One of the beauties of running Javascript related applications is you don’t need to install any client side software, optimize servers and spend tons of time on the core infrastructure. Javascript just work outs of the core browser. In that spirit, there is a lot of increasing momentum on building Machine Learning in Javascript. We have collected a list of resources on Javascript that will be helpful if you are building machine learning applications in Javascript.

A great starting point for learning about Javascript in Machine learning is to go through these slides by Heather Arthur. The library goes through examples of doing image recognition using Javascript.

Here is a compilation of machine learning libraries in Javascript:

1. Encog’s machine learning framework in Javascript : Encog is a machine learning framework available for Java, .Net, and C++. Encog supports different learning algorithms such as Bayesian Networks, Hidden Markov Models and Support Vector Machines. However, its main strength lay in its neural network algorithms. Encog contains classes to create a wide variety of networks, as well as support classes to normalize and process data for these neural networks. Encog trains using many different techniques. Multithreading is used to allow optimal training performance on multicore machines. The C++ version of Encog can offload some processing to an OpenCL compatible GPU for further performance gains. Here is a great case study & example using Encog’s machine learning in Javascript.

2. Deep learning with Java script : The library allows you to formulate and solve Neural Networks in Javascript, and was originally written by Andrej Karpathy (a PhD student at Stanford). However, the library has since been extended by contributions from the community.

3. Brain.js  A neural network library

4. svmjs SVM library

5. Forest.js Javascript library implementing Random trees

6. Numeric.js Performs sophisticated numerical computations.

7. Node-sylvester Linear algebra library

8. Lineareg.js Linear regression with Javascript.

9. Bayesian classifiers including NaturalNode/natural  & harthur/classifier

10. Clustering: harthur/clusterfck

Here is a list of great Algorithms implemented by Buran Kanber in Javascript

1. k-nearest-neighbor (Introduction)
2. k-means clustering (Part 1)
3. Genetic algorithms (Part 1, Part 2)
4. Naive Bayes classifier (Part 1: Document Classification)
5. Sentiment Analysis (Part 1)
6. Full-text Search (Part 1: Relevance Scoring)

Give here is a list of great applications using Javascript Machine learning:

1. https://github.com/rogerbraun/HNBayes - A bayes classifier for the popular Hacker News website. Here is a presentation on how it was done.

2. http://liuliu.me/ccv/js/nss/ - A Face detection application. Just paste an image URL and it circles the part of the image that it considers as a face.

3. nude.js is a JavaScript implementation of a nudity scanner based on approaches from research papers. HTMLCanvas makes it possible to analyse image data and return whether it's nude or not.  

4. OCR solving of Captchas using JS neural net approaches.