Search

Topic analysis represents each document in a text corpus in a low-dimensional latent topic space. In some cases, the desired topic representation is subject to specific requirements or guidelines constituting side information. For instance, sustainability-aware investors might be interested in automatically assessing aspects of firm sustainability based on the textual content of its corporate reports, focusing on the established 17 UN sustainability goals. The main corpus consists of the corporate report texts, while the texts containing the definitions of the 17 UN sustainability goals represent the side information. Under the assumption that both text corpora share a common low-dimensional subspace, we propose representing them in such a space via directed topic extraction using matrix co-factorization. Both the main and the side text corpora are first represented as term–context matrices, which are then jointly decomposed into word–topic and topic–context matrices. The word–topic matrix is common to both text corpora, whereas the topic–context matrices contain specific representations in the shared topic space. A nuisance parameter, which allows us to shift the focus between the error minimization of individual factorization terms, controls the extent to which the side information is taken into account. With our approach, documents from the main and the side corpora can be related to each other in the resulting latent topic space. That is, the corporate reports are represented in the same latent topic space as the descriptions of the 17 UN sustainability goals, enabling a structured automatic sustainability assessment of the textual report's content. We provide an algorithm for such directed topic extraction and propose techniques for visualizing and interpreting the results. [ABSTRACT FROM AUTHOR]

Pyrus communis (P. communis) is the most cultivated and consumed species of pear within the European continent. This fruit has been a staple in Greece since ancient times, hence the name "Gift of the Gods". Given the extensive utilization of this fruit in the industrial sector and the focus on the exploitation of by-products to create new food and beverage products, the present research aimed to enhance the antioxidant activity of the P. communis peel through the implementation of a multifactor extraction system. Increased total polyphenols and ascorbic acid concentration, and enhanced antioxidant activity through radical scavenging and Fe3⁺ to Fe2⁺ reduction, all assist in boosting the health benefits of the extracts. The results indicated that the best conditions for compound yields were a 75% v/v hydroethanolic concentration, an extraction temperature of 80 °C, and 30 min of extraction time. Under the optimal conditions, the total polyphenol content was up to 4.98 mg of gallic acid equivalents (GAE)/g dried weight (dw). The radical scavenging activity by the 2,2-diphenyl-1-picrylhydrazyl (DPPH•) method was expressed as 18.36 μmol ascorbic acid equivalents (AAE)/g dw, while by the ferric-reducing antioxidant power (FRAP) method, it was 35.09 μmol AAE/g dw. Finally, the amount of ascorbic acid was measured at 20.16 mg/100 g dw. In this regard, this study has been conducted to assess and enhance the level of these bioactive compounds in the extract of the P. communis peel, leading to an extract with several applications in different food and beverage industries. [ABSTRACT FROM AUTHOR]

Ross (1967) observed that the coordinate structure constraint can be violated in certain semantically asymmetric structures. In this article we consider one of these structures, namely type A coordination, in detail (the terminology is from Lakoff 1986; an example is Here's the whisky I went to the store and bought). We present experimental evidence showing that the pattern of argument and adjunct extraction from type A coordinate structures matches the pattern of argument and adjunct extraction from structures containing rationale clauses in all crucial respects. This near-perfect parallel behavior suggests that, like rationale clauses, the second conjunct in a type A coordination is an adjunct (see also Brown 2017). We explore the consequences of this finding for both interpretive and syntactic analyses of asymmetric coordination. [ABSTRACT FROM AUTHOR]

Copyright of Linguistische Treffen in Wrocław is the property of Oficyna Wydawnicza ATUT and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Questions with a quantificational subject have readings that seemingly involve quantification into questions (called 'QiQ' for short). In particular, in single-wh questions with a universal quantifier, QiQ-readings call for pair-list answers, similar to pair-list readings of multiple-wh questions. This paper unifies the derivation of QiQ-readings and distinguishes QiQ-readings from pair-list readings of multiple-wh questions. I propose that pair-list multiple-wh questions and QiQ-questions both involve a wh-dependency, namely, that the wh-/quantificational subject stands in a functional dependency with the trace of the wh-object. In particular, in a pair-list multiple-wh question, the wh-subject binds into the trace of the wh-object across an identity operator; in a QiQ-question, the quantificational subject binds into the trace of the wh-object across a predication operator. These operations give rise to distinct definedness requirements, which vary with the quantificational force of the wh-/quantificational subject. The proposed analysis explains a contrast in domain exhaustivity between the pair-list readings of multiple-wh questions and questions with a universal quantifier, while also doing justice to the intuitive similarities between these two types of questions. I further propose that the observed QiQ-effect in a QiQ-question is derived by extracting one of the minimal proposition sets that satisfy the aforementioned quantificational predication condition. The values of these sets determine whether the QiQ-reading is available and whether a QiQ-question admits pair-list answers and/or has a choice flavor. [ABSTRACT FROM AUTHOR]

