A
- Abnormal conditions:
- direction of
- “good” vs. “bad”
- variable set for detecting
- Accuracy (DQ dimension)
- Adjoint of a square matrix
- AlliedSignal
- Analysis(-es):
- association
- correlation
- current state
- DQ capabilities gap
- drill-down
- measurement system
- multiple regression
- network
- Pareto
- principal component
- regression
- return on investment
- root-cause
- signal-to-noise ratios
- Analysis of variance (ANOVA)
- in data tracing
- defined
- elements of
- nested
- two-way
- Analytical insights, in DQPC
- Analytics ascendency model (Gartner)
- Analytics management. See also Data analytics
- Analytics vision
- ANOVA, see Analysis of variance
- Artificial neural networks (ANNs)
- Assessment(s):
- of critical data elements
- of current state
- in DAIC Assess phase
- of data quality
- skilled resources for
- Assess phase (DAIC approach)
- Assignable causes. See also Special cause variation
- Association analysis:
- defined
- for discrete CDEs
- in transaction pattern recognition
- Attribute control charts
- Attribute data
- Attribute profiling
- Average chart
- Average-range (-R) charts
- Average-standard deviation (-S) chart
B
- Backpropagation algorithm, standard
- Backpropagation (feed-forward) method
- Bad data, see Poor-quality data
- Bank of America
- Basel II case study
- CDE rationalization matrix
- correlation and regression analysis
- DQ rules
- signal-to-noise ratios analysis
- Baseline:
- for critical data
- for improvement
- BIDVs (business impacting decision variables)
- Big data:
- Big data analytics
- artificial neural networks
- discrimination and classification method
- examples of projects
- feed-forward (backpropagation) method
- Fisher's discriminant function
- multiple process control
- multiple regression analysis
- operating model for
- principal component analysis
- standard backpropagation algorithm
- stepwise regression
- test of additional information (Rao's test)
- types of
- using Mahalanobis distance
- Big data analytics operating model
- Bramson, Brian, 13n.1, 23n.1
- Business/function-level CDE rationalization matrix
- Business impacting decision variables (BIDVs)
- Business rules, see Data quality (DQ) rules
- Business specification
- Business use, defining
C
- Capabilities gap analysis report
- Cause-related analytics
- c-charts
- CDEs, see Critical data elements
- CDE rationalization matrix
- Basel II case study
- business/function-level
- enterprise
- CDO, see Chief data office
- Central limit theorem
- Chance causes. See also Common cause variation
- Change management processes, formalizing
- Charter:
- Chief data office (CDO):
- components of
- and other functions
- Client experience improvement case study
- Cofactors (matrices)
- Common cause variation
- behaviors of
- uncontrollable variation as
- Communications, program
- Completeness (DQ dimension)
- Confidence interval
- Conformity (DQ dimension)
- Consistency and synchronization (DQ dimension)
- Continuous CDEs:
- correlation analysis for
- regression analysis for
- Control:
- in DAIC Control phase
- in data integration
- defining operating guidelines for
- in DQOM
- Control charts. See also Statistical process control (SPC)
- attribute
- average–standard deviation
- to detect small shifts
- for DQ scorecards
- multivariate process
- sample and sample parameters
- types of data
- value of
- variable
- Controllable variation. See also Special cause variation
- Control limits
- and stable vs. unstable process variation
- upper and lower
- Control phase (DAIC approach)
- COPQD (cost of poor-quality data). See also Quantifying impact of DQ
- Correlation analysis:
- in Basel II case study
- for continuous CDEs
- Correlation coefficient
- Correlation matrix
- Cost implications
- Cost of poor-quality data (COPQD). See also Quantifying impact of DQ
- Cost waterfall
- Critical data elements (CDEs)
- assessment of
- baseline for
- CDE rationalization matrix
- continuous, correlation analysis
- continuous, regression analysis for
- defined
- defining
- discrete, association analysis for
- enterprise
- identification of
- impact of
- prioritization of. See also Funnel approach
- quality levels of
- steady state for
- Cumulative sum (CUSUM) control charts
- Current CDE capability
- Current state, documenting
- Current-state analysis
- Current state assessment, in DQOM
- CUSUM (cumulative sum) control charts
D
- DAIC approach, see Define, Improve, Analyze, and Control approach
- Dashboards, determining set of
- Data. See also Big data
- as key resource
- types of
- variable
- Data analytics
- for big data
- defined
- in DQOM
- executing
- skilled resources for
- types of
- Data architecture, global financial crisis and
- Data assessment, in DQOM
- Data collection plan
- Data consumer business processes
- Data decay (DQ dimension)
- Data dictionary
- Data elements. See also Critical data elements (CDEs)
- in data dictionary
- defined
- paths traveled through organization
- Data governance
- Data innovation
- big data in. See also Big data
- defined
- Data integration, in DQOM
- Data lineage, through data tracing
- Data management function:
- chief data office
- increasing importance of
- Data producer business processes
- Data profiling:
- in Assess phase
- in DQOM
- skilled resources for
- types of
- using DQ rules in
- Data quality (DQ)
- as chief data office component
- and data management function
- dimensions of
