Extracting ontological knowledge from Java source code using Hidden Markov Models

Abstract

Ontologies have become a key element since many decades in information systems such as in epidemiological surveillance domain. Building domain ontologies requires the access to domain knowledge owned by domain experts or contained in knowledge sources. However, domain experts are not always available for interviews. Therefore, there is a lot of value in using ontology learning which consists in automatic or semi-automatic extraction of ontological knowledge from structured or unstructured knowledge sources such as texts, databases, etc. Many techniques have been used but they all are limited in concepts, properties and terminology extraction leaving behind axioms and rules. Source code which naturally embed domain knowledge is rarely used. In this paper, we propose an approach based on Hidden Markov Models (HMMs) for concepts, properties, axioms and rules learning from Java source code. This approach is experimented with the source code of EPICAM, an epidemiological platform developed in Java and used in Cameroon for tuberculosis surveillance. Domain experts involved in the evaluation estimated that knowledge extracted was relevant to the domain. In addition, we performed an automatic evaluation of the relevance of the terms extracted to the medical domain by aligning them with ontologies hosted on Bioportal platform through the Ontology Recommender tool. The results were interesting since the terms extracted were covered at 82.9% by many biomedical ontologies such as NCIT, SNOWMEDCT and ONTOPARON.

Learning Food Composition Knowledge from scientific literature

“…In effect, BioPortal is a biomedical ontology repository containing more than 300 biomedical ontologies, including the following 6 food ontologies: Isotopes for food science (ISO-FOOD) [38], Food-Biomarker Ontology (FOBI) [39], FoodGroupNHNS (FGNHNS) [10] , OntoFood (OF) [11] , The FoodOn Food Ontology (FOODON), Food Interactions with Drugs Evidence Ontology (FIDEO) [40], and Food Matrix for Predictive Microbiology (FMPM) [12] . The Ontology Recommender module was used to find the best ontologies from biomedical texts or set of keywords furnished as input in a text area [41].…”

“…Once found, Food Composition Knowledge are manually extracted from them by copying the elements one by one to build Food Composition Tables [2,8,9]. Manual information extraction from a large number of data sources is a cumbersome task and time consuming [10]. Learning Food Composition Knowledge from tables stored in scientific papers aims to reduce this cost during the knowledge acquisition process.…”

Better Modeling the Programming World with Code Concept Graphs-augmented Multi-modal Learning

“…This is different from other abstractions that have been used previously to improve code comprehension. Among others, some works proposed to create knowledge graphs based on code and documentation [11,35] or code ontologies incorporating many linguistic features of the input programming language [6,22]. On the one hand, the aforementioned works do not intend to integrate their knowledge-based data as part of a deep NLP learning process such as ours.…”

Improved Text Mining Algorithm for Fault Detection using Combined D-Matrix

“…The often occurring phrases i.e. error elements are taken into consideration as key phrases/normal in fields of indications, disappointment modes and different repair movement [15]. The form of such keywords in signs and indications, error modes and different repair motion is considered for verdict frequency.…”