# Similarity Measures ### Key Ideas * Often it is said that the [Dot Product](Dot%20Product.md) measures the similarity of vectors. This is true to a degree, but we must keep in mind that it captures their magnitude as well. This may not be way we want. Consider [this example](https://photos.google.com/photo/AF1QipOWz6DsnocKOEEpVvPaQr0j_4bjOuVOwE_qYarB). * We have other similarity measures that may be more appropriate, such as cosine similarity and euclidean distance.   --- References: * [Measuring Similarity from Embeddings](https://developers.google.com/machine-learning/clustering/similarity/measuring-similarity)