Characterization and Prediction of Issue-Related Risks in Software Projects

Abstract

Abstract-Identifying risks relevant to a software project and planning measures to deal with them are critical to the success of the project. Current practices in risk assessment mostly rely on high-level, generic guidance or the subjective judgements of experts. In this paper, we propose a novel approach to risk assessment using historical data associated with a software project. Specifically, our approach identifies patterns of past events that caused project delays, and uses this knowledge to identify risks in the current state of the project. A set of risk factors characterizing "risky" software tasks (in the form of issues) were extracted from five open source projects: Apache, Duraspace, JBoss, Moodle, and Spring. In addition, we performed feature selection using a sparse logistic regression model to select risk factors with good discriminative power. Based on these risk factors, we built predictive models to predict if an issue will cause a project delay. Our predictive models are able to predict both the risk impact (i.e. the extend of the delay) and the likelihood of a risk occurring. The evaluation results demonstrate the effectiveness of our predictive models, achieving on average 48%-81% precision, 23%-90% recall, 29%-71% F-measure, and 70%-92% Area Under the ROC Curve. Our predictive models also have low error rates: 0.39-0.75 for Macroaveraged Mean Cost-Error and and 0.7-1.2 for Macro-averaged Mean Absolute Error.

Automatic Classifying Self-Admitted Technical Debt Using N-Gram IDF

“…Since the confusion matrix does not consider the multi-class probabilistic classification and cost of each SATD class, we utilize a performance measure namely Macro-Averaged Mean Cost-Error or MMCE proposed in [34]. MMCE assesses how close predicted class probabilities are to the actual classes or MMCE measures how effectively a classification model classify types of SATD.…”

Multi-objective search-based approach to estimate issue resolution time

“…[7,30,33,34] use machine learning techniques (e.g. kNN in [32] or Random Forests in [3,16,21]) to build their prediction models. For example, the work in [32] estimates the fixing time of a bug by finding the previous bugs that have similar description to the given bug (using text similar techniques) and using the known time of fixing those previous bugs.…”

The path to success

“…In knowledge creation and other crowd-sourcing communities, the quality and quantity of user contributions are important [36,53]. Software development communities aim for high quality as well, but also for functionality of the produced software [15], with a focus on timeliness of the results because of delivery deadlines [11]. Similarly, Questions & Answers communities have the primary purpose of providing information seekers with a platform to formulate their questions and receive timely and accurate solutions [17,33].…”