Analisis Daya Tahan Keamanan Data Pada Steganografi Berbasis Teks Dalam Struktur Kalimat Bahasa Indonesia

Abstract

Data security in text-based steganography needs to be analyzed in depth to ensure the sustainability and security of the method. The purpose of this research is to analyze the durability of data security in text-based steganography. The analysis approach consists of steps to identify and evaluate the vulnerability of text steganography methods with Indonesian sentence patterns. The first step is to review previous relevant work in this field to understand the vulnerabilities that have been identified previously. The text insertion model carried out is dictionary-based as many as 1,929 words which are divided into seven-word classes that correspond to sentence patterns in Indonesian, namely adj (adjective), adv (adverb), nom (noun), num (numeral), par (particle), pro (pronominal) and ver (verb). Each word class is organized into sentence patterns and each has the same bit length of eight bits. The results of the durability analysis show that input data with a length of one word still can be hacked if the message insertion process with a choice of short sentence patterns. Whereas with a choice of sentence patterns that are more than two words it will start to be difficult to hack and even cannot be hacked because the number of possibilities is very large and even infinite or infinity.