DOC PREVIEW
UMD CMSC 828G - Lecture 26 Principles of Data Mining

This preview shows page 1-2-3 out of 10 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 10 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

CMSC828G Principles of Data Mining Lecture #26• Today’s Lecture:– What we covered– What we didn’t cover– What next? •Due Today:– Final Project Reports– Course Evaluations• Final Exam: Saturday May 18, 10:30AM –12:30PMData Mining Process• Determining the nature and structure of the representation to be used;• Deciding how to quantify and compare how well different representations fit the data (score function)• Choosing an algorithmic process to optimize the score function; and• Deciding what principles of data management are required to implement the algorithms efficiently.Data Mining Tasks• Exploratory Data Analysis (EDA)– goal: explore data, without clear idea of what we are looking for– techniques: interactive and visual– as p increases, harder to visualize higher dimensions• project, in some intelligent way, onto lower dimensional space– as n increases, harder to visualize large number of individuals• abstract, in some intelligent way, into collectionsData Mining Tasks cont.• Descriptive Modeling– goal: describe all of the data or the process that generated the data – techniques: density estimation, cluster analysis (partition-based, hierarchical, probabilistic model-based) – algorithms: Bayesian networks, k-means, hierarchical agglomerative clustering, mixture modelsData Mining Tasks cont.• Predictive Modeling, classification and regression– goal: build model that will predict the value of one variable from known values of other variables – classification the variable is categorical– regression the variable is quantitative– techniques: decision trees, nearest neighbor, naïve bayes, linear regression, neural networks, SVMs ….Data Mining Tasks cont.• Discovering Patterns and Rules – goal: detect patterns – spot fraudulent behavior, detect unusual stars, find items that occur frequently together in transaction databases– techniques: association rules– algorithms: apriori, FP trees, …Component View• Data mining algorithm components:–task– structure or model– score function– search or optimization method– data management technique3 Algorithms linear scansunspecifiedunspecifiedData Management Techniquebreadth-first with pruninggradient descent on parametersgreedy search over structureSearch Methodsupport/accuracy squared errorcross-validated loss functionScore Functionassociation rulesneural networkdecision treeStructurerule pattern discoveryregressionclassification and regressionTaskA PrioriBackpropagationCARTWhat we didn’t cover, /• Causal Models• Privacy, Security issues• Spatial and Temporal data mining• Visual data mining, Image-based mining• Graph miningWhat next?• In next few weeks, concentrated graph mining reading group. Please email me if you are interested.• In the fall, data mining reading group. Please email me if you are interested.• Many related courses: Databases, Information Retrieval, Semantic Web. With high probability, I will teach machine learning in Spring 2003.• Thanks, and good luck on the final


View Full Document

UMD CMSC 828G - Lecture 26 Principles of Data Mining

Documents in this Course
Lecture 2

Lecture 2

35 pages

Load more
Download Lecture 26 Principles of Data Mining
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Lecture 26 Principles of Data Mining and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Lecture 26 Principles of Data Mining 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?