Data mining finds patterns in huge amounts of data, called big data, usually stored in a data warehouse. Collecting big data involves gathering as much data as possible, which data mining uses to find hidden meanings.
Why Use Data Mining?
Data mining has a range of uses. You could find out customer behaviour and preferences, tailor products to specific needs and make predictions about sales or revenue. However, you need to be aware of your privacy obligations, which a privacy lawyer can advise you on.
Data mining is used by streaming services to recommend movies, which predict what movie you’d be interested in based on:
- your location;
- movies you’ve seen;
- genres you like and dislike;
- movies other customers like; and
- movies other customers watched after the ones you did.
By determining your behaviour, preferences and other features, streaming services can make their platform customer-friendly and more convenient. Their revenue then increases as a result of keeping you on their site longer.
How Does Data Mining Work?
Once you collect and store your data, it needs to be ‘cleaned’ – often by converting it to numbers and filling in missing values – so that you can analyse it. Data analytics refers to the tools you use to find patterns, such as:
- Business Intelligence (BI), a simple interface for analysing and visualising data;
- Machine Learning (ML), which can make predictions and find common customer characteristics; and
- Artificial Intelligence (AI), imitating our brain’s neural pathways to create ML with human-like intelligence.
AI can take context into account and solve more complex problems than ML. For example, ML can find the meaning of words while AI can find the meaning of sentences (and much more) by ‘understanding’ human speech.
After you analyse your data for patterns, you can visualise it to make it easy to understand. This is where BI or other visualisation programs tend to come in. You can display your data and draw conclusions from it.
When Can It Go Wrong?
A large concern of data mining is that it can reveal previously unknown personal information. Target uses ML for advertising purposes based on predicting features of customers, which led to controversy in 2012. A father approached a Target manager and voiced his concerns about maternity product coupons sent to his daughter. He later apologised to the manager after he found out she was, in fact, pregnant.
With the upcoming Consumer Data Right laws, businesses are under more pressure to meet legal and ethical obligations. A best practice approach to ensure you meet these obligations includes:
- using de-identified data where possible;
- taking a privacy-by-design approach;
- conducting a Privacy Impact Assessment;
- being open and transparent about your practices;
- knowing what you’re collecting;
- handling sensitive information with care;
- putting up clear notices; and
- taking steps to minimise risks.
Data mining can benefit your business by revealing more about your customers, products or business. And there are several ways to find these patterns available to you. But you need to consider the ethics of using data and finding patterns in this way.
Unsure where to start? Contact a LawPath consultant on 1800 529 728 to learn more about customising legal documents and obtaining a fixed-fee quote from Australia’s largest legal marketplace.
