Decision modeling : from text to model

Abstract

For a company it is important that decisions are taken in an efficient way. Decision rules are created and by using the input data the output of the decision can be found. Executing decisions with a decision model will ensure more consistency. The decision rules need to be derived from data or company knowledge. This information can be stored in a structured way. Hence there is also the possibility that this information is stored unstructured. There can be concluded that gathering all the required rules is an intensive step for creating a decision model. Information extraction cannot be done in a standardized way. A method to make the information extraction easier is needed. This exploratory research checked if there is a method to get the needed information from the textual data. This method will be created with the help of natural language processing. Through the different levels of a text different analysis can be done. Within natural language processing there is the possibility to search for several structures. These structures are presented by a grammar. When a specific grammar is defined for a decision then in theory these decisions can be separated from the rest of the text. At the end of the research there needs to be examined if the created method will have any added value to the process of modelling decisions. The time that is needed to create a new model by using the method cannot be longer than without using the method.