Modelling Speaker Attribution in Narrative Texts With Biased and Bias-Adjustable Neural Networks.
Ontology highlight
ABSTRACT:
SUBMITTER: Donicke T
PROVIDER: S-EPMC8851310 | biostudies-literature | 2021
REPOSITORIES: biostudies-literature
Publications
Modelling Speaker Attribution in Narrative Texts With Biased and Bias-Adjustable Neural Networks.
Dönicke Tillmann T Varachkina Hanna H Weimer Anna Mareike AM Gödeke Luisa L Barth Florian F Gittel Benjamin B Holler Anke A Sporleder Caroline C
Frontiers in artificial intelligence 20220203
Literary narratives regularly contain passages that different readers attribute to different speakers: a character, the narrator, or the author. Since literary narratives are highly ambiguous constructs, it is often impossible to decide between diverging attributions of a specific passage by hermeneutic means. Instead, we hypothesise that attribution decisions are often influenced by annotator bias, in particular an annotator's literary preferences and beliefs. We present first results on the co ...[more]
Similar Datasets
Project description:Automatic authorship identification is a challenging task that has been the focus of extensive research in natural language processing. Regardless of the progress made in attributing authorship, the need for corpora in under-resourced languages impedes advancing and examining present methods. To address this gap, we investigate the problem of authorship attribution in Albanian. We introduce a newly compiled corpus of Albanian newsroom columns and literary works and analyze machine-learning methods for detecting authorship. We create a set of hand-crafted features targeting various categories (lexical, morphological, and structural) relevant to Albanian and experiment with multiple classifiers using two different multiclass classification strategies. Furthermore, we compare our results to those obtained using deep learning models. Our investigation focuses on identifying the best combination of features and classification methods. The results reveal that lexical features are the most effective set of linguistic features, significantly improving the performance of various algorithms in the authorship attribution task. Among the machine learning algorithms evaluated, XGBoost demonstrated the best overall performance, achieving an F1 score of 0.982 on literary works and 0.905 on newsroom columns. Additionally, deep learning models such as fastText and BERT-multilingual showed promising results, highlighting their potential applicability in specific scenarios in Albanian writings. These findings contribute to the understanding of effective methods for authorship attribution in low-resource languages and provide a robust framework for future research in this area. The careful analysis of the different scenarios and the conclusions drawn from the results provide valuable insights into the potential and limitations of the methods and highlight the challenges in detecting authorship in Albanian. Promising results are reported, with implications for improving the methods used in Albanian authorship attribution. This study provides a valuable resource for future research and a reference for researchers in this domain.
| S-EPMC11495635 | biostudies-literature
Project description:The main sources of information regarding ancient Mesopotamian history and culture are clay cuneiform tablets. Many of these tablets are damaged, leading to missing information. Currently, the missing text is manually reconstructed by experts. We investigate the possibility of assisting scholars, by modeling the language using recurrent neural networks and automatically completing the breaks in ancient Akkadian texts from Achaemenid period Babylonia.
| S-EPMC7502733 | biostudies-literature
Project description:Ancient history relies on disciplines such as epigraphy-the study of inscribed texts known as inscriptions-for evidence of the thought, language, society and history of past civilizations1. However, over the centuries, many inscriptions have been damaged to the point of illegibility, transported far from their original location and their date of writing is steeped in uncertainty. Here we present Ithaca, a deep neural network for the textual restoration, geographical attribution and chronological attribution of ancient Greek inscriptions. Ithaca is designed to assist and expand the historian's workflow. The architecture of Ithaca focuses on collaboration, decision support and interpretability. While Ithaca alone achieves 62% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25% to 72%, confirming the synergistic effect of this research tool. Ithaca can attribute inscriptions to their original location with an accuracy of 71% and can date them to less than 30 years of their ground-truth ranges, redating key texts of Classical Athens and contributing to topical debates in ancient history. This research shows how models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history.
