54 Informatik
Refine
Document Type
- Article (9)
- Doctoral Thesis (2)
- Master's Thesis (1)
Institute
Keywords
- Maschinelles Lernen (6)
- Machine learning (5)
- OA-Publikationsfonds2020 (4)
- Deep learning (3)
- big data (3)
- data science (2)
- random forest (2)
- reinforcement learning (2)
- 3D Interaction Techniques (1)
- Algorithmus (1)
Year of publication
- 2020 (12) (remove)
Multi-user projection systems provide a coherent 3D interaction space for multiple co-located users that facilitates mutual awareness, full-body interaction, and the coordination of activities. The users perceive the shared scene from their respective viewpoints and can directly interact with the 3D content.
This thesis reports on novel interaction patterns for collaborative 3D interaction for local and distributed user groups based on such multi-user projection environments. A particular focus of our developments lies in the provision of multiple independent interaction territories in our workspaces and their tight integration into collaborative workflows. The motivation for such multi-focus workspaces is grounded in research on social cooperation patterns, specifically in the requirement for supporting phases of loose and tight collaboration and the emergence of dedicated orking territories for private usage and public exchange. We realized independent interaction territories in the form of handheld virtual viewing windows and multiple co-located hardware displays in a joint workspace. They provide independent views of a shared virtual environment and serve as access points for the exploration and manipulation of the 3D content. Their tight integration into our workspace supports fluent transitions between individual work and joint user engagement. The different affordances of various displays in an exemplary workspace consisting of a large 3D wall, a 3D tabletop, and handheld virtual viewing windows, promote different usage scenarios, for instance for views from an egocentric perspective, miniature scene representations, close-up views, or storage and transfer areas. This work shows that this versatile workspace can make the cooperation of multiple people in joint tasks more effective, e.g. by parallelizing activities, distributing subtasks, and providing mutual support.
In order to create, manage, and share virtual viewing windows, this thesis presents the interaction technique of Photoportals, a tangible interface based on the metaphor of digital photography. They serve as configurable viewing territories and enable the individual examination of scene details as well as the immediate sharing of the prepared views. Photoportals are specifically designed to complement other interface facets and provide extended functionality for scene navigation, object manipulation, and for the creation of temporal recordings of activities in the virtual scene.
A further objective of this work is the realization of a coherent interaction space for direct 3D input across the independent interaction territories in multi-display setups. This requires the simultaneous consideration of user input in several potential interaction windows as well as configurable disambiguation schemes for the implicit selection of distinct interaction contexts. We generalized the required implementation structures into a high-level software pattern and demonstrated its versatility by means of various multi-context 3D interaction tools.
Additionally, this work tackles specific problems related to group navigation in multiuser projection systems. Joint navigation of a collocated group of users can lead to unintentional collisions when passing narrow scene sections. In this context, we suggest various solutions that prevent individual collisions during group navigation and discuss their effect on the perceived integrity of the travel group and the 3D scene. For collaboration scenarios involving distributed user groups, we furthermore explored different configurations for joint and individual travel.
Last but not least, this thesis provides detailed information and implementation templates for the realization of the proposed interaction techniques and collaborative workspaces in scenegraph-based VR systems. These contributions to the abstraction of specific interaction patterns, such as group navigation and multi-window interaction, facilitate their reuse in other virtual reality systems and their adaptation to further collaborative scenarios.
The K-nearest neighbors (KNN) machine learning algorithm is a well-known non-parametric classification method. However, like other traditional data mining methods, applying it on big data comes with computational challenges. Indeed, KNN determines the class of a new sample based on the class of its nearest neighbors; however, identifying the neighbors in a large amount of data imposes a large computational cost so that it is no longer applicable by a single computing machine. One of the proposed techniques to make classification methods applicable on large datasets is pruning. LC-KNN is an improved KNN method which first clusters the data into some smaller partitions using the K-means clustering method; and then applies the KNN for each new sample on the partition which its center is the nearest one. However, because the clusters have different shapes and densities, selection of the appropriate cluster is a challenge. In this paper, an approach has been proposed to improve the pruning phase of the LC-KNN method by taking into account these factors. The proposed approach helps to choose a more appropriate cluster of data for looking for the neighbors, thus, increasing the classification accuracy. The performance of the proposed approach is evaluated on different real datasets. The experimental results show the effectiveness of the proposed approach and its higher classification accuracy and lower time cost in comparison to other recent relevant methods.
