From Curriculum Vitae parsing to web search and recommendation systems, Word2Vec and other word embedding techniques have an increasing presence in everyday interactions in human society. Biases, such as gender bias, have been thoroughly researched and evidenced to be present in word embeddings. Most of the research focuses on discovering and mitigating gender bias within the frames of the vector space itself. Nevertheless, whose bias is reflected in word embeddings has not yet been investigated. Besides discovering (...) and mitigating gender bias, it is also important to examine whether a feminine or a masculine-centric view is represented in the biases of word embeddings. This way, we will not only gain more insight into the origins of the before mentioned biases, but also present a novel approach to investigating biases in Natural Language Processing systems. Based on previous research in the social sciences and gender studies, we hypothesize that masculine-centric, otherwise known as androcentric, biases are dominant in word embeddings. To test this hypothesis we used the largest English word association test data set publicly available. We compare the distance of the responses of male and female participants to cue words in a word embedding vector space. We found that the word embedding is biased towards a masculine-centric viewpoint, predominantly reflecting the worldviews of the male participants in the word association test data set. Therefore, by conducting this research, we aimed to unravel another layer of bias to be considered when examining fairness in algorithms. (shrink)

The use of ensemble learning, deep learning, and effective document representation methods is currently some of the most common trends to improve the overall accuracy of a text classification/categorization system. Ensemble learning is an approach to raise the overall accuracy of a classification system by utilizing multiple classifiers. Deep learning-based methods provide better results in many applications when compared with the other conventional machine learning algorithms. Word embeddings enable representation of words learned from a corpus as vectors that provide (...) a mapping of words with similar meaning to have similar representation. In this study, we use different document representations with the benefit of word embeddings and an ensemble of base classifiers for text classification. The ensemble of base classifiers includes traditional machine learning algorithms such as naïve Bayes, support vector machine, and random forest and a deep learning-based conventional network classifier. We analysed the classification accuracy of different document representations by employing an ensemble of classifiers on eight different datasets. Experimental results demonstrate that the usage of heterogeneous ensembles together with deep learning methods and word embeddings enhances the classification performance of texts. (shrink)

The number of scientific publications is growing exponentially. Research articles cite other work for various reasons and, therefore, have been studied extensively to associate documents. It is argued that not all references carry the same level of importance. It is essential to understand the reason for citation, called citation intent or function. Text information can contribute well if new natural language processing techniques are applied to capture the context of text data. In this paper, we have used contextualized word (...) embedding to find the numerical representation of text features. We further investigated the performance of various machine-learning techniques on the numerical representation of text. The performance of each of the classifiers was evaluated on two state-of-the-art datasets containing the text features. In the case of the unbalanced dataset, we observed that the linear Support Vector Machine achieved 86% accuracy for the “background” class, where the training was extensive. For the rest of the classes, including “motivation,” “extension,” and “future,” the machine was trained on less than 100 records; therefore, the accuracy was only 57 to 64%. In the case of a balanced dataset, each of the classes has the same accuracy as trained on the same size of training data. Overall, SVM performed best on both of the datasets, followed by the stochastic gradient descent classifier; therefore, SVM can produce good results as text classification on top of contextual word embedding. (shrink)

In this study, we use temporally aligned word embeddings and a large diachronic corpus of English to quantify language change in a data-driven, scalable way, which is grounded in language use. We show a unique and reliable relation between measures of language change and age of acquisition (AoA) while controlling for frequency, contextual diversity, concreteness, length, dominant part of speech, orthographic neighborhood density, and diachronic frequency variation. We analyze measures of language change tackling both the change in lexical representations (...) and the change in the relation between lexical representations and the words with the most similar usage patterns, showing that they capture different aspects of language change. Our results show a unique relation between language change and AoA, which is stronger when considering neighborhood-level measures of language change: Words with more coherent diachronic usage patterns tend to be acquired earlier. The results support theories positing a link between ontogenetic and ethnogenetic processes in language. (shrink)

Deep Learning has been widely used for tackling challenging natural language processing tasks over the recent years. Similarly, the application of Deep Neural Networks in legal analytics has increased significantly. In this survey, we study the early adaptation of Deep Learning in legal analytics focusing on three main fields; text classification, information extraction, and information retrieval. We focus on the semantic feature representations, a key instrument for the successful application of deep learning in natural language processing. Additionally, we share pre-trained (...) legal word embeddings using the word2vec model over large corpora, comprised legislations from UK, EU, Canada, Australia, USA, and Japan among others. (shrink)

Public companies in the US stock market must annually report their activities and financial performances to the SEC by filing the so-called 10-K form. Recent studies have demonstrated that changes in the textual content of the corporate annual filing can convey strong signals of companies’ future returns. In this study, we combine natural language processing techniques and network science to introduce a novel 10-K-based network, named Lazy Network, that leverages year-on-year changes in companies’ 10-Ks detected using a neural network (...) class='Hi'>embedding model. The Lazy Network aims to capture textual changes derived from financial or economic changes on the equity market. Leveraging the Lazy Network, we present a novel investment strategy that attempts to select the least disrupted and stable companies by capturing the peripheries of the Lazy Network. We show that this strategy earns statistically significant risk-adjusted excess returns. Specifically, the proposed portfolios yield up to 95 basis points in monthly five-factor alphas, outperforming similar strategies in the literature. (shrink)

