Collaborative filtering is used to predict the preference or rating of a user for a certain item. Collaborative filtering is based on the notion that similar users rate similarly. A lot of research is done on how to improve this algorithm, mostly with deep learning. A less investigated field for recommender systems is graph signal processing. Graph signal processing is used to reconstruct a graph signal based on the surrounding nodes. The item ratings of a user can be represented as a graph signal. So it is possible to use graph signal processing as a recommender system. In this paper we investigate how the Tikhonov and Sobolev graph regularisers perform for user-based KNN collaborative filtering. We investigated this by comparing the performance of the collaborative filtering algorithm with the two graph regularisers. We found that the Tikhonov regulariser and the Sobolev regulariser performed very similar to user-based KNN collaborative filtering. This means that the added complexity of the graph regularisers did not increase the quality of predictions we can already make with collaborative filtering.

This report details the development process of an open-data-based tool, an extension of the original interface created by Royal HaskoningDHV. The objective was to bridge the gap between geographical data and Architectural, Engineering, and Construction (AEC) industry applications. The tool aimed to transform spatial data for architects, facilitating contextual analysis in Rhinoceros and Grasshopper, ultimately aiding architects and engineers in enhancing designs based on environmental impact. The initial tool focused on Netherlands data, but the ultimate goal was to make it applicable to other countries/regions. The research involved evaluating data availability for different regions, acquiring and aligning relevant data for Grasshopper, and implementing these data workflows into wind and solar analyses. The data evaluation stage revealed challenges due to varying data availability and accessibility across countries. For example, Germany's fragmented data required navigating different portals, while Hong Kong's centralized data via API was more accessible. The lack of standardization hindered automation, necessitating manual data retrieval strategies that could be challenging for non-geomatics experts. Data alignment methods varied, introducing complexities. For instance, Italy required 3D extrusion from 2D shapefiles, leading to unavoidable errors. Spain used a different method, showcasing the difficulty of a universal solution due to data standardization and interoperability issues. Two techniques were envisioned for the open-data tool: TIN-based and Voxel-based methods, each with distinct qualities and limitations. The TIN-method offered high-quality analyses but required rigorous data alignment, while the Voxel-based method allowed flexibility but risked issues with resolution. Limitations of exploratory analysis included a focus on five countries/regions and inherent constraints of Rhinoceros, limiting tool accessibility and requiring alternative approaches. Additionally, language barriers and data platform permeability might have led to overlooked datasets. In conclusion, the report acknowledges the need for future work. Optimization of code for readability and performance is suggested, and the inclusion of additional data types (vegetation, land use, transport) in data workflows is proposed. Input from AEC professionals through methods like questionnaires or testing is recommended for further improvement. This report emphasizes the evolving nature of the tool and the importance of ongoing refinement to meet the needs of diverse AEC professionals.

Mesh data is widely used in engineering for instance for simulations, CAD engineering and visualizations. The accuracy and quality of the meshes influence the reliability and validity of these processes. Besides manual modelling, scanning is becoming increasingly more common due to the increase in devices that have scanning capabilities. Unwanted noise is often present in scanned models. The process of mesh denoising is removing the unwanted noise whilst keeping the features of the mesh. These features are often anisotropic, e.g. sharp edges and corners. The DeltaConv convolution is an anisotropic convolution, Wiersma et al. [24] show the advantage of using the anisotropic DeltaConv convolution for anisotropic tasks over other isotropic convolutions. In this thesis it is investigated if state-of-the-art mesh denoising can benefit from using the DeltaConv convolution. This is done by integrating the DeltaConv convolution in the Dual-DMP [6] algorithm, and tuning this network. In this thesis we found that state-of-the-art mesh denoising can benefit from using the DeltaConv convolution. Due to the expressiveness of the DeltaConv convolution, objects with sharp features are denoised better than the state-of-the-art algorithms and on smooth meshes, DeltaConv works comparable to state-of-the-art algorithms.

In this thesis we present a new idea to objectively assess reconstruction algorithms. Because it is not feasible to completely scan a high-detailed ground truth mesh of large urban objects, the performance of the reconstructed meshes can therefore not be measured objectively. To solve this, we present a new mesh evaluation methodology that can more objectively assess the quality of the generated mesh based on a low detailed ground truth mesh. We achieved this by creating a synthetic dataset based on low-detailed models of large urban buildings and using this as a ground truth mesh and input data for the reconstruction algorithms. Thanks to our new methodology, we were able to compare the output mesh and ground truth mesh using a wireframe model of the meshes. This allows us to give a more objective score to the results without having to look at entire model, which is the usual method. The results of this thesis show that the new methodology has potential to be used for creating a new benchmark, and it opens a new door to using more readily available objects that could not be used before.