Assignment 1: Introduction
Miles A. Cabanos
Data Mining - 10303 CAP4770

Q1: Present an example where data mining is crucial to the success of a business. What data mining function does this business need? Can they be performed alternatively by data query processing or simple statistical analysis?
Best example that I could think of would be Amazon.com. I think one of the functions the business uses is the characterization. This way it can keep track of what type of products customers buy, and then the pop-up windows to suggest similar items. No, I do not think they could be performed by data query processing or simple statistical analysis.
Q2: Define each of the following data mining functionalities: characterization, discrimination, association and correlation analysis, classification, prediction and clustering. Give examples of each data mining functionality, using a real-life database with which you are familiar.
 Characterization: Given data that shares similarities of characteristics and/or other requested specific information. (Example: Data from Amazon.com customers buying superhero books, can then be used to determine what age group buys superhero books. Then can suggest other types of similar books)
 Discrimination: Compares or contrasts data information. (Example: Amazon could find out what customers types buy more superhero book compared to biographies)
 Association and correlation analysis: How certain types of data can be associated with each other. Patterns, relationships, or correlations. (Example: Amazon could also find out what types of items sell together like people who buy cooking with the George Forman grill also buy a George Foreman grill.
 Classification: Data that is can be used to create certain labels for certain classes with certain distinctions or that maybe unknown. (Example: Amazon could create classes of customer of age groups, sex, etc. that buy action or science fiction books)
 Prediction: Data gathered to guess the future of certain outcomes. (Example: Amazon collects data from the past two years of Christmas sales of certain item. They can try to predict sales, amount of items to have on hand in stock.)
 Clustering: Data that may not yet have classifications or labels, raw data gathered together that may share similarities or patterns. (Example: Amazons new data that may not yet be classed, but can be found through clusters of similar attributes.)
Q3: What are the major challenges of mining a huge amount of data (such as billions of tuples) in comparison with mining a small amount of data (such as a few hundred tuple data set)?
It seems that if I understand correctly it would be efficiency and scalability algorithms would have to be efficient, while at the same time be quick enough to be usable. It seems like that is one of the issues of huge amounts of data. Sometimes would have to be mined in partial pieces. Dealing with smaller amounts of data, algorithms are more easily configured to run efficiently for quicker performance.