In the era of Big Data, knowledge integration is key for tasks such as social media aggregation, opinion mining, and cyber-issue detection. The integration of different kinds of knowledge coming from multiple sources, however, is often a problematic issue as it either requires a lot of manual effort in defining aggregation rules or suffers from noise generated by automatic integration techniques. In this work, we propose a method based on conceptual primitives for efficiently integrating pieces of knowledge coming from different common and common-sense resources, which we test in the field of concept-level sentiment analysis.

The objective is to present one important aspect of the European IST-FET project "REV!GIS"1: the methodology which has been developed for the translation (interpretation) of the quality of the data into a "fitness for use" information, that we can confront to the user needs in its application. This methodology is based upon the notion of "ontologies" as a conceptual framework able to capture the explicit and implicit knowledge involved in the application. We do not address the general problem of formalizing such ontologies, instead, we rather try to illustrate this with three applications which are particular cases of the more general "data fusion" problem. In each application, we show how to deploy our methodology, by comparing several possible solutions, and we try to enlighten where are the quality issues, and what kind of solution to privilege, even at the expense of a highly complex computational approach. The expectation of the REV!GIS project is that computationally tractable solutions will be available among the next generation AI tools.

How does the human visual system compute the speed of a coherent motion stimulus that contains motion energy in different spatiotemporal frequency bands? Here we propose that perceived speed is the result of optimal integration of speed information from independent spatiotemporal frequency tuned channels. We formalize this hypothesis with a Bayesian observer model that treats the channel activity as independent cues, which are optimally combined with a prior expectation for slow speeds. We test the model against behavioral data from a 2AFC speed discrimination task with which we measured subjects' perceived speed of drifting sinusoidal gratings with different contrasts and spatial frequencies, and of various combinations of these single gratings. We find that perceived speed of the combined stimuli is independent of the relative phase of the underlying grating components, and that the perceptual biases and discrimination thresholds are always smaller for the combined stimuli, supporting the cue combination hypothesis. The proposed Bayesian model fits the data well, accounting for perceptual biases and thresholds of both simple and combined stimuli. Fits are improved if we assume that the channel responses are subject to divisive normalization, which is in line with physiological evidence. Our results provide an important step toward a more complete model of visual motion perception that can predict perceived speeds for stimuli of arbitrary spatial structure.

Psychophysical experiments have demonstrated that the brain integrates information from multiple sensory cues in a near Bayesian optimal manner. The present study proposes a novel mechanism to achieve this. We consider two reciprocally connected networks, mimicking the integration of heading direction information between the dorsal medial superior temporal (MSTd) and the ventral intraparietal (VIP) areas. Each network serves as a local estimator and receives an independent cue, either the visual or the vestibular, as direct input for the external stimulus. We find that positive reciprocal interactions can improve the decoding accuracy of each individual network as if it implements Bayesian inference from two cues. Our model successfully explains the experimental finding that both MSTd and VIP achieve Bayesian multisensory integration, though each of them only receives a single cue as direct external input. Our result suggests that the brain may implement optimal information integration distributively at each local estimator through the reciprocal connections between cortical regions.

Much of the focus on big data has been on the problem of processing very large sources.   There is an equally hard problem of how to normalize, integrate, and transform the data from many sources into the format required to run large-scale analysis and visualization tools.  We have previously developed an approach to semi-automatically mapping diverse sources into a shared domain ontology so that they can be quickly combined.  In this paper we describe our approach to building and executing integration and restructuring plans to support analysis and visualization tools on very large and diverse datasets.

In this paper, we describe a complex Soar agent that uses and learns multiple types of knowledge while interacting with a human in a real-world domain. Our hypothesis is that a diverse set of memories is required for the different types of knowledge. We first present the agent’s processing, highlighting the types of knowledge used for each phase. We then present Soar’s memories and identify which memory is used for each type of knowledge. We also analyze which properties of each memory make it appropriate for the knowledge it encodes. We conclude with a summary of our analysis.

In this paper we consider optimization as an approach for quickly and flexibly developing hybrid cognitive capabilities that are efficient, scalable, and can exploit knowledge to improve solution speed and quality. In this context, we focus on the Three-Weight Algorithm, which aims to solve general optimization problems. We propose novel methods by which to integrate knowledge with this algorithm to improve expressiveness, efficiency, and scaling, and demonstrate these techniques on two example problems (Sudoku and circle packing).

This paper evaluates heterogeneous information fusion using multi-task Gaussian processes in the context of geological resource modeling. Specifically, it empirically demonstrates that information integration across heterogeneous information sources leads to superior estimates of all the quantities being modeled, compared to modeling them individually. Multi-task Gaussian processes provide a powerful approach for simultaneous modeling of multiple quantities of interest while taking correlations between these quantities into consideration. Experiments are performed on large scale real sensor data.

Traffic sensing is a key baseline input for sustainablecities to plan and administer demand-supplymanagement through better road networks, publictransportation, urban policies etc., Humans sensethe environment frugally using a combination ofcomplementary information signals from differentsensors. For example, by viewing and/or hearingtraffic one could identify the state of traffic on theroad. In this paper, we demonstrate a fusion basedlearning approach to classify the traffic states usinglow cost audio and image data analysis using realworld dataset. Road side collected traffic acousticsignals and traffic image snapshots obtained fromfixed camera are used to classify the traffic conditioninto three broad classes viz., Jam, Mediumand Free. The classification is done on f10sec audio,image snapshot in that 10secg data tuple. Weextract traffic relevant features from audio and imagedata to form a composite feature vector. Inparticular, we extract the audio features comprisingMFCC (Mel-Frequency Cepstral Coefficients)classifier based features, honk events and energypeaks. A simple heuristic based image classifier isused, where vehicular density and number of cornerpoints within the road segment are estimated andare used as features for traffic sensing. Finally thecomposite vector is tested for its ability to discriminatethe traffic classes using Decision tree classifier,SVM classifier, Discriminant classifier and Logisticregression based classifier. Information fusion atmultiple levels (audio, image, overall) shows consistentlybetter performance than individual leveldecision making. Low cost sensor fusion based oncomplementary weak classifiers and noisy featuresstill generates high quality results with an overallaccuracy of 93 - 96%.

Expert finding is an information retrieval task concerned with the search for the most knowledgeable people, in some topic, with basis on documents describing peoples activities. The task involves taking a user query as input and returning a list of people sorted by their level of expertise regarding the user query. This paper introduces a novel approach for combining multiple estimators of expertise based on a multisensor data fusion framework together with the Dempster-Shafer theory of evidence and Shannon's entropy. More specifically, we defined three sensors which detect heterogeneous information derived from the textual contents, from the graph structure of the citation patterns for the community of experts, and from profile information about the academic experts. Given the evidences collected, each sensor may define different candidates as experts and consequently do not agree in a final ranking decision. To deal with these conflicts, we applied the Dempster-Shafer theory of evidence combined with Shannon's Entropy formula to fuse this information and come up with a more accurate and reliable final ranking list. Experiments made over two datasets of academic publications from the Computer Science domain attest for the adequacy of the proposed approach over the traditional state of the art approaches. We also made experiments against representative supervised state of the art algorithms. Results revealed that the proposed method achieved a similar performance when compared to these supervised techniques, confirming the capabilities of the proposed framework.