Competency questions for ontologies are used in a number of ontology development tasks. The questions’ sentences structure have been analysed to inform ontology authoring and validation. One of the problems to make this a seamless process is the hurdle of writing good CQs manually or offering automated assistance in writing CQs. In this paper, we propose an enhanced and automated pipeline where one can trace meticulously through each step, using a mini-corpus, T5, and the SQuAD dataset to generate questions, and the CLaRO controlled language, semantic similarity, and other steps for filtering. This was evaluated with two corpora of different genre in the same broad domain and evaluated with domain experts. The final output questions across the experiments were around 25% for scope and relevance and 45% of unproblematic quality. Technically, it provided ample insight into trade-offs in generation and filtering, where relaxing filtering increased sentence structure diversity but also led to more spurious sentences that required additional processing

Artificial conversational agents are deployed to assist humans in a variety of tasks.Some of these tasks require the capability to communicate numbers as part of their internal and abstract representations of meaning, such as for banking and scheduling appointments.They currently cannot do so for isiZulu, due to the lack of speech and text data and the complexity of the generation due to dependence on noun that is counted.We solved this by extracting and iteratively improving on the rules for speaking and writing numerals as words and creating two algorithms for it.Evaluation of the output by two isiZulu grammarians showed that six out of seven number categories were 90-100% correct.The software was used with an additional set of rules to create a large monolingual text corpus, made up of 771,643 sentences, to enable future data-driven approaches.

Foundational ontologies can be used to enable semantic interoperability in modern information systems. Aligning a domain ontology to a foundational ontology is perceived difficult, however. Reasons include confusing underlying concepts, understanding the philosophical ideologies of foundational ontologies, and lack of alignment guidance. For BFO, there is a BFO Classifier tool for alignment, but users still face challenges. To uncover some of these user challenges, an experiment was performed using 10 BFO-aligned domain ontologies. The alignment of domain entities were analysed, revealing seven different types of mistakes in the alignments. To avoid them, the BFO classifier tool was altered to improve the questions and explanations for the core principles of BFO. Thereafter, the BFO classifier tool was evaluated to measure the effect on alignment with a use-case based approach, using the GORO and AWO ontologies. The evaluation revealed that alterations facilitated alignment, as users felt more confident in their results given the improved understanding of the questions and possible answers.

Roles as argument places in a positionalist account of relations are pervasive in conceptual data modelling and linguistics, known also as components of a relationship and as semantic, thematic, or deep roles as parts of a verb or verb class, respectively. They are also planned to be used in Abstract Wikipedia that seeks to combine them. There is, however, no insight in systematic or ontologically sound usage of such roles, in contradistinction to the ample attention given to aligning classes to nouns and relationships to verbs. Roles, as identifiable argument places, may benefit from similar efforts toward an ontology of roles. We aim to take a first step in that direction in a two-pronged approach. First, we conducted an analysis of a set of 101 conceptual data models on their use of roles. Second, we analysed VerbNet, an authoritative linguistic knowledge base on thematic roles. The results show promise for improvements of naming roles in conceptual data models. VerbNet’s roles are challenging to align to an ontology due to its mixing of the ontological and linguistic layers and flexibility of natural language. The insights obtained also indicate ample avenues for further research.

There has been growing interest in building surface realization systems to support the automatic generation of text in African languages. Such tools focus on converting abstract representations of meaning to a text. Since African languages are low-resourced, economical use of resources and general maintainability are key considerations. However, there is no existing surface realizer architecture that possesses most of the maintainability characteristics (e.g., modularity, reusability, and analyzability) that will lead to maintainable software that can be used for the languages. Moreover, there is no consensus surface realization architecture created for other languages that can be adapted for the languages in question. In this work, we solve this by creating a novel surface realizer architecture suitable for low-resourced African languages that abides by the features of maintainable software. Its design comes after a granular analysis, classification, and comparison of the architectures used by 77 existing NLG systems. We compare our architecture to existing architectures and show that it supports the most features of a maintainable software product.