- linkage between process quality and. See also Issues resolution [IR]
- solution strategy for
- Data quality analyst (DQA)
- in assessment of data quality
- role of
- Data quality issues:
- collecting list of
- compounding of
- implementing management process for
- prioritization matrix for
- resolving, see Issues resolution [IR]
- root-cause analysis of
- sources of
- Data quality operating model (DQOM)
- assessment of current state of DQ in
- current-state analysis in
- data assessment in
- data integration in
- data profiling and analytics in
- discussion on establishing
- DQ projects in
- establishment of program in
- foundational capabilities of
- issues resolution in
- methodology in
- monitoring and control in
- ongoing monitoring and effectiveness measurement in
- skilled resources in
- strategy and governance in
- technology infrastructure and metadata in
- Data quality practices center (DQPC)
- Data quality (DQ) rules
- building/executing
- data profiling using
- specifying
- translating dimensions into
- Data quality scores (DQ scores)
- Data quality (DQ) scorecards
- analytical framework for
- ANOVA
- application of framework
- at CDE level
- creation of
- defined
- determining set of
- development of
- in DQOM
- heat maps
- producing
- SPC charts
- thresholds
- at various levels
- Data standards
- Data strategy:
- in big data analytics operating model
- as data management component
- Data tracing
- case study of
- data lineage through
- defined
- methodology for
- statistical sampling
- Decay, data
- Defaulting customers' behavior patterns case study
- Defects
- Defect-free processes/products
- Defective units
- Define, Improve, Analyze, and Control (DAIC) approach
- Assess phase in
- Control phase in
- Define phase in
- Improve phase in
- in issues resolution
- Six Sigma methodologies
- Define, Measure, Analyze, Design, and Verify (DMADV)
- Define, Measure, Analyze, Improve, and Control (DMAIC) methodology. See also Six Sigma
- Define phase (DAIC approach)
- Degrees of freedom (df)
- Deming, W. Edwards
- Dependency profiling
- Descriptive analytics
- Design for Six Sigma (DFSS) methodology
- Determinant of matrix
- df (degrees of freedom)
- DFSS, see Design for Six Sigma methodology
- Diagnostic analytics
- Dimensions (data quality)
- assessing data quality across
- in DQ assessment
- translating into DQ rules
- Discrete CDEs, association analysis for
- Discrete data
- Discrimination and classification method
- DMADV, see Define, Measure, Analyze, Design, and Verify
- DMAIC (Define, Measure, Analyze, Improve, and Control) methodology. See also Six Sigma
- Documentation, of current state
- DQ, see Data quality
- DQA, see Data quality analyst
- DQOM, see Data quality operating model
- DQPC (data quality practices center)
- DQ projects, in DQOM
- DQ rules, see Data quality rules
- DQ scorecards, see Data quality scorecards
- DQ (data quality) scores
- Drill-down analysis, heat maps for
- Duplicate avoidance (DQ dimension)
- Dynamic signal-to-noise (S/N) ratios
E
- Effectiveness, of DQ improvement
- 80/20 rule
- Enablers, in DQPC
- Enterprise CDEs
- Enterprise CDE rationalization matrix
- Enterprise (data quality) scorecard. See also Data quality (DQ) scorecards
- Error records
- Establishment of DQ program
- EWMA (exponentially weighted moving average) control charts
- Executive sponsor
- Expected variation. See also Common cause variation
- Exponentially weighted moving average (EWMA) control charts
F
- Feed-forward (backpropagation) method
- Finance company case study
- Financial crisis, inadequate information technology and
- Fisher, R. A.
- Fisher's discriminant function
- F-ratio
- Functional variability of process, losses due to
- Funnel approach
- association analysis for discrete CDEs
- Basel II case study
- correlation analysis for continuous CDEs
- regression analysis for continuous CDEs
- signal-to-noise ratios analysis
G
- Gap analysis
- Gartner analytics ascendency model
- Gaussian distribution
- Gear motor assembly case study
- General Electric
- Global financial crisis, inadequate information technology and
- Governance:
- around metadata management
- in big data analytics operating model
- charter for
- data
- in DQPC
- of DQ program
- of metadata
- Gram-Schmidt process (GSP)
- Granese, Bob, 43n.1
H
- Hadoop
- Harmful side effects, losses due to
- Heat maps:
- defined
- for DQ scorecards
I
- IBM
- Improvement:
- commissioning efforts for
- in DAIC Improve phase
- in DQPC
- establishing baseline for
- measuring effectiveness of
- Improve phase (DAIC approach)
- Individual–moving range (X-MR) charts
- Information objects
- Information system testing. See also Mahalanobis-Taguchi Strategy (MTS)
- constructing combination tables
- finance company case study
- method of
- MTS software testing
- orthogonal arrays in
- software company case study
- study of two-factor combinations
- typical arrangement for
- Information technology (IT), global financial crisis and
- Innovation, data, see Data innovation
- Institute of International Finance
- Integration, data
- Integrity (DQ dimension)
- Inverse matrix
- Issues, data quality, see Data quality issues
- Issues resolution (IR)
- DAIC approach for
- in DQOM
- process of
- process quality (Six Sigma) approach
- reengineering of (case study)
- skilled resources for
- IT, global financial crisis and
J
- Jeopardy, Watson on
- Joyce, Ian
- Juran, J. M.
