Online mining the maximal frequent itemsets over data streams is an important problem in data mining. The maximal frequent itemset is the itemset which the support is large or equal to the minimal support and the itemset is not the subset or superse of each itemset. Previous algorithms to mine the maximal frequent itemsets in the traditional database are not suitable for data streams. Because data streams have some characteristics: (1) continuous (2) fast (3) no data limit (4) real time (5) searching once, mining data streams have many new challenges. First, they are unrealistic to keep the entire stream in the main memory or even in a secondary storage area, since a data stream comes continuously and the amount of data is unbounded. Second, traditional methods of mining on stored datasets by multiple scans are infeasible, since the streaming data is passed only once. Third, mining streams requires fast, real-time processing in order to keep up with the high data arrival rate and mining results are expected to be available within short response time. In order to solve mining maximal frequent itemsets from data streams using the landmark window model, Mao et. al. propose the INSTANT algorithm. In the landmark window model, knowledge discovery is performed based on the values between the beginning time and the present. The advantage of using the landmark window model is that the results are correct as compared to the other models. The structure of the INSTANT algorithm is simple and it can save many memory space. But it takes long time in mining the maximal frequent itemsets. When the new transactions comes, the number of comparisons between the old transactions of INSATNT algorithm is too much. In this thesis, we propose the Set-Checking algorithm to mine frequent itemsets from data streams using the landmark window model. We use the structure of lattice to store our information. The structure of lattice records the subset relationship between the child node and the father node. For every node, we can record the itemset and the support. When the new transaction comes, we consider five relations: (1) equivalent (2) superset (3) subset (4) intersection (5) empty relations. According to the lattice structure of the five sets , we can add the transaction and the renew support efficiently. From our simulation result, we find that the process time of our Set-Checking algorithm is faster than that of the INSTANT algorithm.

The importance of developing personalized risk prediction estimates has become increasingly evident in recent years. In general, patient populations may be heterogenous and represent a mixture of different unknown subtypes of disease. When the source of this heterogeneity and resulting subtypes of disease are unknown, accurate prediction of survival may be difficult. However, in certain disease settings the onset time of an observable intermediate event may be highly associated with these unknown subtypes of disease and thus may be useful in predicting long term survival. Throughout this dissertation, we examine an approach to incorporate intermediate event information for the prediction of long term survival: the landmark model. In Chapter 1, we use the landmark modeling framework to develop procedures to assess how a patient’s long term survival trajectory may change over time given good intermediate outcome indications along with prognosis based on baseline markers. We propose time-varying accuracy measures to quantify the predictive performance of landmark prediction rules for residual life and provide resampling-based procedures to make inference about such accuracy measures. We illustrate our proposed procedures using a breast cancer dataset. In Chapter 2, we aim to incorporate intermediate event time information for the prediction of survival. We propose a fully non-parametric procedure to incorporate intermediate event information when only a single baseline discrete covariate is available for prediction. When a continuous covariate or multiple covariates are available, we propose to incorporate intermediate event time information using a flexible varying coefficient model. To evaluate the performance of the resulting landmark prediction rule and quantify the information gained by using the intermediate event, we use robust non-parametric procedures. We illustrate these procedures using a dataset of post-dialysis patients with end-stage renal disease. In Chapter 3, we consider improving efficiency by incorporating intermediate event information in a randomized clinical trial setting. We propose a semi-nonparametric two-stage procedure to estimate survival by incorporating intermediate event information observed before the landmark time. In addition, we present a testing procedure using these resulting estimates to test for a difference in survival between two treatment groups. We illustrate these proposed procedures using an AIDS dataset.

A novel approach is proposed for real-time retrieval of images from a large database of overlapping images of an indoor environment. The procedure extracts visual features from images using selected computer vision techniques, and processes the extracted features to create a reduced list of features annotated with the frame numbers they appear in. This method is named landmark indexing. Unlike some state-of-the-art approaches, the proposed method does not need to consider large image adjacency graphs because the overlap of the images in the map sufficiently increases information gain, and mapping of similar features to the same landmark reduces the search space to improve search efficiency. Empirical evidence from experiments on real datasets shows high (90-100%) accuracy in image retrieval, and improvement in search time from the order of 100-200 milliseconds to the order of 10-30 milliseconds. The image retrieval technique is also demonstrated by integrating it into a 3D real-time navigation system. This system is tested in several indoor environments and all experiments show accurate localization results in large indoor areas with errors in the order of 15-20 centimeters only. / Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2014-04-24 12:44:41.429

