Calculation and optimization of thresholds for sets of software metrics

Abstract

In this article, we present a novel algorithmic method for the calculation of thresholds for a metric set. To this aim, machine learning and data mining techniques are utilized. We define a data-driven methodology that can be used for efficiency optimization of existing metric sets, for the simplification of complex classification models, and for the calculation of thresholds for a metric set in an environment where no metric set yet exists. The methodology is independent of the metric set and therefore also independent of any language, paradigm or abstraction level. In four case studies performed on large-scale open-source software metric sets for C functions, C++, C# methods and Java classes are optimized and the methodology is validated.

Quality indicators for business process models from a gateway complexity perspective

“…However, the need for a binary variable is a limitation, as is the definition of p 0, which is defined by the engineer arbitrarily. It requires an input parameter that affects both the performance and the accuracy of the results Herbold et al [24] Control flow structuredness, coupling, size, method complexity, inheritance, staticness…”

“…Herbold et al [24] used a machine learning algorithm to define an approach for the calculation of the threshold. This utilizes the learning of axis-aligned d-dimensional rectangles for the calculation.…”

Investigating the Strength of Relationships among Software Metrics

“…Thresholds for software metrics are often used in the context of fault-proneness. This means that a measured entity is more fault-prone, if it violates a threshold [18]. Hence, they can act as guidelines for developers to keep them at optimum levels to reduce potential risks.…”

A unique code smell detection and refactoring scheme for evaluating software maintainability

“…An illustration can be: "distinguish classes that have lower dependability" or "classify methods that have a high difficulty". We want to obtain thresholds in a method that can be semantically mapped to these easy necessities, to find out what 'LOW' unity or 'HIGH' difficulty means in terms of the metrics, the unity and difficulty of the software is measured [6].In normal, thresholds may discriminate values. In case threshold gives higher bound, the values that are greater than a threshold value are measured to be difficulties.…”