Combining Machine Learning & Reasoning for Biodiversity Data Intelligence

Abstract

The current crisis in global natural resource management makes it imperative that we better leverage the vast data sources associated with taxonomic entities (such as recognized species of plants and animals), which are known collectively as biodiversity data. However, these data pose considerable challenges for artificial intelligence: while growing rapidly in volume, they remain highly incomplete for many taxonomic groups, often show conflicting signals from different sources, and are multi-modal and therefore constantly changing in structure. In this paper, we motivate, describe, and present a novel workflow combining machine learning and automated reasoning, to discover patterns of taxonomic identity and change - i.e. “taxonomic intelligence” - leading to scalable and broadly impactful AI solutions within the bio-data realm.

A Novel Dataset Towards Extracting Virus-Host Interactions

“…This has been used to align and disambiguate published taxonomies of primates and other species (Franz, N.M. et al, 2016). Further, the approach has the potential to be used in biodiversity conservation applications (Sen, A., Sterner, B., et al, 2021). Such inference may be seen as a generalized form of querying or questionanswering over taxonomic graphs, and moreover provides a highly intuitive and visual representation of taxonomic flux over time.…”

“…Such inference may be seen as a generalized form of querying or questionanswering over taxonomic graphs, and moreover provides a highly intuitive and visual representation of taxonomic flux over time. Further, taxonomic automated reasoning systems have previously been combined (Sen, A., Sterner, B., et al, 2021) with statistical features extracted from biological image repositories (such as citizensourced or herbarium-sourced images) to further facilitate the taxonomic relationship discovery task. While we have only considered textual abstracts in our work so far, further useful context may thus be added by augmenting taxonomic knowledge graphs with images or tables extracted from the full text of the publications.…”