Search

In this paper we present CrumbTrail , an algorithm to clean large and dense knowledge graphs. CrumbTrail removes cycles, out-of-domain nodes and non-essential nodes, i.e., those that can be safely removed without breaking the knowledge graph’s connectivity. It achieves this through a bottom-up topological pruning on the basis of a set of input concepts that, for instance, a user can select in order to identify a domain of interest. Our technique can be applied to both noisy hypernymy graphs – typically generated by ontology learning algorithms as intermediate representations – as well as crowdsourced resources like Wikipedia, in order to obtain clean, domain-focused concept hierarchies. CrumbTrail overcomes the time and space complexity limitations of current state-of-art algorithms. In addition, we show in a variety of experiments that it also outperforms them in tasks such as pruning automatically acquired taxonomy graphs, and domain adaptation of the Wikipedia category graph. [ABSTRACT FROM AUTHOR]

The classification of domain entities into top-level ontology concepts remains an activity performed manually by an ontology engineer. Although some works focus on automating this task by applying machine-learning approaches using textual sentences as input, they require the existence of the domain entities in external knowledge resources, such as pre-trained embedding models. In this context, this work proposes an approach that combines the term representing the domain entity and its informal definition into a single text sentence without requiring external knowledge resources. Thus, we use this sentence as the input of a deep neural network that contains a language model as a layer. Also, we present a methodology used to extract two novel datasets from the OntoWordNet ontology based on Dolce-Lite and Dolce-Lite-Plus top-level ontologies. Our experiments show that by using the transformer-based language models, we achieve promising results in classifying domain entities into 82 top-level ontology concepts, with 94% regarding micro F1-score. [ABSTRACT FROM AUTHOR]

Within the last years, probabilistic causality has become a very active research topic in artificial intelligence and statistics communities. Due to its high impact in various applications involving reasoning tasks, machine learning researchers have proposed a number of techniques to learn Causal Bayesian Networks. Within the existing works in this direction, few studies have explicitly considered the role that decisional guidance might play to alternate between observational and experimental data processing. In this paper, we go further by introducing a serendipitous strategy to elucidate semantic background knowledge provided by the domain ontology to learn the causal structure of Bayesian Networks. We also complement our contribution with an enrichment process by which it will be possible to reuse these causal discoveries, support the evolving character of the semantic background and make an ontology evolution. Finally, the proposed method will be validated through simulations and real data analysis. [ABSTRACT FROM AUTHOR]

An intelligent tutoring system plays vital role in education and its importance is constantly increasing, meanwhile the key challenge in the teaching learning process is assessing students’ learning efficiently. In this paper, a hybrid assessment based-on ACT-R cognitive learning theory, combining ontology knowledge map with skills is proposed. In order to assess how well students master knowledge structure, an ontology knowledge map is constructed to describe declarative knowledge; and in order to assess how well students master knowledge skills, a problem solving process is constructed to describe procedural knowledge based on ACT-R. Finally, a student’s mastery of knowledge is assessed through both the knowledge map and skills in the problem solving process, as well as auxiliary indicators like time usage, prior knowledge level, self-assessment, etc. This method is implemented in a geometric intelligent assessment system and is evaluated in a junior high school. Experiments show that the assessment results are consistent with students’ actual learning levels. The hybrid cognitive assessment method can not only obtain the score of students’ mastery of knowledge points and the structure through knowledge map, but also assess the learning skills in problem solving process through exercises quantitatively. [ABSTRACT FROM AUTHOR]