# Word2Vec: Optimal Hyper-Parameters and Their Impact on NLP Downstream Tasks

arXiv: https://arxiv.org/abs/2003.11645 · HTML: https://mlsanity.com/paper/2003.11645/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Mar 2020.

## Built on by (most cited first, 4 of 4)

- [Potential Idiomatic Expression (PIE)-English: Corpus for Classes of Idioms](https://mlsanity.com/paper/2105.03280/) (2021)
- [HaT5: Hate Language Identification using Text-to-Text Transfer Transformer](https://mlsanity.com/paper/2202.05690/) (2022)
- [Knowledge Enhanced Graph Neural Networks for Graph Completion](https://mlsanity.com/paper/2303.15487/) (2023)
- [NLP Meets RNA: Unsupervised Embedding Learning for Ribozymes with Word2Vec](https://mlsanity.com/paper/2307.05537/) (2023)

## Abstract (arXiv)

Word2Vec is a prominent model for natural language processing (NLP) tasks. Similar inspiration is found in distributed embeddings for new state-of-the-art (SotA) deep neural networks. However, wrong combination of hyper-parameters can produce poor quality vectors. The objective of this work is to empirically show optimal combination of hyper-parameters exists and evaluate various combinations. We compare them with the released, pre-trained original word2vec model. Both intrinsic and extrinsic (downstream) evaluations, including named entity recognition (NER) and sentiment analysis (SA) were carried out. The downstream tasks reveal that the best model is usually task-specific, high analogy scores don't necessarily correlate positively with F1 scores and the same applies to focus on data alone. Increasing vector dimension size after a point leads to poor quality or performance. If ethical considerations to save time, energy and the environment are made, then reasonably smaller corpora may do just as well or even better in some cases. Besides, using a small corpus, we obtain better human-assigned WordSim scores, corresponding Spearman correlation and better downstream performances (with significance tests) compared to the original model, trained on 100 billion-word corpus.