The data analytics industry is littered with a lot of different terminology. Some of which are now part of every day discussions between business leaders and their executive teams. It can be all too easy to get swamped in the dearth of terms and wonder…
What do any of the terms actually mean?
Which ones are relevant?
Why are they relevant?
The following list is by no means exhaustive; but is a list of simple introductory terms that business leaders and their teams should know and understand.
- Big/Small data
- Data lake
- Data pipeline
- Descriptive analytics
- Predictive analytics
- Prescriptive analytics
- Artificial intelligence
- Machine learning
Big/Small data
Big data refers to data that is so large, fast & complex that it isn’t possible to adequately process it using standard techniques. It is categorised by the 3 Vs, as mentioned in our Data Trends article, and is therefore not able to be easily processed and analysed by humans.
In contrast, small data is a subset of big data that is manageable, easily accessible and directly understood by humans.
Data lake
A data lake is one way that organisations can use to store data in a central place. The data can be structured or unstructured and it is stored in its raw format. The idea being that for data that might have multiple use cases, potentially across different departments – the data can be accessed and used without impacting other uses of it.
Data pipeline
A data pipeline is used to describe the process of taking data in its raw format and making it ready for analysing. This is done by applying a series of processing steps that transform the data from its original format into a format that can be easily used for a specific analysis or analyses..
Descriptive analytics
Descriptive analytics is the ability to examine and interpret historical data. It’s purpose is to provide an understanding of what happened during a specific time period and to identify strengths and weaknesses in performance.
Predictive analytics
Predictive analytics is about using current and historical data to forecast what might happen in the future. It applies statistical techniques to data in order to identify patterns so that it can estimate potential future outcomes.
Prescriptive analytics
Prescriptive analytics is the ability to recommend actionable decisions. It uses the potential future outcomes from predictive analytics and delves into the what and why of those outcomes. Models of future scenarios can be created to show the potential upside or downside of decisions.
Artificial intelligence
Weak or narrow AI focuses on simple or single tasks. Apps such as Siri and Alexa use this. They used programmed responses and are only simulations of human behaviour. Strong AI or Artificial General Intelligence (AGI) doesn’t simulate but actually has human cognition. It can learn, reason and make judgements.
Machine learning
Machine learning is an implementation of AI that allows machines to learn and automatically improve from experience without human intervention. Applications of it can be seen in image recognition, face recognition, email spam filters and product recommendations on Amazon and Netflix.
While it would be impossible to discuss all the terms in this piece – the aim has been to uncover some of the mystery and address the above questions for some of these terms.
If you want to learn more about any of these concepts get in touch with us – and we can either take you through it or direct you to further resources to deepen your knowledge.