Multiple ontology languages have been developed over the years, which brings afore two key components: how to select the appropriate language for the task at hand and language design itself. This engineering step entails examining the ontological ‘commitments’ embedded into the language, which, in turn, demands for an insight into what the effects of philosophical viewpoints may be on the design of a representation language. But what are the sort of commitments one should be able to choose from that have an underlying philosophical point of view, and which philosophical stances have a knock-on effect on the specification or selection of an ontology language? In this paper, we provide a first step towards answering these questions. We identify and analyse ontological commitments embedded in logics, or that could be, and show that they have been taken in well-known ontology languages. This contributes to reflecting on the language as enabler or inhibitor to formally characterising an ontology or an ontological investigation, as well as the design of new ontology languages following the proposed design process.

Sharing, downloading, and reusing software is common-place, some of which is carried out legally with open source software. When it is not legal, it is unclear how many infringements have taken place: does an infringement count for the artefact as a whole or for each source file of a computer program? To answer this question, it must first be established whether a computer program should be considered as an integral whole, a collection, or a mere set of distinct files, and why. We argue that a program is a functional whole, availing of, and combining, arguments from mereology, granularity, modularity, unity, and function to substantiate the claim. The argumentation and answer contributes to the ontology of software artefacts, may assist industry in litigation cases, and demonstrates that the notion of unifying relation is operationalisable.

Text generation may require the pluralization of nouns, such as in context-sensitive user interfaces and in natural language generation more broadly. While this has been solved for the widely-used languages, this is still a challenge for the languages in the Bantu language family. Pluralization results obtained for isiZulu and Runyankore showed there were similarities in approach, including the need to combine syntax with semantics, despite belonging to different language zones. This suggests that bootstrapping and generalizability might be feasible. We investigated this systematically for seven languages across three different Guthrie language zones. The first outcome is that Meinhof’s 1948 specification of the noun classes are indeed inadequate for computational purposes for all examined languages, due to non-determinism in prefixes, and we thus redefined the characteristic noun class tables of 29 noun classes into 53. The second main result is that the generic pluralizer achieved over 93% accuracy in coverage testing and over 94% on a random sample. This is comparable to the language-specific isiZulu and Runyankore pluralizers.

Parthood and attendant part-whole relations enjoy interest in ontology authoring for various subject domains, as well as in, e.g., NLP to understand text. The list of common part-whole relations is occasionally slightly modified for languages other than English. For isiZulu, it was shown that there are not always 1:1 mappings and, moreover, dictionaries list many more translations for parthood and part-whole relations. This complicates selecting the semantically appropriate ones for localising ontologies or aligning local ontologies to other ones. It also raises the question whether the ‘common’ part-whole relations are really that common. We aim to investigate the extant part-whole relations in isiZulu and determine their ontological status. We harvested a lexicon of 81 terms from dictionaries, which was reduced to 31 through several iterations of refinement, of which 13 were formalised and aligned to well-known part-whole relations. It showed that in some cases distinctions are made—and for which words exist—that have not been included before in part-whole relations, yet in other cases it is more coarse-grained; e.g., a parthood for portions of cloth, for objects properly contained in the mouth, and for regions with a part-region that has a fiat boundary and objects located in it.

Parthood is used widely in ontologies across subject domains, specified in a multitude of mereological theories, and even more when combined with topology. To complicate the landscape, decidable languages put restrictions on the language features, so that only fragments of the mereo(topo)logical theories can be represented, even though those full features may be needed to check correctness during modelling. We address these issues by specifying a structured network of theories formulated in multiple logics that are glued together by the various linking constructs of the Distributed Ontology Language, DOL. For the KGEMT mereotopology and its five sub-theories, together with the DL-based OWL species and first- and second-order logic, this network in DOL orchestrates 28 ontologies.

