Hybrid metric to reduce labour costs in assessing the complexity of the programs
В В Кукарцев, В С Тынченко, A A Boyko, M A Denisov, Viktor Alekseevich Kukartsev, Е А Чжан · Journal of Physics Conference Series · 2019
Abstract The article deals with the problem of assessing the complexity of software products. A review of widely used metrics (number of lines of code, average number of lines for functions, Halstead’s metrics, Jilb’s metrics, ABC-metric) is being conducted, on the basis of which a conclusion is drawn about their deficiencies. Also, indicators are analyzed, subtracted when assessing the complexity of programs, such as: program saturation with conditional operators or cycle operators, complexity of program understanding, program coding complexity, program content. A hybrid metric based on a composition of already well-known ones is proposed. As a result of the experiment to assess the complexity of the program, it is shown that the proposed hybrid metric allows us to more effectively evaluate the program. The use of such a metric can significantly reduce the computation time.