The unique macroevolutionary dataset of Aze & others has been transferred onto the TimeScale Creator visualisation platform while, as much as practicable, preserving the original unrevised content of its morphospecies and lineage evolutionary trees. This is a “Corrected Version” (not a revision), which can serve as an on-going historical case example because it is now updatable with future time scales. Both macroevolutionary and biostratigraphic communities are now equipped with an enduring phylogenetic database of Cenozoic macroperforate planktonic foraminiferal morphospecies and lineages for which both graphics and content can be visualised together. Key to maintaining the currency of the trees has been specification of time scales for sources of stratigraphic ranges; these scales then locate the range dates within the calibration series. Some ranges or their sources have undergone mostly minor corrections or amendments. Links between lineage and morphospecies trees have been introduced to improve consistency and transparency in timing within the trees. Also, Aze & others’ dual employment of morphospecies and lineage concepts is further elaborated here, given misunderstandings that have ensued. Features displayed on the trees include options for line styles for additional categories for range extensions or degrees of support for ancestor–descendant proposals; these have been applied to a small number of instances as an encouragement to capture more nuanced data in the future. In addition to labeling of eco- and morpho-groups on both trees, genus labels can be attached to the morphospecies tree to warn of polyphyletic morphogenera, and the lineage codes have been decoded to ease their recognition. However, it is the mouse-over pop-ups that provide the greatest opportunity to embed supporting information in the trees. They include details for stratigraphic ranges and their recalibration steps, positions relative to the standard planktonic-foraminiferal zonation, and applications as datums, as well as mutual listings between morphospecies and lineages which ease the tracing of their interrelated contents. The elaboration of the original dataset has been captured in a relational database, which can be considered a resource in itself, and, through queries and programming, serves to generate the TimeScale Creator datapacks.

Predicting the sense of a discourse relation is particularly challenging when connective markers are missing. To address this challenge, we propose a simple deep neural network approach that replaces manual feature extraction by introducing event vectors as an alternative representation, which can be pre-trained using a very large corpus, without explicit annotation. We model discourse arguments as a combination of word and event vectors. Event information is aggregated with word vectors and a Multi-Layer Neural Network is used to classify discourse senses. This work was submitted as part of the CoNLL 2016 shared task on Discourse Parsing. We obtain competitive results, reaching an accuracy of 38%, 34% and 34% for the development, test and blind test datasets, competitive with the best performing system on CoNLL 2015.

No abstract

The unprecedented detail with which contemporary molecular phylogenetics are visualizing infraspecific relationships within living species and species complexes cannot as yet be reliably extended into deep time. Yet paleontological systematics has routinely dealt in (mainly) morphotaxa envisaged in various ways to have been components of past species lineages. Bridging these perspectives can only enrich both. We present a visualization tool that digitally depicts infraspecific diversity within species through deep time. Our integrated species–phenon tree merges ancestor–descendant trees for fossil morphotaxa (phena) into reconstructed phylogenies of lineages (species) by expanding the latter into “species boxes” and placing the phenon trees inside. A key programming strategy to overcome the lack of a simple overall parent–child hierarchy in the integrated tree has been the progressive population of a species–phenon relationship map which then provides the graphical footprint for the overarching species boxes. Our initial case has been limited to planktonic foraminfera via Aze & others’ important macroevolutionary dataset. The tool could potentially be appropriated for other organisms, to detail other kinds of infraspecific granularity within lineages, or more generally to visualize two nested but loosely coupled trees.

No abstract