Download Magic: the Gathering: of Data Data warehousing: Magic Deck Data

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Cluster analysis wikipedia , lookup

Nonlinear dimensionality reduction wikipedia , lookup

Transcript
Magic: the Gathering: of Data Data warehousing: Magic Deck Data Mart Data mining: Magic Deck Archetype Clustering Team Members: Christopher Lawson, Jonathan Tinney, Irwin Yeung Objectives of your proposed projects: In terms of data warehousing, we seek to provide a homogeneous interface to different data sources composed of Magic tournament results . In terms of data mining, our goal is to cluster the unstructured data into specific archetypes, to assist in the creation of decks. We intend to refine upon the methods from our research portion of the project and hopefully obtain a new kernel of strategic information from the clustering of Magic decks types. Any deviation from original project proposal: Thus far, the scope of the project seem appropriate for this class. Our changes are not in a different direction but merely a refinement on the actual execution. Project design done so far: We have gathered the data from numerous credible sources. We have completed an initial schema for our data warehouse but have not yet populated the warehouse or completely nailed down our warehouse design. Performance evaluation methods to be used: Our tables will be normalized with a junction table relating cards and deck information, and this could impact algorithm evaluation performance on Weka as multiple joins increase overhead. Working schedule for remaining tasks: Week 12: Finish population of the data warehouse and develop Data Mart ETL script Week 13: Data Mining with assorted functions in Weka Week 14 | 15: Summary and Presentation of results Initial fact table and schema configuration: This is currently the star schema that will be used in the creation of our data warehouse. We will be using the tournament information provided from the urls listed below in order to populate the database with tournament results from what is currently considered “standard” in magic. With this information, we hope to form a basis for the data warehousing portion of our project. Dataset/algorithms to be used for the project: Sources: http://www.mtgdecks.net/ and http://www.mtgtop8.com/ We will be using data from the “Standard” tournament format with a date range within October.