Natural Language Generation (NLG) can be used to generate personalized health information, which is especially useful when provided in one's own language.However, the NLG technique widely used in different domains and languages-templates-was shown to be inapplicable to Bantu languages, due to their characteristic agglutinative structure.We present here our use of the grammar engine NLG technique to generate text in Runyankore, a Bantu language indigenous to Uganda.Our grammar engine adds to previous work in this field with new rules for cardinality constraints, prepositions in roles, the passive, and phonological conditioning.We evaluated the generated text with linguists and non-linguists, who regarded most text as grammatically correct and understandable; and over 60% of them regarded all the text generated by our system to have been authored by a human being.

There are many domain-specific and language-specific NLG systems, which are possibly adaptable across related domains and languages.The languages in the Bantu language family have their own set of features distinct from other major groups, which therefore severely limits the options to bootstrap an NLG system from existing ones.We present here our first proof-of-concept application for knowledge-to-text NLG as a plugin to the Protégé 5.x ontology development system, tailored to Runyankore, a Bantu language indigenous to Uganda.It comprises a basic annotation model for linguistic information such as noun class, an implementation of existing verbalisation rules and a CFG for verbs, and a basic interface for data entry.

Parthood is used widely in ontologies across subject domains. Some modelling guidance can be gleaned from Ontology, yet it offers multiple mereological theories, and even more when combined with topology, i.e., mereotopology. To complicate the landscape, decidable languages put restrictions on the language features, so that only fragments of the mereo(topo)logical theories can be represented, yet during modelling, those full features may be needed to check correctness. We address these issues by specifying a structured network of theories formulated in multiple logics that are glued together by the various linking constructs of the Distributed Ontology Language, DOL. For the KGEMT mereotopological theory and five sub-theories, together with the DL-based OWL species and first- and second-order logic, this network in DOL orchestrates 28 ontologies. Further, we propose automated steps toward resolution of language feature conflicts when combining modules, availing of the new 'OWL classifier' tool that pinpoints profile violations.

The provision of personalized patient information has been encouraged as a means of complementing information provided during patient-doctor consultations, and linked to better health outcomes through patient compliance with prescribed treatments.The generation of such texts as a controlled fragment of Runyankore, a Bantu language indigenous to Uganda, requires the appropriate tense and aspect, as well as a method for verb conjugation.We present how an analysis of corpora of explanations of prescribed medications was used to identify the simple present tense and progressive aspect as appropriate for our selected domain.A CFG is defined to conjugate and generate the correct form of the verb.

In the highly multilingual setting in South Africa, developing computational tools to support the 11 official languages will facilitate effective communication.The exigency to develop these tools for healthcare applications and doctor-patient interaction is there.An important component in this set-up is generating sentences in the language isiZulu, which involves part-whole relations to communicate, for instance, which part of one's body hurts.From a NLG viewpoint, the main challenge is the fluid use of terminology and the consequent complex agreement system inherent in the language, which is further complicated by phonological conditioning in the linguistic realisation stage.Through using a combined approach of examples and various literature, we devised verbalisation patterns for both meronymic and mereological relations, being structural/general parthood, involvement, containment, membership, subquantities, participation, and constitution.All patterns were then converted into algorithms and have been implemented as a proof-of-concept.

Computerised clinical practice guidelines are a key component of effective clinical decision support systems, especially in low-resource regions such as Malawi. To address shortages in staffing and budgets for training, the practice of task-shifting, the clinical practice guidelines (CPGs) enable health workers with limited training to provide a standardised level of care. However, CPGs are tradition-ally paper-based, with only a few CPGs having been computerised for Malawi's national electronic health record system. These CPGs have been hard-coded into the system, necessitating significant additional work to add support for future and revised CPGs. We further investigate CPG computerisation challenges in order to understand the motivations for the current computerised CPGs implementation. We use semi-structured interviews, code reviews, and observations in Malawi. Most significantly, we extend existing understanding of software engineering principles to the context of low-resource environments, noting that the tensions between conflicting stakeholder requirements, deadline and deliverable expectations, and good software engineering often result in systems that are harder to maintain, further exacerbating potential problems with longevity of ICTD deployments. We further suggest that a component-based approach in conjunction with communities of open source developers might help alleviate this problem by providing more scalable and robust CPG support.

