Software Requirement Specification Compatibility Design Method Through Use Case Diagram Artifacts Using Text Mining on Akuonline Learning Application

Software Requirement Specification (SRS) is a document detailing the requirements for software development created from the Requirements Elicitation process between the Software Developer and the Client.Differences in interpreting requirements can occur between developers and clients.This study aims to determine and calculate the compatibility value of a user requirement with the software being built by measuring the similarity of the Requirement Elicitation results with the software modeling.Compatibility can be achieved through a Text Mining approach involving Text Preprocessing activities on the Requirement Specification results.Data extraction on Requirement Elicitation through interviews is achieved by performing Text Preprocessing and similarity, identifying and finding FR compatibility with the Steps Performed.Four conclusions are drawn from this research.First, based on the SRS named "Akuonline Learning Application," 11 FRs have been identified.Based on Cosine Similarity calculations comparing the FR matrix and the Step matrix performed, and validation results through Gwet's AC1 reliability measurement, the Python version = 0.299, while the questionnaire version = 0.379, showing a difference of 0.08.Referring to Gwet's AC1 Value Interpretation, the validity and reliability values fall into the "Fair agreement" category.