Editor's note: this post is part of an ongoing series exploring developments in artificial intelligence. First, collect huge amounts of training data -- probably more than anyone thought sensible or even possible a decade ago. Second, massage and preprocess that data so the key relationships it contains are easily accessible (the jargon here is "feature engineering"). Finally, feed the result into ludicrously high-performance, parallelized implementations of pretty standard machine-learning methods like logistic regression, deep neural networks, and k-means clustering (don't worry if those names don't mean anything to you -- the point is that they're widely available in high-quality open source packages). Google pioneered this formula, applying it to ad placement, machine translation, spam filtering, YouTube recommendations, and even the self-driving car -- creating billions of dollars of value in the process.

Cluster Analysis is a common technique to group a set of objects in the way that the objects in the same group share certain attributes. It's commonly used in marketing and sales planning to define market segmentations. Here at BigObject we adopt a simple approach to exploring the similarities between objects. We simply calculate the "Feature Vector" based on given attributes and use the score to determine which objects are "alike." This is a simple example to show how to use BigObject to extract product features and then find similar products in your retail data.

Modeling occupant behavior in smart buildings to reduce energy usage in a more accurate fashion has garnered much recent attention in the literature. Predicting occupant comfort in buildings is a related and challenging problem. In some smart buildings, such as NASA AMES Sustainability Base, there are discrepancies between occupants' actual thermal discomfort and sensors based upon a weighted average of wet bulb, dry bulb, and mean radiant temperature intended to characterize thermal comfort. In this paper we attempt to find other contributing factors to occupant discomfort. For our experiment we use a dataset from a Building Automation System (BAS) in NASA Sustainability Base. We choose one conference room for our experiment and empirically establish the thermal discomfort level for the room's temperature sensor. We use various causality metrics and causal graphs to isolate candidate causes of the target room temperature. And we compare these feature sets according to their predictive capability of future instances of discomfort. Moreover, we establish a trade off between computational and statistical performance of adverse event prediction.

Utilities use Demand Response (DR) to balance supply and demand in the electric grid by involving customers in efforts to reduce electricity consumption during peak periods. To implement and adapt DR under dynamically changing conditions of the grid, reliable prediction of reduced consumption is critical. However, despite the wealth of research on electricity consumption prediction and DR being long in practice, the problem of reduced consumption prediction remains largely un-addressed. In this paper, we identify unique computational challenges associated with the prediction of reduced consumption and contrast this to that of normal consumption and DR baseline prediction. We propose a novel ensemble model that leverages different sequences of daily electricity consumption on DR event days as well as contextual attributes for reduced consumption prediction. We demonstrate the success of our model on a large, real-world, high resolution dataset from a university microgrid comprising of over 950 DR events across a diverse set of 32 buildings. Our model achieves an average error of 13.5%, an 8.8% improvement over the baseline. Our work is particularly relevant for buildings where electricity consumption is not tied to strict schedules. Our results and insights should prove useful to the researchers and practitioners working in the sustainable energy domain.

We propose a new model for learning bilingual word representations from non-parallel document-aligned data. Following the recent advances in word representation learning, our model learns dense real-valued word vectors, that is, bilingual word embeddings (BWEs). Unlike prior work on inducing BWEs which heavily relied on parallel sentence-aligned corpora and/or readily available translation resources such as dictionaries, the article reveals that BWEs may be learned solely on the basis of document-aligned comparable data without any additional lexical resources nor syntactic information. We present a comparison of our approach with previous state-of-the-art models for learning bilingual word representations from comparable data that rely on the framework of multilingual probabilistic topic modeling (MuPTM), as well as with distributional local context-counting models. We demonstrate the utility of the induced BWEs in two semantic tasks: (1) bilingual lexicon extraction, (2) suggesting word translations in context for polysemous words. Our simple yet effective BWE-based models significantly outperform the MuPTM-based and context-counting representation models from comparable data as well as prior BWE-based models, and acquire the best reported results on both tasks for all three tested language pairs.

With the advances in the field of medical informatics, automated clinical decision support systems are becoming the de facto standard in personalized diagnosis. In order to establish high accuracy and confidence in personalized diagnosis, massive amounts of distributed, heterogeneous, correlated and high-dimensional patient data from different sources such as wearable sensors, mobile applications, Electronic Health Record (EHR) databases etc. need to be processed. This requires learning both locally and globally due to privacy constraints and/or distributed nature of the multi-modal medical data. In the last decade, a large number of meta-learning techniques have been proposed in which local learners make online predictions based on their locally-collected data instances, and feed these predictions to an ensemble learner,which fuses them and issues a global prediction. However, most of these works do not provide performance guarantees or, when they do,these guarantees are asymptotic. None of these existing works provide confidence estimates about the issued predictions or rate of learning guarantees for the ensemble learner. In this paper, we provide a systematic ensemble learning method called Hedged Bandits, which comes with both long run (asymptotic) and short run (rate of learning) performance guarantees. Moreover, we show that our proposed method outperforms all existing ensemble learning techniques, even in the presence of concept drift.

The hospital readmission rate of patients within 30 days after discharge is broadly accepted as a healthcare quality measure and cost driver in the United States. The ability to estimate hospitalization costs alongside 30 day risk-stratification for such readmissions provides additional benefit for accountable care, now a global issue and foundation for the U.S.~government mandate under the Affordable Care Act. Recent data mining efforts either predict healthcare costs or risk of hospital readmission, but not both. In this paper we present a dual predictive modeling effort that utilizes healthcare data to predict the risk and cost of any hospital readmission (``all-cause''). For this purpose, we explore machine learning algorithms to do accurate predictions of healthcare costs and risk of 30-day readmission.Results on risk prediction for ``all-cause'' readmission compared to the standardized readmission tool (LACE) are promising, and the proposed techniques for cost prediction consistently outperform baseline models and demonstrate substantially lower mean absolute error (MAE).

Multiple concurrent physiological streams generated by various medical devices play important roles in patient condition assessment. However, these physiological streams needto be analyzed together and output in real-time for preciseand timely controlling and management, which poses a non-trivial challenge to existing methods. This paper presents ourresearch on real-time assessing based on this kind of data.To address this problem, we first extract sketches from original data with the help of adaptive sampling and wave splittingalgorithm, then define scalable operators on sketches and propose MUNCA (MUlti-dimensional Nearest Center Analysis)to combine these multiple concurrent data together for anal-ysis. Experiments on real data demonstrate the effectiveness and efficiency of the proposed method.

Electricity disaggregation focuses on classification ofindividual appliances by monitoring aggregate electricalsignals. In this paper we present a novel algorithmto automatically correct labels, discard contaminatedtraining samples, and boost signal to noise ratio throughhigh frequency noise reduction. We also propose amethod for prioritized classification which classifies applianceswith the most intense signals first. When testedon four houses in Kaggles Belkin dataset, these methodsautomatically relabel over 77% of all training samplesand decrease error rate by an average of 45% in bothreal power and high frequency noise classification.

Despite the widespread installation of accelerometers in almost all mobile phones and wearable devices, activity recognition using accelerometers is still immature due to the poor recognition accuracy of existing recognition methods and the scarcity of labeled training data. We consider the problem of human activity recognition using triaxial accelerometers and deep learning paradigms. This paper shows that deep activity recognition models (a) provide better recognition accuracy of human activities, (b) avoid the expensive design of handcrafted features in existing systems, and (c) utilize the massive unlabeled acceleration samples for unsupervised feature extraction. We show substantial recognition improvement on real world datasets over state-of-the-art methods of human activity recognition using triaxial accelerometers.