A Dataset of Textual Functional Requirements for Fried Chicken Processing and Sales Systems and Academic Information Systems
Description
This dataset is compiled to provide structured functional requirements documentation for two system domains that are widely used in Indonesia: processing–sales of geprek chicken, and academic information systems. In the midst of a high requirement for clear documentation of requirements, the availability of public datasets containing functional requirements textually is still very limited. Therefore, this dataset presents 54 lines of data, each containing a complete functional requirement, complete with application attributes, modules, actors, title requirements, text requirements in English, and Indonesian translations. This bilingual presentation makes it easy to use for learning software engineering, SRS creation, and NLP research, such as requirement classification and entity extraction. Some of the data contains the Geprek Resto Management System, which describes general business processes in the culinary industry, including ordering, food processing, cashier-kitchen synchronization, stock management, transaction recording, and financial statement preparation. Requirements are written based on real-world flows, such as the cashier entering an order, the system sending tickets to the kitchen, the chef updating the order status, raw materials automatically being reduced, and notifications of low stock. The dataset also includes specific conditions, such as order cancellations or out-of-stock, providing a realistic context for developing a small to medium business system. The other section outlines the functional requirements of the Academic Information System, which encompass core educational processes, including Study Plan Card, lecture schedule management, attendance tracking, assessment, and academic validation. Actors such as students, lecturers, and academic staff are represented through requirements that describe multi-level interactions, for example, validation of course prerequisites, the process of filling in grades, to recording attendance based on lecture sessions. The requirement patterns in this domain exhibit typical educational complexities, such as tiered data structures, time-based rules, and repetitive evaluation mechanisms. With a consistent table structure and coverage of two different domains, this dataset is a rich resource for requirements engineering research, document analysis automation, and the development and education of restaurant systems. It is hoped that this dataset will encourage digital innovation in culinary SMEs and educational institutions, and also serve as a reference for improving the quality of software requirements documentation.
Files
Steps to reproduce
The dataset was developed through a structured multi-phase methodology that reflects a real software analysis workflow. The process consisted of four integrated stages: eliciting requirements, transforming user stories into formal specifications, documenting them in SRS format, and finally extracting them into a unified dataset. The first stage, requirement elicitation, involved gathering user stories from operational users and domain experts. These stories captured goals, daily tasks, and expected system behaviour in simple narrative form. Information was obtained through interviews with officers, cooks, owners, and other staff; direct observations of real work situations to identify hidden needs; and group discussions where stories were clarified and validated. A typical user story described an action and an expected outcome, such as an officer turning on the Outlet Client device and the system displaying the main menu. In the second stage, transformation, analysts converted the raw user stories into formal textual requirements. Each story was examined to identify user actions, required system responses, and logical flows, including conditional behaviour. The results were written in paragraph form in both English and Indonesian, following a consistent pattern: user initiation, system reaction, alternative conditions, and final system output. This process turned informal narratives into precise, actionable statements suitable for system design. The third stage involved SRS documentation, where formal requirements were compiled into structured documents for each application. Requirements were grouped by functional module—such as Bill Payment, Daily Report, or Payroll Realization—so that developers and stakeholders could easily navigate the system’s functional scope. The narrative format ensured the requirements remained readable for both technical and non-technical audiences. Finally, in the extraction stage, researchers manually reviewed the completed SRS documents and selected 54 finalized requirements to build the dataset. Each selected entry represented one functional requirement and was preserved exactly as documented, ensuring linguistic accuracy and maintaining the original structure. The extracted requirements were then organized into a tabular dataset containing application name, module, actor, requirement title, and the bilingual textual requirement. Through this methodology, the dataset accurately represents real-world documentation practices and provides a reliable resource for research in software engineering, requirement analysis, natural language processing, and information systems development.
Institutions
- Institut Teknologi Sepuluh Nopember