Thesis (Ph. D.)--University of Texas at Austin, 2005. / Vita. Includes bibliographical references.

Thesis (Ph. D.)--University of Texas at Austin, 2005. / Vita. Includes bibliographical references.

Thesis (M.S.) -- Worcester Polytechnic Institute. / Keywords: Intelligent tutoring systems; Behavior; Learning; Visualization. Includes bibliographical references (leaves 37-38).

Thesis (Ph.D.)--University of Melbourne, Dept. of Information Systems, 2009. / Typescript. Includes bibliographical references.

This thesis examines children's arithmetic strategies and their relation to the concepts of commutativity and associativity. Two complementary methods were used in this research: empirical studies and computational models. Empirical studies were carried out to identify the strategies children used for solving problems like 3 + 4, and 3 + 4 + 7, and the conceptual knowledge associated with them. Their understanding of subtraction problems where the minuend is less than the subtrahend (e.g. 6-8) was also considered. A study with 105 subjects revealed a variety of strategies and information about children's knowledge of commutativity and associativity. Four levels of performance of commutativity were also identified. A longitudinal study was carried out with 12 children in order to obtain details of children's changes in strategy, and to double check the results obtained in the main study. The strategies observed to be used by children over a 20 month period parallel those found in previous studies, which show a general transition to more efficient methods. However, the longitudinal study revealed that development of such arithmetic strategies is a slow process. Furthermore, the studies indicated that knowledge of commutativity is a prerequisite for associativity. Models of the observed strategies have been implemented in the form of production rules in a computer program called PALM. The process of implementation highlighted features of children's problem solving that had not been 'detected during the studies. In addition to models that describe the space of strategies, a model of learning has been implemented for the transition from procedural knowledge of commutativity to that of associativity. The model is capable of generalizing its inbuilt knowledge, for instance, its ability to solve 2-term arithmetic expressions, to allow it to solve more complex problems, such as 3-term arithmetic expressions. A further model has been constructed for learning arithmetic strategies that are more efficient than those already represented in the program. It learns specific rules by adding conditions for efficient problem solving to its previous general rules.

As the range of models which tutoring systems can capture is extended, efficient diagnosis becomes more difficult. This thesis describes a solution to this problem based on the generation of 'Critical Problems'; their role in student modelling is analogous to that of the 'Crucial Experiment' in science. We argue that great diagnostic power can be obtained by generating discriminatory problem examples. In general, efficient diagnosis is just not possible without such an hypothesis-testing capability. We describe a program, PO, which given a pair of production rule models and a description of the class of problems which the student must solve, generates an abstract specification of the problems which discriminate between those two hypotheses. Through a process termed 'Abstract Interpretation', PO tips the balance in favour of diagnostic measurement. The key to this problem lies in the realisation that we are only interested in the abstract mapping between a model's inputs and outputs; from the point of view of generating a Critical Problem, the intermediate processing of the model is irrelevant.

In the history of Intelligent Tutoring Systems, SOPHIE (Brown, Burton, and Bell, 1974), now considered a classic, contained many important ideas and features. One of these was its natural language user interface. Today, the trend has moved away from natural language interfaces towards graphical ones although the argument in favour of natural language user interfaces, both from Human Computer Interaction and natural language researchers, still persist. Is this argument correct? This thesis explores this question by investigating how SOPHIE might be re-implemented with a graphical direct manipulation interface instead of a natural language one, with the goal of improving its standard of usability. It begins by analysing the features that seem to have been central to SOPHIE's usability. These, it argues, were not so much an ability to accept well formed complete English sentences, as an ability to accept and interpret correctly a wide range of abbreviated inputs. Two models of interaction, Circuit 1, a pilot, and Circuit II, a fairly full implementation of SOPHIE were implemented and tested. Both employ free-order syntax that allows users to specify the components of a full command in any order. The combination of deixis and free-order syntax supported allows completely general ellipsis which achieves, in extended interaction sequences, the same economy and naturalness that SOPHIE achieved through its use of anaphora and ellipsis. Whilst the free-order syntax. technique is little used at present in user interfaces, the results of observational studies conducted have shown that it saves users time and convenience. Thus, considering key linguistic features of a natural language user interface has shown how novel features can enhance the usability of direct manipulation interfaces. This thesis argues that user interfaces can be improved by employing structures found in natural language or at least conversation which can be constructed within direct manipulation interface styles. This approach was further expanded to support topic shifts between different circuit contexts. Circuit II, like SOPHIE, supports three different topics: normal circuit behaviour, a circuit with an unknown fault, and circuits with user-hypothesised faults. Drawing on Reichman's (1981) work, Circuit II uses natural language cue phrases of the type "by the way", re-implemented in the direct manipulation style, to facilitate shifts between topics in a smoother and more natural way than SOPHIE which , used clumsy explicit commands.

