Penerapan Simhash dan Hamming Distance dalam Deteksi Kemiripan Teks Berita

Jurnal Terapan Teknologi Informasi: JUTEI

article

Mayesti Anggelina¹  Lucia Dwi Krisnawati¹  Danny Sebastian¹ 

[1] Informatika, Universitas Kristen Duta Wacana

关键词: daur ulang teks;    deteksi kemiripan teks;    hamming distance;    simhash;   

DOI  :  10.21460/jutei.2022.62.216

来源: Universitas Kristen Duta Wacana

【 摘 要 】

Text reuse is defined as the reuse of existing written sources for creating a new text. The degree of reuse varies from duplicate, near-duplicate to topically similar text. Though some genres of text reuse are acceptable, their existence causes inefficiency of searching and waste of storage. To overcome this problem, a textual similarity detection system is needed. This study focuses on detecting the text similarity by applying the Simhash algorithm. It is used to create document fingerprints which function as document features through which the degree of text similarity can be compared. The similarity of a suspicious text to the source documents are measured then by Hamming Distance. Focusing on the duplicate and near-duplicate detection, the experiments conducted show that the recall of the duplicate detection  reaches 80%, meaning that the system is capable of retrieving the duplicate sources of the suspicious document.

【 授权许可】

Unknown