1.  BA_EM Electronic Marketing 22. 10. 2013 – Pavel Kotyza @VŠFS

2. Agenda  Effective data mining as a source of relevant data about customer needs

3. What is data mining?  Absolut  Unknown  Useful

4. What is data?

5. Data-mining traditional uses

10. Question  Say an example of data mining?

11. My example

12. What is data mining?  Data mining is the practice of automatically searching large stores of data to discover patterns and trends that go beyond simple analysis.  Data mining uses sophisticated mathematical algorithms to segment the data and evaluate the probability of future events.  Data mining is also known as  Knowledge Discovery (KD) in Data (KDD).

13. The key properties of data mining are  Automatic discovery of patterns  Prediction of likely outcomes  Creation of actionable information  Focus on large data sets and databases

14. Video http://www.youtube.com/watch?v=BjznLJcgSFI

15. Why to use  Data mining can answer questions that cannot be addressed through simple query and reporting techniques.

16. Video example  http://www.ted.com/playlists/56/making_sense_of_too_much_data.html http://www.ted.com/playlists/56/making_sense_of_too_much_data.html

17.  Data Mining types

18. Automatic Discovery  Data mining is accomplished by building models. A model uses an algorithm to act on a set of data. The notion of automatic discovery refers to the execution of data mining models.  Data mining models can be used to mine the data on which they are built, but most types of models are generalizable to new data. The process of applying a model to new data is known as scoring.

19. Prediction  Many forms of data mining are predictive.  E.g. A model might predict income based on education  Predictions have an associated probability (How likely is this prediction to be true?). Prediction probabilities are also known as confidence  How confident can I be of this prediction?  Some forms of predictive data mining generate rules, which are conditions that imply a given outcome.  E.g. A rule might specify that a person who has a bachelor's degree and lives in a certain neighborhood is likely to have an income greater than the regional average. Rules have an associated support  What percentage of the population satisfies the rule?

20. Grouping  Other forms of data mining identify natural groupings in the data.  E.g. A model might identify the segment of the population that has an income within a specified range, that has a good driving record, and that leases a new car on a yearly basis.

21. Actionable Information  Data mining can derive actionable information from large volumes of data.  For example, a town planner might use a model that predicts income based on demographics to develop a plan for low-income housing.  A car leasing agency might a use model that identifies customer segments to design a promotion targeting high-value customers.

22. Why is it important now  Data all around us  Social networks  Search in e-shops  Targeting Advertising  Information overload

23. Social Insight & Personal Advantage  Rent prices  Blogs and News  Movie data  Fashion  Product Prices  Hotties / Adult content categories

24. The beauty of data visualization http://www.ted.com/talks/david_mccandless_the_beauty_of_data_visualization.html

25. Data Mining Process Problem Definition Data Gathering & Preparation Data Access Data Sampling Data Transformation Model Building & Evaluation Create Model Test Model Evaluate & Interpret Model Knowledge Deployment Model Apply Custom Reports External Applicazions

26. Problem definition

29. Data Gathering & Preparation  Data Access  Data Sampling  Data Transformation

30. Model Building & Evaluation  Create Model  Test Model  Evaluate & Interpret Model

31. Knowledge Deployment  Model Apply  Custom Reports  External Applications

32. How predictable are you? http://www.youtube.com/watch?v=DaWcL3oOd-E

33. The End!  Are there in your company/school any assholes? Solution: of the problem:  D-Fenz Tie Test - Extreme example