Arabic text preprocessing for dynamic searchable symmetric encryption: An empirical evaluation and preprocessor selection framework

Kholoud Saad Al-Saleh · 2026

Read Arabic text preprocessing for dynamic searchable symmetric encryption: An empirical evaluation and preprocessor selection framework by Kholoud Saad Al-Saleh (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0351281.txt.

Description

Kholoud Saad Al-Saleh tackles a search problem created by the Arabic language itself, in PLOS ONE in 2026. Encrypted search systems let users query data without exposing it to the server. But Arabic builds dozens of word forms from one root. That bloats the encrypted index and hides matching documents behind spelling variants. The study compares four text preprocessing strategies, three stemmers and analyzers plus a segmenter, against plain normalization inside a working encrypted search scheme.

On a news corpus with benchmark queries, stemming shrank the vocabulary by up to eighty-three percent, with matching savings in index size and transfer cost. A surprise emerged on speed: the plain baseline answered fastest precisely because it retrieved the fewest documents, while better preprocessing found more and paid in time. Scaling tests reached tens of thousands of documents. The result is the first evidence based guide for choosing an Arabic preprocessor.

Catalog record

CREATORKholoud Saad Al-Saleh
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE112612
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0351281

Browse the full library · About the Tower of Babel