No description available.

Roads deteriorate at different rates from weathering and use. Hence, transportation agencies must assess the ride quality of a facility regularly to determine its maintenance needs. Existing models to characterize ride quality produce the International Roughness Index (IRI), the prevailing summary of roughness. Nearly all state agencies use Inertial Profilers to produce the IRI. Such heavily instrumented vehicles require trained personnel for their operation and data interpretation. Resource constraints prevent the scaling of these existing methods beyond 4% of the network. This dissertation developed an alternative method to characterize ride quality that uses regular passenger vehicles. Smartphones or connected vehicles provide the onboard sensor data needed to enable the new technique. The new method provides a single index summary of ride quality for all paved and unpaved roads. The new index is directly proportional to the IRI. A new transform integrates sensor data streams from connected vehicles to produce a linear energy density representation of roughness. The ensemble average of indices from different speed ranges converges to a repeatable characterization of roughness. The currently used IRI is undefined at speeds other than 80 km/h. This constraint mischaracterizes roughness experienced at other speeds. The newly proposed transform integrates the average roughness indices from all speed ranges to produce a speed-independent characterization of ride quality. This property avoids spatial wavelength bias, which is a critical deficiency of the IRI. The new method leverages the emergence of connected vehicles to provide continuous characterizations of ride quality for the entire roadway network. This dissertation derived precision bounds of deterioration forecasting for models that could utilize the new index. The results demonstrated continuous performance improvements with additional vehicle participation. With practical traversal volumes, the achievable precision of forecast is within a few days. This work also quantified capabilities of the new transform to localize roadway anomalies that could pose travel hazards. The methods included derivations of the best sensor settings to achieve the desired performances. Several case studies validated the findings. These new techniques have the potential to save agencies millions of dollars annually by enabling predictive maintenance practices for all roadways, worldwide. / Mountain Plains Consortium (MPC)

Les bâtiments sont de plus en plus équipés avec des réseaux d’automatisation dédiés, visant à réduire la consommation d’énergie ainsi que d’optimiser le confort. D’un autre côté, nous observons l’arrivée de capteurs et actionneurs liés à l’Internet des objets, pouvant naturellement se connecter à des réseaux IP. Dû à des contraintes d’obsolescence ou imposées par les propriétés physiques des bâtiments, il n’est pas rare que différentes technologies doivent cohabiter. Celles-ci fonctionnant avec des modèles et protocoles différents rend le développement de systèmes d’automatisation globaux compliqué. Les modèles classiques de systèmes distribués ne sont pas adaptés aux problématiques des réseaux de capteurs. Le paradigme du Web des objets est basé sur des ressources et vise quand à lui d’uniformiser la couche applicative entre différents objets à l’aide des technologies du Web, essentiellement HTTP et REST. Dans cette thèse, nous nous basons sur le Web des objets afin de créer un framework dédié au bâtiments intelligents, permettant aux développeurs de concevoir des applications composites sans connaissances des différentes technologies sous-jacentes. Grâce aux technologies Web, nous pouvons offrir des services homogènes tout en profitant des ressources disponibles à l’intérieur du réseau (capteurs et actionneurs), formant un nuage auto-géré. Dans le but de doter les bâtiments d’une plus grande intelligence, l’apprentissage automatique, souvent réservé aux experts, est rendu accessible à travers des interfaces Web cachant la complexité des processus. / Buildings are increasingly equipped with dedicated automation networks, aiming to reduce the energy consumption and to optimize the comfort. On the other hand, we see the arrival of sensors and actuators related to the Internet of Things, which can naturally connect to IP networks. Due to constraints imposed by the obsolescence or physical properties of buildings, it is not uncommon that different technologies have to coexist. These networks operate with different models and protocols, making the development of global automation systems difficult. Traditional models of distributed systems are not adapted to the context of sensor networks. The paradigm of the Web of Things is resource-based and strives to standardize the application layer of different objects using Web technologies, primarily HTTP and REST. In this thesis, we use the Web of Things to create a framework dedicated to smart buildings, allowing developers to develop composite applications without knowledge of the underlying technologies. By relying on Web technologies, we can provide seamless service while reusing the available resources within the network (sensors and actuators), forming a self-managed cloud. In order to equip the buildings with a higher-level intelligence, machine learning, often reserved for experts, is made accessible through Web interfaces hiding the complexity of the process.