# MulBot: Unsupervised Bot Detection Based on Multivariate Time Series

arXiv: https://arxiv.org/abs/2209.10361 · HTML: https://mlsanity.com/paper/2209.10361/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 57 references that are in the map.
- First posted on arXiv 21 Sep 2022.

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [TweepFake: about Detecting Deepfake Tweets](https://mlsanity.com/paper/2008.00036/) (2020)
- [The spread of low-credibility content by social bots](https://mlsanity.com/paper/1707.07592/) (2017)
- [Online Human-Bot Interactions: Detection, Estimation, and Characterization](https://mlsanity.com/paper/1703.03107/) (2017)
- [Scalable and Generalizable Social Bot Detection through Data Selection](https://mlsanity.com/paper/1911.09179/) (2019)
- [The paradigm-shift of social spambots: Evidence, theories, and tools for the arms race](https://mlsanity.com/paper/1701.03017/) (2017)
- [A Decade of Social Bot Detection](https://mlsanity.com/paper/2007.03604/) (2020)
- [BotOrNot: A System to Evaluate Social Bots](https://mlsanity.com/paper/1602.00975/) (2016)
- [Heterogeneity-aware Twitter Bot Detection with Relational Graph Transformers](https://mlsanity.com/paper/2109.02927/) (2021)
- [Deep Neural Networks for Bot Detection](https://mlsanity.com/paper/1802.04289/) (2018)
- [Fame for sale: efficient detection of fake Twitter followers](https://mlsanity.com/paper/1509.04098/) (2015)
- [Detection of Novel Social Bots by Ensembles of Specialized Classifiers](https://mlsanity.com/paper/2006.06867/) (2020)
- [RTbust: Exploiting Temporal Patterns for Botnet Detection on Twitter](https://mlsanity.com/paper/1902.04506/) (2019)
- [Bots increase exposure to negative and inflammatory content in online social systems](https://mlsanity.com/paper/1802.07292/) (2018)
- [Twitter Bot Detection Using Bidirectional Long Short-term Memory Neural Networks and Word Embeddings](https://mlsanity.com/paper/2002.01336/) (2020)
- [SATAR: A Self-supervised Approach to Twitter Account Representation Learning and its Application in Bot Detection](https://mlsanity.com/paper/2106.13089/) (2021)
- [LOBO -- Evaluation of Generalization Deficiencies in Twitter Bot Classifiers](https://mlsanity.com/paper/1809.09684/) (2018)
- [Graph-Hist: Graph Classification from Latent Feature Histograms With Application to Bot Detection](https://mlsanity.com/paper/1910.01180/) (2019)
- [Better Safe Than Sorry: An Adversarial Approach to Improve Social Bot Detection](https://mlsanity.com/paper/1904.05132/) (2019)
- [HoloScope: Topology-and-Spike Aware Fraud Detection](https://mlsanity.com/paper/1705.02505/) (2017)
- [The Rise of Social Botnets: Attacks and Countermeasures](https://mlsanity.com/paper/1603.02714/) (2016)
- [GANs for Semi-Supervised Opinion Spam Detection](https://mlsanity.com/paper/1903.08289/) (2019)
- [An Unsupervised Approach to Detect Spam Campaigns that Use Botnets on Twitter](https://mlsanity.com/paper/1804.05232/) (2018)
- [Discovery of the Twitter Bursty Botnet](https://mlsanity.com/paper/1709.06740/) (2017)

## Built on by (most cited first, 3 of 3)

- [Multimodal Detection of Bots on X (Twitter) using Transformers](https://mlsanity.com/paper/2308.14484/) (2023)
- [A Graph Neural Architecture Search Approach for Identifying Bots in Social Media](https://mlsanity.com/paper/2411.16285/) (2024)
- [Multimodal Coordinated Online Behavior: Trade-offs and Strategies](https://mlsanity.com/paper/2507.12108/) (2025)

## Abstract (arXiv)

Online social networks are actively involved in the removal of malicious social bots due to their role in the spread of low quality information. However, most of the existing bot detectors are supervised classifiers incapable of capturing the evolving behavior of sophisticated bots. Here we propose MulBot, an unsupervised bot detector based on multivariate time series (MTS). For the first time, we exploit multidimensional temporal features extracted from user timelines. We manage the multidimensionality with an LSTM autoencoder, which projects the MTS in a suitable latent space. Then, we perform a clustering step on this encoded representation to identify dense groups of very similar users -- a known sign of automation. Finally, we perform a binary classification task achieving f1-score = 0.99, outperforming state-of-the-art methods (f1-score ≤ 0.97). Not only does MulBot achieve excellent results in the binary classification task, but we also demonstrate its strengths in a novel and practically-relevant task: detecting and separating different botnets. In this multi-class classification task we achieve f1-score = 0.96. We conclude by estimating the importance of the different features used in our model and by evaluating MulBot's capability to generalize to new unseen bots, thus proposing a solution to the generalization deficiencies of supervised bot detectors.