Customer churn is one of the most challenging problems for digital retailers. With significantly higher costs for acquiring new customers than retaining existing ones, knowledge about which customers are likely to churn becomes essential. This paper reports a case study where a data-driven approach to churn prediction is used for predicting churners and gaining insights about the problem domain. The real-world data set used contains approximately 200 000 customers, describing each customer using more than 50 features. In the pre-processing, exploration, modeling and analysis, attributes related to recency, frequency, and monetary concepts are identified and utilized. In addition, correlations and feature importance are used to discover and understand churn indicators. One important finding is that the churn rate highly depends on the number of previous purchases. In the segment consisting of customers with only one previous purchase, more than 75% will churn, i.e., not make another purchase in the coming year. For customers with at least four previous purchases, the corresponding churn rate is around 25%. Further analysis shows that churning customers in general, and as expected, make smaller purchases and visit the online store less often. In the experimentation, three modeling techniques are evaluated, and the results show that, in particular, Gradient Boosting models can predict churners with relatively high accuracy while obtaining a good balance between precision and recall.

While consumers value a free and easy return process, the costs to e-tailers associated with returns are substantial and increasing. Consequently, merchants are now tempted to implement stricter policies, but must balance this against the risk of losing valuable customers. With this in mind, data-driven and algorithmic approaches have been introduced to predict if a certain order is likely to result in a return. In this application paper, a novel approach, combining information about the customer and the order, is suggested and evaluated on a real-world data set from a Swedish e-tailer in men’s fashion. The results show that while the predictive accuracy is rather low, a system utilizing the suggested approach could still be useful. Specifically, it is reasonable to assume that an e-tailer would only act on predicted returns where the confidence is very high, e.g., the top 1–5%. For such predictions, the obtained precision is 0.918–0.969, with an acceptable detection rate.

Successful use of probabilistic classification requires well-calibrated probability estimates, i.e., the predicted class probabilities must correspond to the true probabilities. The standard solution is to employ an additional step, transforming the outputs from a classifier into probability estimates. In this paper, Venn predictors are compared to Platt scaling and isotonic regression, for the purpose of producing well-calibrated probabilistic predictions from decision trees. The empirical investigation, using 22 publicly available datasets, showed that the probability estimates from the Venn predictor were extremely well-calibrated. In fact, in a direct comparison using the accepted reliability metric, the Venn predictor estimates were the most exact on every data set.

This paper considers the problem of modelling the energy behaviour of lock-free concurrent queue data structures. Our main contribution is a way to model the energy behaviour of lock-free queue implementations and parallel applications that use them. Focusing on steady state behaviour we decompose energy behaviour into throughput and power dissipation which can be modeled separately and later recombined into several useful metrics, such as energy per operation. Based on our models, instantiated from synthetic benchmark data, and using only a small amount of additional application specific information, energy and throughput predictions can be made for parallel applications that use the respective data structure implementation. To model throughput we propose a generic model forlock-free queue throughput behaviour, based on combination of the dequeuers' throughput and enqueuers' throughput. To model power dissipation we commonly split the contributions from the various computer components into static, activation and dynamic parts, where only the dynamic part depends on the actual instructions being executed. To instantiate the models a synthetic benchmark explores each queue implementation over the dimensions of processor frequency and number of threads. Finally, we show how to make predictions of application throughput and power dissipation for a parallel application using lock-free queue requiring only a limited amount of information about the application work done between queue operations. Our case study on a Mandelbrot application shows convincing prediction results.

Bulk operations on data structures are widely used both on user-level but also on programming language level. Iterations are a good example of such bulk operations. In the sequential setting iterations are easy to design on top of an algorithmic construction of a data structure and is not considered as a challenge. In a concurrent environment, such as a multicore system, the situation is completely different and the issue of extending concurrent data structure designs to support iteration operations opens new research challenges in concurrent algorithmic data structure implementations, with respect to consistency and efficiency. In this paper we take a journey through this young and evolving research topic. More precisely we describe recent advances in the area together with an overview of iteration implementations that have appeared in the research literature as well as in widely-used programming environments and we outline a range of application targets and challenging future directions.

