Beyond the topics: how deep learning can improve the discriminability of probabilistic topic modelling.
Ontology highlight
ABSTRACT:
SUBMITTER: Al Moubayed N
PROVIDER: S-EPMC7924555 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Reinforcement learning tasks are often used to assess participants' tendency to learn more from the positive or more from the negative consequences of one's action. However, this assessment often requires comparison in learning performance across different task conditions, which may differ in the relative salience or discriminability of the stimuli associated with more and less rewarding outcomes, respectively. To address this issue, in a first set of studies, participants were subjected to two versions of a common probabilistic learning task. The two versions differed with respect to the stimulus (Hiragana) characters associated with reward probability. The assignment of character to reward probability was fixed within version but reversed between versions. We found that performance was highly influenced by task version, which could be explained by the relative perceptual discriminability of characters assigned to high or low reward probabilities, as assessed by a separate discrimination experiment. Participants were more reliable in selecting rewarding characters that were more discriminable, leading to differences in learning curves and their sensitivity to reward probability. This difference in experienced reinforcement history was accompanied by performance biases in a test phase assessing ability to learn from positive vs. negative outcomes. In a subsequent large-scale web-based experiment, this impact of task version on learning and test measures was replicated and extended. Collectively, these findings imply a key role for perceptual factors in guiding reward learning and underscore the need to control stimulus discriminability when making inferences about individual differences in reinforcement learning.
| S-EPMC5421749 | biostudies-literature
Project description:K-complexes (KCs) are a recognized electroencephalography marker of sensory processing and a defining feature of sleep stage 2. KC frequency and morphology may also be reflective of sleep quality, aging, and a range of sleep and sensory processing deficits. However, manual scoring of K-complexes is impractical, time-consuming, and thus costly and currently not well-standardized. Although automated KC detection methods have been developed, performance and uptake remain limited. The proposed algorithm is based on a deep neural network and Gaussian process, which gives the input waveform a probability of being a KC ranging from 0% to 100%. The algorithm was trained on half a million synthetic KCs derived from manually scored sleep stage 2 KCs from the Montreal Archive of Sleep Study containing 19 healthy young participants. Algorithm performance was subsequently assessed on 700 independent recordings from the Cleveland Family Study using sleep stages 2 and 3 data. The developed algorithm showed an F1 score (a measure of binary classification accuracy) of 0.78 and thus outperforms currently available KC scoring algorithms with F1 = 0.2-0.6. The probabilistic approach also captured expected variability in KC shape and amplitude within individuals and across age groups. An automated probabilistic KC classification is well suited and effective for systematic KC detection for a more in-depth exploration of potential relationships between KCs during sleep and clinical outcomes such as health impacts and daytime symptomatology.
| S-EPMC7751135 | biostudies-literature
Project description:Scientific research is highly dynamic. New areas of science continually evolve; others gain or lose importance, merge, or split. Due to the steady increase in the number of scientific publications, it is hard to keep an overview of the structure and dynamic development of one's own field of science, much less all scientific domains. However, knowledge of "hot" topics, emergent research frontiers, or change of focus in certain areas is a critical component of resource allocation decisions in research laboratories, governmental institutions, and corporations. This paper demonstrates the utilization of Kleinberg's burst detection algorithm, co-word occurrence analysis, and graph layout techniques to generate maps that support the identification of major research topics and trends. The approach was applied to analyze and map the complete set of papers published in PNAS in the years 1982-2001. Six domain experts examined and commented on the resulting maps in an attempt to reconstruct the evolution of major research areas covered by PNAS.
| S-EPMC387309 | biostudies-other
Project description:Anthropogenic warming has led to an unprecedented year-round reduction in Arctic sea ice extent. This has far-reaching consequences for indigenous and local communities, polar ecosystems, and global climate, motivating the need for accurate seasonal sea ice forecasts. While physics-based dynamical models can successfully forecast sea ice concentration several weeks ahead, they struggle to outperform simple statistical benchmarks at longer lead times. We present a probabilistic, deep learning sea ice forecasting system, IceNet. The system has been trained on climate simulations and observational data to forecast the next 6 months of monthly-averaged sea ice concentration maps. We show that IceNet advances the range of accurate sea ice forecasts, outperforming a state-of-the-art dynamical model in seasonal forecasts of summer sea ice, particularly for extreme sea ice events. This step-change in sea ice forecasting ability brings us closer to conservation tools that mitigate risks associated with rapid sea ice loss.
| S-EPMC8390499 | biostudies-literature
Project description:The transportation sector is a major contributor to greenhouse gas (GHG) emissions and is a driver of adverse health effects globally. Increasingly, government policies have promoted the adoption of electric vehicles (EVs) as a solution to mitigate GHG emissions. However, government analysts have failed to fully utilize consumer data in decisions related to charging infrastructure. This is because a large share of EV data is unstructured text, which presents challenges for data discovery. In this article, we deploy advances in transformer-based deep learning to discover topics of attention in a nationally representative sample of user reviews. We report classification accuracies greater than 91% (F1 scores of 0.83), outperforming previously leading algorithms in this domain. We describe applications of these deep learning models for public policy analysis and large-scale implementation. This capability can boost intelligence for the EV charging market, which is expected to grow to US$27.6 billion by 2027.
