Software
A list of software and knowledge models developed and designed by me or one of my PhD students.
Like our software and want to collaborate? Don't hesitate to contact me!
Data Analytics in Health and Connected Care (DAHCC) assembles both the way and the data to describe the connected care applications and the used sensors to create such care applications together with their link to the people who are involved by or with those care applications (e.g. patients, healthcare professionals etc.).
The DAHCC resource exists out of 3 main components:
A large dataset of daily life activities, bot provided in raw and knowledge graph format.
The DAHCC Ontology capturing care, patient, daily life activity recognition and lifestyle domain knowledge.
Connected Care Applications which shows the potential of combining data with ontological meta-data.
Streaming MASSIF is a Cascading Reasoning platform that allows to perform expressive reasoning over high-velocity streams. It combines 1) RDF Stream Processing, 2) Expressive Reasoning and 3) Complex Event Processing in a unifying way. Streaming MASSIF provides a Web Interface to aid in rapid prototyping of Stream Reasoning applications.
INK is a data structure for representing a knowledge graph (KG), taking into account nodes and their neighborhoods till a certain, specified depth.
INK can be used for multiple semantic tasks, such as semantic rule mining and node classification.
Python implementation and extension of RDF2Vec to numerically represent nodes in a Knowledge Graph to be used for further downstream ML tasks.
PowerShap is a feature selection method that uses statistical hypothesis testing and power calculations on Shapley values, enabling fast and intuitive wrapper-based feature selection.
RSP4J is a library to build RDF Stream Processing (RSP) Engines according to the reference model RSP-QL. RSP4J is inspired by the OWL API, and other work that aim at spreading the Semantic Web (Stream Reasoning) research by means of practical and usable software tools.
RR-GCN is an extension of Relational Graph Convolutional Networks (R-GCN) in which the weights are randomly initialized and kept frozen (i.e. no training step is required). As such, our technique is unsupervised and the produced embeddings can be used for any downstream ML task/model. Surprisingly, empirical results indicate that the embeddings produced by our RR-GCN can be competitive to, and even sometimes outperform, end-to-end R-GCNs.
An algorithm that efficiently mines for a specific type of walks that maximize information gain. These walks can be used to create an interpretable feature representation for nodes in a Knowledge Graph for classification tasks.
DIVIDE is designed as a component of a semantic IoT platform, with the ability to automatically derive and configure the queries of the platform's stream processing components. This is done adaptively based on the actual application context. DIVIDE also eliminates the need the perform semantic reasoning while evaluating continuous queries on streaming data, allowing for increased performance in complex IoT contexts that need to deal with high-volume & high-frequency data streams.
C-Sprite is a very efficient RDF Stream Processing engine that optimizes query answering over hierarchical concepts in RDF data streams.
FOLIO stands for Failure Mode and Effect Analysis combined with Anomaly Ontology. Folio captures all application-independent concepts that occur within Failure Mode and Effect Analyses (FMEA), Fault Tree Analysis (FTA) and anomaly detection methods. The GitHub also contains the mapping rules that can be used to generate an instantiation of this ontology for any created FMEA table or FTA tree.
Code and docker container to use the on query reasoning capabilities of Stardog in a streaming/ sensor environment.
The full blogpost was made available on Stardog Labs: https://www.stardog.com/labs/blog/stream-reasoning-with-stardog/
Github: https://github.com/IBCNServices/StardogStreamReasoning/
An algorithm that searches for a set of shapelets in a genetic fashion. Shapelets are small subseries that are discriminative for a certain class. It has been shown that by projecting the original dataset to a distance space, where each axis corresponds to the distance to a certain shapelet, classifiers are able to achieve state-of-the-art results on a plethora of datasets.
An interactive web application that allows to create a Markov Chain of user behavior on a mobile application and to cluster similar user sessions together.
An innovative technique that constructs an ensemble of decision trees and converts this ensemble into a single, interpretable decision tree with an enhanced predictive performance.
Ontologies modelling all the concepts required to support ambient-aware continuous care applications.
GitHub: https://github.com/IBCNServices/Accio-Ontology/tree/gh-pages
Knowledge Graph from the published Kaggle dataset about COVID-19 literature, commonly known as CORD-19 (https://www.kaggle.com/allen-institute-for-ai/CORD-19-research-challenge).
This Knowledge Graph contains different information for each paper such as author information, content information and citation information in a linked format.
More information about the construction, the application and how to use this dataset can be found at www.covid-kg.tools.
A generator that allows to generate an ontology (T-Box and A-Box) by specifying a wide variety of a parameters indicating, e.g., the size, the amount and kind of axioms, number of properties, how connected the ontology should be, etc. OTAGen also generates queries, of which the characteristics can be specified.
The generator can be used to benchmark ontology processing techniques, e.g. reasoning or querying.