Information Extraction on Weather Forecasts with Semantic Technologies

Semantic Web technologies and bias in artificial intelligence: A systematic literature review

“…An ontology defines a set of classes, attributes and relationships that model a knowledge domain with varying levels of expressivity [35]. For example, these formal and explicit specifications can be of a shared conceptualisation of meteorological variables (temperature, precipitation, visibility) to capture the domain of weather forecasting [31], the feelings and emotions conveyed by visual features to capture the psychology of human affect [43], or the reasoning steps of tasks involving problem-solving in specific domains (writing a risk assessment in industrial, insurance, health or environmental domains [53]). In particular, these forms of knowledge representation can capture terms or statements about the real world at different levels of domain specialisation [73].…”

“…This scope gives rise to foundational or top-level, general or core-reference ontologies. The most commonly used in the surveyed papers are domain ontologies (e.g., for the travel and tourism domain [76]) and application ontologies (e.g., for the extraction of information from a weather forecast written in natural language [31]).…”

“…Therefore, we investigate the combination of SW technologies with bias mitigation methods. Bias mitigation has generally been divided into three groups [56,62]: those focusing on changing the training data [2,6,19,21,24,39,41,46,47,57,85], the learning algorithm during the model generation [3,31,51,54,76,88], or the model outcomes according to the results in a holdout dataset which was not involved during the training phase [29]. Such methods may mitigate undesirable associations of specific demographic groups with hateful connotations.…”

“…In machine learning (ML), the absence of the context about the domain of the text has shown to leave annotators in an indecisive state so that their annotations incorrectly shift towards the most frequent sense of a word [17]. The use of subjective text and opinions or any data arising from human interpretation can also have challenging impacts if used to develop AI applications [3,31,46,51].…”

Profile generation from web sources: an information extraction system

“…It extracts information from tables in clinical literature (Milosevic et al 2019), Gogar T. et al use convolutional neural networks to extract information from web pages (Gogar et al 2016), Garrido A.L. et al take advantage of structure and language conventions to obtain required information from weather forecasts (Garrido et al 2016), Arulanandam R. et al employ named entity recognition (NER) and conditional random field (CRF) to extract crime information from news articles (Arulanandam et al 2014).…”