Download Mathew Rogers - Marine biotoxin database

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

Nonlinear dimensionality reduction wikipedia , lookup

Transcript
Marine biotoxin database data mining from large
geo-spatial data sets
Mathew Rogers
Senior Advisor – Food Assurance Programmes
MAF
What is MAF trying to do for you?
Getting the biotoxin data into a relational format
Ensure that historical data and any future data is “clean”,
accurate and secured
Provide industry with easier access to the data and in a way
which is more flexible and user driven
What do we need to achieve that?
A data source (or data warehouse) with “clean” data
• MAF’s Sample Attribute Management Database (SAMD)
A Business Intelligence (BI) tool
• Due to be implemented mid 2012
What is SAMD?
Sample Attribute Management Database (SAMD)
MAF’s relational database for the collection and storage of food sampling
attribute data.
Contains most of MAF’s food monitoring programmes
Common sample attributes are shared (i.e. have a consistent format
across the entire database)
Monitoring Programme specific attributes included
SAMD – The shellfish page
What is Business Intelligence?
Business Intelligence (BI)
Provides a user interface to data sets
Produces quick visual data summaries and customised reports
Allows users to query data sets
Business Intelligence Reports
Business Intelligence Reports
Business Intelligence Reports
What is Data Mining?
Data Mining
Data mining is the discovery of trends or patterns from large
data sets.
Analysing what has happened and what is likely to happen in
the future
It involves looking at data in new ways or from a different
perspective.
What are the tools needed for data mining?
Data mining tools
The tools we already have or planning to develop, namely:
• Data sets (i.e. SAMD)
• A Business Intelligence (BI) tool
How do you conduct data mining?
Open your mind
We have the ability to ask questions of the data
But the power is in combining this data with other data sets linked by a
common attribute (such as date, geo-codes etc.)
What can be found?
Information that will support effective and robust sampling plans and data
models
Real time information that can be used to efficiently manage farms and
harvesting
So where are we up to?
SAMD has been configured for the biotoxin data
The historical data up to early 2009 has been reviewed and
“cleaned”
Development work has started to investigate reporting and data
display options from the BI tool
So in summary
How is MAF storing the biotoxin data and ensuring it is clean, accurate and
protected?
• By using MAF’s Sample Attribute Management database (SAMD)
How are we going to provide user friendly access to the data?
• Through a suitable BI tool
What could be the potential benefits of all this work and data mining be?
• Greater efficiencies and smart decision making
Final thoughts
Good policy comes from good data
Good practice comes from good data
Good data leads to good information
Good information allows industry to operate efficiently