We propose APTHunter, a system for prompt detection of Advanced and Persistent Threats (APTs) in early stages. We provide an approach for representing the indicators of compromise that appear in the cyber threat intelligence reports and the relationships among them as provenance queries that capture the attacker's malicious behavior. We use the kernel audit log as a reliable source for system activities and develop an optimized whole system provenance graph that provides the causal relationships and information flows among system entities in a compact format. Then, we model the threat hunting as a behavior match problem by applying provenance queries to the optimized provenance graph to find any hits as indicators of an APT attack. We evaluate APTHunter on adversarial engagements from DARPA over different OS platforms, as well as real-world APT campaigns. Based on our experimental results, APTHunter promptly and reliably detects attack artifacts in early stages. [ABSTRACT FROM AUTHOR]

Enterprise Internet of Things (E-IoT) systems allow users to control audio, video, scheduled events, lightning fixtures, door access, and relays in complex smart installations. These systems are widely used in government or smart private offices, smart buildings/homes, conference rooms, schools, hotels, and similar professional settings. However, even with their widespread use, the security of many E-IoT systems and components has not been researched in the literature. To address this research gap, we focus on E-IoT communication buses, one of the core components used to connect E-IoT devices, and introduce LightningStrike attacks that demonstrate several weaknesses with E-IoT proprietary communication protocols used in E-IoT communication buses. Specifically, we show that popular E-IoT proprietary communication protocols are susceptible to Denial-of-Service (DoS), eavesdropping, impersonation, and replay attacks. As such threats cannot be mitigated through traditional defense mechanisms due to the limitations posed by E-IoT, we propose LGuard, a defense system to protect E-IoT systems against the attacks over communication buses. LGuard uses closed-circuit television footage and computer vision techniques to detect replay attacks. For impersonation and DoS attacks, LGuard utilizes traffic analysis. Finally, LGuard obfuscates the E-IoT traffic via inserting redundant traffic to the bus against eavesdropping attacks. We evaluated the performance of LGuard in a realistic E-IoT deployment, and our detailed evaluations show that LGuard achieves an overall accuracy and precision of 99% in detecting DoS, impersonation, and replay attacks while effectively increasing the difficulty of extracting valuable information for eavesdroppers. In addition, LGuard does not incur any operational overhead or modification to the existing E-IoT system. [ABSTRACT FROM AUTHOR]

The production capacity of compact reactors with micrometer or millimeter-scale channels or tubes is increased by numbering-up. In previous studies, a multi-channel plate type reactor and a multi-tube type reactor (MTR) were developed and applied to extraction and reaction operations. Fluid distribution has often been evaluated to design these reactors, but temperature control, which is critical to the reaction, has not been fully considered. It is important to solve this problem and establish a design method. In this study, computational fluid dynamics (CFD)-based design was performed so as to achieve the uniform flow and temperature distributions among the reaction tubes in the MTR, where an exothermic reaction proceeds in each tube with an immobilized catalyst and the reaction temperature is controlled by a coolant flowing outside the tubes. Effects of multi-tube arrangement of lattice, concentric circles and single circle, shell cross-sectional shape of circle, rectangle and ring, and reaction tubes with or without catalyst-free inert sections on the reactor performance were investigated by CFD. The usefulness of a two-step approach of designing the MTR after designing the double-tubular reactor was confirmed through a case study on parallel reactions. [ABSTRACT FROM AUTHOR]

Duplicate Bug Report Detection (DBRD) is one of the famous problems in software triage systems like Bugzilla. There are two main approaches to this problem, including information retrieval and machine learning. The second one is more effective for validation performance. Duplicate detection needs feature extraction, which is a time-consuming process. Both approaches suffer runtime issues, because they should check the new bug report to all bug reports in the repository, and it takes a long time for feature extraction and duplicate detection. This study proposes a new two-step classification approach which tries to reduce the search space of the bug repository search space in the first step and then check the duplicate detection using textual features. The Mozilla and Eclipse datasets are used for experimental evaluation. The results show that overall, 87.70% and 89.01% validation performance achieved averagely for accuracy and F1-measure, respectively. Moreover, 95.85% and 87.65% of bug reports can be classified in step one very fast for Eclipse and Mozilla datasets, respectively, and the other one needs textual feature extraction until it can be checked by the traditional DBRD approach. An average of 90% runtime improvement is achieved using the proposed method. [ABSTRACT FROM AUTHOR]

The extraction process of plant-based pharmaceutic raw material Sambucus Niger, is related to many factors such as the size of grind raw material, extraction temperature, number of spinning of the mixer, maceration time, concentration of the solvent, as well as the amount of the solvent in relation with the raw material. Each of these factors has a direct relation with the type of pharmaceutic plant which undergoes extraction. The content of the raw material which may be grounded flowers, stalk mixed with flowers, roots of raw material, or fruit of raw material indicated the method that should be used for extraction, such as extraction with various solvents-Maceration, extraction with CO2, and percolation extraction. In this case, when the raw material is a fruit of a pharmaceutics plant, the extraction is made by maceration, a method that also has its complexity in the technological process for such raw material. The grinding degree of raw material indicates optimization of the technological process, on the other side presents maintaining e high value of the extraction coefficient as well as reaching maximum values of basic substances that are present in the raw material. [ABSTRACT FROM AUTHOR]

