AutoTutor: Incorporating Back-Channel Feedback and Other Human-Like Conversational Behaviors into an Intelligent Tutoring System

This paper describes our recent attempts to incorporate human-like conversational behaviors into the dialog moves delivered by an animated pedagogical agent that simulates human tutors. We first present a brief overview of the modules comprising AutoTutor, an intelligent tutoring system. The second section describes a set of conversational behaviors that are being incorporated into AutoTutor. The behaviors of interest involve variations in intonation, head movements, arm and hand movements, facial expressions, eye blinking, gaze direction, and back-channel feedback. The final section presents a recent empirical study concerned with back-channel feedback events during human-to-human tutoring sessions. The back-channel feedback events emitted by tutors are mostly positive (63%), mostly verbal (77%), and immediately follow speech-act boundaries or noun-phrase boundaries (83%). Tutors also deliver back-channelevents at a very high rate when students are emitting dialog, about 13 events per minute. Conversely, 88% of students' back-channel feedback events are head nods, and they occur at unbounded locations (63%).     

Routing of a hazmat truck in the presence of weather systems

This paper focuses on the effects of weather systems on hazmat routing. We start by analyzing the effects of a weather system on a vehicle traversing a single link. This helps characterize the time-dependent attributes of a link due to movement of the weather systems. This analysis is used as a building block for the problem of finding a least risk path for hazmat transportation on a network exposed to such weather systems. Several methods are offered to solve the underlying problem, and computational results are reported. We draw two conclusions from this paper. First, it is possible to determine the time-dependent attributes for links on a network provided that some assumptions on the nature of the weather system are made. Second, heuristics can provide effective solutions for practical size problems while allowing for parking the vehicle to avoid weather system effects.     

Information‐driven robotic sampling in the coastal ocean

Efficient sampling of coastal ocean processes, especially mechanisms such as upwelling and internal waves and their influence on primary production, is critical for understanding our changing oceans. Coupling robotic sampling with ocean models provides an effective approach to adaptively sample such features. We present methods that capitalize on information from ocean models and in situ measurements, using Gaussian process modeling and objective functions, allowing sampling efforts to be concentrated to regions with high scientific interest. We demonstrate how to combine and correlate marine data from autonomous underwater vehicles, model forecasts, remote sensing satellite, buoy, and ship‐based measurements, as a means to cross‐validate and improve ocean model accuracy, in addition to resolving upper water‐column interactions. Our work is focused on the west coast of Mid‐Norway where significant influx of Atlantic Water produces a rich and complex physical–biological coupling, which is hard to measure and characterize due to the harsh environmental conditions. Results from both simulation and full‐scale sea trials are presented.     

Dielectric properties of the system Pb2 −x Nd x Li0·5Nb1·5 O6+δ

The dielectric properties of the ceramic defect pyrochlore-structured system Pb_2−x Nd _x Li_0·5Nb_1·5 O_6+δ has been studied as a function of concentrationx, frequency and temperature. The results have been analyzed for relaxation process and conduction mechanism. Now deceased.     

Automatic classification of Tamil documents using vector space model and artificial neural network

Automatic text classification based on vector space model (VSM), artificial neural networks (ANN), K-nearest neighbor (KNN), Naives Bayes (NB) and support vector machine (SVM) have been applied on English language documents, and gained popularity among text mining and information retrieval (IR) researchers. This paper proposes the application of VSM and ANN for the classification of Tamil language documents. Tamil is morphologically rich Dravidian classical language. The development of internet led to an exponential increase in the amount of electronic documents not only in English but also other regional languages. The automatic classification of Tamil documents has not been explored in detail so far. In this paper, corpus is used to construct and test the VSM and ANN models. Methods of document representation, assigning weights that reflect the importance of each term are discussed. In a traditional word-matching based categorization system, the most popular document representation is VSM. This method needs a high dimensional space to represent the documents. The ANN classifier requires smaller number of features. The experimental results show that ANN model achieves 93.33% which is better than the performance of VSM which yields 90.33% on Tamil document classification.     

Cellular network configuration with co-channel and adjacent-channel interference constraints

Design of cellular networks has drawn much recent interest from the OR scientific community. A challenging issue is the handling of channel interference constraints. Co-channel interference occurs when the same channel is reused within a threshold distance. Adjacent-channel interference occurs when two channels with adjacent or nearby frequencies are used in the same cell tower. We present a mathematical programming formulation for this channel allocation problem with both types of interference constraints—it also includes decisions on location of cell towers. Our focus is on the special case where a cell tower and/or channel can interfere with at most two other towers/channels. By establishing theoretical properties for channel allocation amongst towers under this circumstance, we develop an efficient solution procedure. An iteration of the procedure uses a heuristic to locate the cell towers, then allocates the channels to the towers using a polynomial-time algorithm, and finally improves this allocation using a simulated annealing procedure. The iterative steps are embedded within an external simulated annealing method. This nested simulated annealing procedure provides encouraging computational results compared to a standard commercial solver like ILOG CPLEX 8.1. The major contribution of the work is the simultaneous consideration of co-channel and adjacent-channel interference constraints.     

Studies on gas–solid heat transfer during pneumatic conveying

Interactions between solids and gas during pneumatic conveying can be utilized for variety of applications including flash drying, solids preheating etc. Experiments on air–solid heat transfer were carried out in a vertical pneumatic conveying heat exchanger of 54 mm inside diameter, using gypsum as the solid material. The effect of solids feed rate (0.6–9.9 g/s), air velocity (4.21–6.47 m/s) and particle size (231–722.5 μm) on air–solid heat transfer rate, heat transfer area and air–solid heat transfer coefficient has been studied. Empirical correlations have been proposed for the prediction of Nusselt number based on the present experimental data. The proposed correlations predict Nusselt number within an error of ±15% for the present data.     

Growth of low dislocation density single crystals of nickel

Low dislocation density single crystals of nickel have been grown at high ambient pressure by the Czochralski method. X-ray Laue picture shows that the crystals are strain-free. The dislocation density was determined to be <10³/cm² by the etching procedure. It was found that the necking and cone regions are very critical in the dislocation introduction in the crystals. An increase in the ambient pressure used during the growth seems to aid the crystal quality.     

Damage scenario analysis of bridges using crowdsourced smartphone data from passing vehicles

Bridge wear and deterioration occur over time under typical operations. Inspections can be resource intensive, infrequent, and sometimes require bridge closure. Instrumented passing vehicles may be used to record vibration responses and extract damage-sensitive bridge frequency response characteristics. This paper presents a methodology for extracting bridge frequencies from crowdsourced dynamic response data recorded within passing vehicles. Detected bridge frequency changes and a novel point cloud methodology are used to identify critical damage scenarios and estimate remaining life. The use of a novel crowdsourced Welch methodology allows for spectra averaging that reduces the influence of noise and transient events. Numerical and experimental studies illustrate that bridge frequencies can be accurately estimated and that changes in bridge frequency caused by different damage scenarios can be characterized. Numerical examples for the damage scenario analysis, including estimation of remaining life, are provided for two bridge types and three specific damage scenarios.