Deep Learning has been widely used for tackling challenging natural language processing tasks over the recent years. Similarly, the application of Deep Neural Networks in legal analytics has increased significantly. In this survey, we study the early adaptation of Deep Learning in legal analytics focusing on three main fields; text classification, information extraction, and information retrieval. We focus on the semantic feature representations, a key instrument for the successful application of deep learning in natural language processing. Additionally, we share pre-trained (...) legal word embeddings using the word2vec model over large corpora, comprised legislations from UK, EU, Canada, Australia, USA, and Japan among others. (shrink)

Automatic synonym extraction plays an important role in many natural language processing systems, such as those involving information retrieval and question answering. Recently, research has focused on extracting semantic relations from word embeddings since they capture relatedness and similarity between words. However, using word embeddings alone poses problems for synonym extraction because it cannot determine whether the relation between words is synonymy or some other semantic relation. In this paper, we present a novel solution for this problem by (...) proposing the SynoExtractor pipeline, which can be used to filter similar word embeddings to retain synonyms based on specified linguistic rules. Our experiments were conducted using KSUCCA and Gigaword embeddings and trained with CBOW and SG models. We evaluated automatically extracted synonyms by comparing them with Alma’any Arabic synonym thesauri. We also arranged for a manual evaluation by two Arabic linguists. The results of experiments we conducted show that using the SynoExtractor pipeline enhances the precision of synonym extraction compared to using the cosine similarity measure alone. SynoExtractor obtained a 0.605 mean average precision for the King Saud University Corpus of Classical Arabic with 21% improvement over the baseline and a 0.748 MAP for the Gigaword corpus with 25% improvement. SynoExtractor outperformed the Sketch Engine thesaurus for synonym extraction by 32% in terms of MAP. Our work shows promising results for synonym extraction suggesting that our method can also be used with other languages. (shrink)

Programming artificial intelligence to make fairness assessments of texts through top-down rules, bottom-up training, or hybrid approaches, has presented the challenge of defining cross-cultural fairness. In this paper a simple method is presented which uses vectors to discover if a verb is unfair or fair. It uses already existing relational social ontologies inherent in Word Embeddings and thus requires no training. The plausibility of the approach rests on two premises. That individuals consider fair acts those that they would be (...) willing to accept if done to themselves. Secondly, that such a construal is ontologically reflected in Word Embeddings, by virtue of their ability to reflect the dimensions of such a perception. These dimensions being: responsibility vs. irresponsibility, gain vs. loss, reward vs. sanction, joy vs. pain, all as a single vector. The paper finds it possible to quantify and qualify a verb as fair or unfair by calculating the cosine similarity of the said verb’s embedding vector against FairVec—which represents the above dimensions. We apply this to Glove and Word2Vec embeddings. Testing on a list of verbs produces an F1 score of 95.7, which is improved to 97.0. Lastly, a demonstration of the method’s applicability to sentence measurement is carried out. (shrink)

Programming artificial intelligence to make fairness assessments of texts through top-down rules, bottom-up training, or hybrid approaches, has presented the challenge of defining cross-cultural fairness. In this paper a simple method is presented which uses vectors to discover if a verb is unfair or fair. It uses already existing relational social ontologies inherent in Word Embeddings and thus requires no training. The plausibility of the approach rests on two premises. That individuals consider fair acts those that they would be (...) willing to accept if done to themselves. Secondly, that such a construal is ontologically reflected in Word Embeddings, by virtue of their ability to reflect the dimensions of such a perception. These dimensions being: responsibility vs. irresponsibility, gain vs. loss, reward vs. sanction, joy vs. pain, all as a single vector. The paper finds it possible to quantify and qualify a verb as fair or unfair by calculating the cosine similarity of the said verb’s embedding vector against FairVec—which represents the above dimensions. We apply this to Glove and Word2Vec embeddings. Testing on a list of verbs produces an F1 score of 95.7, which is improved to 97.0. Lastly, a demonstration of the method’s applicability to sentence measurement is carried out. (shrink)

We examined whether variations in contextual diversity, spacing, and retrieval practice influenced how well adults learned new words from reading experience. Eye movements were recorded as adults read novel words embedded in sentences. In the learning phase, unfamiliar words were presented either in the same sentence repeated four times (same context) or in four different sentences (diverse context). Spacing was manipulated by presenting the sentences under distributed or non‐distributed practice. After learning, half of the participants were asked to retrieve the (...) new words, and half had an extra exposure to the new words. Although words experienced in diverse contexts were acquired more slowly during learning, they enjoyed a greater benefit of learning at immediate posttest. Distributed practice also slowed learning, but no benefit was observed at posttest. Although participants who had an extra exposure showed the greatest learning benefit overall, learning also benefited from retrieval opportunity, when words were experienced in diverse contexts. These findings demonstrate that variation in the content and structure of the learning environment impacts on word learning via reading. (shrink)

The principal argument suggested here is that we are all morally embedded selves: We have no control over the abilities that make us moral agents nor can we control the degree to which we have these abilities; in other words, we are not responsible for our good or bad qualities as moral agents. This, I believe, calls for the adoption of embedded compatibilism (EC). According to EC, people have control over their conduct; this control, however, is embedded within prerequisites, which (...) they cannot control and hence are not responsible for having or lacking. On the one hand, EC enables us to explain why a lack of control at the ultimate level does not eliminate moral judgment altogether. However on the other hand, EC ought to change our understanding of moral responsibility; interalia, it supports a hybrid notion of punishment, indicates the incomplete nature of guilt and reintroduces the problem of moral luck. (shrink)

In this chapter, I am going to discuss a very interesting case brought to our attention by Saul and references therein: NP-related substitution failure in simple sentences. Whereas it is well known that opacity occurs in intensional contexts and that in such contexts it is not licit to replace an NP with a co-referential one, one would not expect that substitution failure should also be exhibited by simple sentences in the context of stories about Superman. The suggested explanation of these (...) cases is to posit an embedding explicature, that is to say the insertion of structure that ipso facto creates an intensional context capable of blocking substitution. I consider various complications to this story in the light of important objections by García-Carpintero and, finally, I consider how this story fares when one applies constraints on explicatures along the lines of those proposed by Hall in an interesting paper.In general, this chapter exploits interesting considerations by Norrick on the structural similarities between stories and indirect reports. Norrick believes there are important differences, but he is inclined to concede that we could study structural similarities. An important similarity, brought out by the examples discussed by Saul, is that the narrative frame, once it is inserted into the interaction, can be left implicit and, during the act of narrating or referring to the story, one need not repeat the words ‘the story says’ or ‘we are told that…’ every time. Although implicit, these words are heard because they do some work at the structural level, as is shown by this attempt to resolve an otherwise intractable philosophical problem. The explicatures of simple sentences are perceived because they are integrated into the speakers’/hearers’ perception of the overall plan of discourse, as Haugh most interestingly notes:As Haugh and Jaszczolt note, this means that any putative “communicative intention of A is embedded within his higher-order intention”. In other words, to figure out the implicature that evidently arises here, the participants are necessarily making inferences about some kind of overall aim. According to this view, then, inferences about the intended implicature arise concomitant with inferences about the overall aim of the speaker.It follows from the considerations by Haugh that, since the explicature connected with simple sentences depends on the perception of the overall aim or plan of the conversation, it is not easily cancelled. Readers can check by themselves that the explicatures due to simple sentences cannot be cancelled, as cancelling them would involve returning to illogical discourses.. Haugh’s considerations about the overall aim of the discourse are precious in explaining how the embedding explicatures I posit are calculated once and for all for the whole stretch of the discourse framed by the narrative act. (shrink)

With two eye movements tracking experiments, present study sought to establish whether predictability of target words facilitate characters processing in Chinese reading. Target Words embedded in sentences in both experiments were composed by 2 Chinese characters. Predictability of target words were manipulated in both experiments, beyond that, frequency of targets' first characters were manipulated in Experiment 1, frequency of targets' second characters were manipulated in Experiment 2. Pervasive predictability effects were observed on almost all of eye movement measures in both (...) experiments. It also yield reliable characters frequency effect in Experiment 1 & 2, on some measures. None reliable interaction effects of predictability and first character frequency were observed in Experiment 1, but predictability and second character frequency interactively impacted on some measures of eye movement. Theoretical implications of the data are discussed. (shrink)

The pervasive use of distributional semantic models or word embeddings for both cognitive modeling and practical application is because of their remarkable ability to represent the meanings of words. However, relatively little effort has been made to explore what types of information are encoded in distributional word vectors. Knowing the internal knowledge embedded in word vectors is important for cognitive modeling using distributional semantic models. Therefore, in this paper, we attempt to identify the knowledge encoded in (...) class='Hi'>word vectors by conducting a computational experiment using Binder et al.'s (2016) featural conceptual representations based on neurobiologically motivated attributes. In an experiment, these conceptual vectors are predicted from text‐based word vectors using a neural network and linear transformation, and prediction performance is compared among various types of information. The analysis demonstrates that abstract information is generally predicted more accurately by word vectors than perceptual and spatiotemporal information, and specifically, the prediction accuracy of cognitive and social information is higher. Emotional information is also found to be successfully predicted for abstract words. These results indicate that language can be a major source of knowledge about abstract attributes, and they support the recent view that emphasizes the importance of language for abstract concepts. Furthermore, we show that word vectors can capture some types of perceptual and spatiotemporal information about concrete concepts and some relevant word categories. This suggests that language statistics can encode more perceptual knowledge than often expected. (shrink)

From an early age, exposure to a spoken language has allowed us to implicitly capture the structure underlying the succession of speech sounds in that language and to segment it into meaningful units. Statistical learning, the ability to pick up patterns in the sensory environment without intention or reinforcement, is thus assumed to play a central role in the acquisition of the rule-governed aspects of language, including the discovery of word boundaries in the continuous acoustic stream. Although extensive evidence (...) has been gathered from artificial languages experiments showing that children and adults are able to track the regularities embedded in the auditory input, as the probability of one syllable to follow another syllable in the speech stream, the developmental trajectory of this ability remains controversial. In this work, we have collected Event-Related Potentials while 5-year-old children and young adults were exposed to a speech stream made of the repetition of eight three-syllable nonsense words presenting different levels of predictability to mimic closely what occurs in natural languages and to get new insights into the changes that the mechanisms underlying auditory statistical learning might undergo through the development. The participants performed the aSL task first under implicit and, subsequently, under explicit conditions to further analyze if children take advantage of previous knowledge of the to-be-learned regularities to enhance SL, as observed with the adult participants. These findings would also contribute to extend our knowledge of the mechanisms available to assist SL at each developmental stage. Although behavioral signs of learning, even under explicit conditions, were only observed for the adult participants, ERP data showed evidence of online segmentation in the brain in both groups, as indexed by modulations in the N100 and N400 components. A detailed analysis of the neural data suggests, however, that adults and children rely on different mechanisms to assist the extraction of word-like units from the continuous speech stream, hence supporting the view that SL with auditory linguistic materials changes through development. (shrink)

We have two polynomial time results for the uniform word problem for a quasivariety Q: The uniform word problem for Q can be solved in polynomial time iff one can find a certain congruence on finite partial algebras in polynomial time. Let Q* be the relational class determined by Q. If any universal Horn class between the universal closure S and the weak embedding closure S̄ of Q* is finitely axiomatizable then the uniform word problem for (...) Q is solvable in polynomial time. This covers Skolem's 1920 solution to the uniform word problem for lattices and Evans' 1953 applications of the weak embeddability property for finite partial V algebras. (shrink)

Parts of the body are often embedded in the structure of compound words, such as heartbreak and brainchild. We explored the relationships between the semantics of compounds and their constituent body parts, asking whether these relationships are largely arbitrary or instead reflect deeper metaphorical mappings shared across languages and cultures. In three studies, we found that U.S. English speakers associated the English translation equivalents of Chinese compounds with their constituent body parts at rates well above chance, even for compounds with (...) highly abstract meanings and even when accounting for the semantic relatedness of the compounds and body parts. English speakers in India and Chinese speakers in Hong Kong showed similar intuitions about these associations. Our results suggest that the structure of compound words can provide insight into cross‐culturally shared ways of connecting meaning to the body. (shrink)

Among the myriad of technical approaches and abstract guidelines proposed to the topic of AI bias, there has been an urgent call to translate the principle of fairness into the operational AI reality with the involvement of social sciences specialists to analyse the context of specific types of bias, since there is not a generalizable solution. This article offers an interdisciplinary contribution to the topic of AI and societal bias, in particular against the poor, providing a conceptual framework of the (...) issue and a tailor-made model from which meaningful data are obtained using Natural Language Processing word vectors in pretrained Google Word2Vec, Twitter and Wikipedia GloVe word embeddings. The results of the study offer the first set of data that evidences the existence of bias against the poor and suggest that Google Word2vec shows a higher degree of bias when the terms are related to beliefs, whereas bias is higher in Twitter GloVe when the terms express behaviour. This article contributes to the body of work on bias, both from and AI and a social sciences perspective, by providing evidence of a transversal aggravating factor for historical types of discrimination. The evidence of bias against the poor also has important consequences in terms of human development, since it often leads to discrimination, which constitutes an obstacle for the effectiveness of poverty reduction policies. (shrink)

This article presents and interprets a number of neglected paradoxes in early Chinese philosophical texts (ca. 500-100 B.C.). Looking beyond well-known paradoxes put forward by masters such as Hui Shi and Gongsun Long, it intends to complement our picture of Warring States and early Western Han paradoxical statements. The first section contrasts the neglected paradoxes with the well-known ones. It is contended here that our understanding of these latter paradoxes is hampered by a lack of context and that the neglected (...) paradoxes possess an interpretative advantage by virtue of their being context-embedded. The second section presents an overview of three groups of neglected paradoxes, showing that the paradoxical effect of these paradoxes results from a challenge to the semantics of their central terms. The third section discusses the distribution of the paradoxes throughout the early literature and concludes that they typically appear in "Daoist" writings. The final section proposes a semantic-rhetorical interpretation. Placing the paradoxes against the background of the features and use of important terms, it is argued that they constitute unorthodox redefinitions and are formulated to influence the behavior and values of their intended audience. (shrink)

For a quasi variety of algebras K, the Higman Theorem is said to be true if every recursively presented K-algebra is embeddable into a finitely presented K-algebra; the Generalized Higman Theorem is said to be true if any K-algebra which is recursively presented over its finitely generated subalgebra is embeddable into a K-algebra which is finitely presented over this subalgebra. We suggest certain general conditions on K under which the Higman Theorem implies the Generalized Higman Theorem; a finitely generated K-algebra (...) A is embeddable into every existentially closed K-algebra containing a finitely generated K-algebra B if and only if the word problem for A is Q-reducible to the word problem for B. The quasi varieties of groups, torsion-free groups, and semigroups satisfy these conditions. (shrink)

This article presents and interprets a number of neglected paradoxes in early Chinese philosophical texts . Looking beyond well-known paradoxes put forward by masters such as Hui Shi and Gongsun Long, it intends to complement our picture of Warring States and early Western Han paradoxical statements. The first section contrasts the neglected paradoxes with the well-known ones. It is contended here that our understanding of these latter paradoxes is hampered by a lack of context and that the neglected paradoxes possess (...) an interpretative advantage by virtue of their being context-embedded. The second section presents an overview of three groups of neglected paradoxes, showing that the paradoxical effect of these paradoxes results from a challenge to the semantics of their central terms. The third section discusses the distribution of the paradoxes throughout the early literature and concludes that they typically appear in "Daoist" writings. The final section proposes a semantic-rhetorical interpretation. Placing the paradoxes against the background of the features and use of important terms, it is argued that they constitute unorthodox redefinitions and are formulated to influence the behavior and values of their intended audience. (shrink)

To make a business case for corporate sustainability, firms must be able to sell their sustainable products. The influence that firm engagement with non-governmental organizations may have on consumer adoption of sustainable products has been neglected in previous research. We address this by embedding corporate sustainability in a cosmopolitan framework that connects firms, consumers, and civil society organizations based on the understanding of responsibility for global humanity that underlies both the sustainability and cosmopolitanism concepts. We hypothesize that firms’ sustainability (...) engagement and their NGO engagement influence consumer adoption of sustainable products. Empirically, we investigate the adoption of sustainable Eco-circle products made from recycled fibers marketed by Li Ning, a China-based global sportswear brand. We apply a stepwise regression approach to test our hypotheses with paper-and-pencil survey data from 217 Chinese consumers. We find adoption to be positively associated with consumers’ sustainability attitude but not with firms’ sustainability engagement. For firm–NGO engagement, these relationships are reversed: Adoption is positively associated with firm–NGO engagement, but not with consumers’ related attitude. Our results present a picture of the Chinese context in which there is a business case for corporate sustainability if firms’ words about sustainable product strategies are supported by signals from civil society about firm deeds. The results imply that in a Chinese context, firms need to be particularly aware of the role of NGOs when hoping to be rewarded for sustainability. (shrink)

To make a business case for corporate sustainability, firms must be able to sell their sustainable products. The influence that firm engagement with non-governmental organizations (NGOs) may have on consumer adoption of sustainable products has been neglected in previous research. We address this by embedding corporate sustainability in a cosmopolitan framework that connects firms, consumers, and civil society organizations based on the understanding of responsibility for global humanity that underlies both the sustainability and cosmopolitanism concepts. We hypothesize that firms’ (...) sustainability engagement and their NGO engagement influence consumer adoption of sustainable products. Empirically, we investigate the adoption of sustainable Eco-circle products made from recycled fibers marketed by Li Ning, a China-based global sportswear brand. We apply a stepwise regression approach to test our hypotheses with paper-and-pencil survey data from 217 Chinese consumers. We find adoption to be positively associated with consumers’ sustainability attitude but not with firms’ sustainability engagement. For firm–NGO engagement, these relationships are reversed: Adoption is positively associated with firm–NGO engagement, but not with consumers’ related attitude. Our results present a picture of the Chinese context in which there is a business case for corporate sustainability if firms’ words about sustainable product strategies are supported by signals from civil society about firm deeds. The results imply that in a Chinese context, firms need to be particularly aware of the role of NGOs when hoping to be rewarded for sustainability. (shrink)

The current study aims to separate conscious and unconscious behaviors by employing both online and offline measures while the participants were consciously performing a task. Using an eye-movement tracking paradigm, we observed participants’ response patterns for distinguishing within-word-boundary and across-word-boundary reverse errors while reading Chinese sentences . The results showed that when the participants consciously detected errors, their gaze time for target words associated with across-word-boundary reverse errors was significantly longer than that for targets words associated with (...) within-word-boundary reverse errors. Surprisingly, the same gaze time pattern was found even when the readers were not consciously aware of the reverse errors. The results were statistically robust, providing converging evidence for the feasibility of our experimental paradigm in decoupling offline behaviors and the online, automatic, and unconscious aspects of cognitive processing in reading. (shrink)

In the leadership literature of the past 100 years or so, rhetoric has been a topic for a long time and ethics was introduced some 30 years ago. However, the three topics, leadership, rhetoric, and ethics, have not been connected. This is astonishing because when ethical leadership made its comeback, scholars acknowledged the debt that ethical leadership owes to Aristotelian ideas. For Aristotle, leadership, ethics, and rhetoric were inseparable: without ethics, there could neither be good leadership nor rhetoric, and rhetoric (...) was the bridge from leader to follower and back. Aristotle is known as the practical philosopher and we argue that closing this major lacuna in leadership theory, would be beneficial for practice too and would help prevent Theranos-like scandals. The lacuna lies in what some call a Machiavellian view of leadership and rhetoric: both are regarded as amoral tools to achieve given ends. Aristotle would bristle at that idea. Both, leadership and rhetoric, to be good, must aim at the noble. The noble is not dictated by a heroic despot nor by a selfless saint but is understood by a leader because s/he is a socially embedded individual. We argue that leadership should be understood as personal development, that organisational rules must not inhibit virtue, that leadership is more about listening than talking, that synthesising that which is heard is central, and finally, that the common, not the collective, good is the goal of leadership. (shrink)

A distributed neural network model called SPEC for processing sentences with recursive relative clauses is described. The model is based on separating the tasks of segmenting the input word sequence into clauses, forming the case‐role representations, and keeping track of the recursive embeddings into different modules. The system needs to be trained only with the basic sentence constructs, and it generalizes not only to new instances of familiar relative clause structures but to novel structures as well. SPEC exhibits plausible (...) memory degradation as the depth of the center embeddings increases, its memory is primed by earlier constituents, and its performance is aided by semantic constraints between the constituents. The ability to process structure is largely due to a central executive network that monitors and controls the execution of the entire system. This way, in contrast to earlier subsymbolic systems, parsing is modeled as a controlled high‐level process rather than one based on automatic reflex responses. (shrink)

This contribution defends Ripstein's attempt to reconstruct Kant's political philosophy as entirely and consistently grounded on the idea of people's innate right to freedom as independence, in particular with respect to charges of circularity raised by other contributors to this symposium. However, it also argues that, if the concept of freedom as independence is to provide a foundation for a full-blown account of political justice, a richer interpretation of it should be provided. In other words, we must be willing to (...) make controversial and empirically informed claims about what counts as a threat to our freedom as independence under specific circumstances. We must have a more embedded account of freedom as independence, one that engages with the contingencies of politics and of the human condition. (shrink)

The linguistic input children receive across early childhood plays a crucial role in shaping their knowledge about the world. To study this input, researchers have begun applying distributional semantic models to large corpora of child‐directed speech, extracting various patterns of word use/co‐occurrence. Previous work using these models has not measured how these patterns may change throughout development, however. In this work, we leverage natural language processing methods—originally developed to study historical language change—to compare caregivers' use of words when talking (...) to younger versus older children. Some words' usage changed more than others; this variability could be predicted based on the word's properties at both the individual and category levels. These findings suggest that caregivers' changing patterns of word use may play a role in scaffolding children's acquisition of conceptual structure in early development. (shrink)

Regular polysemes are sets of ambiguous words that all share the same relationship between their meanings, such as CHICKEN and LOBSTER both referring to an animal or its meat. To probe how a distributional semantic model, here exemplified by bidirectional encoder representations from transformers (BERT), represents regular polysemy, we analyzed whether its embeddings support answering sense analogy questions similar to “is the mapping between CHICKEN (as an animal) and CHICKEN (as a meat) similar to that which maps between LOBSTER (as (...) an animal) to LOBSTER (as a meat)?” We did so using the LRcos model, which combines a logistic regression classifier of different categories (e.g., animal vs. meat) with a measure of cosine similarity. We found that (a) the model was sensitive to the shared structure within a given regular relationship; (b) the shared structure varies across different regular relationships (e.g., animal/meat vs. location/organization), potentially reflective of a “regularity continuum;” (c) some high-order latent structure is shared across different regular relationships, suggestive of a similar latent structure across different types of relationships; and (d) there is a lack of evidence for the aforementioned effects being explained by meaning overlap. Lastly, we found that both components of the LRcos model made important contributions to accurate responding and that a variation of this method could yield an accuracy boost of 10% in answering sense analogy questions. These findings enrich previous theoretical work on regular polysemy with a computationally explicit theory and methods, and provide evidence for an important organizational principle for the mental lexicon and the broader conceptual knowledge system. (shrink)

The commentaries provide a multitude of perspectives on the theory of lexical access presented in our target article. We respond, on the one hand, to criticisms that concern the embeddings of our model in the larger theoretical frameworks of human performance and of a speaker's multiword sentence and discourse generation. These embeddings, we argue, are either already there or naturally forgeable. On the other hand, we reply to a host of theory-internal issues concerning the abstract properties of our feedforward spreading (...) activation model, which functions without the usual cascading, feedback, and inhibitory connections. These issues also concern the concrete stratification in terms of lexical concepts, syntactic lemmas, and morphophonology. Our response stresses the parsimony of our modeling in the light of its substantial empirical coverage. We elaborate its usefulness for neuroimaging and aphasiology and suggest further cross-linguistic extensions of the model. (shrink)

In this study bibliometric mapping method was employed to visualise the current research trends and the impact of the two most influential models of working memory, namely: A. D. Baddeley and G. J. Hitch’s multicomponent working memory model and N. Cowan’s embedded-processes model of working memory. Using VOSviewer software two maps were generated based on the index-term words extracted from the research papers citing Baddeley and Cowan, respectively. The maps represent networks of co-occurrences of index terms and can be interpreted (...) as an indication of the main research fields related to the examined models of WM. The results of the analysis revealed that the spheres of influence of the two main conceptualisations of WM are rather different than similar. Although the first two clusters, i.e. “brain mapping” and “higher-level cognition and development” are present in both maps, their relative importance varies. The remaining clusters are unique to each map. Baddeley’s theory seems to have a greater influence on “neuropsychology”, while Cowan’s theory - on basic research on “biological systems”, including the nervous system in humans and animals. The second difference between these theories concerns their relations to functions and dysfunctions associated with particular sensory modalities: in Baddelay’s theory with the “auditory modality” cluster, and in Cowan’s - with the “visual modality” one. (shrink)

Computational models of infant word‐finding typically operate over transcriptions of infant‐directed speech corpora. It is now possible to test models of word segmentation on speech materials, rather than transcriptions of speech. We propose that such modeling efforts be conducted over the speech of the experimental stimuli used in studies measuring infants' capacity for learning from spoken sentences. Correspondence with infant outcomes in such experiments is an appropriate benchmark for models of infants. We demonstrate such an analysis by applying (...) the DP‐Parser model of Algayres and colleagues to auditory stimuli used in infant psycholinguistic experiments by Pelucchi and colleagues. The DP‐Parser model takes speech as input, and creates multiple overlapping embeddings from each utterance. Prospective words are identified as clusters of similar embedded segments. This allows segmentation of each utterance into possible words, using a dynamic programming method that maximizes the frequency of constituent segments. We show that DP‐Parse mimics American English learners' performance in extracting words from Italian sentences, favoring the segmentation of words with high syllabic transitional probability. This kind of computational analysis over actual stimuli from infant experiments may be helpful in tuning future models to match human performance. (shrink)

The main result is that for every recursively enumerable existential consistent theory Γ , there exists a finitely presented SQ-universal group H such that Γ is satisfied in every nontrivial quotient of H. Furthermore if Γ is satisfied in some group with a soluble word problem, then H can be taken with a soluble word problem. We characterize the finitely generated groups with soluble word problem as the finitely generated groups G for which there exists a finitely (...) presented group H all of the nontrivial quotients of which embed G. We prove also that for every countable group G, there exists a 2-finitely generated SQ-universal group H such that every nontrivial quotient of H embeds G. (shrink)

i) We show for each context-free language L that by considering each word of L as a structure in a natural way, one turns L into a finite union of classes which satisfy a finitary analog of the characteristic properties of complete universal first order classes of structures equipped with elementary embeddings. We show this to hold for a much larger class of languages which we call free local languages. ii) We define local languages, a class of languages between (...) free local and context-sensitive languages. Each local language L has a natural extension L ∞ to infinite words, and we prove a series of "pumping lemmas", analogs for each local language L of the "uvxyz theorem" of context free languages: they relate the existence of large words in L or L ∞ to the existence of infinite "progressions" of words included in L, and they imply the decidability of various questions about L or L ∞ . iii) We show that the pumping lemmas of ii) are independent from strong axioms, ranging from Peano arithmetic to ZF + Mahlo cardinals. We hope that these results are useful for a model-theoretic approach to the theory of formal languages. (shrink)

A Darwinian theory of the evolution of language must be incremental: to explain the transition from a hominin baseline with great ape grade communicative capacities to language-equipped hominins as a series of small steps. This paper takes up that project for the special case of words, giving an incremental model of the call to word transition. The model is embedded in a general conception of human social evolution with independent empirical support, but it also depends on some more specific (...) assumptions about language and about the earlier forms of hominin communication. Given these assumptions, the paper can be no more than a working draft of a more complete theory. (shrink)

This is a defence of the authenticity of Plato’s Epistula vii against the recent onslaught by Frede and Burnyeat (2015). It focusses on what Ep. vii has to say about writing and the embedded philosophical Digression and evaluates this in the context of other mainly late dialogues. In the Cratylus, Socrates ends with resignation regarding the potential of language study as a source of truth. This is also the case in Ep. vii, where the four means of knowledge (names, definitions, (...) images (diagrams) and knowledge/insight/true opinion) do not offer the essence of reality due to the weakness of language, in this case owing to a diagram with contrary properties. Consequently, name and definition are impermanent and conventional. Contra Burnyeat, definition is not impossible but useful in the acquisition of knowledge. Only after dialectical examination comes a flash of insight into reality. So, reality (truth) must, as in the Cratylus, be studied directly. Significantly for Plato (as opposed to Socrates), the epistemology is illustrated by a mathematical case. Hence it is relevant to look at mathematical procedure in middle and late dialogues. Moreover, the possible role of division is considered. Finally, the critique of writing justified by the Digression is placed in the context of the general aversion to the written word. The epistemology of the Digression is shown to be a fair and competent synopsis of the later Plato’s epistemology. Hence, there seems to be no reason thus far to doubt that Ep. vii is a genuine work of Plato. (shrink)

Marie was 15 when her abdominal pain began. After two years of negative work-ups, countless visits to gastroenterologists, and over 70 days of high school missed, she found herself readmitted to the hospital. “Refractory abdominal pain” was her ostensible diagnosis; “troubled teen” who was “going to be difficult” was embedded in the emergency department’s sign-out. When the medical team arrived to meet Marie, she was huddled in the corner of her hospital bed, silent and withdrawn. Her intern noted the numerous (...) scars cutting across her forearms as she felt for her radial pulses. Marie squirmed in pain when the intern palpated her abdomen.The ED’s work-up had been negative for abdominal pathology. When the... (shrink)

In his last writings, Wittgenstein repeatedly addresses the question of how our concepts relate to general facts of nature or human nature and how they are embedded in our lives. In doing so, he uses the term “pattern of life”, characterizing the complicated relationship between concepts and our lives and how our concepts “are connected with what interests us, with what matters to us” (LWPP II, 46). But who is this “us”, and whose interests manifest in the concepts we use (...) to designate patterns of life? What if certain concepts—or their absence—are exclusionary, discriminatory, or otherwise unjust to those who are not “us”? In this paper, I want to discuss Wittgenstein’s notion of “pattern of life” in its interweaving with facts, human life, and concepts, as well as its political implications. To this end, I will first outline the relationship between facts and concept formation as Wittgenstein drew it in his last writings. Based on this, I will argue that he uses the concept of pattern of life to capture the complicated relationship between concepts and human nature or “social facts”. Going beyond Wittgenstein and drawing on recent feminist epistemology, I will raise the question of the political implications of our patterns of life and concomitant social “conceptual injustices”. Finally, I will show how imagining facts otherwise and other conceptual worlds can help us to reveal the prejudices and injustices of our concepts and can lead to conceptual change and new patterns of life that may ultimately even change “things”, i.e., our thinking, judging and acting in the world. (shrink)

In everyday language, concepts appear alongside related concepts. Societal biases often emerge in these collocations; e.g., female names collocate with art- related concepts, and African American names collocate with negative concepts. It is unknown whether such collocations merely reflect societal biases or contribute to them. Concepts that are themselves neutral in valence but nevertheless collocate with valenced concepts provide a unique opportunity to address this question. For example, when asked, most people evaluate the concept “cause” as neutral, but “cause” is (...) frequently followed by negative concepts. We use such semantically prosodic concepts to test the influence of collocation on the emergence of implicit bias: do neutral concepts that frequently collocate with valenced concepts have corresponding implicit bias? In evaluative priming tasks, participants evaluated positive/negative nouns or pictures after seeing verb primes that were strongly valenced, neutral in valence but collocated with valenced concepts in corpora, or neutral in valence and not collocated with valenced concepts in corpora. Throughout, neutral primes with positive collocates facilitated the evaluation of positive targets much like strongly valenced primes, whereas neutral primes without valenced collocates did not. That neutral concepts with valenced collocates parallel the influence of valenced concepts suggests that their collocations in natural language may be sufficient for fostering implicit bias. Societal implications of the causal embedding hypothesis are discussed. (shrink)

Among theories of human language comprehension, cue‐based memory retrieval has proven to be a useful framework for understanding when and how processing difficulty arises in the resolution of long‐distance dependencies. Most previous work in this area has assumed that very general retrieval cues like [+subject] or [+singular] do the work of identifying (and sometimes misidentifying) a retrieval target in order to establish a dependency between words. However, recent work suggests that general, handpicked retrieval cues like these may not be enough (...) to explain illusions of plausibility (Cunnings & Sturt, 2018), which can arise in sentences like The letter next to the porcelain plate shattered. Capturing such retrieval interference effects requires lexically specific features and retrieval cues, but handpicking the features is hard to do in a principled way and greatly increases modeler degrees of freedom. To remedy this, we use well‐established word embedding methods for creating distributed lexical feature representations that encode information relevant for retrieval using distributed retrieval cue vectors. We show that the similarity between the feature and cue vectors (a measure of plausibility) predicts total reading times in Cunnings and Sturt’s eye‐tracking data. The features can easily be plugged into existing parsing models (including cue‐based retrieval and self‐organized parsing), putting very different models on more equal footing and facilitating future quantitative comparisons. (shrink)