Content Data Mining

At data extractions we can uncover patterns in your data using anomaly detection, association rules, clustering and comparative analysis. Our data mining techniques can discover what’s really in your data and use that to help you make better business decisions.

Data mining is a relatively young field that involves the process of discovering new patterns from data sets.  At data extractions we can help with your data mining needs.

  • Prospecting/Prediction – making calculated guesses about the information.   
  • Anomaly detection – The identification of unusual data records, that might be interesting or data errors and require further investigation.   
  • Association rules – Dependency modeling that searches for relationships between variables. ex. “Customers who bought this also bought…”
  • Clustering – discovering groups and structures in the data that are in some way or another similar
  • Classification – generalizing known structure to apply to new data.   
  • Regression– Attempts to find a function which models the data with the least error.

Contact Us for your data mining needs.

An example of some of our past projects:

Case 1 – Content Catalogs:

Our customer had multiple content product catalogs and wanted to make sure that their master content catalog had all records from both catalogs and had the best “version” of that content in the master catalog.  At first glance we realized that neither catalog had a unique key to which we could compare the two catalogs.  After we prepped and scrubbed the data, we identified an algorithm that would help us match unique products from both catalogs.  Once we matched both catalogs, we found the products that were in catalog 1 and not in catalog 2 and vice-versa.  Those products made it into the master catalog.  Our second process included developing an algorithm to help determine which catalog would we choose the overlapping products from (products that matched in catalog 1 and catalog 2).  Once we applied the algorithm to the matched products, we extracted unique products from that source and inserted those into the master.

Case 2 – Web Analytics:

Our client owned a small e-commerce website that had a little over 6000 unique products on it.  Obviously, they were able to tell which products sold well and which products customers most frequently added to their cart.  What they didn’t know was how often people looked at those products.  We were able to harness the data from their web analytics application and merge it with their cart and ordering system data.  From there, we were able to glean a number of insights including: poor product content variations, poor cart conversion and poor checkout conversion.  Once these issues were corrected, the customer noticed an immediate increase in sales and visitors.

Case 3 – The Merger:

Our customer had recently acquired a new company that was a former competitor of theirs.  They wanted to merge both of their customer data sets into a single source without duplicates and merge prospective customer data together in a similar fashion.  Through the use of our data mining practices, we were able to merge both sets together quickly to give our client the comprehensive customer and prospect list that they were looking for.


Contact Us for your data mining needs.