Interactions Between Decision Goals Applied to the Calculation of Context Dependent Re-rankings of Results of Internet Search Engines

Search keywords can be considered as decision criteria (goals), whereas the documents contained in the original result list of a search engine can be considered as decision alternatives in the sense of adequate search results. The paper shows how a decision making model based on interactions between decision goals can be used for re-ranking of the search results obtained by classical search engines. The decision making model was previously applied to real world problems in production optimization and business process management. In contrast to other approaches, the interactive structure of decision goals for each decision situation is calculated explicitly based on fuzzy types of interaction. In this paper, after a brief description of the model an application to the calculation of context depending re-rankings of search results of internet search engines is presented. It is shown that keywords of search queries can be understood as decision goals and that the decision making model can be successfully applied for context dependent re-ranking of the search results. Using the model, popularity based rankings can be re-ranked making use of context dependent information derived from the query keywords.