We propose a novel MLaaS Dataset Generator (MDG) framework that creates configurable and reproducible datasets for evaluating Machine Learning as a Service (MLaaS) selection and composition. MDG simulates realistic MLaaS behaviour by training and evaluating diverse model families across multiple real-world datasets and data distribution settings. It records detailed functional attributes, quality of service metrics, and composition-specific indicators, enabling systematic analysis of service performance and cross-service behaviour. Using MDG, we generate more than ten thousand MLaaS service instances and construct a large-scale benchmark dataset suitable for downstream evaluation. We also implement a built-in composition mechanism that models how services interact under varied Internet of Things conditions. Experiments demonstrate that datasets generated by MDG enhance selection accuracy and composition quality compared to existing baselines. MDG provides a practical and extensible foundation for advancing data-driven research on MLaaS selection and composition.

The dynamic nature of Internet of Things (IoT) environments challenges the long-term effectiveness of Machine Learning as a Service (MLaaS) compositions. The uncertainty and variability of IoT environments lead to fluctuations in data distribution, e.g., concept drift and data heterogeneity, and evolving system requirements, e.g., scalability demands and resource limitations. This paper proposes an adaptive MLaaS composition framework to ensure a seamless, efficient, and scalable MLaaS composition. The framework integrates a service assessment model to identify underperforming MLaaS services and a candidate selection model to filter optimal replacements. An adaptive composition mechanism is developed that incrementally updates MLaaS compositions using a contextual multi-armed bandit optimization strategy. By continuously adapting to evolving IoT constraints, the approach maintains Quality of Service (QoS) while reducing the computational cost associated with recomposition from scratch. Experimental results on a real-world dataset demonstrate the efficiency of our proposed approach.

We propose a novel change detection framework to identify changes in the long-term performance behavior of an Infrastructure as a Service (IaaS). An IaaS’s long-term performance behavior is represented by an IaaS performance signature. The proposed framework leverages time series similarity measures and a sliding window technique to detect changes in IaaS performance signatures. We introduce a new IaaS performance noise model that enables the proposed framework to distinguish between performance noise and actual changes in performance. The proposed framework utilizes a novel Signal-to-Noise Ratio-based approach to detect changes when prior knowledge about performance noise is available. A set of experiments is conducted using real-world datasets to demonstrate the effectiveness of the proposed change detection framework.

We propose a novel framework to select IaaS providers according to a consumer’s long-term performance requirements. The proposed framework leverages free short-term trials to discover the unknown QoS performance of IaaS providers. We design a temporal skyline-based filtering method to select candidate IaaS providers for the short-term trials. A novel cooperative long-term QoS prediction approach is developed that utilizes past trial experiences of similar consumers using a workload replay technique. We propose a new trial workload generation model that estimates a provider’s long-term performance in the absence of past trial experiences. The confidence of the prediction is measured based on the trial experience of the consumer. A set of experiments are conducted based on real-world datasets to evaluate the proposed framework.

We propose a novel Infrastructure-as-a-Service composition framework that selects an optimal set of consumer requests according to the provider’s qualitative preferences on long-term service provisions. Decision variables are included in the temporal conditional preference networks to represent qualitative preferences for both short-term and long-term consumers. The global preference ranking of a set of requests is computed using a k -d tree indexing-based temporal similarity measure approach. We propose an extended three-dimensional Q-learning approach to maximize the global preference ranking. We design the on-policy-based sequential selection learning approach that applies the length of request to accept or reject requests in a composition. The proposed on-policy-based learning method reuses historical experiences or policies of sequential optimization using an agglomerative clustering approach. Experimental results prove the feasibility of the proposed framework.

We propose a novel approach to select IaaS cloud services for a long-term period where the service providers offer limited QoS information. The proposed approach leverages free short-term trials to obtain the previously undisclosed QoS information. A new significance-based trial scheme is proposed using frequency distribution analysis to test a consumer's long-term workloads in a short trial. We introduce a novel IaaS signature technique to uniquely identify the variability of a provider's QoS performance. A Signature-based QoS Performance Discovery (SPD) algorithm is proposed which leverages the combination of free trials and IaaS signatures. A set of exhaustive experiments with real-world datasets is conducted to evaluate the proposed approach.

We propose a novel approach to select privacy-sensitive IaaS providers for a long-term period. The proposed approach leverages a consumer's short-term trial experiences for long-term selection. We design a novel equivalence partitioning based trial strategy to discover the temporal and unknown QoS performance variability of an IaaS provider. The consumer's long-term workloads are partitioned into multiple Virtual Machines in the short-term trial. We propose a performance fingerprint matching approach to ascertain the confidence of the consumer's trial experience. A trial experience transformation method is proposed to estimate the actual long-term performance of the provider. Experimental results with real-world datasets demonstrate the efficiency of the proposed approach.