THE SECRETS OF DATA MINING FOR YOUR MARKETING STRATEGY
1) CLUSTER ANALYSIS TO IDENTIFY SINGLE TARGET GROUPS
Cluster analysis enables identifying a given user group according to common features within a database. These features can include age, geographic location, education level and so on. It is a data mining technique that is useful in marketing to segment the database and, for example, send a promotion to the right target for that product or service (young people, mothers, pensioners, etc.). The variable combinations are endless and make cluster analysis more or less selective according to the search requirements.
Resources:
– Cluster Analysis by Jmp [ENG]
– La Cluster Analysis [PDF ITA]
– Tutorial Cluster Analysis by Origine Lab [ENG]
– Data Mining – Cluster Analysis by Tutorials Point
– Cluster Analysis for Market Segmentation [SLIDE ENG]
2) REGRESSION ANALYSIS TO MAKE MARKETING FORECASTS
To be able to tell the future is the dream of any marketing professional. So without having to resort to a crystal ball, we have a data mining technique in our regression analysis that enables us to study changes, habits, customer satisfaction levels and other factors linked to criteria such as advertising campaign budget, or similar costs. When one of these criterion is changed you will have a pretty clear idea of what will happen to your user basin.
Resources:
– Regression Analysis – predicting the future by Michael Pawlicki [ENG]
– Regression Analysis – by b2binternational [ENG]
– TECHNIQUE #9: Regression Analysis by Marketing Profs [ENG]
– The Use Of Regression Analysis In Marketing Research by IDEAS [ENG]
3) CLASSIFICATION ANALYSIS TO IDENTIFY SPAMS AND MORE BESIDES
How can you classify an email reply from a customer? And how can you recognize any links between potential customers for your products before and after running an advertising campaign? There is just one answer: classification analysis, the data mining technique that enables recognizing the patterns (recurring schemes) inside a database. An effective solution to improve your marketing strategy performance, to delete any superfluous information and to create improved subarchives.
Resources:
– Classification Analysis by Berkley [ENG]
– Principal Components & Classification Analysis by Statistica [ENG]
– Data mining: classification and analysis [ENG]
4) ANOMALY DETECTION TO RECOGNIZE ANY ABNORMALITIES
Every day each business, large or small, has to handle the consequences of any mistakes that are made by employees, suppliers or even customers. A simple mistake in data entry or product purchase is as bothersome as a stone in your shoe. Not life threatening, but very uncomfortable. To eliminate any database inconsistencies or anomalies at source, a special data mining technique is used called anomaly detection. Again, our software will handle the search as it is programmed to perform complex operations in databases containing up to thousands of records (addresses, names, etc.).
Resources:
– Machine Learning Anomaly Detection Service by Microsoft [ENG]
– Survey on Anomaly Detection using Data Mining Technique by ScienceDirect [ENG]
– Outlier and Anomaly Detection by KDD Topics
– 6 Benefits Of Anomaly Detection Software For IT Ops/APM by Prelert
5) INTRUSION DETECTION FOR GREATER SYSTEM SECURITY
Marketing and security are two aspects that may not seem related, but they go (or should go) hand in hand. Imagine the disastrous effects of a DEM (Direct email marketing) campaign conducted using a contaminated database. To avoid using databases infected by intruders (individual values added by hackers, or even viruses that duplicate the data) it is sufficient to search for the intruders, a data mining technique that decontaminates the database and guarantees greater security for the entire system.
Resources:
– Data mining for network security and intrusion detection by R-bloggers [ENG]
– Data Mining Approaches for Intrusion Detection by Wenke Lee and Salvatore J. Stolfo [ENG]
– Effective approach toward Intrusion Detection System using data mining technique by Science Direct [ENG]
6) ASSOCIATION RULE LEARNING TO DISCOVER LINKS BETWEEN DATA
Association rule learning is used for all product sale activities, especially when large volumes are concerned. Online using ecommerce or personally in a shop or mall, we may have to create interesting relations between the data we have available. Relations that you may not even have suspected or imagined. An example? 90% of customers who buy a product online then by another, and always the same one. Details that enable us to create pinpointed marketing proposals, special promotions and winning formulas.
Resources:
– Association Rule Learning – Wikipedia [ENG]
– Association Rule Mining – Not Your Typical Data Science Algorithm [ENG]
– Association rules (in data mining) by Search Business [ENG]
– Association Rule Learning and the Apriori Algorithm by R-Bloggers [ENG]
7) DECISION TREES TO OPTIMIZE PROJECT RISK MANAGEMENT
Every time you make a decision you are facing a crossroads. When there are a lot of options, the crossroads become a decision tree. Initially it may be confusing having to handle a decision tree, but if we have the right computer tool that organizes the tree and submits definitive choices complete with costs/benefits, then it is a different story and the tree becomes a valuable tool for Project Risk Management. Once again, the extent of the analysis mainly depends on the available technology: the more advanced the software the better your tree will indicate the best path to follow.
Resources:
– Using a Decision Trees Example in Project Risk by Bright Hubpm [ENG]
– Decision Tree (CART) – Retail Case Study Example (Part 5) by YOU CANanalytics [ENG]
– Decision Trees for Decision Making by Harvard Business Review [ENG]
– Decision Trees Choosing by Projecting “Expected Outcomes by Mind Tools [ENG]
8) NEURAL NETWORKS TO AUTOMATE LEARNING
To complement clustering and decision trees is the neural network concept. It is one of the latest data mining applications whereby the means you use for marketing operations, i.e. the computer managing your database, “learns” to identify a certain pattern containing elements with precise relationships with each other. The outcome of this learning is the recognition and storing of patterns that will be useful, perhaps not immediately, but in the future to decide whether and how to pursue a goal. The same neural network can also help to recognize the composition of the product or service target more precisely.
Resources:
– Neural Network Analysis by Ecommerce Digest [ENG]
– Neural networks – are you ready for the rise of the machines? By Beyond [ENG]
– Expert Systems with Applications by Semantic Scholar [ENG]
9) INDUCTION RULE FOR PREDICTIVE DATA BASED ANALYSIS
If a given circumstance occurs, then another and another again, we have this result. That is basically how the induction rule works. Which is quite something: with this data mining technique you can process sophisticated predictive analyses inside your database with thousands and thousands of records of order numbers. To be able to identify concealed recurrence means saving time and acting informed, something your competitors often forget how to do. Am I right?
Resources:
– Rule Induction Method by DMS.IRB [ENG]
– Rule Induction by Semantic Scholar [ENG]
– An Experimental Study of Using Rule Induction Algorithm in Combiner Multiple Classifier by IJCIR [PDF ENG]
10) DATA WAREHOUSING FOR BIG DATA PROCESSING
The last, essential data mining technique, or should I say application, is data warehousing. We are now in the sphere of customer (and not only) profiling, especially regarding Big Data processing. To choose software such as Egon for your data warehousing means simplifying your database, extracting the most interesting data about your customers, simplifying the creation of detailed reports and much more besides. When you have to migrate programs and systems, being able to count on data warehousing software is even more important, not just for marketing but for the evolution of the business itself. Trying is believing!
Resources:
– A Data Warehouse [YOUTUBE ENG]
– What’s the difference between data mining and data warehousin? By Programmer Interview [ENG]
