Web-Based Economic Activity Classification: Comparing semi-supervised text classification methods to deal with noisy labels

In order to provide accurate statistics for industries, the classification of enterprises by economic activity is an important task for national statistical institutes. The economic activity codes in the Dutch business register are less accurate for small enterprises since small enterprises are not labelled manually. To increase the quality of...

Enhance the search process of educational material by automating quality assessments of OERs

The role of educators nowadays has changed from that of sole providers of information to that of information managers. Their job requires them to sift through the vast amount of available educational material on the internet in order to find the ones that are of highest quality and are more useful for their purposes. In this thesis I propose a...

Language-consistent Open Relation Extraction: from Multilingual Text Corpora

Open Relation Extraction (ORE) aims to find arbitrary relation tuples between entities in unstructured texts. Even though recent research efforts yield state-of-the-art results for the ORE task by utilizing neural network based models, these works are solely focused on the English language. Methods were proposed to tackle the ORE task for...

Choosing from skyline sets

The skyline operator has been proposed to bridge the gap between traditional and multimedia database systems by finding the optimal objects according to the notion of Pareto dominance. According to the notion of Pareto dominance an object dominates another if it is better in one attribute and equal in all others. Skyline sets end up being pretty...

Coner: A Collaborative Approach for Long-Tail Named Entity Recognition in Scientific Publications

Named Entity Recognition (NER) for rare long-tail entities as e.g. often found in domain-specific scientific publications is a challenging task, as typically the extensive training data and test data for fine-tuning NER algorithms is lacking. Recent approaches presented promising solutions relying on training NER algorithms in a iterative...

Learning Object Metadata Workflows for Description, Findability and Reusability Improvement

With the increase of online education, a good description of learning resources has become vital for educational resource sharing and reuse. Resource description has been under the spotlight in recent years. Educational platforms can benefit from good resource organisation and description, thereby providing a higher quality of services and...

Creating Video Sequence Alternatives Across MOOCs Based on Document Similarity

Since the introduction of Massive Online Open Courses (MOOCs) in 2008, the number of MOOCs offered by universities has increased enormously. Over 700 universities offer a total of 5924 MOOCs. Each MOOC holds a sequence of 10 to 140 videos and are meant to help online learners understand a given topic. These videos are intended to be watched in a...

Identifying memes and their interactions in online communities

Memes are theorized to be the building blocks of culture. Due to a lack of empirical validation, however, the theory of memes — memetics — remains in its infancy. We argue that one of the missing components for such empirical validation is a method for the large-scale identification of memes.<br/><br/>In this thesis, we develop a method for the...

Modelling and Mining of Data mining Workflows in Data Science Scripts

Knowledge discovery and data mining (KDD) is the process that extracts new knowledge from data. Data scientists program their solutions of their KDD challenges in data mining programming scripts. Data mining programming scripts contain multiple or all steps of the KDD process. In other words, these scripts start with data and end with knowledge....

Adopting Argumentation Mining for Claim Extraction from TED Talks

Engagement is critical for academic learning. It's commonly believed that motivating students to learn is crucial in education. We think that by providing students some interesting content based on what they are learning is a good idea. Since TED Talks share attractive new ideas, we are planning to motivate students by recommending TED Talks...

Building a workbench to improve sharing and reproducibility of MOOC experiments

This thesis improves sharing of code and reproducibility (S&amp;R) in research for massive open online courses (MOOCs). Reproducibility is recreating an experiment by a different researcher. Science in general struggles with repro- ducibility. MOOC experiments often contain useful code that could be used by other researchers, but that code is...

Aspect-based Review Extraction for E-Commerce Products

Nowadays, more and more products are sold online. Under popular products, there are normally hundreds or even thousands of reviews left by the previous customers. These reviews help potential buyers understand the products better and make the purchase decision. However, most shopping websites only give an overview score (e.g. 3 star out of 5) of...

CV Scanning Tool

Transaction Monitoring

Being a bank, bunq deals with transaction fraud on a regular basis. All transactions that are handled by bunq are monitored for these cases of fraud by a transaction monitoring system. When this system flags a transaction as being possibly fraudulent, a bunq employee has to manually check this transaction. The problem with the current system is...

Semantic enrichment and exploration process on domain specific digital libraries

With the growing number of scientific publications, the concept<br/>of navigating effectively and searching for domain specific information<br/>is rather significant and highly important for the scientific<br/>community [2]. For instance, to search by topics, research methods,<br/>used datasets, or scientific objectives. Such deep meta-data...