Data analysis in JavaScript
- 1. Data analysis in JavaScript Yes it can be done
- 2. Aims ● Show that data analysis can be done in JavaScript (live coding!) ● Demonstrate practical techniques you can use immediately ● Show off some tools that I’ve created it
- 3. Twitter Please Tweet @BrisJS and @ashleydavis75 Also include #javascript #nodejs #visualization
- 4. My story ● Long-time software developer ● Mostly in games development ● A few years in the financial world ● Last 7 years: virtual reality, Node.js and web tech ● Systematic stock trader
- 5. My book ● Data Wrangling with JavaScript ● Available on Manning Early Access Program (MEAP) ● Pre-order on Amazon ● Soon to be printed 40% discount voucher: datawrangjs http://www.datawranglingwithjavascript.com/
- 6. Today’s tools ● Data-Forge ● Data-Forge Plot
- 7. Data-Forge ● Open-source ● Mature ● Swiss army knife for data ● Inspired by LINQ and Pandas ● Works in Node.js and the browser ● Please star the repo! github.com/data-forge/data-forge-ts www.npmjs.com/package/data-forge
- 8. Data-Forge Plot ● Open source ● Very new, early stages, looking for contributors ● Extension to Data-Forge for visualizations ● Super easy to produce charts from your data ● A range of charting libraries ● Static and interactive visualizations ● Please star the repo! github.com/data-forge/data-forge-plot www.npmjs.com/package/data-forge-plot
- 9. Reasons to use JavaScript for data analysis ● You already work in JavaScript ● Your production environment is JavaScript ● JavaScript is a mature and capable language ● Interactive visualization always ends up in JavaScript anyway ● The JavaScript ecosystem is growing
- 10. Reasons not to use JS for data analysis ● Maturity and experience ○ Missing tools (we are fixing this) ○ Missing native libraries (this is more difficult) ● Algorithm stability and consistency ○ Consistency been implementations ○ Consistency between Node.js and browsers ● Lack of experienced data scientists
- 11. Data analysis in JavaScript
- 12. Data analysis questions? Any data analysis starts with questions. What do you want to know? ● How much weight have I lost ● How much weight have I lost on average per day / week ● Which days are better/worse for weight loss ● How do I know when I’ve gained or lost significant weight
- 13. Data analysis techniques ● Loading a data file ● Plotting a chart ● Group and summarize
- 14. Live coding What could go wrong? https://github.com/ashleydavis/brisjs-data-analysis-talk
- 15. Resources ● Contact ○ Email: ashley@codecapers.com.au ○ Twitter: @ashleydavis75 ○ GitHub: ■ ashleydavis ■ data-forge ● Data Wrangling with JavaScript ○ datawranglingwithjavascript.com ● The Data Wrangler ○ the-data-wrangler.com Your support ● Star my repos ● Help promote my work ● Contribute to open source ● Buy Data Wrangling with JavaScript