<![CDATA[Penyakit pada orang lanjut usia (lansia) sering sulit dikenali sejak dini akibat keterbatasan pengetahuan masyarakat serta akses layanan kesehatan yang terbatas. Berbagai penelitian sebelumnya telah menerapkan sistem pakar untuk diagnosis penyakit, namun sebagian besar belum secara khusus memfokuskan pada diagnosis penyakit lansia dengan pendekatan tingkat keyakinan berbasis ketidakpastian yang mudah diakses masyarakat umum. Kesenjangan penelitian (research gap) terletak pada masih terbatasnya sistem diagnosis mandiri berbasis web yang mampu mengakomodasi ketidakpastian gejala lansia menggunakan metode Certainty Factor. Penelitian ini bertujuan mengembangkan sistem pakar berbasis web untuk diagnosis awal penyakit pada lansia menggunakan metode Certainty Factor. Sistem dikembangkan melalui tahapan analisis, perancangan, implementasi menggunakan PHP dan MySQL, serta pengujian sistem. Evaluasi dilakukan melalui pengujian fungsional black box dan pengujian akurasi dengan membandingkan hasil diagnosis sistem dan pakar menggunakan data 50 lansia. Hasil pengujian menunjukkan bahwa 46 data diagnosis sesuai dan 4 data tidak sesuai, sehingga diperoleh tingkat akurasi sebesar 92%. Hasil tersebut menunjukkan bahwa sistem memiliki performa yang baik sebagai alat bantu diagnosis awal penyakit pada lansia. Meskipun jumlah sampel masih terbatas, hasil ini sejalan dengan penelitian sistem pakar berbasis Certainty Factor sebelumnya yang menunjukkan akurasi pada kisaran menengah hingga tinggi. Sistem yang dikembangkan memberikan kontribusi berupa platform diagnosis awal yang mudah diakses dan mendukung pengambilan keputusan kesehatan awal bagi lansia.  Diseases in older adults (elderly) are often difficult to identify at an early stage due to limited public knowledge and restricted access to healthcare services. Previous studies have implemented expert systems for disease diagnosis; however, most have not specifically focused on elderly disease diagnosis using uncertainty-based confidence approaches that are easily accessible to the general public. The research gap lies in the limited availability of web-based self-diagnosis systems capable of accommodating uncertainty in elderly symptoms using the Certainty Factor method. This study aims to develop a web-based expert system for early diagnosis of diseases in elderly individuals using the Certainty Factor method. The system was developed through stages of analysis, design, implementation using PHP and MySQL, and system testing. Evaluation was conducted through functional testing using the black-box method and accuracy testing by comparing system diagnosis results with expert diagnoses using data from 50 elderly patients. The testing results showed that 46 diagnoses were consistent with expert evaluations while 4 diagnoses were inconsistent, resulting in an accuracy rate of 92%. These results indicate that the developed system performs well as a supporting tool for early disease diagnosis in elderly individuals. Although the sample size remains limited, the findings are consistent with previous Certainty Factor–based expert system studies reporting moderate to high accuracy levels. The developed system contributes a web-based early diagnosis platform that is easily accessible and supports preliminary health decision-making for elderly individuals.]]>
