13 research outputs found
Using Knowledge Graphs for Machine Learning in Smart Home Forecasters
Get PDFInternet of Things (IoT) brings together heterogeneous data from smart devices in smart homes. Smart devices operate within different platforms, but ontologies can be used to create a common middle ground that allows communications between these smart devices outside of those platforms. The data communicated by the smart devices can be used to train the prediction algorithms used in forecasters. This research will first focus on the creation of a mapping to transform IoT data into a knowledge graph than can be used in the common middle ground and investigate the effect of using that IoT knowledge graph data as input for prediction algorithms. Experiments to determine the impact of incorporating other related information in the training of the prediction algorithms will be performed by using external datasources that can be linked to the knowledge graph and by using federated learning over IoT data from other smart homes. Initial results on the transformation mapping of IoT data to an ontology is presented.</p
OfficeGraph
No full text<p>OfficeGraph is a large, real world knowledge graph containing measurements taken by 444 IoT devices, over 11 months, in a seven story office building. The devices are made up of 17 different sensor models, which make measurements of many different properties.</p><p>The <i>devices in room</i> enrichment adds more information about which devices are located in which rooms, and on which floor those rooms are.<br>The <i>Wikidata days</i> enrichment provides a link to Wikidata, by matching the dates of the measurements to those dates' entities in Wikidata.<br>The <i>graph learning</i> enrichment provides additional properties that have beneficial effects on the learning process when using graph embedding models. <br>The enrichments are located in separate files, with the graph learning enrichment folder only containing the enrichments for the devices on the 7th floor, which were used in the machine learning experiment (the code of this experiment is available on <a href="https://github.com/RoderickvanderWeerdt/semantic-enrichment-of-IoT-graphs/tree/main/OfficeGraph">GitHub</a>.</p><p>This is a zipped version of OfficeGraph. Instead of one file containing the entire knowledge graph the device folders contains a separate file for each individual device. Each file contains all the measurements made by the device. The enrichments are included in separate files.</p><p>The resource paper describing this dataset is currently under submission.</p><p> </p>
OfficeGraph
No full text<p>OfficeGraph is a large, real world knowledge graph containing measurements taken by 444 IoT devices, over 11 months, in a seven story office building. The devices are made up of 17 different sensor models, which make measurements of many different properties.</p><p>The <i>devices in room</i> enrichment adds more information about which devices are located in which rooms, and on which floor those rooms are.<br>The <i>Wikidata days</i> enrichment provides a link to Wikidata, by matching the dates of the measurements to those dates' entities in Wikidata.<br>The <i>graph learning</i> enrichment provides additional properties that have beneficial effects on the learning process when using graph embedding models. <br>The enrichments are located in separate files, with the graph learning enrichment folder only containing the enrichments for the devices on the 7th floor, which were used in the machine learning experiment (the code of this experiment is available on <a href="https://github.com/RoderickvanderWeerdt/semantic-enrichment-of-IoT-graphs/tree/main/OfficeGraph">GitHub</a>.</p><p>This is a zipped version of OfficeGraph. Instead of one file containing the entire knowledge graph the device folders contains a separate file for each individual device. Each file contains all the measurements made by the device. The enrichments are included in separate files.</p><p>The resource paper describing this dataset is currently under submission.</p><p> </p>
IoT Measurement Knowledge Graphs:Constructing, Working and Learning with IoT Measurement Data as a Knowledge Graph
Get PDFIoT devices generate substantial amounts of measurements and use many different formats to store this data. In order to make IoT devices interoperable, a number of ontologies have been developed, such as the Smart Applications REFerence (SAREF) ontology, to create knowledge graphs that can represent all kinds of measurements from IoT devices. We call the resulting knowledge graphs: IoT measurement knowledge graphs. In this thesis, we set out to investigate what differentiates IoT measurement knowledge graphs from regular knowledge graphs. In the first half, we describe in detail how to create IoT measurement knowledge graphs from real-world measurement data. In the second half, we investigate what happens when we use the graphs in different scenarios, with different methods and applications. In order to generate an IoT measurement knowledge graph, we create a mapping to transform the measurement data coming from IoT