Open acs opened 6 years ago
Some references about learning ML:
and before deciding the tech adoption:
Infographic for Python, Scala and R technologies Top 20 libs for Python
A pretty interesting site based on ML for providing contents:
And once the moment arrives to learn the basic algorithms for ML: https://www.quora.com/What-are-the-top-10-data-mining-or-machine-learning-algorithms/answer/Xavier-Amatriain
I am starting to read the book Data Science from Scratch Joel Grus (examples) and it starts talking about this venn diagram about what is a data science. I have started also the ML in coursera (this one is recommend also in the Joel's book)
Probably a good start is to use the Nearest neighbors model (and it has supervised and unsupervised types).
From Joel's book:
Nearest neighbors model, one of the simplest predictive models there is. It makes no mathematical assumptions, and it doesn’t require any sort of heavy machinery. The only things it requires are:
Scikit learn supports them so it also a great choice to start playing this library.
And there are samples using it: https://github.com/dabbabi-zayani/Twitter-Sentiment-Classifier https://www.analyticsvidhya.com/blog/2018/03/introduction-k-neighbours-algorithm-clustering/
After 9 months I am more interested yet in this field. Yesterday I was in:
https://www.meetup.com/es-ES/MachineLearningSpain/events/259864319/
presenting our devroom about Machine Learning and Big Data and I attended a presentation in which Asier Arranz (https://twitter.com/asierarranz) showed the cool stuff that can be done with:
And recently Tensorflow 2.0 was published and it has been integrated with Keras. So it is a great moment to start using this new version.
An interesting team working in applying AI: https://openai.com/
The Kaggle micro course about ML is great: few time needed to learn the basic concepts.
In this use case I plan to use the perceval twitter backend to collect data from Twitter and then use machine learning algorithms to show interesting facts. The idea is to use Python for the development. There are a lot of work around this use case, so the first step is the state of the art analysis.
For example:
https://www.quora.com/What-are-some-real-world-applications-of-applying-Machine-Learning-Algorithms-to-Twitter-Data-Set-I-have-tried-Recommendation-Algorithm-and-I-want-to-know-more-so-that-it-gives-me-more-practical-exposure-to-Applied-Machine-Learning