The classical Internet of things routing and wireless sensor networks can provide more precise monitoring of the covered area due to the higher number of utilized nodes. Because of the limitations in shared transfer media, many nodes in the network are prone to the collision in simultaneous transmissions. Medium access control protocols are usually more practical in networks with low traffic, which are not subjected to external noise from adjacent frequencies. There are preventive, detection and control solutions to congestion management in the network which are all the focus of this study. In the congestion prevention phase, the proposed method chooses the next step of the path using the Fuzzy decision-making system to distribute network traffic via optimal paths. In the congestion detection phase, a dynamic approach to queue management was designed to detect congestion in the least amount of time and prevent the collision. In the congestion control phase, the back-pressure method was used based on the quality of the queue to decrease the probability of linking in the pathway from the pre-congested node. The main goals of this study are to balance energy consumption in network nodes, reducing the rate of lost packets and increasing quality of service in routing. Simulation results proved the proposed Congestion Control Fuzzy Decision Making (CCFDM) method was more capable in improving routing parameters as compared to recent algorithms.
Abstract In the first part of this research, the utilization of tuned mass dampers in the vibration control of tall buildings during earthquake excitations is studied. The main issues such as optimizing the parameters of the dampers and studying the effects of frequency content of the target earthquakes are addressed.
Abstract The non-dominated sorting genetic algorithm method is improved by upgrading generic operators, and is utilized to develop a framework for determining the optimum placement and parameters of dampers in tall buildings. A case study is presented in which the optimal placement and properties of dampers are determined for a model of a tall building under different earthquake excitations through computer simulations.
Abstract In the second part, a novel framework for the brain learning-based intelligent seismic control of smart structures is developed. In this approach, a deep neural network learns how to improve structural responses during earthquake excitations using feedback control.
Abstract Reinforcement learning method is improved and utilized to develop a framework for training the deep neural network as an intelligent controller. The efficiency of the developed framework is examined through two case studies including a single-degree-of-freedom system and a high-rise building under different earthquake excitation records.
Abstract The results show that the controller gradually develops an optimum control policy to reduce the vibrations of a structure under an earthquake excitation through a cyclical process of actions and observations.
Abstract It is shown that the controller efficiently improves the structural responses under new earthquake excitations for which it was not trained. Moreover, it is shown that the controller has a stable performance under uncertainties.
This research aims to model soil temperature (ST) using machine learning models of multilayer perceptron (MLP) algorithm and support vector machine (SVM) in hybrid form with the Firefly optimization algorithm, i.e. MLP-FFA and SVM-FFA. In the current study, measured ST and meteorological parameters of Tabriz and Ahar weather stations in a period of 2013–2015 are used for training and testing of the studied models with one and two days as a delay. To ascertain conclusive results for validation of the proposed hybrid models, the error metrics are benchmarked in an independent testing period. Moreover, Taylor diagrams utilized for that purpose. Obtained results showed that, in a case of one day delay, except in predicting ST at 5 cm below the soil surface (ST5cm) at Tabriz station, MLP-FFA produced superior results compared with MLP, SVM, and SVM-FFA models. However, for two days delay, MLP-FFA indicated increased accuracy in predicting ST5cm and ST 20cm of Tabriz station and ST10cm of Ahar station in comparison with SVM-FFA. Additionally, for all of the prescribed models, the performance of the MLP-FFA and SVM-FFA hybrid models in the testing phase was found to be meaningfully superior to the classical MLP and SVM models.
Coronary Artery Disease Diagnosis: Ranking the Significant Features Using a Random Trees Model
(2020)
Heart disease is one of the most common diseases in middle-aged citizens. Among the vast number of heart diseases, coronary artery disease (CAD) is considered as a common cardiovascular disease with a high death rate. The most popular tool for diagnosing CAD is the use of medical imaging, e.g., angiography. However, angiography is known for being costly and also associated with a number of side effects. Hence, the purpose of this study is to increase the accuracy of coronary heart disease diagnosis through selecting significant predictive features in order of their ranking. In this study, we propose an integrated method using machine learning. The machine learning methods of random trees (RTs), decision tree of C5.0, support vector machine (SVM), and decision tree of Chi-squared automatic interaction detection (CHAID) are used in this study. The proposed method shows promising results and the study confirms that the RTs model outperforms other models.