Two experiments examined the role of various cues on children’s performance in a well-known object-displacement task. In this task, children observed a toy rolling down a ramp whose trajectory was occluded by an opaque screen with doors. A barrier was placed along the ramp, behind one of the doors, to stop the toy. The top portion of the barrier was visible above the screen. To search successfully, children had to retrieve the hidden toy by opening the correct door. Previous work had found that the barrier was an ineffective cue among children less than three years of age. According to a landmark-based account, this was because the barrier was only an indirect cue to object location. If a cue directly marked the location, then it would be more likely attended and utilized. This model underscores the spatial relation between cue and the target. Other cue properties are important in so far that they modify this spatial relation. In Experiment 1, a cue’s distance from the target object was manipulated (i.e., short vs. long), but the location marked by the cue was kept constant (i.e., correct door was directly below). The search performances of 24- and 30-month old children were compared under no cue, short-cue/short-door, and long-cue/long-door conditions. Both age groups performed equally well under both cued conditions. In Experiment 2, a cue’s movement (i.e., coincident with the car vs. not coincident with the car) down the ramp was manipulated. The performance of 24- and 30-month old children were compared under attached-direct cue and unattached-direct cue conditions. Both age groups performed well under both conditions. Collectively, the results provide support for the landmark-based account. The spatial relation between cue and target underlies toddlers’ search. Properties of the cue matter to the extent that they impact how well the cue marks its target.

Two experiments examined the role of various cues on children’s performance in a well-known object-displacement task. In this task, children observed a toy rolling down a ramp whose trajectory was occluded by an opaque screen with doors. A barrier was placed along the ramp, behind one of the doors, to stop the toy. The top portion of the barrier was visible above the screen. To search successfully, children had to retrieve the hidden toy by opening the correct door. Previous work had found that the barrier was an ineffective cue among children less than three years of age. According to a landmark-based account, this was because the barrier was only an indirect cue to object location. If a cue directly marked the location, then it would be more likely attended and utilized. This model underscores the spatial relation between cue and the target. Other cue properties are important in so far that they modify this spatial relation. In Experiment 1, a cue’s distance from the target object was manipulated (i.e., short vs. long), but the location marked by the cue was kept constant (i.e., correct door was directly below). The search performances of 24- and 30-month old children were compared under no cue, short-cue/short-door, and long-cue/long-door conditions. Both age groups performed equally well under both cued conditions. In Experiment 2, a cue’s movement (i.e., coincident with the car vs. not coincident with the car) down the ramp was manipulated. The performance of 24- and 30-month old children were compared under attached-direct cue and unattached-direct cue conditions. Both age groups performed well under both conditions. Collectively, the results provide support for the landmark-based account. The spatial relation between cue and target underlies toddlers’ search. Properties of the cue matter to the extent that they impact how well the cue marks its target.

The thesis challenges how the existing fabric can be adapted and transformed to generate similar benefits to that of a new construction. / Dissertation (MArch(Prof))--University of Pretoria, 2009. / Architecture / unrestricted