L
- Larger-the-better type S/N ratio
- Levels (of factors)
- Lineage of data
- Linear discriminant function
- Loss due to poor data quality
- categories of
- forms of
- impact of. See also Quantifying impact of DQ
- Loss function
- Loss to society
- Lost opportunities
- Lower control limits
M
- McKinsey & Company
- Mahalanobis, P. C.
- Mahalanobis distance (MD)
- basic calculation of
- in big data analytics
- calculated with Gram-Schmidt process
- purpose of
- Mahalanobis space (MS)
- Mahalanobis-Taguchi Strategy (MTS)
- client experience improvement case study
- defaulting customers' behavior patterns case study
- direction of abnormals in
- gear motor assembly case study
- GSP calculation in
- marketing case study
- medical diagnosis example of
- orthogonal arrays in
- S/N ratios in
- software testing with
- stages in
- variable set for detecting abnormal conditions in
- Management:
- analytics
- in big data analytics operating model
- change
- data
- metadata
- MapReduce
- Marketing case study
- Marketing costs
- Matrix(-ces):
- adjoint of a square matrix
- cofactors
- correlation
- defined
- determinant of
- inverse
- nonsingular
- singular
- square
- transpose of
- Matrix theory
- MD, see Mahalanobis distance
- Mean square (MS)
- Measurement:
- in Control phase (DAIC approach)
- with DQ scorecards
- of effectiveness
- in Six Sigma
- Measurement system analysis. See also Information system testing
- Metadata:
- formalizing change management process for
- means for gathering, managing, and updating
- resources for defining, gathering, and managing
- updating
- Methodology (in DQOM)
- assess current state of DQ
- conducting current-state analysis
- data quality projects
- discussion on establishing DQ program
- establishing DQ program
- monitoring and measurement
- Monitoring:
- in Control phase (DAIC approach)
- defining operating guidelines for
- in DQOM
- in DQPC
- ongoing
- relevance of SPC in
- skilled resources for
- Monitoring and reporting (M&R) function. See also Data quality (DQ) scorecards
- Motorola
- Moving range (MR)
- M&R (monitoring and reporting) function. See also Data quality (DQ) scorecards
- MS (Mahalanobis space)
- MS (mean square)
- MTS, see Mahalanobis-Taguchi Strategy
- Multiple process control
- Multiple regression analysis
- Multivariate process control charts
- Multivariate systems:
- correlations in
- defined
- design and development of, see Mahalanobis-Taguchi Strategy [MTS]
N
- Nested analysis of variance (nested ANOVA)
- Network analysis
- Nominal-the-best type S/N ratio
- Nondynamic signal-to-noise (S/N) ratios
- Nonsingular matrix
- Nonuniformity
- Normal distribution
- Normal group (in MTS)
- NoSQL databases
- np-charts
O
- OAs, see Orthogonal arrays
- Ongoing monitoring:
- control processes for
- in DQOM
- Operating model, see Data quality operating model (DQOM)
- Operations leads
- Orthogonal arrays (OAs)
- in information system testing
- for minimizing number of test combinations
- in MTS
- three-level
- two-level
- Oversampling
P
- Pareto analysis
- CDE prioritization through
- in data tracing
- Pareto diagram
- Pareto principle
- PCA (principal component analysis)
- p-charts
- Percent defective charts
- Plan:
- Poor-quality data:
- cost of
- loss due to
- managing indicators of
- quantifying impact of, see Quantifying impact of DQ
- Predictive analytics
- Preparatory analytics
- Prescriptive analytics
- Principal components
- Principal component analysis (PCA)
- Prioritization matrix:
- CDE, using Pareto analysis
- CDE rationalization matrix
- for data quality issues
- for reengineering IR process
- Prioritization of CDEs, see Funnel approach
- Process CDE potential index
- Process quality:
- linkage between data quality and. See also Issues resolution [IR]
- measuring, see Six Sigma
- Process variation, stable vs. unstable
- Production environment, monitoring
- Profiling, see Data profiling
- Program charter
- Program communications
- Program management professions
- Program strategy
- Project charter
- Project management professions
- Project manager, role of
- Project methodologies. See also individual methodologies
- Project plan, creating
- Project sponsor, role of
- Proportion charts (p-charts)
- p-value
Q
- Quality loss function (QLF)
- Quantifying impact of DQ
- building quantification framework
- cost waterfall
- prioritization matrix
- remediation and ROI
- trading office example of
R
- Random sample approach
- Range chart
- Rao, C. R.
