Code clones detection using machine learning technique: Support vector machine

Metric level based code clone detection using optimized code manager

“…5 projects, each having 2 versions are used in this study. The results of this study shows that iclone provides best results as compare to traditional clone detection approaches [18]. Sushma et al presented a technique for detection of module level code clones in projects.…”

“…Optimized Support Vector Machine is two special properties are that OSVM achieve: (i) High-generalization by maximizing the margin and (ii) Support an effective learning of non-linear methods by kernel trick. Optimized SVM helps to perform the mapping between the inputspace and feature-space to support non-linear classification issues [3], [18]. The kernel function is harmful for responsibility this by permitting the absence of the exact formulation of the method of mapping which can cause the issue of curse of dimensions.…”

Detección Automática de Similitudes de Código Fuente Utilizando Técnicas de Aprendizaje Automático

“…Dentro de este contexto, se propone la construcción de un sistema, empleando de técnicas de aprendizaje automático supervisado, tales como Redes Neuronales Artificiales (RNA) y Máquinas de Vectores de Soporte (SVM-Support Vector Machines), comúnmente utilizadas para minería de datos (Jadon, 2016). Dichas técnicas serán realimentadas con métricas de código fuente orientadas al procesamiento de lenguaje natural (Flores Sàez, 2017), con el objetivo de poder determinar un valor indicativo del nivel de similitud de un código fuente con respecto a un corpus, de manera tal de brindarle al usuario una medida cuantitativa que lo ayude en la identificación de fragmentos de reutilización de código.…”

A Survey on Trending Algorithms for Software Code Clone Detection

“…Even though, the tactics of cut-copy and paste are measured not to be good from the aspect of software maintenance, lots of programmers uses this technique for their coding [5]. Here is a list of reasons that exist behind the code cloning:…”

“…Boreas purposed a novel checking based strategy for qualities grids, which depicts the program fragments individually in a powerful way to recognize the clones. Boreas had presented three terms numbering conditions (CE), tally vectors (CV) and check framework (CM) [5]. CE portrays the sample for the variables that was isolated into three phases.…”