Paul Kishchuk from Vector Research did an introductary session on Data Basics for Business Owners on March 10th. Here are some key takeaways:
The DIKW Pyramid:
| Data | Raw facts & figures without context. |
| Information | Contextualized data. Processed, cleaned, summarized, aggregated, or validated. Useful for answering who, what, where, when. |
| Knowledge | Analyzed and interpreted information to identify patterns, trends, or relationships. Helps us with understanding how or why things occur and inform decisions that aid with achieving our goals. |
| Wisdom | Understanding of the underlying knowledge, with context, history, and experience. Helps with making informed decisions. |
- Most of what we’re presented with is information, e.g. you bank statement with summary of transactions & your total balance.
- It’s useful to access raw data in order to draw meaningful relationships and understand patterns in behaviour, e.g. spending by category.
Data
Is the gateway to analytics and visualization.
Structured Data
- Alphanumeric
- Fixed in format
- In rows & columns
- Consistent
- No nested data
- No totals or summarization
- Machine-readable
Could be collected from a single point in time, over time, or in real-time (also known as Streaming Data).
Data Union: When you add new data to an existing dataset (needs to have same structure), used to lengthen a data series.
Sentiment Analysis: Analyzing qualitative data (text) to determine the emotional tone of the messages. i.e. Associating keywords with positive, neutral, or negative sentiments and identifying how many times it is mentioned.
Tools:
- Tableau: Data visualization tool.
- Microsoft Power BI: Tableau alternative.
- Mockaroo: Generates random data based on a data schema you define. Free for 1000 rows.
