Use Case Elicitation Method Using “When” Sentences from User Reviews
Neunghoe Kim, Chan-ki Hong
Abstract
Neunghoe Kim, Chan-ki Hong
Abstract
User review sites are spaces where users can freely post and share their opinions, which are trusted by many people and directly influence sales. In addition, they overcome the limitations arising from existing requirements collection and are able to gather the needs of large numbers of different people at a low cost. Therefore, such sites are attracting attention as new spaces for understanding user needs. In a previous study, a user review analysis was attempted using 5W and 1H, and we inferred that a sentence containing “when” has special information based on the user experience. In addition, the requirements of the derivative activities in a user review can identify more user needs than the general requirements of derivative activities. In this paper, we propose a systematic method of deriving “when” sentences contain meaningful information from user reviews and converting them into use cases, which is one of the requirements of a specification method. This method converts unstructured data into structured data such that it can be included as the user requirements during software development from user comments expressed in natural language. This method will reduce project failures and increase the likelihood of success by enabling an efficient collection and analysis of user needs from valuable user reviews.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
User review sites are spaces where users can freely post and share their opinions, which are trusted by many people and directly influence sales. In addition, they overcome the limitations arising from existing requirements collection and are able to gather the needs of large numbers of different people at a low cost. Therefore, such sites are attracting attention as new spaces for understanding user needs. In a previous study, a user review analysis was attempted using 5W and 1H, and we inferred that a sentence containing “when” has special information based on the user experience. In addition, the requirements of the derivative activities in a user review can identify more user needs than the general requirements of derivative activities. In this paper, we propose a systematic method of deriving “when” sentences contain meaningful information from user reviews and converting them into use cases, which is one of the requirements of a specification method. This method converts unstructured data into structured data such that it can be included as the user requirements during software development from user comments expressed in natural language. This method will reduce project failures and increase the likelihood of success by enabling an efficient collection and analysis of user needs from valuable user reviews.
Key concepts: Computer science, User requirements document, User story, Requirements elicitation, User needs, Sentence, User modeling, User Research