Machine learning
summary
Naive copying of libraries and Excel formulas fails but a data‑first workflow succeeds: raw URL lists turn into feature vectors, experts label samples, and models trained in R caret (with PHP fetching and orchestration) reliably separate good links from spam after moving from one model to many per site type. The rescue task handles 1,000,000+ backlinks, demo code uses the PHP-ML library and C extensions for speed, and a recommendation system already serves billions of pageviews per month.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:48Backlink classification case
- 8:01Manual to ML approach
- 11:18First ML failure
- 13:28Data-oriented machine learning workflow
- 16:00ML tasks and algorithms
- 20:14Code review price prediction
- 25:43Data understanding
- 28:20Exploratory analysis
- 30:37Neural networks in PHP
- 31:42PHP for ML
- 33:11ML libraries and tools
- 36:21ML as a process
related talks
45mMachine learning and trend analysis in PHP
37mLearning Machine Learning
21mMachine Learning and Cyber Security - Detecting malicious URLs in the haystack
42mMachine learning in your hobby project: Obstacles, entry points and solutions
22mReal-world applications of machine learning: Beyond the hype
54mMachine Learning explained with Examples
36mClosing Keynote: Machine Learning APIs with PHP
50mMachine Learning powered regression