Measurement of Similarity Between Requirement Elicitation and Requirement Specification Using Text Pre-Processing in the Cinemaloka Application

There are differences in perceptions between Clients and Developers regarding software requirements specifications therefore, research is needed to determine the perceived similarity between software requirements specifications and requirements elicitation results. The SRS document used in this study is called the Cinemaloka application. This document contains business processes and Requirements Specifications related to website-based cinema ticket reservations. This study aims to measure the suitability of perceptions between developers and clients regarding the specification of software requirements that will be or are being built. There are methods that are combined in this research, namely: determining the similarity of the requirement specification with the elicitation of requirements, analyzing the text contained in the elicitation results, Text Pre-processing, and validation through Gwet's AC1. The results of the measurement of the similarity between the elicitation of requirements and the requirement specification carried out in this study resulted in a match between the applications made and the wishes of potential users/clients. Through stages such as CountVectorizer, PorterStemmer, and Cosine Similarity, resulting in a match of 0.717144. Kappa Score from Gwet's AC1 formula using Python is -0.2222, which means “Less than Chance-agreement,” while the Kappa Score value using a questionnaire filled out by the Expert is 0.5378, which means “Moderate Aggregation.”.