[EN] Automated temporal planning deals with the construction of plans for problems specified with durative actions of possibly different duration. The goal of temporal planning is to select and schedule the actions so as to achieve the problem goals, and the optimality criteria is the plan makespan. However, many real-word application define goals with time constraints which may not be satisfied with the plan of optimal makespan. The 2006 International Planning Competition introduced the PDDL3.0 language and organized the first and only track in planning with state trajectory constraints, including time restrictions, soft constraints and preferences. Particularly, four domains that feature deadline constraints were tested at the IPC and two planners participated in such competition although no one exhibited a good performance or fully-correction in the fulfillment of deadlines. An early detection of unsolvability in temporal planning problems, specifically those with strict deadlines, is crucial for avoiding an unfruitful exploration of the search space. And this is precisely the objective of this PhD dissertation. This work contributes with a temporal landmark-based model embedded into a temporal planner that allows for a rapid and eficient identification of unsolvable problems with deadline constraints. Our model incorporates the basic model operators of PDDL3.0 for defining temporal constraints and extends the STRIPS landmark concept to a temporal context. A temporal landmark is associated to three intervals that denote the time frame of the generation, validity and necessity of the landmark in the plan, respectively. The set of temporal landmarks of a problem along with their ordering relations and temporal restrictions define a temporal landmark graph which comprises the relationships that must exist between the literals of a solution plan and, therefore, between the corresponding actions. As long as new data is incorporated in the graph, the landmarks intervals are updated and propagated accordingly, thus reflecting a more accurate picture of a solution plan. The temporal landmark model is integrated in a domain-independent temporal planner named TempLM. During the search process, TempLM uses the temporal landmark graph to prune those partial plans of the search tree that are not compliant with the information of the graph. Additionally, we present a further improvement by which a feedback routine between the landmarks graph and the plan of a node tree is established. This process is used to refine the information of the graph and likewise narrow the search. In the experimental evaluation, we show the effectiveness of the proposed approach for detecting unsolvability and solving temporal planning problems with tight deadline constraints. / [ES] El objetivo de la planificación temporal automática es la construcción de planes con acciones de diferente duración que necesitan ser programadas adecuadamente con el fin de conseguir los objetivos del problema. En planificación temporal, la optimalidad se mide como la duración del plan más corto. Sin embargo, en muchos problemas del mundo real es necesario gestionar restricciones temporales asociadas a los objetivos del problema que pueden no satisfacerse con el plan de menor duración. En la Competición Internacional de Planificación del año 2006 se presentó el lenguaje PDDL3.0 y se realizó la primera y única competición de planificadores con gestión de restricciones de trayectorias de estado y preferencias. Concretamente, en esta IPC se probaron cuatro dominios con restricciones temporales donde los objetivos deben satisfacerse dentro de un límite de tiempo establecido o deadline. Dos planificadores participaron en esta competición aunque ninguno presentó un buen comportamiento respecto al cumplimiento de los deadlines. En este tipo de problemas, especialmente cuando se trata de problemas altamente restringidos, es crucial detectar la irresolubilidad de los mismos lo antes posible, y en este punto es donde se establece nuestro objetivo. Este trabajo de tesis presenta un modelo de landmarks temporales que permite identificar rápida y eficientemente la irresolubilidad de problemas de planificación con restricciones. Nuestro modelo incorpora las restricciones temporales del lenguaje PDDL3.0 y extiende el concepto de landmark STRIPS al contexto temporal. A cada landmark temporal se le asocian tres tipos de intervalos que se actualizan y propagan de acuerdo a las relaciones de orden y restricciones temporales entre ellos. Los landmarks temporales junto con sus relaciones de orden y restricciones temporales forman un grafo de landmarks temporales donde se sintetiza las relaciones que existen entre los literales de un plan solución y, consecuentemente, entre sus correspondientes acciones. Cuando se añade nueva información al grafo, se actualizan y propagan los intervalos de los landmarks, reflejando una imagen más precisa del plan solución. Posteriormente, el modelo de landmarks temporales se integra en un planificador temporal heurístico independiente del dominio denominado TempLM. TempLM utiliza toda la información extraída del problema para podar los planes parciales en el árbol de búsqueda que no son compatibles con la información del grafo de landmarks temporales. Además, se dispone de un proceso de retroalimentación entre el grafo de landmarks temporales y el proceso de búsqueda de un plan solución que permite enriquecer el grafo y, asimismo, acotar el espacio de búsqueda. Los resultados experimentales muestran que esta aproximación detecta rápidamente problemas irresolubles y también es muy efectiva para resolver problemas con restricciones muy ajustadas. / [CAT] L'objectiu de la planificació temporal automàtica és la construcció de plans amb accions de diferent durada que necessiten ser programades adequadament amb la finalitat d'aconseguir els objectius del problema. En planificació temporal, l'optimitat es mesura com la durada del pla més curt. No obstant açò, en molts problemas del món real és necessari gestionar restriccions temporals associades als objectius del problema que poden no satisfer-se amb el pla de menor durada. En la Competició Internacional de Planificació de l'any 2006 es va presentar el llenguatge PDDL3.0 i es va realitzar la primera i única competició de planificadors amb gestió de restriccions de trajectòries d'estat i preferències. Concretament, en aquesta IPC es van provar quatre dominis amb restriccions temporals on els objectius han de satisfer-se dins d'un límit de temps establit o deadline. Dos planificadors van participar en aquesta competició encara que cap va presentar un bon comportament respecte al compliment dels deadlines. En aquest tipus de problemes, especialment quan es tracta de problemes altament restringits, és crucial detectar la irresolubilitat dels mateixos el més prompte possible, i en aquest punt és on s'estableix el nostre objectiu. Aquest treball de tesi presenta un model de landmarks temporals que permet identificar ràpida i eficientment la irresolubilitat de problemes de planificació amb restriccions. El nostre model incorpora les restriccions temporals del llenguatge PDDL3.0 i estén el concepte de landmark STRIPS al context temporal. A cada landmark temporal se li associen tres tipus d'intervals que s'actualitzen i propaguen d'acord a les relacions d'ordre i restriccions temporals entre ells. Els landmarks temporals juntament amb les seues relacions d'ordre i restriccions temporals formen un graf de landmarks temporals on se sintetitza les relacions que existeixen entre els literals d'un pla solució i, conseqüentment, entre les seues corresponents accions. Quan s'afig nova informació al graf, s'actualitzen i propaguen els intervals dels landmarks, reflectint una imatge més precisa del pla solució. Posteriorment, el model de landmarks temporals s'integra en un planificador temporal heurístic independent del domini denominat TempLM. TempLM utilitza tota la informació extreta del problema per a podar els plans parcials en l'arbre de cerca que no són compatibles amb la informació del graf de landmarks temporals. A més, es disposa d'un procés de retroalimentació entre el graf de landmarks temporals i el procés de cerca d'un pla solució que permet enriquir el graf i, així mateix, limitar l'espai de cerca. Els resultats experimentals mostren que aquesta aproximació detecta ràpidament problemes irresolubles i també és molt efectiva per a resoldre problemes amb restriccions molt ajustades. / Marzal Calatayud, EJ. (2016). Modelo de landmarks temporales para problemas de planificación temporal altamente restringidos [Tesis doctoral no publicada]. Universitat Politècnica de València. https://doi.org/10.4995/Thesis/10251/61460 / TESIS

