Summary of Evaluating Rocket and Catch22 Features For Calf Behaviour Classification From Accelerometer Data Using Machine Learning Models, by Oshana Dissanayake et al.
Evaluating ROCKET and Catch22 features for calf behaviour classification from accelerometer data using Machine Learning models
by Oshana Dissanayake, Sarah E. McPherson, Joseph Allyndree, Emer Kennedy, Padraig Cunningham, Lucile Riaboff
First submitted to arxiv on: 28 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Signal Processing (eess.SP)
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 paper investigates the application of machine learning models to automatically classify calf behavior in dairy farms using accelerometer data from neck collars. The authors compare the performance of ROCKET, Catch22, and hand-crafted features for time-series classification problems. They analyzed 27.4 hours of annotated behaviors from 30 pre-weaned calves, computing additional time-series features and splitting them into train, validation, and test sets. Three machine learning models (Random Forest, eXtreme Gradient Boosting, and RidgeClassifierCV) were trained to classify six behaviors indicative of calf welfare, achieving the best performance with ROCKET features (0.70 +/- 0.07). The results demonstrate the potential of precision livestock farming in advancing animal welfare on a larger scale. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This research aims to develop a way to automatically monitor and understand how calves behave in dairy farms using special collars that track their movements. By analyzing this data, scientists can identify behaviors that affect calf well-being, like weaning or dehorning. The study compares three different methods for analyzing the data: ROCKET, Catch22, and hand-crafted features. They found that the ROCKET method works best (70% accurate) and could be used to improve animal welfare on a larger scale. |
Keywords
» Artificial intelligence » Classification » Extreme gradient boosting » Machine learning » Precision » Random forest » Time series
