DOC PREVIEW
UIC CS 583 - Chapter 2 Data Preprocessing

This preview shows page 1-2-15-16-17-32-33 out of 33 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 33 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 33 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 33 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 33 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 33 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 33 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 33 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 33 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Chapter 2 Data PreprocessingData Types and FormsChapter 2: Data PreprocessingWhy Data Preprocessing?Why Is Data Preprocessing Important?Multi-Dimensional Measure of Data QualityMajor Tasks in Data PreprocessingSlide 8Data CleaningMissing DataHow to Handle Missing Data?Noisy DataHow to Handle Noisy Data?Binning Methods for Data SmoothingOutlier RemovalSlide 16Data IntegrationData TransformationData Transformation: NormalizationSlide 20Data Reduction StrategiesDimensionality ReductionHistogramsClusteringSamplingSlide 26Slide 27DiscretizationDiscretization and Concept HierarchyBinningEntropy-based (1)Entropy-based (2)SummaryUIC - CS 594 1Chapter 2Data PreprocessingUIC - CS 594 2Data Types and FormsAttribute-value data:Data typesnumeric, categorical (see the hierarchy for its relationship) static, dynamic (temporal)Other kinds of datadistributed datatext, Web, meta dataimages, audio/videoUIC - CS 594 3Chapter 2: Data PreprocessingWhy preprocess the data?Data cleaning Data integration and transformationData reductionDiscretizationSummaryUIC - CS 594 4Why Data Preprocessing?Data in the real world is dirtyincomplete: missing attribute values, lack of certain attributes of interest, or containing only aggregate datae.g., occupation=“”noisy: containing errors or outlierse.g., Salary=“-10”inconsistent: containing discrepancies in codes or namese.g., Age=“42” Birthday=“03/07/1997”e.g., Was rating “1,2,3”, now rating “A, B, C”e.g., discrepancy between duplicate recordsUIC - CS 594 5Why Is Data Preprocessing Important?No quality data, no quality mining results!Quality decisions must be based on quality datae.g., duplicate or missing data may cause incorrect or even misleading statistics.Data preparation, cleaning, and transformation comprises the majority of the work in a data mining application (90%).UIC - CS 594 6Multi-Dimensional Measure of Data QualityA well-accepted multi-dimensional view:AccuracyCompletenessConsistencyTimelinessBelievabilityValue addedInterpretabilityAccessibilityUIC - CS 594 7Major Tasks in Data PreprocessingData cleaningFill in missing values, smooth noisy data, identify or remove outliers and noisy data, and resolve inconsistenciesData integrationIntegration of multiple databases, or filesData transformationNormalization and aggregationData reductionObtains reduced representation in volume but produces the same or similar analytical resultsData discretization (for numerical data)UIC - CS 594 8Chapter 2: Data PreprocessingWhy preprocess the data?Data cleaning Data integration and transformationData reductionDiscretizationSummaryUIC - CS 594 9Data CleaningImportance“Data cleaning is the number one problem in data warehousing”Data cleaning tasksFill in missing valuesIdentify outliers and smooth out noisy data Correct inconsistent dataResolve redundancy caused by data integrationUIC - CS 594 10Missing DataData is not always availableE.g., many tuples have no recorded values for several attributes, such as customer income in sales dataMissing data may be due to equipment malfunctioninconsistent with other recorded data and thus deleteddata not entered due to misunderstandingcertain data may not be considered important at the time of entrynot register history or changes of the dataUIC - CS 594 11How to Handle Missing Data?Ignore the tuple Fill in missing values manually: tedious + infeasible?Fill in it automatically witha global constant : e.g., “unknown”, a new class?! the attribute meanthe most probable value: inference-based such as Bayesian formula, decision tree, or EM algorithmUIC - CS 594 12Noisy DataNoise: random error or variance in a measured variable.Incorrect attribute values may due tofaulty data collection instrumentsdata entry problemsdata transmission problemsetcOther data problems which requires data cleaningduplicate records, incomplete data, inconsistent dataUIC - CS 594 13How to Handle Noisy Data?Binning method:first sort data and partition into (equi-depth) binsthen one can smooth by bin means, smooth by bin median, smooth by bin boundaries, etc.Clusteringdetect and remove outliersCombined computer and human inspectiondetect suspicious values and check by human (e.g., deal with possible outliers)UIC - CS 594 14Binning Methods for Data SmoothingSorted data for price (in dollars): 4, 8, 9, 15, 21, 21, 24, 25, 26, 28, 29, 34Partition into (equi-depth) bins:Bin 1: 4, 8, 9, 15Bin 2: 21, 21, 24, 25Bin 3: 26, 28, 29, 34Smoothing by bin means:Bin 1: 9, 9, 9, 9Bin 2: 23, 23, 23, 23Bin 3: 29, 29, 29, 29Smoothing by bin boundaries:Bin 1: 4, 4, 4, 15Bin 2: 21, 21, 25, 25Bin 3: 26, 26, 26, 34UIC - CS 594 15Outlier RemovalData points inconsistent with the majority of dataDifferent outliersValid: CEO’s salary, Noisy: One’s age = 200, widely deviated pointsRemoval methodsClusteringCurve-fittingHypothesis-testing with a given modelUIC - CS 594 16Chapter 2: Data PreprocessingWhy preprocess the data?Data cleaning Data integration and transformationData reductionDiscretizationSummaryUIC - CS 594 17Data IntegrationData integration: combines data from multiple sourcesSchema integrationintegrate metadata from different sourcesEntity identification problem: identify real world entities from multiple data sources, e.g., A.cust-id  B.cust-#Detecting and resolving data value conflictsfor the same real world entity, attribute values from different sources are different, e.g., different scales, metric vs. British unitsRemoving duplicates and redundant dataUIC - CS 594 18Data TransformationSmoothing: remove noise from dataNormalization: scaled to fall within a small, specified rangeAttribute/feature constructionNew attributes constructed from the given onesAggregation: summarizationGeneralization: concept hierarchy climbingUIC - CS 594 19Data Transformation: Normalizationmin-max normalizationz-score normalizationnormalization by decimal scalingAAAAAAminn ewminn ewmaxn ewminmaxminvv _)__(' AAdevstandmeanvv_'jvv10'Where j is the smallest integer such that Max(| |)<1'vUIC - CS 594 20Chapter


View Full Document

UIC CS 583 - Chapter 2 Data Preprocessing

Download Chapter 2 Data Preprocessing
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 Chapter 2 Data Preprocessing 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 Chapter 2 Data Preprocessing 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?