The longitudinal dispersion coefficient (LDC) plays an important role in modeling the transport of pollutants and sediment in natural rivers. As a result of transportation processes, the concentration of pollutants changes along the river. Various studies have been conducted to provide simple equations for estimating LDC. In this study, machine learning methods, namely support vector regression, Gaussian process regression, M5 model tree (M5P) and random forest, and multiple linear regression were examined in predicting the LDC in natural streams. Data sets from 60 rivers around the world with different hydraulic and geometric features were gathered to develop models for LDC estimation. Statistical criteria, including correlation coefficient (CC), root mean squared error (RMSE) and mean absolute error (MAE), were used to scrutinize the models. The LDC values estimated by these models were compared with the corresponding results of common empirical models. The Taylor chart was used to evaluate the models and the results showed that among the machine learning models, M5P had superior performance, with CC of 0.823, RMSE of 454.9 and MAE of 380.9. The model of Sahay and Dutta, with CC of 0.795, RMSE of 460.7 and MAE of 306.1, gave more precise results than the other empirical models. The main advantage of M5P models is their ability to provide practical formulae. In conclusion, the results proved that the developed M5P model with simple formulations was superior to other machine learning models and empirical models; therefore, it can be used as a proper tool for estimating the LDC in rivers.
In this study, machine learning methods of artificial neural networks (ANNs), least squares support vector machines (LSSVM), and neuro-fuzzy are used for advancing prediction models for thermal performance of a photovoltaic-thermal solar collector (PV/T). In the proposed models, the inlet temperature, flow rate, heat, solar radiation, and the sun heat have been considered as the input variables. Data set has been extracted through experimental measurements from a novel solar collector system. Different analyses are performed to examine the credibility of the introduced models and evaluate their performances. The proposed LSSVM model outperformed the ANFIS and ANNs models. LSSVM model is reported suitable when the laboratory measurements are costly and time-consuming, or achieving such values requires sophisticated interpretations.
Image Analysis Using Human Body Geometry and Size Proportion Science for Action Classification
(2020)
Gestures are one of the basic modes of human communication and are usually used to represent different actions. Automatic recognition of these actions forms the basis for solving more complex problems like human behavior analysis, video surveillance, event detection, and sign language recognition, etc. Action recognition from images is a challenging task as the key information like temporal data, object trajectory, and optical flow are not available in still images. While measuring the size of different regions of the human body i.e., step size, arms span, length of the arm, forearm, and hand, etc., provides valuable clues for identification of the human actions. In this article, a framework for classification of the human actions is presented where humans are detected and localized through faster region-convolutional neural networks followed by morphological image processing techniques. Furthermore, geometric features from human blob are extracted and incorporated into the classification rules for the six human actions i.e., standing, walking, single-hand side wave, single-hand top wave, both hands side wave, and both hands top wave. The performance of the proposed technique has been evaluated using precision, recall, omission error, and commission error. The proposed technique has been comparatively analyzed in terms of overall accuracy with existing approaches showing that it performs well in contrast to its counterparts.
In conjunction with the improved methods of monitoring damage and degradation processes, the interest in reliability assessment of reinforced concrete bridges is increasing in recent years. Automated imagebased inspections of the structural surface provide valuable data to extract quantitative information about deteriorations, such as crack patterns. However, the knowledge gain results from processing this information in a structural context, i.e. relating the damage artifacts to building components. This way, transformation to structural analysis is enabled. This approach sets two further requirements: availability of structural bridge information and a standardized storage for interoperability with subsequent analysis tools. Since the involved large datasets are only efficiently processed in an automated manner, the implementation of the complete workflow from damage and building data to structural analysis is targeted in this work. First, domain concepts are derived from the back-end tasks: structural analysis, damage modeling, and life-cycle assessment. The common interoperability format, the Industry Foundation Class (IFC), and processes in these domains are further assessed. The need for usercontrolled interpretation steps is identified and the developed prototype thus allows interaction at subsequent model stages. The latter has the advantage that interpretation steps can be individually separated into either a structural analysis or a damage information model or a combination of both. This approach to damage information processing from the perspective of structural analysis is then validated in different case studies.