An official website of the United States government
Cross-linguistically, morphological material that expresses comparison (e.g. more) appears to be colexified with aspectual (“phasal”) adverbs that, under negation, encode the termination of some eventuality (CESSATIVEs, e.g. *(not)...anymore). Using data drawn from the Diyari language of central Australia, we propose a diachronic trajectory for the lexical item marla ‘very, truly’. This word first developed a comparative semantics and, subsequently, a cessative reading restricted to negative polar contexts. This proposal moves us towards a lexical entry that permits for the unification of comparative and aspectual readings for items which exhibit this polysemy and—on the basis of robust pragmatic principles— predicts their polarity-sensitive distribution cross-linguistically.
more »
« less
Understanding the Polarity of Events in the Biomedical Literature: Deep Learning vs. Linguistically-informed Methods
An important task in the machine reading of biochemical events expressed in biomedical texts is correctly reading the polarity, i.e., attributing whether the biochemical event is a promotion or an inhibition. Here we present a novel dataset for studying polarity attribution accuracy. We use this dataset to train and evaluate several deep learning models for polarity identification, and compare these to a linguistically-informed model. The best performing deep learning architecture achieves 0.968 average F1 performance in a five-fold cross-validation study, a considerable improvement over the linguistically informed model average F1 of 0.862.
more »
« less
- Award ID(s):
- 1740858
- PAR ID:
- 10111684
- Date Published:
- Journal Name:
- Proceedings of the Workshop on Extracting Structured Knowledge from Scientific Publications
- Page Range / eLocation ID:
- 21 to 30
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
We describe a deep learning system for satellite image segmentation. Our CNN model embeds contextual feature dependencies in both spatial and frequency domains. Its Spatial Weighting Module uses a multi-scale pooling layer to represent correlations at longer length scales in the spatial domain. Its Frequency Weighting Module uses frequency-domain information to better discriminate between object classes. Experimental results on the Potsdam dataset demonstrate that our model has a 1.9% higher average F1 accuracy than previous methods.more » « less
-
Metabolic pathways are a human-defined grouping of life sustaining biochemical reactions, metabolites being both the reactants and products of these reactions. But many public datasets include identified metabolites whose pathway involvement is unknown, hindering metabolic interpretation. To address these shortcomings, various machine learning models, including those trained on data from the Kyoto Encyclopedia of Genes and Genomes (KEGG), have been developed to predict the pathway involvement of metabolites based on their chemical descriptions; however, these prior models are based on old metabolite KEGG-based datasets, including one benchmark dataset that is invalid due to the presence of over 1500 duplicate entries. Therefore, we have developed a new benchmark dataset derived from the KEGG following optimal standards of scientific computational reproducibility and including all source code needed to update the benchmark dataset as KEGG changes. We have used this new benchmark dataset with our atom coloring methodology to develop and compare the performance of Random Forest, XGBoost, and multilayer perceptron with autoencoder models generated from our new benchmark dataset. Best overall weighted average performance across 1000 unique folds was an F1 score of 0.8180 and a Matthews correlation coefficient of 0.7933, which was provided by XGBoost binary classification models for 11 KEGG-defined pathway categories.more » « less
-
Cell polarity refers to the asymmetric distribution of proteins and other molecules along a specified axis within a cell. Polarity establishment is the first step in many cellular processes. For example, directed growth or migration requires the formation of a cell front and back. In many cases, polarity occurs in the absence of spatial cues. That is, the cell undergoes symmetry breaking. Understanding the molecular mechanisms that allow cells to break symmetry and polarize requires computational models that span multiple spatial and temporal scales. Here, we apply a multiscale modeling approach to examine the polarity circuit of yeast. In addition to symmetry breaking, experiments revealed two key features of the yeast polarity circuit: bistability and rapid dismantling of the polarity site following a loss of signal. We used modeling based on ordinary differential equations (ODEs) to investigate mechanisms that generate these behaviors. Our analysis revealed that a model involving positive and negative feedback acting on different time scales captured both features. We then extend our ODE model into a coarse-grained reaction–diffusion equation (RDE) model to capture the spatial profiles of polarity factors. After establishing that the coarse-grained RDE model qualitatively captures key features of the polarity circuit, we expand it to more accurately capture the biochemical reactions involved in the system. We convert the expanded model to a particle-based model that resolves individual molecules and captures fluctuations that arise from the stochastic nature of biochemical reactions. Our models assume that negative regulation results from negative feedback. However, experimental observations do not rule out the possibility that negative regulation occurs through an incoherent feedforward loop. Therefore, we conclude by using our RDE model to suggest how negative feedback might be distinguished from incoherent feedforward regulation.more » « less
-
Background: Widespread dementia detection could increase clinical trial candidates and enable appropriate interventions. Since the Clock Drawing Test (CDT) can be potentially used for diagnosing dementia-related disorders, it can be leveraged to develop a computer-aided screening tool. Objective: To evaluate if a machine learning model that uses images from the CDT can predict mild cognitive impairment or dementia. Methods: Images of an analog clock drawn by 3,263 cognitively intact and 160 impaired subjects were collected during in-person dementia evaluations by the Framingham Heart Study. We processed the CDT images, participant’s age, and education level using a deep learning algorithm to predict dementia status. Results: When only the CDT images were used, the deep learning model predicted dementia status with an area under the receiver operating characteristic curve (AUC) of 81.3% ± 4.3%. A composite logistic regression model using age, level of education, and the predictions from the CDT-only model, yielded an average AUC and average F1 score of 91.9% ±1.1% and 94.6% ±0.4%, respectively. Conclusion: Our modeling framework establishes a proof-of-principle that deep learning can be applied on images derived from the CDT to predict dementia status. When fully validated, this approach can offer a cost-effective and easily deployable mechanism for detecting cognitive impairment.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Conference Paper:
- https://doi.org/10.18653/v1/W19-2603
