Question: What is data mining?
Answer: Data mining tools are used to examine large, complex databases. The steps of data mining are:
- Develop an understanding of the application
- Create a target data set
- Clean and preprocess the data (missing data, time series, transpose, concatenating where necessary)
- Reduce the number of variables and find invariant representations of data if possible
- Choose the data mining task and algorithm(s)
- Search for patterns of interest
- Interpret the pattern mined
- Consolidate the knowledge discovered
Question: How can data mining help my business?
Answer: A business has to collect considerable amounts of data about clients, purchases, and vendors. Data mining can turn that data into business intelligence. Data mining finds patterns and relationships in the data that are not obvious.
Question: What are examples of data mining?
Answer: Many retail outlets are now using shopper’s cards to track purchases. Then a market basket analysis is performed to examine combinations of items that are purchases by the same customer. A grocery store or other retail store uses that information to position items in the store. It is also used to customize advertisements. Online stores also use data mining for similar purposes.
The communications industry was one of the first to use data mining techniques to track connections with the purpose of predicting future use, and the need for equipment installation.
Another example of data mining is to use it to predict the likelihood of a customer to “churn”; that is, to change vendors. Once predicted, the business can intervene to prevent the churn.
Question: What kind of data can be analyzed using data mining?
Answer: There are different data mining tools that can be used with just about any kind of data. One of the more recent, and interesting, tools, is text analysis. Text analysis can examine competitor ads-and can automatically extract ads from web documents.