With several excellent stories published in November and December, we update
our list of

top 2015 stories, using data from the full year 2015, and using unique page views for ranking.

Here are the top 15 most viewed KDnuggets stories in 2015 and a word cloud of

top stories.

### 15 Most Viewed from 2015

- R vs Python for Data Science: The Winner is ...
- 60+ Free Books on Big Data, Data Science, Data Mining, Machine Learning, Python, R, and more
- Top 20 Python Machine Learning Open Source Projects
- 50+ Data Science and Machine Learning Cheat Sheets
- TensorFlow Disappoints - Google Deep Learning falls shallow
- The Grammar of Data Science: Python vs R
- Top 30 Social Network Analysis and Visualization Tools
- The one language a Data Scientist must master
- 7 Steps to Mastering Machine Learning With Python
- Gartner 2015 Magic Quadrant for Advanced Analytics Platforms: who gained and who lost
- Top 5 arXiv Deep Learning Papers, Explained
- 5 Best Machine Learning APIs for Data Science
- Top 20 Data Science MOOCs
- Data Lake vs Data Warehouse: Key Differences
- Deep Learning for Text Understanding from Scratch

To compute the number of shares, we used excellent

BuzzSumo Chrome Extension.

### 15 Most shared stories from 2015

- 60+ Free Books on Big Data, Data Science, Data Mining, Machine Learning, Python, R, and more
- Top 20 Python Machine Learning Open Source Projects
- 50+ Data Science and Machine Learning Cheat Sheets
- Poll: What Predictive Analytics, Data Mining, Data Science software/tools you used in the past 12 months?
- Which Big Data, Data Mining, and Data Science Tools go together?
- R vs Python for Data Science: The Winner is ...
- The one language a Data Scientist must master
- 30 Can't miss Harvard Business Review articles on Data Science, Big Data and Analytics
- Top 5 arXiv Deep Learning Papers, Explained
- R leads RapidMiner, Python catches up, Big Data tools grow, Spark ignites
- 7 common mistakes when doing Machine Learning
- Gartner 2015 Hype Cycle: Big Data is Out, Machine Learning is in
- Deep Learning for Text Understanding from Scratch
- 5 Best Machine Learning APIs for Data Science
- 10 things statistics taught us about big data analysis

Analysis of views vs shares shows that there is little correlation (R

^{2}=0.38, but only 0.05 after excluding the 3 most viewed posts), but also a wide divergence, from 1.9 shares/100 views (lowest among top posts) to 23 shares/100 views.

Since shares occur after viewing, we can use shares/100 unique views as a measure of most shareable content.

Here are 5 most shareable posts

- Poll: What Predictive Analytics, Data Mining, Data Science software/tools you used in the past 12 months?, 23.0 shares/ 100 unique views
- 30 Can't miss Harvard Business Review articles on Data Science, Big Data and Analytics, 17.0
- Which Big Data, Data Mining, and Data Science Tools go together?, 14.8
- Top 10 Machine Learning Projects on Github, 11.0
- R leads RapidMiner, Python catches up, Big Data tools grow, Spark ignites, 10.5

The least shareable posts were