Afifah Trista Ayunda

PERANCANGAN EXPERT SYSTEM MENGGUNAKAN TEKNIK BACKWARD CHAINING BERBASIS WEB

Introduction

Perancangan expert system menggunakan teknik backward chaining berbasis web. Rancang sistem pakar berbasis web dengan backward chaining untuk diagnosis penyakit sapi. Membantu dokter hewan & peternak identifikasi penyakit ternak lebih cepat & akurat.

Abstract

Kemajuan teknologi saat ini membuat proses pengambilan keputusan menjadi lebih cepat dan tepat, terutama dalam permasalahan yang kompleks. Teknologi komputer, khususnya kecerdasan buatan, telah banyak dimanfaatkan untuk membantu proses diagnosis penyakit melalui sistem pakar. Sistem pakar bekerja dengan meniru pengetahuan dan cara berpikir seorang ahli agar mampu menyelesaikan masalah seperti yang dilakukan pakar tersebut. Tujuannya bukan menggantikan peran manusia, tetapi menyimpan dan memanfaatkan pengetahuan pakar dalam bentuk sistem yang dapat diakses oleh banyak pengguna kapan pun dan di mana pun. Penelitian ini bertujuan merancang sistem pakar untuk diagnosa penyakit pada sapi ternak dengan menerapkan metode backward chaining. Pendekatan ini dianggap efektif karena menggunakan penalaran berbasis tujuan sehingga mampu menangani proses pemilihan keputusan secara terstruktur. Hasil penelitian ini diharapkan memudahkan dokter dipuskeswan dan peternak sapi di daerah pasar ternak pelangki dalam mengidentifikasi penyakit sapi, sedangkan secara teoritis penelitian ini memberikan pemahaman mengenai perancangan sistem pakar berbasis backward chaining.

Review

This paper, titled "PERANCANGAN EXPERT SYSTEM MENGGUNAKAN TEKNIK BACKWARD CHAINING BERBASIS WEB," addresses a highly relevant issue in the domain of animal husbandry and agricultural technology. The abstract effectively highlights the increasing importance of artificial intelligence and expert systems in facilitating faster and more accurate decision-making for complex problems. Specifically, the proposed research aims to leverage expert systems for the diagnosis of diseases in cattle, a critical area that can significantly impact livestock health and productivity. By developing a system that mimics human expert knowledge, the authors seek to provide a valuable tool for veterinarians at puskeswan (animal health centers) and cattle farmers in Pelangki livestock markets, thereby democratizing access to specialized diagnostic assistance. The methodological choice of backward chaining for the expert system design is clearly articulated and justified. The abstract explains that this goal-oriented reasoning approach is well-suited for structured decision-making processes, which is crucial for accurate disease diagnosis where a specific conclusion (diagnosis) is sought based on observed symptoms. The emphasis on a "web-based" implementation, as suggested by the title, implies a focus on accessibility and ease of deployment, allowing users to access the diagnostic system anytime and anywhere. This design-centric approach, rather than an implementation or evaluation study, sets the scope for the paper, focusing on the conceptual framework and architecture of such a system. The anticipated outcomes of this research are twofold. Practically, the system is expected to significantly ease the burden of disease identification for both professional veterinarians and cattle farmers, potentially leading to earlier intervention and better health outcomes for livestock. Theoretically, the study promises to contribute to a deeper understanding of designing web-based expert systems utilizing backward chaining, providing a valuable reference for future research in artificial intelligence applications. While the abstract effectively outlines the design goals and potential impact, a full paper would benefit from elaborating on the knowledge acquisition process, the validation strategy for the expert rules, and how the system's performance will be evaluated against real-world diagnostic challenges. Further details on the system's architecture and user interface design would also strengthen the overall contribution.

Full Text

Comments

You need to be logged in to post a comment.