- Rao's test
- Rationalization matrix, CDE, see CDE rationalization matrix
- Rational subgrouping
- Reengineering issues resolution case study
- Reference group (in MTS)
- Regression analysis:
- in Basel II case study
- for continuous CDEs
- multiple regression
- need for
- purpose of
- stepwise regression
- Relationship profiling
- Reliability-based analytics
- Remediation:
- commissioning efforts for
- in quantification framework
- Reports, determining set of
- Reporting:
- capabilities gap analysis report
- monitoring and reporting function. See also Data quality [DQ] scorecards
- relevance of SPC in
- Resources, in DQOM
- Return on investment (ROI):
- in analysis of solutions
- in quantification framework
- Review:
- in issues resolution
- operational-level
- Rework costs
- Risk-weighted assets (RWAs)
- ROI, see Return on investment
- Role clarity
- Root-cause analysis
- Run charts
- RWAs (risk-weighted assets)
S
- Sample:
- Sample size:
- Sampling, statistical, see Statistical sampling
- Sampling frequency
- Sampling parameters
- S-chart
- Scope, defining
- Scorecards. See also Data quality (DQ) scorecards
- Service-level agreements, establishing
- Shewhart, Walter
- Shi, Chuan
- Sigma
- Sigma levels
- Signal-to-noise (S/N) ratios
- Signal-to-noise (continued)
- larger-the-better type
- in MTS
- nominal-the-best type
- nondynamic
- smaller-the-better type
- Signal-to-noise ratios analysis
- in Basel II case study
- in MTS
- Singular matrix
- Six Sigma
- Design for Six Sigma
- development of methodologies
- importance of measurement in
- and issues resolution
- magnitude of sigma levels in
- Six Sigma processes
- Skilled resources, in DQOM
- Smaller-the-better type S/N ratio
- Small shifts, control charts to detect
- SMEs, see Subject-matter experts
- S/N ratios, see Signal-to-noise ratios
- Software testing:
- Solution strategy (for DQ)
- SPC, see Statistical process control
- Special cause variation
- Specification, business
- SPM (statistical process monitoring)
- Sponsors:
- Square matrix
- SS (sum of squares)
- Standards, data
- Standardized distance
- Standardized variables
- State-transition model profiling
- Statistical process control (SPC)
- common and special cause variation in
- control charts
- in data tracing
- in DQ monitoring and reporting
- in establishing thresholds
- goal of
- Statistical process monitoring (SPM)
- Statistical sampling:
- in data tracing
- in reducing number of CDEs. See also Funnel approach
- Steady state:
- implementation of
- monitoring and control activities in
- Stepwise regression
- Strategy. See also Mahalanobis-Taguchi Strategy (MTS)
- data
- for data quality
- of DQ program
- technology
- Subgroups
- Subject-matter experts (SMEs):
- in assessment of data quality
- in cost assessment
- in establishing thresholds
- Sum of squares (SS)
- System performance, see Information system testing
T
- Taguchi, Genichi
- Target
- Target quality values
- Technology and operations analyst, role of
- Technology environments, configuring
- Technology infrastructure, in DQOM
- Technology leads
- Technology strategy, in big data analytics operating model
- Terminologies, for big data processing
- Test of additional information (Rao's test)
- Three-level orthogonal arrays
- Thresholds:
- defined
- for DQ scorecards
- Timeliness (DQ dimension)
- Tracing, see Data tracing
- Transpose of a matrix
- Triage, in issues resolution
- Twin charts. See also Variable control charts
- Two-level orthogonal arrays
- Two-way ANOVA
U
- u-charts
- Uncontrollable variation
- Undersampling
- Uniformity
- Updating metadata
- Upper control limits
V
- Validity (DQ dimension)
- Variables, standardized
- Variable control charts
- Variable data
- Variation:
- common vs. special cause
- in Six Sigma
- sources of
- in statistical process control
- Vision, for analytics
W
- Waste
- Watson (IBM computer)
- Western Electric Run rules
- Work streams:
- conducting current-state analysis
- data quality projects
- establishing DQ program
- monitoring and measurement
X
- X-MR (individual–moving range) charts
- -R (average-range) charts
- -S (average-standard deviation) chart
..................Content has been hidden....................
You can't read the all page of ebook, please click
here login for view all page.