Sensor fusion is the use of sensory data from multiple sources, combined into one comprehensive result for more accurate, complete, and dependable, business intelligence and security applications. "IoT devices, security cameras and other data capture sensors are practically everywhere," said Nadir Ali, CEO of Inpixon. "The challenge is to filter the information captured by those devices, so it can be processed and analyzed in a meaningful way. Inpixon has extensive experience in radio frequency data fusion, which we believe we can leverage due to the similarities between radio waves and the light waves captured by CCTV cameras. "Location is the lynchpin," continued Mr. Ali. "To know what's going on in your building -- for security purposes, for sales or customer service purposes, or for applications like location-based marketing or augmented reality -- you must know the location of persons and electronic devices in your space.

Due to its high computational speed and accuracy compared to ab-initio quantum chemistry and forcefield modeling, the prediction of molecular properties using machine learning has received great attention in the fields of materials design and drug discovery. A main ingredient required for machine learning is a training dataset consisting of molecular features\textemdash for example fingerprint bits, chemical descriptors, etc. that adequately characterize the corresponding molecules. However, choosing features for any application is highly non-trivial. No "universal" method for feature selection exists. In this work, we propose a data fusion framework that uses Independent Vector Analysis to exploit underlying complementary information contained in different molecular featurization methods, bringing us a step closer to automated feature generation. Our approach takes an arbitrary number of individual feature vectors and automatically generates a single, compact (low dimensional) set of molecular features that can be used to enhance the prediction performance of regression models. At the same time our methodology retains the possibility of interpreting the generated features to discover relationships between molecular structures and properties. We demonstrate this on the QM7b dataset for the prediction of several properties such as atomization energy, polarizability, frontier orbital eigenvalues, ionization potential, electron affinity, and excitation energies. In addition, we show how our method helps improve the prediction of experimental binding affinities for a set of human BACE-1 inhibitors. To encourage more widespread use of IVA we have developed the PyIVA Python package, an open source code which is available for download on Github.

Abstract--Vehicle location prediction or vehicle tracking is a significant topic within connected vehicles. This task, however, is difficult if only a single modal data is available, probably causing bias and impeding the accuracy. With the development of sensor networks in connected vehicles, multimodal data are becoming accessible. Therefore, we propose a framework for vehicle tracking with multimodal data fusion. Images, being processed in the module of vehicle detection, provide direct information about the features of vehicles, whereas velocity estimation can further evaluate the possible location of the target vehicles, which reduces the number of features being compared, and decreases the time consumption and computational cost. Vehicle detection is designed with a color-faster R-CNN, which takes both the shape and color of the vehicles into consideration. Meanwhile, velocity estimation is through the Kalman filter, which is a classical method for tracking. Finally, a multimodal data fusion method is applied to integrate these outcomes so that vehicle-tracking tasks can be achieved. Experimental results suggest the efficiency of our methods, which can track vehicles using a series of surveillance cameras in urban areas. ITH technological advancements in vehicles and transportation system, motorists require comfort and intelligent driving, not only mobility. Thus, there has been a great deal of research which mainly falls into one of two directions. On one hand, researchers tend to develop more intelligent vehicles, or devices that can be attached to vehicles, bringing up several popular topics such as autonomous vehicles or driverless vehicles [1].

When it comes to content intelligence, it is applied in the form of data recognition and tracking. For example, this could involve a CRM that anticipates a contact's likelihood to close or a platform anticipating why your content is driving engagement. This information can then be used to inform decisions and strategize more effectively.

Two years ago when I reported from Strata NYC I shared how new data prep technologies were bringing data integration and data quality capabilities to business users. Instead of waiting for IT to ETL experimental data into the data warehouse or data lake, business users apply tools like Tableau Prep, Trifacta Wrangler, or Talend Data Preparation, to perform profiling, cleansing, and integrating data on their own. This trend continued at last week's Strata Data Conference as vendors are expanding their capabilities and services inby applying their technologies to business specific use cases. Informatica demoed me Informatica Data Catalog embedded with Claire, an "AI inside" that self classifies data and makes it easier for end users to find data sources and subject matter experts. The CEO of MemGraph demonstrated a real time graph database used to find anomalies and customer relationships.

Another use case that is worth pointing out is knowledge graph. This entails building a graph representation to integrate all of the data to support a specific use case. Knowledge graphs can be narrow and support a specific application or they can be as wide as a data lake or data warehouse that attempts to keep track of everything. The problem with knowledge graphs is how to get all of the data identified and then do an ETL process that adds it to the knowledge graph in a coherent way. It is nearly impossible to do this without understanding the relationships between data and identifying when new data is relevant to the data you have in the graph.

In this Book we argue that the fruitful interaction of computer vision and belief calculus is capable of stimulating significant advances in both fields. From a methodological point of view, novel theoretical results concerning the geometric and algebraic properties of belief functions as mathematical objects are illustrated and discussed in Part II, with a focus on both a perspective 'geometric approach' to uncertainty and an algebraic solution to the issue of conflicting evidence. In Part III we show how these theoretical developments arise from important computer vision problems (such as articulated object tracking, data association and object pose estimation) to which, in turn, the evidential formalism is able to provide interesting new solutions. Finally, some initial steps towards a generalization of the notion of total probability to belief functions are taken, in the perspective of endowing the theory of evidence with a complete battery of estimation and inference tools to the benefit of all scientists and practitioners.

Connection between the theory of aggregation functions and formal concept analysis is discussed and studied, thus filling a gap in the literature by building a bridge between these two theories, one of them living in the world of data fusion, the second one in the area of data mining. We show how Galois connections can be used to describe an important class of aggregation functions preserving suprema, and, by duality, to describe aggregation functions preserving infima. Our discovered method gives an elegant and complete description of these classes. Also possible applications of our results within certain biclustering fuzzy FCA-based methods are discussed.

Sensor fusion is a key technology that integrates various sensory inputs to allow for robust decision making in many applications such as autonomous driving and robot control. Deep neural networks have been adopted for sensor fusion in a body of recent studies. Among these, the so-called netgated architecture was proposed, which has demonstrated improved performances over the conventional convolutional neural networks (CNN). In this paper, we address several limitations of the baseline negated architecture by proposing two further optimized architectures: a coarser-grained gated architecture employing (feature) group-level fusion weights and a two-stage gated architectures leveraging both the group-level and feature level fusion weights. Using driving mode prediction and human activity recognition datasets, we demonstrate the significant performance improvements brought by the proposed gated architectures and also their robustness in the presence of sensor noise and failures.

Retention marketing is about preventing your valuable customers from churning. Reducing customer churn requires you to know two things: 1) which customers are about to churn and 2) which remedies will keep them from churning. In this paper, I show you how marketers can improve their customer retention efforts by 1) integrating disparate data silos and 2) employing machine learning predictive analytics. Your marketing strategy is only as good as your ability to deliver measurable results. In our world of Big Data, marketers no longer need to simply rely on their gut instincts to make marketing decisions.