Improving the Elicitation of Delightful Context-Aware Features: A Data-Based Approach

Abstract

The pursuit of innovation in the mobile software industry frequently requires coming up with new features -and not just any feature, but startling and unexpected delightful features. Despite the potential of context awareness to provide a system with delightfulness, current requirements elicitation techniques do not cope with an essential aspect: comprehension of the relationships among the numerous contextual elements of a certain domain and how they can positively influence the user task. As a result, solution providers continuously miss the opportunity to add more value to their software solutions by identifying context-aware behaviors that will lead to better recommendations or adaptations. This paper discusses this problem and proposes a process for tackling it by taking the task to be improved as input and semi-automatically generating (1) a conceptual context model that reveals and represents both the relevance of contextual elements and the relationships among them, and (2) candidate requirements. The expected scientific contributions of the ongoing research are delineated as well.

xPACE and TASC Modeler: Tool support for data-driven context modeling

“…Figure 2a shows the component diagram of the solution. TASC Modeler and xPACE are containerized and can be deployed and executed in any machine running the Docker Engine 4 , an open-source containerization technology. The Docker container for starting TASC Modeler requires approx.…”

“…Context-aware functionalities are functionalities that consider context to produce a certain system behavior, typically an adaptation or recommendation. From a requirements engineering point of view, the elicitation of context-aware functionalities requires context modeling, which is an early step that involves identification of contextual elements, analysis of accessibility (i.e., which contextual elements have available sources from which their values can be read), analysis of the relevance of these contextual elements for user tasks of interests, and analysis of combinations of contextual elements for these user tasks [4]. The analysis of relevance and combinations can be challenging, though: In a scenario with dozens of contextual elements, how to figure out which contextual elements, particularly in combination with each other, may influence a given user task?…”

“…We approached this problem by automating context modeling for the elicitation of context-aware functionalities. For this purpose, we designed a data-driven context modeling process (first introduced in [4]) that has three parts. First, the user task of interest is analyzed; then, contextual data that relates to the user task is collected; and finally, at the core of the data-driven context modeling process, the contextual data is processed to identify relevant combinations of contextual elements -and finally a context model is generated.…”

Designing a Conversational Requirements Elicitation System for End-Users

“…Rietz, Tim (2019) Proceedingsbeitrag (1000096482) tools may be used for continuous RE to maintain and incrementally improve existing systems. However, for the development of entirely new systems, disruptive changes in software or arguably also to identify delightful features, feedback or annotations may not provide a high enough level of detail [16]. Furthermore, while FAME provides minimal guidance on how to provide feedback through using a feedback form, AnnotatePro only allows free-form annotations, complicating the subsequent analysis of user input.…”