Summary of Classifying Spam Emails Using Agglomerative Hierarchical Clustering and a Topic-based Approach, by F. Janez-martino et al.
Classifying spam emails using agglomerative hierarchical clustering and a topic-based approach
by F. Janez-Martino, R. Alaiz-Rodriguez, V. Gonzalez-Castro, E. Fidalgo, E. Alegre
First submitted to arxiv on: 7 Feb 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research paper presents a novel approach to classify spam emails into multiple categories, focusing on the needs of cybersecurity units. The authors propose two datasets, SPEMC-15K-E (English) and SPEMC-15K-S (Spanish), each containing approximately 15,000 labeled emails using agglomerative hierarchical clustering into 11 classes. The paper evaluates 16 pipelines combining different text representation techniques (TF-IDF, Bag of Words, Word2Vec, BERT) with various classifiers (Support Vector Machine, Naive Bayes, Random Forest, Logistic Regression). Results show that TF-IDF and Logistic Regression achieve the highest performance for both English and Spanish datasets, with F1 scores of 0.953 and 0.945, respectively, along with high accuracy rates. The paper also highlights the processing time, revealing that TF-IDF with Logistic Regression leads to the fastest classification. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Imagine getting rid of annoying spam emails that can harm your computer or steal personal information. This research aims to make it easier for cybersecurity teams to identify and block these unwanted messages. The team created two large datasets of labeled spam emails in English and Spanish, with 15,000 examples each. They tested different ways to analyze the emails using various algorithms. Results show that one method (TF-IDF and Logistic Regression) performs exceptionally well on both datasets, accurately identifying spam emails most of the time. This breakthrough could help keep your digital life safer! |
Keywords
* Artificial intelligence * Bag of words * Bert * Classification * Hierarchical clustering * Logistic regression * Naive bayes * Random forest * Support vector machine * Tf idf * Word2vec
