retail design
DESCRIPTION
Recom Systems Provides Business Intelligence solutions to retail industry at a very low price.TRANSCRIPT
Recom Retail Solution
Introduction
• Retail Model is centered around manufacturer in countries like India
• This model is designed for forward integration and backward integration
• Model driver is supply and demand• Major Control is in hands of
Government
Business Intelligence
• Financial Parameters
• Economic Parameters
• Customer Parameters
• Sale Parameters
• Human Resource Parameters
• Internal Processes Parameters
Customized Solutions
• Solutions for Quick response to changing market conditions
• Critical Reports available at Intranet, email and desktop applications
• Integration of Corporate strategy and performance
• Increases overall efficiency, profitability and market share
• Information Security maintained over the value chain
Critical Business Information Processing
• Critical business information is processed quickly which otherwise can take days and even months to consolidate
• Large amount of data analyzed and processed
• Forecasting tools using Artificial Intelligence have uncanny predicting abilities
Analyzing Dimensions
• Business information is analyzed on country-wise, region-wise, state-wise or city-wise
• Analysis using Branch wise on hundreds of parameters• Branch-wise and year-wise, quarter-wise, month-wise
or date-wise• Customer-wise sales and trends• Forecasting trends on multiple parameters and
dimensions• Associating Customers for sale trends• Creating Cluster • Up-Selling and Cross-selling• Arranging Display Racks of Products using AI Tools • Creating Sales Campaign using AI tools
Resource Optimization
• Optimizing Manpower Deployment• Optimizing Sales• Optimizing Cash Flow• Optimizing Machinery Investments and
Deployment• Controlling Waste and Rejects• Optimization of Quality Checks• Optimization of Time Delays
Dimensional Measure and Aggregation
Naïve Bayes
Sales Vs Geography Vs Time
Customer Vs Geographical Dimension
Customer Dimension
Geographical Dimension
Employee VS Sales Hierarchical Dimension
Employee Vs Sales Dimension
Setting BucketProperty
Product Dimension
Time Dimension
Relationship Diagram
Dimensions and Measure Groups
Many to Many Relationships
Applying Calculations to Dimensions
Yearly Gross Profit Margin on Sales
Expanding Product Category
Creating Sub Cubes within Multi Dimensional Cubes
Key Performance Indicators
Time Series Prediction
Input column content typesContinuous, Cyclical, Discrete, Discretized, Key, Table, and Ordered
Predictable column content typesContinuous, Cyclical, Discrete, Discretized, Table, and Ordered
Modeling flagsMODEL_EXISTENCE_ONLY, NOT NULL, and REGRESSOR
IsDescendant PredictNodeId
IsInNode PredictProbability
PredictAdjustedProbability PredictStdev
Decision Trees
Clustering
Market Basket Analysis
• Consider shopping cart filled with severalitems
• Market basket analysis tries to answer thefollowing questions:
– Who makes purchases?– What do customers buy together?
– In what order do customers purchase items?• Given a database of customer transactions,each transaction is a set of items – deduce
association rules.
Examples of Market BasketAnalysis
• Co- ocurrences – 80% of all customerspurchase items a, b, and c together.
• Association Rules – 60% of all customerswho purchase X and Y also buy Z.
• Sequential Patterns – 60% of customerswho first buy X also purchase Y within two
weeks.
Confidence and Support• We prune the set of all possible association
rules using two measures of interest:– Confidence of a rule: X -> Y has confidence c if
P( Y| X)= c.– Support of a rule: X-> Y has support s ifP( XY)= s. Also, support of an itemset XY.
• Direct Marketing• Fraud Detection for Medical Insurance• Floor/ Shelf Planning• Web Site Layout• Cross- selling
Applications
Frequent Itemsets Applications
– Classification– Seeds for constructing Bayesian networks– Web log analysis– Collaborative filtering
Association Rules Approaches• Problem Reduction• Breadth- First Search• Depth- First Search
EnvironmentInteroperability: key components (client/network/server) work together.Salability: any of the key elements may be replaced when the need to either grow or reduce processing for that element dictates, without major impact on the other elements.Adaptability: new technology (multi-media, broad band networks, distributed database, etc.) may be incorporated into the system.Affordability: using less expensive insures cost effectiveness MISs which available on each platform.Data Integrity: entity, domain and referential integrity are maintained on the database server.Accessibility: data may be accessed from WANs and multiple client applications.Perform: performance may optimize by hardware and process.Security: data security is centralized on the server.
Data Warehouse
One or more tools to extract fields from any kind of data structure (flat, hierarchical, relational, or object) including external data.
The synthesis of the data into a nonvolatile, integrated, subject oriented database with a metadata “catalog.”
All AI applications on Data Warehouse
Data Warehouse Advantages
• Improves product inventory turns • Costs of product introduction are decreased, with
improved selection of target markets• Separating query processing from running against
operational databases enables most cost-effective decision making
• Improved productivity, by keeping all required data in single location and eliminates rekeying of data
• Reduced redundant processing, support, and software to support overlapping decision support applications
• Enhanced customer relations through improved knowledge of individual requirements and trends, through customization, improved communications
Data Ware House Design Considerations
• Heterogeneity of data sources, which affects data conversion, equality, timeless
• Use of historical data, which implies that data may be “old”
• Tendency of databases to grow very large • The database management system that supports the
warehouse database • The communications infrastructure that connects the
ware- house, data marks, operational system, and end users
• The systems management framework that enables centralized management and administration of the entire environment
Data Mining
• Identification of Patterns to guide marketing• Discovering meaningful new correlation, patterns, and
trends by digging into (mining) large amounts of data • Using artificial intelligence (AI) and statistical and
mathematical techniques • Data mining uses well-established statistical and
machine learning techniques to build model that predict customer behavior
• Automating the mining process, integrating it with commercial data warehouses, and presenting it in relevant way for business users
Data Mining Continued
• Data mining techniques can yield the benefits of automation when implemented on existing software and hardware platform and can be implemented on new systems as existing upgraded and new products developed
• When data mining tools are implemented on high performance parallel processing systems, they can analyze massive database in minutes
• Data mining software can help to find the high-profits gems buried in mountains of information
Database Selection and Preparation
• This step includes the identification of databases and factors to be explored
• Whenever possible, required records can be retrieved using a live data dictionary
• Data preparation includes filling in missing values and removing errors
Analysis
• The large database groups defined during the preparation phase are further divided using clustering techniques
• Determine which factors are involved in the maximization of particular goals
Row Labels Internet Order Count Internet Average Sales Amount Internet Average Unit Price Internet Extended Amount Internet Freight Cost
CY Q1 6,984 1072.287903 490.9439303 $7,488,858.71 $187,222.15
Accessories 4,619 37.57320416 19.11140073 $173,550.63 $4,339.16
Accessories 4,619 37.57320416 19.11140073 $173,550.63 $4,339.16
Bikes 3,853 1876.266344 1876.266344 $7,229,254.22 $180,731.53
Bikes 3,853 1876.266344 1876.266344 $7,229,254.22 $180,731.53
Clothing 1,930 44.58749223 37.09218103 $86,053.86 $2,151.45
Clothing 1,930 44.58749223 37.09218103 $86,053.86 $2,151.45
CY Q2 8,021 1131.331797 511.7246006 $9,074,412.34 $226,861.10
Accessories 5,171 38.62985496 19.55889357 $199,754.98 $4,994.32
Accessories 5,171 38.62985496 19.55889357 $199,754.98 $4,994.32
Bikes 4,883 1797.313654 1797.313654 $8,776,282.57 $219,407.29
Bikes 4,883 1797.313654 1797.313654 $8,776,282.57 $219,407.29
Clothing 2,170 45.33400461 37.30557072 $98,374.79 $2,459.49
Clothing 2,170 45.33400461 37.30557072 $98,374.79 $2,459.49
CY Q3 5,851 964.884227 451.4985295 $5,645,537.61 $141,138.99
Accessories 3,906 39.02180492 19.56851586 $152,419.17 $3,810.81
Accessories 3,906 39.02180492 19.56851586 $152,419.17 $3,810.81
Bikes 2,774 1953.86977 1953.86977 $5,420,034.74 $135,501.00
Bikes 2,774 1953.86977 1953.86977 $5,420,034.74 $135,501.00
Clothing 1,564 46.72870844 37.65260175 $73,083.70 $1,827.18
Clothing 1,564 46.72870844 37.65260175 $73,083.70 $1,827.18
CY Q4 6,803 1050.987587 479.6316196 $7,149,868.55 $178,747.38
Accessories 4,512 38.79325798 19.42892441 $175,035.18 $4,376.27
Accessories 4,512 38.79325798 19.42892441 $175,035.18 $4,376.27
Bikes 3,695 1865.37838 1865.37838 $6,892,573.11 $172,314.50
Bikes 3,695 1865.37838 1865.37838 $6,892,573.11 $172,314.50
Clothing 1,797 45.77643851 37.34010894 $82,260.26 $2,056.61
Clothing 1,797 45.77643851 37.34010894 $82,260.26 $2,056.61
Grand Total 27,659 1061.451145 486.0869105 $29,358,677.22 $733,969.61
Typical Sales Report for Internet Sales
Data Mining as an Data Mining as an Application PlatformApplication Platform
What is Data Mining Anyway?