Concurrent data structures provide the means to multi-threaded applications to share data. Data structures come with a set of predefined operations, specified by the semantics of the data structure. In the literature and in several contemporary commonly used programming environments, the notion of iteration has been introduced for collection data structures, as a bulk operation enhancing the native set of operations. Iterations in several of these contexts have been treated as sequential in nature and may provide weak consistency guarantees when running concurrently with the native operations of the data structures. In this work we study iterations in concurrent data structures in the context of concurrency with the native operations and the guarantees that they provide. Besides linearizability, we propose a set of consistency specifications for such bulk operations, including also concurrency-aware properties by building on Lamport’s systematic definitions for registers. Furthermore, by using queues and composite registers as case-studies of underlying objects, we provide a set of constructions of iteration operations, satisfying the properties and showing containment relations. Besides the trade-off between consistency and throughput, we demonstrate the trade-off between the overhead of the bulk operation and possible support (helping) by the native operations of the data structure. We describe a set of algorithms that demonstrate these and study the implications on the efficiency of the implementations.

Group communication is deployed in many evolving Internet-scale cooperative applications such as multiplayer online games and virtual worlds to efficiently support interaction on information relevant to a potentially very large number of users or objects. Especially peer-to-peer based group communication protocols have evolved as a promising approach to allow intercommunication between many distributed peers. Yet, the delivery semantics of robust and scalable protocols such as gossiping is not sufficient to support consistency semantics beyond eventual consistency because no relationship on the order of events is enforced. On the other hand, traditional consistency models provided by reliable group communication providing causal or even total order are restricted to support only small groups. This article proposes the cluster consistency model which bridges the gap between traditional and current approaches in supporting both scalability and ordered event delivery. We introduce a dynamic and fault tolerant cluster management method that can coordinate concurrent access to resources in a peer-to-peer system and can be used to establish fault-tolerant configurable cluster consistency with predictable reliability, running on top of decentralised probabilistic protocols supporting scalable group communication. This is achieved by a general two-layered architecture that can be applied on top of the standard Internet communication layers and offers a modular, layered set of services to the applications that need them. Further, we present a fault-tolerant method implementing causal cluster consistency with predictable reliability, running on top of decentralised probabilistic protocols supporting group communication. This paper provides analytical and experimental evaluation of the properties regarding the fault tolerance of the approach. Furthermore, our experimental study, conducted by implementing and evaluating the two-layered architecture on top of standard Internet transport services, shows that the approach scales well, imposes an even load on the system, and provides high-probability reliability guarantees.

A lock-free bag data structure supporting unordered buffering is presented in this paper. The algorithm supports multiple producers and multiple consumers, as well as dynamic collection sizes. To handle concurrency efficiently, the algorithm was designed to thrive for disjoint-access-parallelism for the supported semantics. Therefore, the algorithm exploits a distributed design combined with novel techniques for handling concurrent modifications of linked lists using double marks, detection of total emptiness, and efficient memory management with hazard pointer handover. Experiments on a 24-way multi-core platform show significantly better performance for the new algorithm compared to previous algorithms of relevance.

A lock-free FIFO queue data structure is presented in this paper. The algorithm supports multiple producers and multiple consumers and weak memory models. It has been designed to be cache-aware and work directly on weak memory models. It utilizes the cache behavior in concert with lazy updates of shared data, and a dynamic lock-free memory management scheme to decrease unnecessary synchronization and increase performance. Experiments on an 8-way multi-core platform show significantly better performance for the new algorithm compared to previous fast lock-free algorithms.

Efficient, scalable memory allocation for multithreaded applications on multiprocessors is a significant goal of recent research. In the distributed computing literature it has been emphasized that lock-based synchronization and concurrency-control may limit the parallelism in multiprocessor systems. Thus, system services that employ such methods can hinder reaching the full potential of these systems. A natural research question is the pertinence and the impact of lock-free concurrency control in key services for multiprocessors, such as in the memory allocation service, which is the theme of this work. We show the design and implementation of NBmalloc, a lock-free memory allocator designed to enhance the parallelism in the system. The architecture of NBmalloc is inspired by Hoard, a well-known concurrent memory allocator, with modular design that preserves scalability and helps avoiding false-sharing and heap-blowup. Within our effort to design appropriate lock-free algorithms for NBmalloc, we propose and show a lock-free implementation of a new data structure, flat-set, supporting conventional “internal” set operations as well as “inter-object” operations, for moving items between flat-sets. The design of NBmalloc also involved a series of other algorithmic problems, which are discussed in the paper. Further, we present the implementation of NBmalloc and a study of its behaviour in a set of multiprocessor systems. The results show that the good properties of Hoard w.r.t. false-sharing and heap-blowup are preserved.