Rekomendasi Perbaikan Pernyataan Kebutuhan yang Rancu dalam Spesifikasi Kebutuhan Perangkat Lunak Menggunakan Teknik Berbasis Aturan

Penulis

Depandi Enda, Daniel Siahaan

Abstrak

Tahap awal dalam pengembangan perangkat lunak ialah menelusuri, mengumpulkan dan menyajikan segala kebutuhan pengguna ke dalam sebuah dokumen spesifikasi kebutuhan perangkat lunak (SKPL). Latar belakang akademik yang beragam, pengalaman yang berbeda, dan keterbatasan pengetahuan yang dimiliki oleh perekayasa kebutuhan memungkinkan adanya kesalahan dalam pembuatan dokumen SKPL. Salah satu kesalahan yang sering muncul pada sebuah dokumen SKPL ialah terdapatnya penggunaan kata-kata yang rancu. Hal ini tentunya dapat menyebabkan kesalahan penafsiran dan kesulitan dalam memahami kebutuhan perangkat lunak yang hendak dibangun bagi pemangku kepentingan dalam proses pengembangan perangkat lunak. Penelitian ini bertujuan mengusulkan sebuah pendekatan untuk memberikan rekomendasi perbaikan pernyataan kebutuhan perangkat lunak yang rancu. Adapun metode yang diusulkan adalah teknik berbasis aturan dengan menggunakan model bahasa n-gram. Realibilitas metode usulan di-evaluasi menggunakan indeks statistik Gwet’s AC1. Hasil analisis metode rekomendasi yang diusulkan memiliki tingkat proporsi kesepakatan yang lebih baik dibandingkan dengan metode rekomendasi menggunakan teknik statistik berbasis frekuensi n-gram. Metode rekomendasi yang diusulkan memiliki nilai indeks statistik Gwet’s AC1 tertinggi sebesar 0.5263 dengan tingkat proporsi kesepakatan sedang.

 

Abstract

The first stage in software development is to investigate, collect and provide all user requirements into a software requirements specification document (SRS’s). Diverse academic background, different experiences, and the limitations of knowledge possessed by the requirement engineer make possible mistakes in the creation of SRS’s documents. One of the most common mistakes in SRS’s document is the use of ambiguous words. This can certainly lead to misinterpretation and difficulties in understanding the software requirement that stakeholders to built in the software development process. The purpose of this research is to build an approach that gives recommendation improvement of ambiguous software requirement statement. The proposed method is a rule-based technique using n-gram language model. The reliability of the proposed method is evaluated using Gwet's AC1 statistical index. The analysis results of the proposed recommendation method have a better level of agreement proportion than the recommendation method using the n-gram frequency-based statistical technique. The proposed recommendation method has the highest Gwet's AC1 statistic value of 0.5263 with a moderate agreement proportion rate.

Kata Kunci


Metode Rekomendasi; Pernyataan Kebutuhan Perangkat Lunak; Kata Rancu; Teknik Berbasis Aturan; Indeks Statistik Gwet’s AC1

Teks Lengkap:

PDF

Referensi


ATHANASELIS, T., MAMOURAS, K., BAKAMIDIS, S. dan DOLOGLOU, I., 2011. a Corpus Based Technique for Repairing Ill-Formed Sentences With Word Order Errors Using Co-Occurrences of N-Grams. International Journal on Artificial Intelligence Tools, [daring] 20(3), hal.401–424.

BATCH POSTER SYSTEM, 2005. Detailed Business Requirements. Technical report, EDS MySC, Malaysia

BHATIA, M., KUMAR, A. BENIWAL, R., 2016. Ontology based Framework for Detecting Ambiguities in Software Requirement Specification. In: 2016 International Conference on Computing for Sustainable Global Development (INDIACom). IEEE, hal.3572–3575.

BRAY, I.K., 2002. An Introduction To Requirements Engineering, © Pearson Education Limited

BUSSEL, D. VAN, 2009. Detecting ambiguity in requirements specifications. [daring] Tilburg University.

CASK LOADER SOFTWARE, 1999. General Requirements Document Draft, © Electric Power Research Institute, Inc.

GWET, K., 2002. Kappa Statistic is not satisfactory for assessing the extent of agreement between raters. Statistical Methods For Inter-Rater Reliability Assessmen, [daring] (1), hal.1–5.

HENRICH, V. & REUTER, T., 2009. LISGrammarChecker: Language Independent Statistical Grammar Checking. [daring] Hochschule Darmstadt & Reykjavík University.

HUSSAIN, H.I., 2007. Using text classification to automate ambiguity detection in SRS documents. [daring] Concordia University.

MULIAWAN, I.W., SIAHAAN, D.O., SARWOSARI, 2011. Analisis ambiguitas kebutuhan perangkat lunak berdasarkan acuan. In: Prosiding Seminar Nasional Manajemen Teknologi XIV. Surabaya: Program Studi MMT-ITS, Surabaya 23 Juli 2011, hal.C28(1-7).

NABER, D., 2003. A Rule-Based Style and Grammar Checker. [daring] Bielefeld University.

NELBACH, F.J., 2002. Software Requirements Document For the Data Cycle System (DCS) Of The SOFIA Project, Universities Space Research Association.

PRESSMAN, R. S., 2008. Software Engineering, A Practioner Approach, Eight Edition, Mc Graw Hill

R. DUBOIS, 2002. Gamma-ray Large Area Space Telescope (GLAST) Large area telescope (LAT) Science Analysis Software Level III Specification, Technical report, GE- 0000X-DO.

R. MOESER, & P. PERLEY, 2003. Expanded Very Large Array (EVLA) operations interface, software requirements. Technical report, EVLA-SW-003 Revision: 2.5

S. GEORGE, 2001. PESA high-level trigger selection software requirements. Technical report, Centre for Particle Physics at Royal Holloway University.

SABRIYE, A. & ZAINON, W., 2017. A Framework For Detecting Ambiguity In Software Requirement Specification. In: 2017 8th International Conference on Information Technology (ICIT). Penang: IEEE, hal.209–213.

SIAHAAN, D. O., 2012. Analisa Kebutuhan Dalam Rekayasa Perangkat Lunak, Edisi 1. Yogyakarta : Andi

SINGH, S.P., KUMAR, A., SINGH, L., BHARGAVA, M., GOYAL, K., SHARMA, B., 2016. Frequency based spell checking and rule based grammar checking. International Conference on Electrical, Electronics, and Optimization Techniques, ICEEOT 2016, hal.4435–4439.

STEVENSON, M, M. HARTLEY, H. IACOVOU, A. TAN, L. PHAN, 2005. Software Requirements Specification for Sort Algorithm Demonstration Program, Version 2.4, SDPM

TJONG, S.F., 2008. Avoiding ambiguitiy in requirements specifications. Faculty of Engineering & Computer Science.

VIMALRAJ, T.J. & SEEMA, B., 2016. Identification of Ambiguity in Requirement Specification using Multilingual Word Sense. 5(6), hal.386–388.

WU, J., CHANG, J. & CHANG, S.J., 2013. Correcting Serial Grammatical Errors based on N-grams and Syntax. International Journal of Computational Linguistics & Chinese Language Processing, Volume 18, Number 4, December 2013-Special Issue on Selected Papers from ROCLING XXV, 18(4), hal.31–44.




DOI: http://dx.doi.org/10.25126/jtiik.201852627