• Machine learning of patterns in data
• Application of patterns to new data
What is Data Mining Anyway?
• Machine learning of patterns in data
• Application of patterns to new data
Comparative BenefitsPredictive Projects versus Nonpredictive
Projects
“Data Mining is Hard”• “White-coats” only need apply• How do you
– … define problem?– … select data?– … choose inputs?– … choose outputs?– … interpret results?– … validate results?
What Does Data Mining Do?
Explores Explores Your DataYour Data
Finds Finds PatternsPatterns
Performs Performs PredictionPrediction
ss
Mining Model
What does Data Mining do?Illustrated
DMEngine
Data To Predict
DMEngine
Predicted Data
Training Data
Mining Model
Mining Model
DB dataDB dataClient dataClient dataApplication dataApplication data
DB dataDB dataClient dataClient dataApplication dataApplication data““Just one rowJust one row””
Analysis ServicesAnalysis ServicesServerServer
Mining ModelMining Model
Data Mining AlgorithmData Mining Algorithm DataDataSourceSource
Server Mining Architecture
Your ApplicationYour Application
OLE DB/ ADOMD/ XMLAOLE DB/ ADOMD/ XMLA
DeployDeploy
BI Dev BI Dev StudioStudio (Visual (Visual Studio)Studio)
AppAppDataData
““Putting Data Putting Data Mining to Work”Mining to Work”
““Doing Data Doing Data Mining”Mining”
Business Business UnderstandingUnderstanding
Data Data UnderstandingUnderstanding
Data Data PreparationPreparation
ModelingModeling
EvaluationEvaluation
DeploymentDeployment
DataData
Data Mining ProcessCRISP-DM
www.crisp-dm.orgwww.crisp-dm.org
SSASSSAS(Data(Data
Mining)Mining)
SSAS SSAS (OLAP)(OLAP)DSVDSV
SSISSSISSSAS(OLAP)SSAS(OLAP)SSRSSSRSFlexible APIsFlexible APIs
SSISSSISSSASSSAS(OLAP(OLAP))
Business Business UnderstandingUnderstanding
Data Data UnderstandingUnderstanding
Data Data PreparationPreparation
ModelingModeling
EvaluationEvaluation
DeploymentDeployment
DataData
Data Mining Process in SQLCRISP-DM
www.crisp-dm.orgwww.crisp-dm.org
DataData
What Do Data Mining Applications Do?
Explores Explores Your DataYour Data
Finds Finds PatternsPatterns
Performs Performs PredictionPrediction
ss
Automatic Automatic MiningMining
Pattern Pattern ExplorationExploration
Perform Perform PredictionPrediction
ss
Algorithm Training
Algorithm Module
Case Processor
(generates and prepares all
training cases)
StartCases
Process One Case
Converged/complete?NoYes
Done!
Persist patterns
DM data flow
Cube
HistoricalDataset
NewDataset
Data Transform (DTS)Reporting
Mining Models
ModelBrowsing
Prediction
LOBApplication
Cube
PredictionParser Validation-I &
InitializationAST
Binding &Validation-II
DMX treeExecution Planning
DMX tree
Input data
Read / Evaluate one row
Push response
Untokenize results
Income Gender
$50,000 F
1 2
50000 2
1 2 3
50000 2 1
Income Gender Plan
$50,000 F Attend