No abstract

There exist a number of large legacy systems that still undergo continuous maintenance and enhancement. Due to the sheer size and complexity of the software systems and limited resources, managers are confronted with crucial decisions regarding allocation and training of new engineers, intelligent allocation of testing personnel, assessment of release readiness of the software and so on. While the area of bug prediction by mining software repositories holds promise, and is a worthwhile endeavor, the current state of the art techniques are not accurate enough in predicting bugs and hence are of limited usefulness to managers. So instead of predicting files as buggy or not we take a different viewpoint and focus on providing decision support for managers. In this paper we present a set of metrics to guide the managers in taking these decisions. These metrics are evaluated using 4 open source systems and 2 proprietary systems.

Occurrence of multiple seizures is a common phenomenon observed in patients with epilepsy: a neurological malfunction that affects approximately 50 million people in the world. Seizure prediction is widely acknowledged as an important problem in the neurological domain, as it holds promise to improve the quality of life for patients with epilepsy. A noticeable number of clinical studies showed evidence of symptoms (patterns) before seizures and thus, there is large research on predicting seizures. There is very little existing literature that systematically illustrates the steps in machine learning for seizure prediction, limited training data and class imbalance are a few challenges. In this paper, we propose a novel way to overcome these challenges. We present the improved results for various classification algorithms. An average of 21.71% improvement in accuracy is attained using our approach.

No abstract