Galactic Air Improves Ancillary Revenues with Dynamic Personalized Pricing

Ancillaries are a rapidly growing source of revenue for airlines, yet their prices are currently statically determined using rules of thumb and are matched only to the average customer or to customer groups. Offering ancillaries at dynamic and personalized prices based on flight characteristics and customer needs could greatly improve airline revenue and customer satisfaction. Through a start-up (Deepair) that builds and deploys novel machine learning techniques to introduce such dynamically priced ancillaries to airlines, we partnered with a major European airline, Galactic Air (pseudonym), to build models and algorithms for improved pricing. These algorithms recommend dynamic personalized ancillary prices for a stream of features (called context) relating to each shopping session. Our recommended prices are restricted to be lower than the human-curated prices for each customer group. We designed and compared multiple machine learning models and deployed the best-performing ones live on the airline’s booking system in an online A/B testing framework. Over a six-month live implementation period, our dynamic pricing system increased the ancillary revenue per offer by 25% and conversion rate by 15% compared with the industry standard of human-curated rule-based prices.

Stress-Testing Framework for Urban Water Systems: A Source to Tap Approach for Stochastic Resilience Assessment

Optimizing the design and operation of an Urban Water System (UWS) faces significant challenges over its lifespan to account for the uncertainties of important stressors that arise from population growth rates, climate change factors, or shifting demand patterns. The analysis of a UWS’s performance across interdependent subsystems benefits from a multi-model approach where different designs are tested against a variety of metrics and in different times scales for each subsystem. In this work, we present a stress-testing framework for UWSs that assesses the system’s resilience, i.e., the degree to which a UWS continues to perform under progressively increasing disturbance (deviation from normal operating conditions). The framework is underpinned by a modeling chain that covers the entire water cycle, in a source-to-tap manner, coupling a water resources management model, a hydraulic water distribution model, and a water demand generation model. An additional stochastic simulation module enables the representation and modeling of uncertainty throughout the water cycle. We demonstrate the framework by “stress-testing” a synthetic UWS case study with an ensemble of scenarios whose parameters are stochastically changing within the UWS simulation timeframe and quantify the uncertainty in the estimation of the system’s resilience.

Internet of Things Testing Framework, Automation, Challenges, Solutions and Practices

The internet of things (IoT) is aimed at modifying the life of people by adopting the possible computing techniques to the physical world, and thus transforming the computing environment from centralized form to decentralized form. Most of the smart devices receive the data from other smart devices over the network and perform actions based on their implemented programs. Thus, testing becomes an intensive process in the IoT that will require some normalization too. The composite architecture of IoT systems and their distinctive characteristics require different variants of testing to be done on the components of IoT systems. This chapter will discuss the necessity for IoT testing in terms of various criteria of identifying and fixing the problems in the IoT systems. In addition, this chapter examines the core components to be focused on IoT testing and testing scope based on IoT device classification. It also elaborates the various types of testing applied on healthcare IoT applications, and finally, this chapter summarizes the various challenges faced during IoT testing.

Rank-sensitive proportional aggregations in dynamic recommendation scenarios

In this paper, we focus on the problem of rank-sensitive proportionality preservation when aggregating outputs of multiple recommender systems in dynamic recommendation scenarios. We believe that individual recommenders may provide complementary views on the user’s preferences or needs, and therefore, their proportional (i.e. unbiased) aggregation may be beneficial for the long-term user satisfaction. We propose an aggregation framework (FuzzDA) based on a modified D’Hondt’s algorithm (DA) for proportional mandates allocation. Specifically, we adjusted DA to register fuzzy membership of items and modified the selection procedure to balance both relevance and proportionality criteria. Furthermore, we propose several iterative votes assignment strategies and negative implicit feedback incorporation strategies to make FuzzDA framework applicable in dynamic recommendation scenarios. Overall, the framework should provide benefits w.r.t. long-term novelty of recommendations, diversity of recommended items as well as overall relevance. We evaluated FuzzDA framework thoroughly both in offline simulations and in online A/B testing. Framework variants outperformed baselines w.r.t. click-through rate (CTR) in most of the evaluated scenarios. Some variants of FuzzDA also provided the best or close-to-best iterative novelty (while maintaining very high CTR). While the impact of the framework variants on user-wise diversity was not so extensive, the trade-off between CTR and diversity seems reasonable.

Security Testing Framework for Web Applications

The growth of web-based applications has increased tremendously from last two decades. While these applications bring huge benefits to society, yet they suffer from various security threats. Although there exist various techniques to ensure the security of web applications, still a large number of applications suffer from a wide variety of attacks and result in financial loses. In this article, a security-testing framework for web applications is proposed with an argument that security of an application should be tested at every stage of software development life cycle (SDLC). Security testing is initiated from the requirement engineering phase using a keyword-analysis phase. The output of the first phase serves as input to the next phase. Different case study applications indicate that the framework assists in early detection of security threats and applying appropriate security measures. The results obtained from the implementation of the proposed framework demonstrated a high detection ratio with a less false-positive rate.

Statistical tests for intra-tumour clonal co-occurrence and exclusivity

Tumour progression is an evolutionary process in which different clones evolve over time, leading to intra-tumour heterogeneity. Interactions between clones can affect tumour evolution and hence disease progression and treatment outcome. Intra-tumoural pairs of mutations that are overrepresented in a co-occurring or clonally exclusive fashion over a cohort of patient samples may be suggestive of a synergistic effect between the different clones carrying these mutations. We therefore developed a novel statistical testing framework, called GeneAccord, to identify such gene pairs that are altered in distinct subclones of the same tumour. We analysed our framework for calibration and power. By comparing its performance to baseline methods, we demonstrate that to control type I errors, it is essential to account for the evolutionary dependencies among clones. In applying GeneAccord to the single-cell sequencing of a cohort of 123 acute myeloid leukaemia patients, we find 1 clonally co-occurring and 8 clonally exclusive gene pairs. The clonally exclusive pairs mostly involve genes of the key signalling pathways.

Automated Testing of Data Survivability and Restorability

Regular data backups are fundamental for protection against cyber-attacks and damage to infrastructure. To ensure a successful restoration, backed up data must be tested regularly for restorability to the company’s current environment. Cloud providers generally test their backedup data, but a testing framework is also required for locally stored files and databases. The paper proposes an automated test framework that validates the continued usability of backed up data for target restoration environments. The framework tests backups of Excel files, MySQL and Postgres databases, PDF documents and flat files.

Robustness Testing Framework For Neural Network Image Classifier

Neural network has made remarkable achievements in the field of image classification, but they are threatened by adversarial examples in the process of application, making the robustness of neural network classifiers face danger. Programs or software based on neural network image classifiers need to undergo rigorous robustness testing before use and promotion, in order to effectively reduce losses and security risks. To comprehensively test the robustness of neural network image classifiers and standardize the test process, starting from the two aspects of generated content and interference intensity, a variety of robustness test sets are constructed, and a robustness testing framework suitable for neural network classifiers is proposed. And the feasibility and effectiveness of the test framework and method are verified by testing LENET-5 and the model reinforced by the adversavial training.