Cheat Sheet: Get started with Data Science and Python

This blog post is a short cheat sheet when it comes to getting started with data science and Python and is not a full list. The post lists some useful main topics to get familiar with:

• Basic tooling
• Using existing models
• Build your own models and machine learning
• Deep learning
• Useful learning resources

1. Basic tooling

• Python – Open-Source – Programming language
  • Widely used by data scientists and very easy to consume when you work with strings, listings, and handle various file resources. I like this blog post for more details Top 7 Reasons Why You Need to Learn Python as a Data Scientist
• Jupyter Notebook – Open-Source Workbench for Data Science uses Python.
• Here is a useful blog post autocomplete in jupyter notebooks, when you are going to work with Jupyter Notebook.
• Natural Language Toolkit _(NLTK)_ which is Open-Source and is a leading platform for building Python programs to work with human language data.
  • Stemming – “Where does a word come from?” (For details, please visit Sample usage for stem within NLTK)
  • Valence Aware Dictionary for Sentiment Reasoning (VADER) is a lexicon and rule-based sentiment analysis tool that is specifically attuned to sentiments expressed in social media. (open-sourced)
• spaCy an Open-Source Natural Language Processing library (NLP) model for Python

2. Existing models

Maybe it’s a good choice to start with Natural Language Processing. Therefore you can start with spaCy an Open-Source Natural Language Processing library for Python and you can check out the following topics:

• Token/Tokenization – What is the category/label of a word/character/word combination? A token is an instance of a sequence of characters in some particular document that is grouped as a useful semantic unit for processing. from the Cambridge University Press. (Details in spaCY)
• Lemmatization is closely related to stemming.
• Stop words are the words that are filtered out before or after the processing of natural language data (text). (Details in spaCY)
• Matching finding matched tokens in a context.
• Parts of Speech PoS and tagging. (Details in spaCY)
• Visualizing some dependencies with Displacy. (Details in spaCY)
• Named entity recognition locates and classifies named entities mentioned in unstructured text. (Details in spaCY)
• Word2vec two-layer network
• SciPy contains fundamental algorithms for scientific computing in Python.

3. Build models and machine learning.

Machine learning algorithms build a model based on sample data, known as training data, to make predictions or decisions without being explicitly programmed to do so.

• Scikit learn which does contain simple and efficient tools for predictive data analysis.
• Supervised learning using labeled data for historical data (for example spam filters, what can be future spams?).
• Unsupervised learning is a type of algorithm that learns patterns from untagged data.
  • term frequency–inverse document frequency (tf-idf) which is a numerical statistic that is intended to reflect how important a word is to a document in a collection or corpus. Used in context with LDA and NMF.
  • Latent Dirichlet allocation LDA is a generative statistical model that explains a set of observations through unobserved groups, and each group explains why some parts of the data are similar.
  • Non-negotiable matrix factorization NMF is a group of algorithms in multivariate analysis and linear algebra where a matrix V is factorized into (usually) two matrices W and H, with the property that all three matrices have no negative elements.
• Accuracy and precision how often do the model deliver the right result?
• A confusion matrix is a specific table layout that allows visualization of the performance of an algorithm, typically a supervised learning one (in unsupervised learning it is usually called a matching matrix).

4. Get started with deep learning.

• Deep learning is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised, or unsupervised.
  • Long short-term memory (LSTM) is an artificial neural network used in the fields of artificial intelligence and deep learning.
  • Rectifier helps to provide flexible handling of input for the activation function.
  • Keras is an open-source software library that provides a Python interface for artificial neural networks.
  • A recurrent neural network is a class of artificial neural networks where connections between nodes can create a cycle, allowing output from some nodes to affect subsequent input to the same nodes.

5. Useful learning resources

• Natural Language Processing with Python an Udemy course
• Getting Started with Python for Data Science a free course by codecademy
• How to Learn Python for Data Science in 5 Steps A blog post from DATAQUEST.
• Python for Data Science (Ultimate Quickstart Guide) A short free guide at Elite Data Science.
• Free learning on Cognitive Class.ai

---

I hope this was useful to you and let’s see what’s next?

Greetings,

Thomas

#ai, #cheatsheet, #datascience