# How to Generate a Good Word Embedding?

arXiv: https://arxiv.org/abs/1507.05523 · HTML: https://mlsanity.com/paper/1507.05523/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jul 2015.

## Built on by (most cited first, 9 of 9)

- [Detecting of a Patient's Condition From Clinical Narratives Using Natural Language Representation](https://mlsanity.com/paper/2104.03969/) (2021)
- [Exploring the Potential of Large Language Models for Improving Digital Forensic Investigation Efficiency](https://mlsanity.com/paper/2402.19366/) (2024)
- [Document Embedding for Scientific Articles: Efficacy of Word Embeddings vs TFIDF](https://mlsanity.com/paper/2107.05151/) (2021)
- [Fusing Domain-Specific Content from Large Language Models into Knowledge Graphs for Enhanced Zero Shot Object State Classification](https://mlsanity.com/paper/2403.12151/) (2024)
- [LSTM-RASA Based Agri Farm Assistant for Farmers](https://mlsanity.com/paper/2204.09717/) (2022)
- [Leveraging knowledge graphs to update scientific word embeddings using latent semantic imputation](https://mlsanity.com/paper/2210.15358/) (2022)
- [Multi-criteria recommendation systems to foster online grocery](https://mlsanity.com/paper/2312.08393/) (2023)
- [Simplicity within biological complexity](https://mlsanity.com/paper/2405.09595/) (2024)
- [The Impact of Dimensionality on the Stability of Node Embeddings](https://mlsanity.com/paper/2604.08492/) (2026)

## Abstract (arXiv)

We analyze three critical components of word embedding training: the model, the corpus, and the training parameters. We systematize existing neural-network-based word embedding algorithms and compare them using the same corpus. We evaluate each word embedding in three ways: analyzing its semantic properties, using it as a feature for supervised tasks and using it to initialize neural networks. We also provide several simple guidelines for training word embeddings. First, we discover that corpus domain is more important than corpus size. We recommend choosing a corpus in a suitable domain for the desired task, after that, using a larger corpus yields better results. Second, we find that faster models provide sufficient performance in most cases, and more complex models can be used if the training corpus is sufficiently large. Third, the early stopping metric for iterating should rely on the development set of the desired task rather than the validation loss of training embedding.