Digest of articles on data analysis No. 4 (06.23.2014 —06.07.2014)

Good afternoon, dear readers.
Flew 2 weeks and I present to you a new collection of materials on the topic of data analysis. This time, the material turned out quite a bit, but from this article and notes in the digest does not become less interesting.
So, from today's selection you will learn how to classify text using naive Bayes, learn how to increase the speed of the gradient descent. In addition, there will be another article about what the analyst should use and what skills to possess and much more interesting.

Theory

• Cluster analysis (using consumer segmentation as an example) Part 1 (RU) (thanks to great_boba for the translation)
• Text classification using the Baysev classifier (EN)
  Instructions for use of naive Baysevskogo classifier to the text
• How banks can add value to customer data
  Quite an interesting presentation on this topic.
• Operating Environment Analysis (EN)
  A small instruction on the methodology for the comparative analysis of complex technical, economic and social systems.
• Choosing a learning rate factor in gradient descent (EN)
  
• Modern aspects of the presentation of texts in the analysis of natural language: classical and alternative approaches (RU)
  
• Hard and Soft data analytics skills
  Another note about what skills an analyst should possess
• 7 lessons from McGraw-Hill data analysis
  Useful tips from the analysis gurus
• Use component selection methods to classify text.
  

')

Practice using different tools

• Development of naive Bayes classifier on JAVA (EN)
  
• Deploying a Scoring System for Prediction (EN)
  Deploying a system using OpenCPU and R
• Analyzing Google Analytics data with Big Query and R

Educational videos

• Determine cybercrime with big data analysis
  

Various related articles

• If you think big data is numeric, think again
• Transformation of data analysis into data science
  Reasoning on the topic that includes the concept of Data Science
• Visualization helps you understand dirty data.
  How various visualization tools will help to understand noisy data.
• Big Challenge for Big Data: Climate Change
  A note on how big data is used in weather forecasting.
• What is machine learning? It depends on who you ask.
  What can people understand by the term "machine learning"
• Using big data to optimize digital marketing
• How big data can improve your customer relationships
  In the article, the CEO of SugarCRM answers a series of questions on the designated topic.
• How to work with Big Data. 10 rules (RU)
• Effective median score (RU)