Algorithm for Categorizing Fish Species at Risk

Abstract

The paper presents an algorithmic approach for analysis of statistical data on quantity of fish catches stored in time series datasets. The developed algorithm applies trend modeling and categorizing rules for processing total data on fish species catches as well as data on fish species catches by areas. This algorithm finds out the fish species that might be at risk and groups them accordingly into the following four categories: a) economical, b) biological, c) biodiversity and d) biological and biodiversity. The analysis of these categories supports planning for future activities referring to the sustainability of the fishery ecosystem in Greece. The presented algorithm is applied on the sea fishery time series data from Greece, but it can also be applied on the same data from other countries or on the same type of integrated data from many countries belonging to big fishing areas (e.g. the Mediterranean Sea) towards data mining of fish species at risk.

Keywords: time series datasets, categorizing rules, economical, biological and biodiversity risk, fishery ecosystem