Pradip Mainali

Pradip Mainali, Ismini Psychoula, Fabien A. P. Petitcolas

In this paper, we present a novel method to compute decision rules to build a more accurate interpretable machine learning model, denoted as ExMo. The ExMo interpretable machine learning model consists of a list of IF...THEN... statements with a decision rule in the condition. This way, ExMo naturally provides an explanation for a prediction using the decision rule that was triggered. ExMo uses a new approach to extract decision rules from the training data using term frequency-inverse document frequency (TF-IDF) features. With TF-IDF, decision rules with feature values that are more relevant to each class are extracted. Hence, the decision rules obtained by ExMo can distinguish the positive and negative classes better than the decision rules used in the existing Bayesian Rule List (BRL) algorithm, obtained using the frequent pattern mining approach. The paper also shows that ExMo learns a qualitatively better model than BRL. Furthermore, ExMo demonstrates that the textual explanation can be provided in a human-friendly way so that the explanation can be easily understood by non-expert users. We validate ExMo on several datasets with different sizes to evaluate its efficacy. Experimental validation on a real-world fraud detection application shows that ExMo is 20% more accurate than BRL and that it achieves accuracy similar to those of deep learning models.

Ismini Psychoula, Andreas Gutmann, Pradip Mainali et al.

The application of machine learning to support the processing of large datasets holds promise in many industries, including financial services. However, practical issues for the full adoption of machine learning remain with the focus being on understanding and being able to explain the decisions and predictions made by complex models. In this paper, we explore explainability methods in the domain of real-time fraud detection by investigating the selection of appropriate background datasets and runtime trade-offs on both supervised and unsupervised models.

Pradip Mainali, Carlton Shepherd

Motion-based fall detection systems are concerned with detecting falls from vulnerable users, which is typically performed by classifying measurements from a body-worn inertial measurement unit (IMU) using machine learning. Such systems, however, necessitate the collection of high-resolution measurements that may violate users' privacy, such as revealing their gait, activities of daily living (ADLs), and relative position using dead reckoning. In this paper, we investigate the application of multi-party computation (MPC) to IMU-based fall detection for protecting device measurement confidentiality. Our system is evaluated in a cloud-based setting that precludes parties from learning the underlying data using multiple, disparate cloud instances deployed in three geographical configurations. Using a publicly-available dataset, we demonstrate that MPC-based fall detection from IMU measurements is practical while achieving state-of-the-art error rates. In the best case, our system executes in 365.2 milliseconds, which falls well within the required time window for on-device data acquisition (750ms).

Pradip Mainali, Carlton Shepherd, Fabien A. P. Petitcolas

This paper proposes a new privacy-enhancing, context-aware user authentication system, ConSec, which uses a transformation of general location-sensitive data, such as GPS location, barometric altitude and noise levels, collected from the user's device, into a representation based on locality-sensitive hashing (LSH). The resulting hashes provide a dimensionality reduction of the underlying data, which we leverage to model users' behaviour for authentication using machine learning. We present how ConSec supports learning from categorical and numerical data, while addressing a number of on-device and network-based threats. ConSec is implemented subsequently for the Android platform and evaluated using data collected from 35 users, which is followed by a security and privacy analysis. We demonstrate that LSH presents a useful approach for context authentication from location-sensitive data without directly utilising plain measurements.