master Assignment
MACHINE LEARNING TO quantify e3value business models
Type: Master CS
Student: Unassigned
Duration: TBD
If you are interested please contact:
Description:
Background
The e3value modeling language is used to develop business models for digital business ecosystems and platforms. It is a graphical language (e.g. similar to the UML), that represents the parties in an ecosystem or platform and what they exchange of value with each other. The method is used in practice and software tooling is available (see e.g. https://cms.thevalueengineers.nl/uploads/Executable_business_modeling_for_ecosystems_with_e3value_eb32345a82.pdf for an accessible introduction).
Problem
An e3value model can be quantified with prices and other market data, and if done correctly, several calculations can be done, for example computing the net revenue of each participant, or estimating payback periods of an investment. However, finding the data for these calculations is a manual task, requires a significant amount of time, and is inefficient.
Solution orientation
The student will develop a software prototype that uses machine learning to find data on the Internet that can be used to quantify e3value model. Two kinds of queries must be implemented:
- Questions about data available on the web: Given this report about the market, find similar reports.
- Specific questions about the market: Given these reports about the market, what is the average price of this item? What is the price range?
We make available a number of these e3value models for which the quantification is already done and/or for which the quantification needs to be done. A number of scenarios will be developed that demonstrate how data collection for e3value models can be done in a semi-automated way.
Organizational context:
This project is performed in collaboration with The Value Engineers (see www.thevalueengineers.nl), a spin-off of the VU and UT.
Required:
An independent student, familiar with machine learning, graphical conceptual modelling techniques such as the UML, and willing to learn e3value.