Investigation of the genetic algorithm possibilities for retrieving relevant cases from big data in the decision support systems
Konstantin Serdyukov, Tatiana Vladimirovna Avdeenko · 2017
In present paper we consider the advantages and disadvantages of case-based reasoning (CBR) approach for knowledge representation of the application domain. One of the CBR shortcomings is the insufficient speed of real-time retrieval of cases, as well as the insufficient relevance of the retrieved cases to the current situation. To solve these problems, we offer the use of genetic algorithm. We propose formal statement of the genetic algorithm to the CBR retrieving stage. The results of the investigation are presented. The major advantage of the genetic algorithm is that it gives a more compact set of retrieved cases extracted which possesses, however, characteristic features of the current situation. This property can be very useful when extracting such cases from big data In conclusion, the perspectives of applying the method for adaptation of cases have been given.