devices. We explore multiple ways to create this mapping and evaluate the resulting IoT measurement knowledge graph with competency questions. Using our mapping, we create OfficeGraph, by transforming IoT measurement data recorded over a year from 444 IoT devices located in an office building. This IoT measurement knowledge graph is validated by answering competency questions created with the support of the building owners, showing that we can answer questions we were not able to answer without it. Besides the entities, literals, and relations in knowledge graphs, the combination of these, the context of entities, provides additional knowledge. Therefore, if we want to use entities in knowledge graphs to train machine learning models, it would be a waste to take only the entities from the graph, because this would leave out information. Representation models, such as RDF2Vec and Graph Convolutional Networks (GCNs), can be used to learn (embedding) representations for entities that take context into account, creating a representation based on which relations, entities, and literals occur near the entity in the graph. We learn over IoT measurement knowledge graphs using multiple representation models and experiment with the effect of making more knowledge available to the representation models through semantic enrichment. This is done by making implicit information, such as (e.g.) consecutive measurements, explicitly available, by adding a relation between two measurements. Results show that the semantic enrichment has a positive effect on the learnability, based on the evaluation score, of the entity representations. Due to the dynamic nature of IoT measurement knowledge graphs, it is to be expected that new measurements are made after the initial IoT measurement knowledge graphs are created. Therefore, we investigate the possibility of estimating representations for new entities that are based on the existing entity representations. We introduce the embedding estimation method, which used the numerical attributes of entities, the measurement values, to find entities similar to the new entity. Then by averaging the embedding representations of those entities, it creates a new, estimated, embedding for the new entity. We perform experiments to test the embedding estimation method, and show that there is a trade-off between the amount of time saved by not re-training the entire pipeline, but also a decrease in accuracy, with the estimated embeddings. This thesis offers a focused research into IoT measurement knowledge graph. We provide insight into how IoT measurement knowledge graph can be created and how they differ from regular knowledge graphs, specifically in being numerical, dynamic, and shallow knowledge graphs. We suggest and validate remedies to the shallowness of IoT measurement knowledge graphs, through the semantic enrichments. Furthermore we show how the numerical aspect can be a strength, through the embedding estimation method, which can help with the dynamic nature of IoT measurement knowledge graph
Supporting Smart Home Scenarios Using OWL and SWRL Rules
Get PDFDespite the pervasiveness of IoT domotic devices in the home automation landscape, their
potential is still quite under-exploited due to the high heterogeneity and the scarce expressivity of
the most commonly adopted scenario programming paradigms. The aim of this study is to show
that Semantic Web technologies constitute a viable solution to tackle not only the interoperability
issues, but also the overall programming complexity of modern IoT home automation scenarios. For
this purpose, we developed a knowledge-based home automation system in which scenarios are
the result of logical inferences over the IoT sensors data combined with formalised knowledge. In
particular, we describe how the SWRL language can be employed to overcome the limitations of the
well-known trigger-action paradigm. Through various experiments in three distinct scenarios, we
demonstrated the feasibility of the proposed approach and its applicability in a standardised and
validated context such as SARE
Making heterogeneous smart home data interoperable with the SAREF ontology
Get PDFSAREF is an ontology created to enable interoperability between smart devices, but there is a lack in the literature of practical examples to implement SAREF in real applications. We validate the practical implementation of SAREF through two approaches. We first examine two methods to map the IoT data available in a smart home into linked data using SAREF: (1) by creating a template-based mapping to describe how SAREF can be used and (2) by using a mapping language to demonstrate it can be simple to map, while still using SAREF. The second approach demonstrates the communication capabilities of IoT devices when they share knowledge represented using SAREF and describes how SAREF enables interoperability between different devices. The two approaches demonstrate that all the information from various data sets of smart devices can successfully be transformed into the SAREF ontology and how SAREF can be applied in a concrete interoperability framework.</p
