination of GloVe and Word2Vec models. To improve the F1 score, we have taken steps to address class imbalance in the competitive model through various approaches, such as under-sampling, over-sampling, and class-weight adjustments. Our report, found within this repository, provides a comprehensive overview of all methodologies and their detailed implementations. The repository includes the data folder and an instructions file for easy use.