Risk, a potential occurrence of some undesirable event, can be dangerous if not adequately identified and dealt with early on during software development. However, identifying risks can be difficult, hence oftentimes resulting in a particular software system that is unable to address risks, especially critical ones adequately. This paper proposes an ontology-based framework for performing risk analysis with the Augmented Reference Model -The Reference Model augmented with risk analysis. The Reference Model emphasizes that the user requirements are met through the collaboration between the system and the events occurring in its environment -i.e., not by the system alone, hence the term "collaborative system." We also offer an activity-oriented ontology to carry out risk analysis by identifying risks from negating the events in the environment and system. Such negations of the requirements, specifications, and domain events generate a graph-like representation, called Risk Analysis Graph (RAG), to help perform risk analysis. To validate our framework, we have performed two experiments using questionnaires to identify risks and use the risk analysis tool to generate RAG for performing risk analysis. We feel that at least these experiments show that RAG helps identify risks -especially the critical and uncommon ones that we would not have thought of.

A risk is an undesirable event that can result in mishaps if not identified early on during requirements engineering adequately. However, identifying risks can be challenging, and requirements engineers may not always be aware if risks are ignored. In this paper, we present Murphy -a framework for performing risk analysis. Murphy adopts the Reference Model, in which requirements are supposed to be met not by the projected software system behavior alone but through collaboration between the system and events occurring in its environment, hence the term Collaborative System. Murphy provides risk analysis facilities that include an activity-oriented ontology for carrying out risk analysis by systematically identifying risky activities in the system and in the environment, thereby obtaining a Risk Analysis Graph (RAG) and towards devising risk mitigation strategies later. In order to see both the strengths and weaknesses of Murphy, we experimented on developing a smartphone app involving a group of Ph.D. and senior-level graduate students -one group using Murphy and the other not using Murphy. Our observation, we feel, shows that the risks identified by the group using Murphy were able to identify more critical risks and those risks were comprehensive and relevant as. well. The results also showed that incorporating risk mitigation strategies for the risks identified can indeed help avoid them to some extent.

Preparing a dataset representing business problems is an essential task in Machine Learning (ML). A suitable dataset is critical to accurate ML algorithms, which helps validate business problems. For example, preparing a dataset for predicting loan default in one bank would be vital in the ML project as bank staff may take some actions to mitigate the problem. However, preparing a dataset for identifying potential business problems is challenging. Some challenges might include determining possible events leading to problems, identifying testable factors of the events, and mapping a testable factor to data features to extract relevant data from source data. ML models using irrelevant or unimportant data may give incorrect predictions, negatively impacting problem validation, consequently not solving business problems. We present a goal-oriented approach for preparing an ML dataset to address this challenge. The approach provides an ontology and a process for guiding data preparation. In addition, it helps capture problematic business events, refine a business event to find a testable factor, map a testable factor to a database entity and features, and extract data from a database or Big data. We illustrate the approach using a retail banking application and a Financial database. The experimental results, we believe at least, show that the approach supports preparing a relevant ML dataset, helping validate business problems.

Validating an elicited problem to hinder a business goal is often more important than finding solutions in general. For example, validating the impact of a client's account balance toward an unpaid loan would be critical as a bank can take some actions to mitigate the problem. However, business organizations face difficulties confirming whether some business events or phenomena are causing a problem against a business goal. Some challenges to validate a problem are identifying testable factors for the identified problem, preparing data to validate, analyzing relationships between the factors and a problem, and reasoning the relationships towards high-level problems. Information systems developed to solve unconfirmed problems frequently tackle an erroneous problem, leading to some dissatisfying systems, consequently not achieving business goals. This paper proposes a goal-oriented and machine learningbased approach, Gomphy, for validating a business problem. The Gomphy presents an ontology and a process, a problem-related entity modeling method to identify relevant data features, a data preparation method, and an evaluation method of a problem for high-level problems. To illustrate our approach, we have validated problems behind an unpaid loan in one bank as an empirical study. We feel that at least the proposed approach helps validate business events negatively contributing to a goal, giving some insights about the validated problem.

No abstract