Goal model analysis helps individual select between alternatives. Given the amount of information contained within these models, it can be difficult for users to make selections over larger models. Recent work has demonstrated the value of coloring model elements to communicate the level of satisfaction over goals and tasks of interest to users. This work is limited because our choice of color palette may not be applicable to an international audience. In this workshop paper, we describe our efforts to make color visualizations adaptable and appropriate for international users, as well as those with a color vision deficiency. We describe a set of color palettes we developed and how we integrate them into existing tooling. Additionally, we allow individuals to create their own custom palettes
Goal models are an effective mechanism for elicitation and analysis in early Requirements Engineering, improving communication with stakeholders. However, in real scenarios, goal models become a complex network of actors and evaluating them can be difficult. Some tools assess the level of achievement of actors and intentional elements, which alleviate this limitation. However, using tools to evaluate models has some limitations, like the difficulty of integrating goal models with other models, the training required to use these tools, and the impossibility of assessing models when tools become outdated. In this work, we propose a formalization of iStar 2.0 models in the form of linear constraints that makes it possible to evaluate and optimize models using linear programming. This formalization allows the evaluation of goal models on existing tools and facilitates integration with other approaches.
Goal models proved their importance in several disciplines and fields of application. However, most goal modeling support tools are mainly biased toward requirements engineering. This paper presents a new conceptual tool “iTactic.” iTactic supports decision-making and tactics based on goal modeling. The tool defines a strategy in terms of a goal model using i* terminology. The employed evaluation method is based on the analogy to system models of control systems, accounting for input and output values at each point in time. iTactic goes beyond goal model evaluation and provides room for costs in its data model, providing a basis for comparing the expected success from different strategy implementation alternatives to the cost of these alternatives.
Software analytic tools consume big amounts of data coming from either (or both) the software development process or the system usage and aggregate them into indicators which are rendered to different types of stakeholders, also offering them a portfolio of techniques and capabilities such as what-if analysis, prediction and alerts. Precisely, the variety of stakeholders and the different goals they pursue justifies the convenience of performing an intentional analysis of the use of software analytics tools. With this aim, we first enumerate the different stakeholders and identify their intentional relationships with software analytics tools in the form of dependencies. Then, we focus on one particular stakeholder, namely the requirements engineer, and identify further intentional elements represented in a strategic rationale model. The resulting model provides an abstract view of the domain which may help stakeholders when deciding on the adoption of software analytic tools in their particular context. we aim at providing an intentional view to software analytic tools using the i* language [6]. We present an overview of actors, intentional elements and dependencies characterizing these tools and their context of use, with the purpose of facilitating their understanding and fit-for-purpose in the data-driven software development cycle.
Reinforcement learning (RL) is a much studied branch of machine learning and one in which substantial progress has taken place over the past few years. In RL, intelligent agents repeatedly interact with their environment and learn from the consequences of their actions. A key to effective RL is often the presence of a simulated environment that mimics the one that the agent is supposed to be optimized against. Such simulators allow for great numbers of training iterations in a cost-effective manner and without affecting the real environment. A systematic modeling and design process that allows efficient development of maintainable and comprehensible simulators can, hence, be beneficial for effective RL. We propose an approach for model-driven generation of RL environment simulations with discrete action spaces using goal models. The proposal utilizes earlier work for model-driven development of decision-theoretic action theories, whereby the standard iStar 2.0 notation is extended to include preconditions, stochastic effects, and reward modeling. Models in the extended notation can be translated into a formal specification for model-based reasoning based on Markov Decision Processes (MDPs). To also allow for model-free RL we introduce a module that queries the action-theoretic, stochastic action and reward structure aspects of the generated formal specification in order to guide episodic simulations of the modeled domain. The module is wrapped by a popular framework for building RL training and testing environments, making it accessible by popular RL agent frameworks.
Capturing requirements in machine learning projects is a challenging task. It requires domain knowledge as well as experience in the machine learning field. The i* framework is a popular high abstraction-layer requirements capturing tool. However, the use of i* directly in the machine learning field (ML) is unfeasible due to it cannot capture all the restrictions and relationships of ML elements. In previous works we have extended i* to better capture machine learning requirements. In this paper, we apply the i* for machine learning extension to a real machine learning case study, in the context of a project focused on the diagnosis and treatment of Attention-Deficit/Hyperactivity Disorder (ADHD). The results show that the use of the i* for machine learning extension provides insights about the correct path to follow, aiding in the definition and selection of machine learning solutions that better fulfill the project requirements. Moreover, it facilitates faster development of the machine learning solution in a more structured way, avoiding errors and making the application of i* an effective tool for managing machine learning requirements.
