evaluation
Differences
This shows you the differences between two versions of the page.
| Both sides previous revisionPrevious revisionNext revision | Previous revision | ||
| evaluation [2010/05/06 02:05] – mark | evaluation [2010/05/06 14:15] (current) – bil | ||
|---|---|---|---|
| Line 1: | Line 1: | ||
| + | ====== Evaluation of Clustering Algorithms ====== | ||
| + | |||
| Software clustering researchers have developed several evaluation methods for software clustering algorithms. This research is important because | Software clustering researchers have developed several evaluation methods for software clustering algorithms. This research is important because | ||
| Line 11: | Line 13: | ||
| ====== Evaluation based on authoritative decomposition ====== | ====== Evaluation based on authoritative decomposition ====== | ||
| - | |||
| The main principle of evaluation based on an authoritative decomposition is that a clustering produced by an algorithm should resemble the clustering produced by some authority. Therefore, such evaluation methods provide the means to calculate the quality of an automatic decomposition by comparing it to the authoritative one. | The main principle of evaluation based on an authoritative decomposition is that a clustering produced by an algorithm should resemble the clustering produced by some authority. Therefore, such evaluation methods provide the means to calculate the quality of an automatic decomposition by comparing it to the authoritative one. | ||
| Line 29: | Line 30: | ||
| * Evaluation of software clustering algorithms that produce flat decompositions | * Evaluation of software clustering algorithms that produce flat decompositions | ||
| * Evaluation of software clustering algorithms that produce nested decompositions | * Evaluation of software clustering algorithms that produce nested decompositions | ||
| + | |||
| + | ====== Evaluation methods not based on an authoritative decomposition ====== | ||
| + | |||
| + | The main drawback of methods that require an authoritative decomposition is that they assume that such a decomposition exists. To construct such a decomposit | ||
| + | ion for a middle-size software system is a challenging task. Various research studies deal with the construction of an authoritative decomposition. This work addresses two questions: | ||
| + | |||
| + | * What is the right process for the construction of an authoritative decomposition? | ||
| + | * Why is the constructed decomposition authoritative? | ||
| + | |||
| + | |||
| + | Several evaluation methods that evaluate a software clustering approach based solely on its output have been developed. Jingwei et al. in "// | ||
| + | * Stability | ||
| + | * Extremity of Cluster Distribution | ||
| + | |||
| + | The main criticize of both methods is that they can only identify ``bad'' | ||
| + | |||
| + | ===== Stability ===== | ||
| + | Stability reflects how sensitive is a clustering approach to perturbations of the input data. The intuition behind stability in software clustering is that similar clustering decompositions should be produced for similar versions of a software system. A stability function measures the percentage of changes between produced decompositions of successive versions of an evolving software system. Under conditions of small changes between consecutive versions, an algorithm should produce similar clustering. | ||
| + | |||
| + | ===== Extremity of Cluster Distribution ===== | ||
| + | An interesting property of a clustering algorithm is the size of the clusters it produces. The cluster size distribution of a clustering algorithm should not exhibit extremity. In other words, a clustering algorithm should avoid the following situations: | ||
| + | |||
| + | * The majority of files are grouped into one or few huge clusters | ||
| + | * The majority of clusters are singletons | ||
| + | |||
| + | ====== See also ====== | ||
| + | * [[UpMoJo]] | ||
| + | * [[refsim|LimSim]] | ||
| + | * [[StructureEvaluation|Structure Evaluation]] | ||
evaluation.1273111506.txt.gz · Last modified: 2010/05/06 02:05 by mark