1 / 14

Model-oriented: Simulation Optimization Data-oriented OLAP (Online Analytical Processing) Data Mining Expert Syste

تکنیک های مورد استفاده در پشتیبانی تصمیم. Model-oriented: Simulation Optimization Data-oriented OLAP (Online Analytical Processing) Data Mining Expert Systems Fuzzy Logic Neural Systems Case-based Reasoning . Simulation. Model-oriented: Simulation

tale
Download Presentation

Model-oriented: Simulation Optimization Data-oriented OLAP (Online Analytical Processing) Data Mining Expert Syste

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. تکنیک های مورد استفاده در پشتیبانی تصمیم • Model-oriented: • Simulation • Optimization • Data-oriented • OLAP (Online Analytical Processing) • Data Mining • Expert Systems • Fuzzy Logic • Neural Systems • Case-based Reasoning

  2. Simulation • Model-oriented: • Simulation • Manager’s Mental Model (Variables influence the profit) - Tentative assumptions • Formalized in DSS as a mathematical model • - Using What-if to try out different assumptions • - MATLAB

  3. Optimization • Model-oriented: • Optimization • Starts with Manager’s optimization criteria • Using mathematical model to determine optimal decisions based on the criteria • - MATLAB

  4. OLAP • Data-oriented: • OLAP • Exploring transaction data from transactional Databases • Dimensions of Data • OLAP raise from difficulties analyzing databases – Slowed down processing • Using data warehouse: Extraction, consolidation, and filtering • So: Transaction processing and OLAP without mutual interference • - Targit - Logixml

  5. OLAP

  6. Data Mining • Data-oriented: • Data mining • Using analytical tools to find patterns (Correlation) in transaction databases such as customer receipts • Wonderful for Marketing Research • Such as correlation between two different product sold in specific hours • Such as a bank finding that customers with multiple accounts were unprofitable • - Rapid Miner

  7. Expert System • Expert Systems: • Supporting professionals engaging in design, diagnosis, or evaluation of complex situations requiring knowledge • Converting data into recommendations • Less experienced people doing similar task as experts do • Represent knowledge in an explicit form – So-called: knowledge-based systems • Representing knowledge as If-then rules • - CLIPS

  8. Expert System • Example: Whether to grant a business loan: • If: The applicant is current on all debts, and the applicant has been profitable for two years, and the applicant has strong market position, • Then: The applicant is an excellent credit risk

  9. Expert System • Four major components of an expert system: • Knowledge-base: Set of facts and if-then rules • Database of facts • The inference engine: Using rules in the knowledge-base and facts in the database to infer new facts • The interface: interacting with the user • Explanation module: how a particular fact wasinferred

  10. Expert system

  11. Expert System • Forward chaining: starting with data and try to draw conclusions • Backward chaining: Starting with tentative conclusions and then look for facts in the database • Risky to trust expert system: not truly experts, different situations • Unless used in the same situations

  12. Fuzzy Logic • Minimizing problems with expert systems: just true or false • In expert systems: Trivial difference between 1$ profit and 1$ lost, affecting decision as much as difference between 100$ million profit and a 1$ lost • Combining a number of different rules based on conditions such as: • Very profitable • Profitable • slightly profitable • and so on... • - FuzzyTECH

  13. Neural Network • Neural Networks: • Using Statistical models to find patterns in data • Model the way human brain works • Recognizes patterns based on examples that have been trained to it • Each training example is a set of characteristics and a result: such as whether or not a loan was repaid • Applying numerical weights to characteristics based on examples • - MATLAB

  14. Neural Network

More Related