More Related Content
Similar to Visual and analytical mining of sales transaction data for production planning and marketing (20)
Visual and analytical mining of sales transaction data for production planning and marketing
- 1. Visual and analytical mining of transactions data for production planning and marketing Gurdal Ertek, Can Kuruca, Cenk Aydin, Besim Ferit Erel, Harun Dogan, Mustafa Duman, Mete Ocal, Zeynep Damla Ok SABANCI UNIVERSITY
- 2. Introduction Motivation Large amounts of enterprise data available Data mining Deriving necessary and meaningful information out of data Framework combining visual and analytical data mining Filtering --- Interactive pie charts Clustering --- k-means algorithm Comparison --- Parallel coordinate plot
- 3. Motivation: Explosion of Data Data from marketing Barcode systems, accounting software, ERP software, e-commerce data (B2B and B2C) Data from manufacturing CIM systems, barcode, radio frequency technologies
- 4. Data Mining Effective collection, management, reporting, interpretive analysis and mining of enterprise data: Establishing effective control of manufacturing activities Achieving effective production planning and increased sales, and consequently increasing the firm’s profitability Increasing customer satisfaction by offering and timely delivering them products that they are willing to purchase. CRM: Customer Relationship Management
- 5. Sales Transaction Data Collected and archieved in almost every firm Essential input for both marketing and production planning Framework and prototype implementation CuReMa
- 6. Literature Review … -1990’s: Scatterplot, boxplot, … 1990-2000’s: Information visualization
- 7. Proposed Framework Filtering Clustering Comparison
- 8. Implementation of the Proposed Framework: CuReMa
- 9. Filtering: Interactive Visual Querying
- 10. Clustering: Analytical Data Mining
- 11. Comparison: Visual Data Mining
- 12. Comparison: Visual Data Mining
- 13. Future Work Other visual metaphors and analytical approaches can be used to extend the framework Ex: Drawing association rules Other data fields can be incorporated Ex: Ages and income levels of customers Other clustering algorithms can be used Ex: Self-organizing maps Other criteria for clustering can be implemented Ex: Recency and frequency of purchases Localization issues Ex: Inflation and local holidays