| S-EPMC7892356 | biostudies-literature
Project description:Hateful commenting, also known as 'toxicity', frequently takes place within news stories in social media. Yet, the relationship between toxicity and news topics is poorly understood. To analyze how news topics relate to the toxicity of user comments, we classify topics of 63,886 online news videos of a large news channel using a neural network and topical tags used by journalists to label content. We score 320,246 user comments from those videos for toxicity and compare how the average toxicity of comments varies by topic. Findings show that topics like Racism, Israel-Palestine, and War & Conflict have more toxicity in the comments, and topics such as Science & Technology, Environment & Weather, and Arts & Culture have less toxic commenting. Qualitative analysis reveals five themes: Graphic videos, Humanistic stories, History and historical facts, Media as a manipulator, and Religion. We also observe cases where a typically more toxic topic becomes non-toxic and where a typically less toxic topic becomes "toxicified" when it involves sensitive elements, such as politics and religion. Findings suggest that news comment toxicity can be characterized as topic-driven toxicity that targets topics rather than as vindictive toxicity that targets users or groups. Practical implications suggest that humanistic framing of the news story (i.e., reporting stories through real everyday people) can reduce toxicity in the comments of an otherwise toxic topic.
| S-EPMC7034861 | biostudies-literature
Project description:A large variety of misconceptions about learning, teaching, and other educational topics is prevalent in the public but also among educational professionals. Such misconceptions may lead to ill-advised judgments and actions in private life, professional practice, and policymaking. Developing effective correction strategies for these misconceptions hinges on a better understanding of the factors that make individuals susceptible to or resilient against misconceptions. The present study surveyed students from educational and non-educational fields of study to investigate whether the endorsement of four typical educational misconceptions can be predicted by study-related variables (i.e., field of study and study progress) and by students' cognitive ability (i.e., numeracy), epistemic orientations, general world views (i.e., conservative orientation), and education-related values (i.e., educational goals). A sample of N = 315 undergraduates in teacher education and education- and non-education-related fields of study completed an online survey. Results from structural equation models showed that the pattern of effects strongly varied across the specific misconceptions. The two misconceptions related to teaching factors (i.e., class size and effectiveness of direct instruction as a teaching method) were the most strongly affected by the field of study and had an association with conservative orientation. In contrast, the misconception about the effectiveness of grade retention as an educational intervention was more prevalent among the students emphasizing conventional educational goals, such as discipline. None of the investigated explanatory variables proved predictive of the misconception about the "feminization" of education as an educational-equity topic. Moreover, neither numeracy nor epistemic orientation was found to have any effect on the endorsement of educational misconceptions. These findings emphasize the topic dependency of the factors that make individuals susceptible to misconceptions. Future research and intervention approaches need to consider the topic specificity of educational misconceptions.
| S-EPMC8635341 | biostudies-literature
Project description:We present cisTopic, a probabilistic framework to simultaneously discover co-accessible enhancers and stable cell states from sparse single-cell epigenomics data (http://github.com/aertslab/cistopic). On a compendium of single-cell ATAC-seq datasets from differentiating hematopoietic cells, brain, and transcription-factor perturbation dynamics, we demonstrate that topic modelling can be exploited for a robust identification of cell types, enhancers, and relevant transcription factors. cisTopic provides insight into the mechanisms underlying regulatory heterogeneity within cell populations.
2019-02-18 | GSE114557 | GEO
Project description:Antimicrobial peptides (AMPs) are a valuable source of antimicrobial agents and a potential solution to the multi-drug resistance problem. In particular, short-length AMPs have been shown to have enhanced antimicrobial activities, higher stability, and lower toxicity to human cells. We present a short-length (?30 aa) AMP prediction method, Deep-AmPEP30, developed based on an optimal feature set of PseKRAAC reduced amino acids composition and convolutional neural network. On a balanced benchmark dataset of 188 samples, Deep-AmPEP30 yields an improved performance of 77% in accuracy, 85% in the area under the receiver operating characteristic curve (AUC-ROC), and 85% in area under the precision-recall curve (AUC-PR) over existing machine learning-based methods. To demonstrate its power, we screened the genome sequence of Candida glabrata-a gut commensal fungus expected to interact with and/or inhibit other microbes in the gut-for potential AMPs and identified a peptide of 20 aa (P3, FWELWKFLKSLWSIFPRRRP) with strong anti-bacteria activity against Bacillus subtilis and Vibrio parahaemolyticus. The potency of the peptide is remarkably comparable to that of ampicillin. Therefore, Deep-AmPEP30 is a promising prediction tool to identify short-length AMPs from genomic sequences for drug discovery. Our method is available at https://cbbio.cis.um.edu.mo/AxPEP for both individual sequence prediction and genome screening for AMPs.
| S-EPMC7256447 | biostudies-literature
Project description:Knowing protein function is crucial to advance molecular and medical biology, yet experimental function annotations through the Gene Ontology (GO) exist for fewer than 0.5% of all known proteins. Computational methods bridge this sequence-annotation gap typically through homology-based annotation transfer by identifying sequence-similar proteins with known function or through prediction methods using evolutionary information. Here, we propose predicting GO terms through annotation transfer based on proximity of proteins in the SeqVec embedding rather than in sequence space. These embeddings originate from deep learned language models (LMs) for protein sequences (SeqVec) transferring the knowledge gained from predicting the next amino acid in 33 million protein sequences. Replicating the conditions of CAFA3, our method reaches an Fmax of 37?±?2%, 50?±?3%, and 57?±?2% for BPO, MFO, and CCO, respectively. Numerically, this appears close to the top ten CAFA3 methods. When restricting the annotation transfer to proteins with?<?20% pairwise sequence identity to the query, performance drops (Fmax BPO 33?±?2%, MFO 43?±?3%, CCO 53?±?2%); this still outperforms naïve sequence-based transfer. Preliminary results from CAFA4 appear to confirm these findings. Overall, this new concept is likely to change the annotation of proteins, in particular for proteins from smaller families or proteins with intrinsically disordered regions.
| S-EPMC7806674 | biostudies-literature