Validating SAREF in a Smart Home Environment
Get PDFSAREF is an ontology created to enable interoperability between smart devices. While the IoT community has shown interest and understanding of SAREF as a means for interoperability, there is a lack in the literature of practical examples to implement SAREF in real applications. In order to validate the practical implementation of SAREF we perform two experiments. First we map IoT data available in a smart home into RDF using SAREF. In the second part of the paper an IoT environment is created by using the Knowledge Engine, a framework created to allow communication between smart devices, operating on Raspberry Pi’s emulating IoT devices, where the communication of the IoT devices is performed by sharing knowledge represented with SAREF. These experiments demonstrate that SAREF is an ontology that is successfully applicable in different situations, with data-mapping showing that SAREF is able to represent the information of different smart devices and by using the Knowledge Engine showing that SAREF can enable interoperability between smart devices.</p
Updating Knowledge Graph Embeddings by Intermediate Estimations on Numerical Attributes
No full textGraph embedding methods are used to create representations of knowledge graph entities in a high-dimensional vector space. These embeddings can be used in downstream tasks such as classification or link prediction. However, typically graph embedding methods require training on the entire knowledge graph, making them less efficient when a knowledge graph is expected to be dynamic, such as with IoT measurements graphs, which are updated throughout the day with new numerical measurements. This paper introduces a method for efficiently creating embeddings for new knowledge graph entities, without retraining the embedding model. The proposed method estimates an embedding for new entities, by averaging the embedding of the k nearest neighbors, where nearness is based on numerical attributes associated with the entities. We investigate the performance of this method, both on a synthetic knowledge graph, and five real-world knowledge graphs. In these experiments, we employ RDF2vec as the embedding method, and classification as the downstream task. We compare three distance measures for determining nearest entities. We observe a trade-off between the accuracy and efficiency of estimating the embeddings and retraining the embedding model. Resulting in a significant decrease of training time, compared to retraining the full model, with only a relatively small reduction in precision. Results show that in cases where the attributes are representative enough our method is an effective and efficient method to incrementally adjust graph embeddings.</p
Learning and Reasoning over Smart Home Knowledge Graphs
No full textVendors of Smart Home devices and research institutes come together in
the EU funded InterConnect project to develop a shared data model and accompanying infrastructure that facilitates seamless integration where all devices and services communicate via a human and machine interpretable language: SAREF. The
Linked Data approach where a knowledge graph captures the shared information
about Smart Devices and services in the network offers the opportunity for novel
AI research. In this paper we outline four research directions currently being undertaken by the authors and present the results of using the SAREF knowledge graphs
to 1) improve the explainability of existing machine learning (ML) approaches, 2)
uniform transformation of graph data to tabular input for these ML approaches, 3)
applying graph Deep Learning to derive missing classes and values and 4) reasoning with SWRL rules for Smart Home control
Representation Learning on IoT Knowledge Graphs
No full textIn order to make the large amounts of messages generated by IoT devices in Smart Buildings interoperable, ontologies are used to represent the data as knowledge graphs (KGs). Learning over these IoT KGs can be used for various tasks, such as prediction or classification. Existing methods for KG representation learning are often evaluated on benchmark KGs and it is not explored how such methods perform on IoT KGs. The specific structure of the IoT KGs is likely to influence the quality of the representations. In this study, we investigate how the structure of IoT KGs affects the effectiveness of representation learning methods. Additionally, we look at the effect on representation quality of enriched IoT KGs, with for example temporal sequences or measurement value similarity, and the effect of the size of the IoT KGs. We perform experiments on three IoT KGs, with two representation learning methods (RDF2Vec and GCN) and two evaluation tasks (classification and value prediction). The results show that models trained with representations from enriched KGs outperform models trained with representations from original KGs on the evaluation tasks.(This article is a revised and extended version of [24]. It constitutes a significant extension with regards to the number and scale of experiments, embedding methods and evaluation tasks.)</p
