# A Comparison of Semantic Similarity Methods for Maximum Human Interpretability

arXiv: https://arxiv.org/abs/1910.09129 · HTML: https://mlsanity.com/paper/1910.09129/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Oct 2019.

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

- [Eliciting Latent Predictions from Transformers with the Tuned Lens](https://mlsanity.com/paper/2303.08112/) (2023)
- [CONFLARE: CONFormal LArge language model REtrieval](https://mlsanity.com/paper/2404.04287/) (2024)
- [BanglaAutoKG: Automatic Bangla Knowledge Graph Construction with Semantic Neural Graph Filtering](https://mlsanity.com/paper/2404.03528/) (2024)
- [Revisiting Cosine Similarity via Normalized ICA-transformed Embeddings](https://mlsanity.com/paper/2406.10984/) (2024)
- [Natural Language Generation Using Link Grammar for General Conversational Intelligence](https://mlsanity.com/paper/2105.00830/) (2021)
- [App Review Driven Collaborative Bug Finding](https://mlsanity.com/paper/2301.02818/) (2023)
- [Rethinking Word Similarity: Semantic Similarity through Classification Confusion](https://mlsanity.com/paper/2502.05704/) (2025)
- [Conical Classification For Computationally Efficient One-Class Topic Determination](https://mlsanity.com/paper/2111.00375/) (2021)
- [Basis Vector Metric: A Method for Robust Open-Ended State Change Detection](https://mlsanity.com/paper/2509.07308/) (2025)

## Abstract (arXiv)

The inclusion of semantic information in any similarity measures improves the efficiency of the similarity measure and provides human interpretable results for further analysis. The similarity calculation method that focuses on features related to the text's words only, will give less accurate results. This paper presents three different methods that not only focus on the text's words but also incorporates semantic information of texts in their feature vector and computes semantic similarities. These methods are based on corpus-based and knowledge-based methods, which are: cosine similarity using tf-idf vectors, cosine similarity using word embedding and soft cosine similarity using word embedding. Among these three, cosine similarity using tf-idf vectors performed best in finding similarities between short news texts. The similar texts given by the method are easy to interpret and can be used directly in other information retrieval applications.