Comparative Study of Machine Learning Classifiers and Guidance for Algorithm Selection
The article summarizes a JMLR 2014 study that evaluated 179 classifiers across 121 UCI datasets, finding Random Forests and Gaussian‑kernel SVMs to be top performers, provides a review of supervised learning algorithms, and includes visual guidance for selecting appropriate machine‑learning methods.
Author: Yu Fei. Link: https://www.zhihu.com/question/27306416/answer/36701217 . Source: Zhihu. Copyright belongs to the author; please contact the author for permission to reproduce.
JMLR 2014 October issue published a notable paper titled “Do we Need Hundreds of Classifiers to Solve Real‑World Classification Problems?” which evaluated 179 classification models on all 121 UCI datasets and found that Random Forests and SVMs with Gaussian kernels (using LibSVM) performed best. Do we Need Hundreds of Classifiers to Solve Real World Classification Problems?
I first encountered this on Liu Zhiyuan’s Weibo, referencing the Sina Visitor System; JMLR is a top journal in the machine‑learning field.
2. Review of Algorithms The paper provides a review of several supervised learning algorithms. https://s3-us-west-2.amazonaws.com/mlsurveys/54.pdf At the end of the document there is a table comparing the properties of various algorithms.
3. Algorithm Selection In addition, the following diagram is provided to help you choose a machine‑learning algorithm.
Architects Research Society
A daily treasure trove for architects, expanding your view and depth. We share enterprise, business, application, data, technology, and security architecture, discuss frameworks, planning, governance, standards, and implementation, and explore emerging styles such as microservices, event‑driven, micro‑frontend, big data, data warehousing, IoT, and AI architecture.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.