Missing features can negatively impact the performance of machine learning solutions and past research has been focused on how to acquire the most predictive features in static scenarios. Active Feature Acquisition (AFA) for data streams extends the conventional static paradigm by taking into account that the importance of a feature may change as the stream drifts.In this study, we propose an AFA method that takes the cost of the features for each arriving instance into account and, at the same time, allows multiple features to be acquired at once. This reflects the fact that labels often depend on multiple features, so that acquiring many low-cost features may result in higher improvement than acquiring a single feature, as has been proposed in our earlier work [1]. We evaluated our approach on 7 real and 8 synthetic data sets. We investigated three different budget sizes and three different feature cost configurations for 7 different percentages of missing data on each of our data sets. We use the results of our experiments to elaborate on when the acquisition of multiple features is more or less beneficial than acquiring a single feature. Finally, we show the need for more sophisticated metrics to estimate a feature’s predictive quality.

Missing data is a common occurrence in the time series domain, for instance due to faulty sensors, server downtime or patients not attending their scheduled appointments. One of the best methods to impute these missing values is Multiple Imputations by Chained Equations (MICE) which has the drawback that it can only model linear relationships among the variables in a multivariate time series. The advancement of deep learning and its ability to model non-linear relationships among variables make it a promising candidate for time series imputation. This work proposes a modified Convolutional Denoising Autoencoder (CDA) based approach to impute multivariate time series data in combination with a preprocessing step that encodes time series data into 2D images using Gramian Angular Summation Field (GASF). We compare our approach against a standard feed-forward Multi Layer Perceptron (MLP) and MICE. All our experiments were performed on 5 UEA MTSC multivariate time series datasets, where 20 to 50% of the data was simulated to be missing completely at random. The CDA model outperforms all the other models in 4 out of 5 datasets and is tied for the best algorithm in the remaining case.

Stream classification algorithms traditionally treat arriving observations as independent. However, in many applications the arriving examples may depend on the "entity" that generated them, e.g. in product reviewing or in the interactions of users with an application server. In this study, we investigate the potential of this dependency by partitioning the original stream of observations into entity-centric substreams and by incorporating entity-specific information into the learning model. We propose a k Nearest Neighbour inspired stream classification approach (kNN), in which the label of an arriving observation is predicted by exploiting knowledge on the observations belonging to this entity and to entities similar to it. For the computation of entity similarity, we consider knowledge about the observations and knowledge about the entity, potentially transferred from another domain. To distinguish between cases where this kind of knowledge transfer is beneficial for stream classification and cases where the knowledge on the entities does not contribute to classifying the observations, we also propose a heuristic approach based on random sampling of substreams using k Random Entities (kRE). Our learning scenario is not fully supervised: after acquiring labels for the initial few observations of each entity, we assume that no additional labels arrive, and attempt to predict the labels of near-future and far-future observations from that initial seed. We report on our findings from three datasets.