Generic, reusable ontology elements, such as a foundational ontology's categories and part-whole relations, are essential for good and interoperable knowledge representation. Ontology developers, which include domain experts and novices, face the challenge to figure out which category or relationship to choose for their ontology authoring task. To reduce this bottleneck, there is a need to have guidance to handle these Ontology-laden entities. We solve this with a generic approach and realize it with the Foundational Ontology and Reasoner-enhanced axiomatiZAtion (FORZA) method, containing DOLCE, a decision diagram for DOLCE categories, part-whole relations, and an automated reasoner that is used during the authoring process to propose feasible axioms. This fusion has been integrated in the MoKi ontology development tool to validate its implementability.

An approach in achieving semantic interoperability among heterogeneous systems is to offer infrastructure to assist with linking and integration using a foundational ontology. Due to the creation of multiple foundational ontologies, this also means linking and integrating those ones. In order to achieve this, we have selected the widely used foundational ontologies DOLCE, BFO, and GFO, and their related modules, on which to perform ontology mediation (alignment, mapping, and merging). The foundational ontologies were aligned by identifying correspondences between ontology entities using seven tools, documentation, and our manual alignments, and comparing their effectiveness. Thereafter, based on the alignments, we created correspondences in the ontology files resulting in entity mappings and merged ontologies. However, during the mapping process, it was found that differences in foundational ontologies, such as their hierarchical structure, conflicting axioms due to complement and disjointness, and incompatible domain and range restriction, cause logical inconsistencies in foundational ontology alignments, thereby greatly reducing the number of mappings. We analyse and present these logical inconsistencies with possible solutions to some of them. 1

A growing number of ontologies are already available thanks to development initiatives in many different fields. In such ontology developments, developers must tackle a wide range of difficulties and handicaps, which can result in the appearance of anomalies in the resulting ontologies. Therefore, ontology evaluation plays a key role in ontology development projects. OOPS! is an on-line tool that automatically detects pitfalls, considered as potential errors or problems, and thus may help ontology developers to improve their ontologies. To gain insight in the existence of pitfalls and to assess whether there are differences among ontologies developed by novices, a random set of already scanned ontologies, and existing well-known ones, data of 406 OWL ontologies were analysed on OOPS!’s 21 pitfalls, of which 24 ontologies were also examined manually on the detected pitfalls. The various analyses performed show only minor differences between the three sets of ontologies, therewith providing a general landscape of pitfalls in ontologies.

Generic, reusable ontology elements, such as a foundational ontology?s categories and part-whole relations, are essential for good and interoperable knowledge representation. Ontology developers face the challenge to figure out which category their class belongs to and which relationship to choose for their ontology authoring tasks. To reduce this bottleneck, there is a need to have guidance to handle these Ontology-laden entities. We solve this with the generic approach GENERATOR: Guided ENtity reuse and class Expression geneRATOR that enables easy reuse of already represented knowledge such that it guides the modeller toward the comparatively best options of possible axioms to add. This has been realised with DOLCE, part-whole relations, and an automated reasoner that is used during the authoring process to propose feasible axioms, and the software has been integrated in the MoKi ontology development tool.

A purpose of a foundational ontology is to solve interoperability issues among ontologies. Many foundational ontologies have been developed, re-introducing the ontology interoperability problem. We address this with the new online foundational ontology repository ROMULUS, in which DOLCE, BFO and GFO have been aligned. We summarise the alignments, mappings, and logical inconsistencies of the foundational ontologies, and ROMULUS?s features.

Bio-ontology development is a resource-consuming task despite the many open source ontologies available for reuse. Various strategies and tools for bottom-up ontology development have been proposed from a computing angle, yet the most obvious one from a domain expert perspective is unexplored: the abundant diagrams in the sciences. To speed up and simplify bio-ontology development, we propose a detailed, micro-level, procedure, DiDOn, to formalise such semi-structured biological diagrams availing also of a foundational ontology for more precise and interoperable subject domain semantics. The approach is illustrated using Pathway Studio as case study.