Quick and Dirty Data Science with Incanter
summary
Quick, concrete recipe for scraping and analysing time series with Clojure: use Incanter (+ encanter.io) to read CSV/JSON logs, clj-time to parse dates and coerce to milliseconds, then dollar-map/dollar-rollup and dollar-join to compute seasonal means and standard-deviation bands. The live code builds a JFreeChart time series with dashed high/low thresholds and alert points; it flags nine low-beer days in April and shows a roughly flat overall trend (mean ≈ 8,922 pints/day), so the pipeline works end-to-end.
AI-written summary · about
chapters (17)
- 0:00Quick and dirty data science
- 1:09Understanding data needs
- 3:18ETL for data transformation
- 5:24Incanter for time series
- 7:50Loading beer dataset
- 9:16Converting dates
- 11:10Rolling means calculation
- 14:03Time series charts
- 17:34Interpreting charts
- 21:26Weekly cycles
- 25:02Seasonal means by day
- 28:20Standard deviation for alerts
- 31:09Joining seasonal data
- 33:30Alert thresholds
- 36:17Visualizing alerts
- 39:09April anomalies analysis
- 40:32Conclusion
related talks
26mDoing data science with Clojure: the ugly, the sad, the joyful
43mDoing Data Science with Clojure
40mClojure Emerging Data Ecosystem
44mData Science in Clojure
18mNavigating Clojure: A Fresh Look From a Data Analysis Background
33mClojure Data Science
01h 26mData Driven Action: a primer on Data Science
27mData and Clojure: Using Clojure for more than software development