ABSTRACT Introduction: Medical imaging continues to play an increasing role in health care and is an integral part of medicine and dentistry. Recent technological advancements have led to the development of fully automated landmark identification (ALI) systems capable of tracing Cone-Beam Computed Tomography (CBCT). The purpose of this study was to evaluate the accuracy and reliability of an ALI system as a tool for automatic landmark location compared to human judges. Methods: One hundred subjects’ CBCT volumes from multiple imaging centers were traced by two human judges who were calibrated and had an ICC close to 1. Fifty-three landmarks were identified in the x, y, and z coordinate planes using Checkpoint Software (Stratovan Corporation, Davis, CA). The ground truth was created by calculating the mean values of the x, y, and z coordinates for each landmark across both judges’ landmark identification. To evaluate the accuracy of ALI, the mean absolute error at each coordinate and mean error distance (mm) between the human landmark identification (ground truth) and the ALI were determined, and a successful detection rate (SDR) was calculated. Results: Overall, the ALI system was as successful at landmarking as the human judges with the exception of a few landmarks. The mean error distance for all 53 landmarks was 4.04 mm ± 6.5. Forty-nine out of 53 landmarks were located within a mean error of 4mm when the average for the coordinates of human judges was considered as a ground truth. Conclusion: Across all three coordinate planes, 96% of the landmarks had a mean absolute error of less than 4mm when compared to the ground truth. The ALI was more precise than humans when identifying landmarks on the same image at different times. This study demonstrates the promise of ALI in aiding orthodontists with landmark identification on CBCTs in the future.

No description available.

The effective application of mobile robotics requires that robots be able to perform tasks with an extended degree of autonomy. Simultaneous localisation and mapping (SLAM) aids automation by providing a robot with the means of exploring an unknown environment while being able to position itself within this environment. Vision-based SLAM benefits from the large amounts of data produced by cameras but requires intensive processing of these data to obtain useful information. In this dissertation it is proposed that, as the saliency content of an image distils a large amount of the information present, it can be used to benefit vision-based SLAM implementations. The proposal is investigated by developing a new landmark for use in SLAM. Image keypoints are grouped together according to the saliency content of an image to form the new landmark. A SLAM system utilising this new landmark is implemented in order to demonstrate the viability of using the landmark. The landmark extraction, data filtering and data association routines necessary to make use of the landmark are discussed in detail. A Microsoft Kinect is used to obtain video images as well as 3D information of a viewed scene. The system is evaluated using computer simulations and real-world datasets from indoor structured environments. The datasets used are both newly generated and freely available benchmarking ones. / Dissertation (MEng)--University of Pretoria, 2013. / gm2014 / Electrical, Electronic and Computer Engineering / unrestricted