Kakas bantu perhitungan nilai kopling menggunakan conceptual coupling metrics

Laras Husna Aulia
Fajar Pradana
Bayu Priyambadha

Abstract


ABSTRAK

Kualitas perangkat lunak dapat diukur dengan nilai kopling pada fase perancangan. Tingkat kopling pada perancangan mengindikasikan seberapa besar hubungan antar komponen dalam sebuah sistem. Untuk menentukan nilai kopling dapat menggunakan empat parameter metrik kopling konseptual antara lain Conceptual Similarity be-tween Methods (CSM), Conceptual Similarity between a Method and a Class (CSMC), Conceptual Similarity between two Classes (CSBC) dan Conceptual Coupling of Class (CoCC). Perhitungan nilai kopling dapat dilakukan dengan cara mengambil method-method pada klas kemudian dihitung nilai kecocokan antar dokumen kode sumber dengan menggunakan Latent Semantic Indexing (LSI). Metode LSI digunakan untuk menghitung kecocokan antar dokumen kode sumber berdasarkan kesamaan kata maupun kesamaan makna kata, kemudian nilai tersebut digunakan untuk menghitung nilai metrik kopling konseptual. Pengujian akurasi pada kakas bantu ini dilakukan dengan membandingkan dengan kakas bantu lain sebagai pembanding. Pengujian dikelompokan ke dalam 3 (tiga) nilai toleransi yang berbeda yaitu 0,05; 0,1 dan 0,20. Penggunaan nilai toleransi yang berbeda disebabkan karena terdapat selisih antara kakas bantu yang dikembangan dengan kakas  bantu pembanding. Hasil pengujian akurasi dengan nilai toleransi sebesar 0,05; didapatkan nilai akurasi sistem sebesar 24,83%. Jika nilai toleransi sebesar 0,1 maka nilai akurasi sistem sebesar 43,41%. Sedangkan, jika nilai toleransi sebesar 0,20 maka nilai akurasi yang diperoleh diatas 50% yaitu 69,78%.

Kata kunci:   kopling, kualitas, Latent Semantic Indexing (LSI), metrik kopling konseptual parsing method, perangkat lunak.

 

ABSTRACT

   Coupling is one of the parameter of quality of software. Coupling is a paramater for measuring how much the relationship between components of a system. To determine the value of coupling can use four of the parameters of conceptual coupling metrics among other Conceptual Similarity between Methods (CSM), Conceptual Similarity be-tween a Method and a Class (CSMC), Conceptual Similarity between two Classes (CSBC) and Conceptual Coupling of Class (CoCC). Before counting the value of the coupling, parsing method carried out using spoon library and com-putation macth between source code documents to be done by using a method of information retrieval that is Latent Semantic Indexing (LSI). The use of LSI method because this research requires the calculation of a match between the document's source code based on common words or similarity of meaning of the word. After that, the LSI value is used to calculate the Conceptual Coupling Metric Testing accuracy of these tools is done by comparing with other systems as a comparison.. Testing is classified into 3(three) different tolerance values are 0.05, 0.1 and 0.20. The use tolerance values differ, because there is a difference between systems developed and the comparator system.  If tolerance value is 0.05 then the accuracy value is 24,83%. If tolerance value is 0.1 then the accuracy value is 43,41%. Meanwhile, If tolerance value is 0.2 then the accuracy value is up to 50%, it is 69,78%.

Keywords: Conceptual coupling metrics, coupling, Latent Semantic Indexing, LSI, parsing method.


References


D. Poshyvanyk and A. Marcus, "The Conceptual Coupling Metrics for Object-Oriented Systems," in 22nd IEEE International Conference on Software Maintenance, Philadelphia, 2006.

G. Gui and P. D. Scott, "Ranking reusability of software components using coupling metrics," Journal of Systems and Software, vol. 80, no. 9, pp. 1450-1459, 2007.

A. Parashar and J. K. Chhabra, "Clustering Dynamic Class Coupling Data to Measure Class Reusability Pattern," in Communications in Computer and Information Science, High Performance Architecture and Grid Computing ed., vol. 169, Heidelberg, Springer, 2011, pp. 126-130.

A. Dennis, B. H. Wixom and R. M. Roth, Systems Analysis and Design, 6 ed., Hoboken: Wiley, 2014.

A. Aloysius and L. Arockiam, "Coupling Complexity Metric: A Cognitive Approach," International Journal of Information Technology and Computer Science (IJITCS), vol. 4, no. 9, pp. 29-35, 2012.

H. Bunyamin, "Information Retrieval System Dengan Metode Latent Semantic Indexing," Institut Teknologi Bandung, Bandung, 2005.

M. Riley, E. Heinen and J. Ghosh, "A Text Retrieval Approach To Content-Based Audio Retrieval," in ISMIR 2008: Proceedings of the 9th International Conference of Music Information Retrieval, Philadelphia, 2008.




DOI: https://doi.org/10.26594/teknologi.v6i2.598

Article metrics

Abstract views : 372 | views : 155


 

Index by:
                     

Powered by:


Plagiarism checker:

ISSN:

ISSN (Print): 2087-8893 and ISSN (Online): 2527-3671


Visit:

Free counters!

 

 TEKNOLOGI on Facebook: