2005 SENG Grad Course 609.13
Instructor: Dr. Michael M. Richter:
Title: Knowledge Management for E-Commerce
Description
The objective of this course is to introduce into essential aspects of dealing with customers in e-commerce. An essential aspect is the sales process that has to be realized by an intelligent support system.
There are three main concepts: Customers, products and sellers. With each concept (complex) knowledge is associated that has to be managed.
The task for the sellers is to supply customers with the best available products. To do this in an intelligent way three formal modelling tasks have to be done: Modelling
the customers
the products
the sales process
We model the sale as a dynamic, knowledge intensive process which takes place between the customer and an electronic agent.
For this reason some principal techniques will be presented. The techniques deal with three aspects:
Precise descriptions, e.g. for products
Vague descriptions, e.g. for customer demands
Description and execution of the sales process.
First a short introduction of the sales process from an economical view is given. Then anintroduction into necessary principles of logic, object orientation and uncertainty reasoning will be given. This includes rules and constraints. Uncertainty techniques are more deeply analysed with respect to similarity and case based reasoning. These techniques are essential for dealing with customer demands that are imprecise, incomplete, personally weighted, and can only be satisfied to a certain degree.
The difference to data based approaches are investigated by presenting in detail similarity based retrieval techniques that are central to approximative solutions.
These foundations are used to describe customer classes approximately, to represent and customize products and to perform conversations with the customers successfully.
Some emphasis is put on the dialogs between customers and sales agents.
Knowledge management activities are introduced in order to realize the sales process from the supplier´s point of view.
In order to get a feeling for the intended application areas there will be practical work in the assignment as part of the whole lecture. For this purpose the commercial tool CBR-Works will be provided and an introduction into this tool will given. The task of the practical exercise using CBR-Works is to develop an Intelligent Product Catalog.
The temporal order of the course with number of hours is :
•1.Introduction into E-Commerce and Economic Models (1 hour)
•2. Introduction into Similarity-Based and Case-Based Reasoning (1)
•3. Representation Formalisms: Exact and inexact Methods (1)
•4. Basic Issues of Customer Relationship Management (1)
•5. Getting Acquainted with the Tool CBR-Works (2)
•6/7. Similarity Formalismsand retrieval formalisms (2)
•8. Product Representation (2)
•9. Customer Characterization (1)
•10. From Catalogs to Dialogs (2)
•11. Knowledge Management: Pro-active Actions, Notification and Information Assistants (3)
Sometimes, however, there will be some changes in nthe ordering of the material because the students get those contents related to the assignment early so that they can start to work on it after the first day of class.
The assignment will be in the form of a project.
The task of the assignment is:
To model an e- commerce shop.
The choice of the shop is up to the students. For the principalstructure of the shop the commercial product CBR-Works is provided.
Suggestions for the shop are selling:
-Houses (advices are giving in the CBR-Works introduction)
-Laptops, notebooks, PCs
-Music ( a music ontology is provided)
A customer class with preferences has to be described like
-Students
-Retired couples
-Etc.
(ppt.presentation)
The product catalog for the shop has to bebuilt. Part of this are the preferences of the customer class. (CBR-Works presentation)
The dialog has to be modelled (CBR-Works and ppt. Presentation)
The assignment can be done in groups of two persons.
At the second day of class the groups have to outline their project in the class.
The grade assessment is purely based an the project. The project has to be presented at the end of the course in the form of a demo and an oral presentation.
The whole course will be available before the course starts.
The instructor will be available for extensive discussions.