| S-EPMC8907065 | biostudies-literature
Project description:Effective natural communication requires listeners to incorporate not only very general linguistic principles which evolved during a lifetime but also other information like the specific individual language use of a particular interlocutor. Traditionally, research has focused on the general linguistic rules, and brain science has shown a left hemispheric fronto-temporal brain network related to this processing. The present fMRI research explores speaker-specific individual language use because it is unknown whether this processing is supported by similar or distinct neural structures. Twenty-eight participants listened to sentences of persons who used more easy or difficult language. This was done by manipulating the proportion of easy SOV vs. complex OSV sentences for each speaker. Furthermore, ambiguous probe sentences were included to test top-down influences of speaker information in the absence of syntactic structure information. We observed distinct neural processing for syntactic complexity and speaker-specific language use. Syntactic complexity correlated with left frontal and posterior temporal regions. Speaker-specific processing correlated with bilateral (right-dominant) fronto-parietal brain regions. Finally, the top-down influence of speaker information was found in frontal and striatal brain regions, suggesting a mechanism for controlled syntactic processing. These findings show distinct neural networks related to general language principles as well as speaker-specific individual language use.
| S-EPMC8153046 | biostudies-literature
Project description:IntroductionSpiking neural networks (SNNs) are a model of computation that mimics the behavior of biological neurons. SNNs process event data (spikes) and operate more sparsely than artificial neural networks (ANNs), resulting in ultra-low latency and small power consumption. This paper aims to adapt and evaluate gradient-based explainability methods for SNNs, which were originally developed for conventional ANNs.MethodsThe adapted methods aim to create input feature attribution maps for SNNs trained through backpropagation that process either event-based spiking data or real-valued data. The methods address the limitations of existing work on explainability methods for SNNs, such as poor scalability, limited to convolutional layers, requiring the training of another model, and providing maps of activation values instead of true attribution scores. The adapted methods are evaluated on classification tasks for both real-valued and spiking data, and the accuracy of the proposed methods is confirmed through perturbation experiments at the pixel and spike levels.Results and discussionThe results reveal that gradient-based SNN attribution methods successfully identify highly contributing pixels and spikes with significantly less computation time than model-agnostic methods. Additionally, we observe that the chosen coding technique has a noticeable effect on the input features that will be most significant. These findings demonstrate the potential of gradient-based explainability methods for SNNs in improving our understanding of how these networks process information and contribute to the development of more efficient and accurate SNNs.
| S-EPMC10565802 | biostudies-literature
Project description:Studies on attribution biases in schizophrenia have produced mixed results, whereas such biases have been more consistently reported in people with anxiety disorders. Anxiety comorbidities are frequent in schizophrenia, in particular social anxiety disorder, which could influence their patterns of attribution biases. The objective of the present study was thus to determine if individuals with schizophrenia and a comorbid social anxiety disorder (SZ+) show distinct attribution biases as compared with individuals with schizophrenia without social anxiety (SZ-) and healthy controls. Attribution biases were assessed with the Internal, Personal, and Situational Attributions Questionnaire in 41 individual with schizophrenia and 41 healthy controls. Results revealed the lack of the normal externalizing bias in SZ+, whereas SZ- did not significantly differ from healthy controls on this dimension. The personalizing bias was not influenced by social anxiety but was in contrast linked with delusions, with a greater personalizing bias in individuals with current delusions. Future studies on attribution biases in schizophrenia should carefully document symptom presentation, including social anxiety.
| S-EPMC5506709 | biostudies-other
Project description:Individuals often align their emotional states during conversation. Here, we reveal how such emotional alignment is reflected in synchronization of brain activity across speakers and listeners. Two "speaker" subjects told emotional and neutral autobiographical stories while their hemodynamic brain activity was measured with functional magnetic resonance imaging (fMRI). The stories were recorded and played back to 16 "listener" subjects during fMRI. After scanning, both speakers and listeners rated the moment-to-moment valence and arousal of the stories. Time-varying similarity of the blood-oxygenation-level-dependent (BOLD) time series was quantified by intersubject phase synchronization (ISPS) between speaker-listener pairs. Telling and listening to the stories elicited similar emotions across speaker-listener pairs. Arousal was associated with increased speaker-listener neural synchronization in brain regions supporting attentional, auditory, somatosensory, and motor processing. Valence was associated with increased speaker-listener neural synchronization in brain regions involved in emotional processing, including amygdala, hippocampus, and temporal pole. Speaker-listener synchronization of subjective feelings of arousal was associated with increased neural synchronization in somatosensory and subcortical brain regions; synchronization of valence was associated with neural synchronization in parietal cortices and midline structures. We propose that emotion-dependent speaker-listener neural synchronization is associated with emotional contagion, thereby implying that listeners reproduce some aspects of the speaker's emotional state at the neural level.
| S-EPMC6865790 | biostudies-literature
Project description:A critical component of textbooks is fair representation of the material they cover. Within conservation biology, fair coverage is particularly important given Earth's breadth of species and diversity of ecosystems. However, research on species tends to be biased towards certain taxonomic groups and geographic areas and their associated ecosystems, so it is possible that textbooks may exhibit similar biases. Considering the possibility of bias, our goal was to evaluate contemporary conservation biology textbooks to determine if they are representative of Earth's biodiversity. We found that textbooks did not accurately reflect Earth's biodiversity. Species, ecosystems, and continents were unevenly represented, few examples mentioned genetic diversity, and examples of negative human influence on the environment outweighed positive examples. However, in terms of aquatic versus terrestrial representation, textbooks presented a representative sample. Our findings suggest that modern conservation biology textbooks are biased in their coverage, which could have important consequences for educating our next generation of scientists and practitioners.
| S-EPMC7351170 | biostudies-literature
Project description:Verbal communication is a joint activity; however, speech production and comprehension have primarily been analyzed as independent processes within the boundaries of individual brains. Here, we applied fMRI to record brain activity from both speakers and listeners during natural verbal communication. We used the speaker's spatiotemporal brain activity to model listeners' brain activity and found that the speaker's activity is spatially and temporally coupled with the listener's activity. This coupling vanishes when participants fail to communicate. Moreover, though on average the listener's brain activity mirrors the speaker's activity with a delay, we also find areas that exhibit predictive anticipatory responses. We connected the extent of neural coupling to a quantitative measure of story comprehension and find that the greater the anticipatory speaker-listener coupling, the greater the understanding. We argue that the observed alignment of production- and comprehension-based processes serves as a mechanism by which brains convey information.
| S-EPMC2922522 | biostudies-literature
Project description:Our voice provides salient cues about how confident we sound, which promotes inferences about how believable we are. However, the neural mechanisms involved in these social inferences are largely unknown. Employing functional magnetic resonance imaging, we examined the brain networks and individual differences underlying the evaluation of speaker believability from vocal expressions. Participants (n = 26) listened to statements produced in a confident, unconfident, or "prosodically unmarked" (neutral) voice, and judged how believable the speaker was on a 4-point scale. We found frontal-temporal networks were activated for different levels of confidence, with the left superior and inferior frontal gyrus more activated for confident statements, the right superior temporal gyrus for unconfident expressions, and bilateral cerebellum for statements in a neutral voice. Based on listener's believability judgment, we observed increased activation in the right superior parietal lobule (SPL) associated with higher believability, while increased left posterior central gyrus (PoCG) was associated with less believability. A psychophysiological interaction analysis found that the anterior cingulate cortex and bilateral caudate were connected to the right SPL when higher believability judgments were made, while supplementary motor area was connected with the left PoCG when lower believability judgments were made. Personal characteristics, such as interpersonal reactivity and the individual tendency to trust others, modulated the brain activations and the functional connectivity when making believability judgments. In sum, our data pinpoint neural mechanisms that are involved when inferring one's believability from a speaker's voice and establish ways that these mechanisms are modulated by individual characteristics of a listener. Hum Brain Mapp 38:3732-3749, 2017. © 2017 Wiley Periodicals, Inc.
| S-EPMC6866933 | biostudies-literature