C2. Business problems and data science solutions

C2. Business problems and data science solutions

1. Data mining process

   • Its task is to find patterns from data through a set of subtasks.
   • 1. Business understanding
     • 2. Data preparation
       • 3. Modelling
         • 4. Evaluation
           • 5. Deployment

2. Data mining tasks

   • Classification and class probability estimation - classify individual among a small set of classes
     • "Among all the customers of MegaTelCo, which are likely to respond to a given offer?”
   • Regression - predict numerical value (numeric target)
     • “How much will a given customer use the service?”
   • Similarity matching - identify similar individuals based on data known
     • finding companies similar to their best business customers
   • Clustering - group individuals by similarity
     • Other people similar to you bought this
   • Co-ocurrence grouping - co-ocurrence
     • Items commonly purchased together
   • Profiling - behavior description
     • Typical cell phone usage of this customer
   • Link (connection) prediction
     • Fb friends suggestions
   • Data reduction - replace large with small dataset
     • Viewer genre preferences
   • Causal modeling - explain what events influence other events
     • advertisements to consumers that buy more

3. Other analytics techniques and Technologies

   • Statistics
   • Database querying
     • Difference with data mining: there is no discovery of patterns or models
   • Data Warehousing
   • Regression analysis
   • Machine learning

4. A critical skill in data science is the ability to decompose a data-analytics problem into pieces such that each piece matches a known task for which tools are available.

5. Datamining problem

   • Supervised - target
     • Must have data on the target
   • Unsupervised - no target