Search

In this paper, we propose SCANING, an unsupervised framework for paraphrasing via controlled noise injection. We focus on the novel task of paraphrasing algebraic word problems having practical applications in online pedagogy as a means to reduce plagiarism as well as ensure understanding on the part of the student instead of rote memorization. This task is more complex than paraphrasing general-domain corpora due to the difficulty in preserving critical information for solution consistency of the paraphrased word problem, managing the increased length of the text and ensuring diversity in the generated paraphrase. Existing approaches fail to demonstrate adequate performance on at least one, if not all, of these facets, necessitating the need for a more comprehensive solution. To this end, we model the noising search space as a composition of contextual and syntactic aspects and sample noising functions consisting of either one or both aspects. This allows for learning a denoising function that operates over both aspects and produces semantically equivalent and syntactically diverse outputs through grounded noise injection. The denoising function serves as a foundation for learning a paraphrasing function which operates solely in the input-paraphrase space without carrying any direct dependency on noise. We demonstrate SCANING considerably improves performance in terms of both semantic preservation and producing diverse paraphrases through extensive automated and manual evaluation across 4 datasets., 12 pages (main}; 22 pages in total

Conceptual models are artifacts representing conceptualizations of particular domains. Hence, multi-domain model catalogs serve as empirical sources of knowledge and insights about specific domains, about the use of a modeling language’s constructs, as well as about the patterns and anti-patterns recurrent in the models of that language crosscutting different domains. However, to support domain and language learning, model reuse, knowledge discovery for humans, and reliable automated processing and analysis by machines, these catalogs must be built following generally accepted quality requirements for scientific data management. Especially, all scientific (meta)data—including models—should be created using the FAIR principles (Findability, Accessibility, Interoperability, and Reusability). In this paper, we report on the construction of a FAIR model catalog for Ontology-Driven Conceptual Modeling research, a trending paradigm lying at the intersection of conceptual modeling and ontology engineering in which the Unified Foundational Ontology (UFO) and OntoUML emerged among the most adopted technologies. In this initial release, the catalog includes over a hundred models, developed in a variety of contexts and domains. The paper also discusses the research implications for (ontology-driven) conceptual modeling of such a resource.

In Risk Management, security issues arise from complex relations among objects and agents, their capabilities and vulnerabilities, the events they are involved in, and the value and risk they ensue to the stakeholders at hand. Further, there are patterns involving these relations that crosscut many domains, ranging from information security to public safety. Understanding and forming a shared conceptualization and vocabulary about these notions and their relations is fundamental for modeling the corresponding scenarios, so that proper security countermeasures can be devised. Ontologies are instruments developed to address these conceptual clarification and terminological systematization issues. Over the years, several ontologies have been proposed in Risk Management and Security Engineering. However, as shown in recent literature, they fall short in many respects, including generality and expressivity - the latter impacting on their interoperability with related models. We propose a Reference Ontology for Security Engineering (ROSE) from a Risk Treatment perspective. Our proposal leverages on two existing Reference Ontologies: the Common Ontology of Value and Risk and a Reference Ontology of Prevention, both of which are grounded on the Unified Foundational Ontology (UFO). ROSE is employed for modeling and analysing some cases, in particular providing clarification to the semantically overloaded notion of Security Mechanism.

With the global metamorphosis of the beauty industry and the rising demand for beauty products worldwide, the need for an efficacious makeup recommendation system has never been more. Despite the significant advancements made towards personalised makeup recommendation, the current research still falls short of incorporating the context of occasion in makeup recommendation and integrating feedback for users. In this work, we propose BeautifAI, a novel makeup recommendation system, delivering personalised occasion-oriented makeup recommendations to users while providing real-time previews and continuous feedback. The proposed work's novel contributions, including the incorporation of occasion context, region-wise makeup recommendation, real-time makeup previews and continuous makeup feedback, set our system apart from the current work in makeup recommendation. We also demonstrate our proposed system's efficacy in providing personalised makeup recommendation by conducting a user study., Withdrawing due to issues with training the Makeup Style Transfer (section about style transfer). This renders the current methodology invalid

Online educational platforms organize academic questions based on a hierarchical learning taxonomy (subject-chapter-topic). Automatically tagging new questions with existing taxonomy will help organize these questions into different classes of hierarchical taxonomy so that they can be searched based on the facets like chapter. This task can be formulated as a flat multi-class classification problem. Usually, flat classification based methods ignore the semantic relatedness between the terms in the hierarchical taxonomy and the questions. Some traditional methods also suffer from the class imbalance issues as they consider only the leaf nodes ignoring the hierarchy. Hence, we formulate the problem as a similarity-based retrieval task where we optimize the semantic relatedness between the taxonomy and the questions. We demonstrate that our method helps to handle the unseen labels and hence can be used for taxonomy tagging in the wild. In this method, we augment the question with its corresponding answer to capture more semantic information and then align the question-answer pair's contextualized embedding with the corresponding label (taxonomy) vector representations. The representations are aligned by fine-tuning a transformer based model with a loss function that is a combination of the cosine similarity and hinge rank loss. The loss function maximizes the similarity between the question-answer pair and the correct label representations and minimizes the similarity to unrelated labels. Finally, we perform experiments on two real-world datasets. We show that the proposed learning method outperforms representations learned using the multi-class classification method and other state of the art methods by 6% as measured by Recall@k. We also demonstrate the performance of the proposed method on unseen but related learning content like the learning objectives without re-training the network., 16 pages, accepted at ECML-PKDD 2021

Instructional Systems Design is the practice of creating of instructional experiences that make the acquisition of knowledge and skill more efficient, effective, and appealing. Specifically in designing courses, an hour of training material can require between 30 to 500 hours of effort in sourcing and organizing reference data for use in just the preparation of course material. In this paper, we present the first system of its kind that helps reduce the effort associated with sourcing reference material and course creation. We present algorithms for document chunking and automatic generation of learning objectives from content, creating descriptive content metadata to improve content-discoverability. Unlike existing methods, the learning objectives generated by our system incorporate pedagogically motivated Bloom's verbs. We demonstrate the usefulness of our methods using real world data from the banking industry and through a live deployment at a large pharmaceutical company., Proceedings of the 11th International Conference on Education Data Mining (EDM 2018)

Also part of the Transactions on Large-Scale Data- and Knowledge-Centered Systems book sub series (TLDKS, volume 9670); International audience; Contains extended and revised version of the best papers selected from DaWaK 2014

International audience

International audience

Data mining has as goal to extract knowledge from large databases. A database may be considered as a search space consisting of an enormous number of elements, and a mining algorithm as a search strategy. In general, an exhaustive search of the space is infeasible. Therefore, efficient search strategies are of vital importance. Search strategies on genetic-based algorithms have been applied successfully in a wide range of applications. We focus on the suitability of genetic-based algorithms for data mining. We discuss the design and implementation of a genetic-based algorithm for data mining and illustrate its potentials.