This paper provides an overview of categorical and gradient effects in islands, with a focus on English, and argues that most islands are gradient. In some cases, the island is circumvented by the construction type in which the extraction takes place, and there is growing evidence that the critical factor is pragmatic in nature, contrary to classic and categorical accounts of island effects that are favored in generative circles to this day. In other cases, the island effect is malleable and can weaken with increased exposure to the extraction pattern, a phenomenon traditionally referred to as 'syntactic satiation'. However, it is not clear what satiation consists of. Some argue that it is nothing more than task adaptation (mere increased familiarity with the experimental paradigm, impacting difficult sentences more than easy ones), whereas others propose that it consists of a form of error-driven structure-dependent form of learning. The present paper discusses this controversy, and the broader adaptation debate, and argues that both task adaptation and grammatical adaptation are taking place during the processing of complex sentences, and that both frequency and attention are plausible factors to stimulate adaptation. [ABSTRACT FROM AUTHOR]

Adjunct clauses are traditionally assumed to be strong islands for extraction across languages. However, the universal island status of adjunct clauses has been challenged by studies showing that extraction is possible from finite adjunct clauses in the Mainland Scandinavian languages. The possibility of extraction in these languages appears to be affected by various factors, including the type of adjunct clause, the type of extraction dependency, and the presence of contextual facilitation. These findings call for a re-evaluation of the islandhood of adjunct clauses in English. We conducted an acceptability judgment study on relativization from three types of finite adjunct clauses in English (if-, when-, and because-clauses) in the presence of supporting context. We found that the three clause types showed rather non-uniform acceptability patterns: extraction from when- and because-clauses both yielded significantly lower ratings than extraction from if-clauses, which patterned with non-island that-clauses. Our results suggest that at least for relativization, if- and when-adjuncts are not invariably strong islands in English, and that extra-grammatical factors may be key in understanding island structures traditionally assumed to be purely syntactic in nature. [ABSTRACT FROM AUTHOR]

Recent research suggests that island effects may vary as a function of dependency type, potentially challenging accounts that treat island effects as reflecting uniform constraints on all filler-gap dependency formation. Some authors argue that cross-dependency variation is more readily accounted for by discourse-functional constraints that take into account the discourse status of both the filler and the constituent containing the gap. We ran a judgment study that tested the acceptability of wh-extraction and relativization from nominal subjects, embedded questions (EQs), conditional adjuncts, and existential relative clauses (RCs) in Norwegian. The study had two goals: (i) to systematically investigate cross-dependency variation from various constituent types and (ii) to evaluate the results against the predictions of the Focus Background Conflict constraint (FBCC). Overall we find some evidence for cross-dependency differences across extraction environments. Most notably wh-extraction from EQs and conditional adjuncts yields small but statistically significant island effects, but relativization does not. The differential island effects are potentially consistent with the predictions of the FBCC, but we discuss challenges the FBCC faces in explaining finer-grained judgment patterns. [ABSTRACT FROM AUTHOR]

In this article, I will argue that many of the theoretical approaches to extraction from participle adjunct islands suffer from the fact that the focus of investigation lies on perceived grammaticality differences in interrogative structures. Following approaches which make an explicit connection between extraction asymmetries and properties of the underlying proposition, I will argue that there is good evidence for the existence of similar differences in declarative adjunct constructions which can explain most of the grammaticality patterns observed for interrogatives. A crucial distinction to the majority of previous theories is the focus on acceptability rather than grammaticality, and the assumption that acceptability in declaratives is determined by a variety of semantic and syntactic complexity factors which do not influence how strongly extraction degrades the structure. This line of argumentation is more compatible with approaches to island phenomena that explain the low acceptability of some extractions by independent effects such as processing complexity and discourse function instead of syntactic principles blocking the extraction. I will also discuss a partially weighted, multifactorial model for the acceptability of declarative and interrogative participle adjunct constructions, which explains the judgment patterns in the literature without the need for additional, complex licensing conditions for extraction. [ABSTRACT FROM AUTHOR]

Extraction from relative clauses is generally taken to be unacceptable in Icelandic, unlike in the Mainland Scandinavian languages. Recent studies on Mainland Scandinavian show that the type of dependency as well as the embedding predicate matters for the acceptability of such extractions, and the study of spontaneously produced examples has improved our ability to create felicitous extraction contexts. The studies of Icelandic extraction predate these findings, and there is to date no study which systematically compares parallel sentences in Icelandic and Mainland Scandinavian. This article presents such a study, using two acceptability judgment experiments, one in Icelandic and one in Swedish, drawing on newly gained insights about fronting conditions in the two languages to create plausible contexts. The Icelandic participants rated extraction from relative clauses as unnatural, with a very large acceptability cost compared to in situ versions and good fillers. Extraction from að-clauses received mixed ratings, and local fronting was rated on a par with the in situ versions. In Swedish, extraction from relative clauses was rated as natural a majority of the time. There was no extraction cost in local fronting, extraction from att-clauses, or extraction from relative clauses in existential sentences, while extraction with other embedding predicates incurred some cost. No differences relating to the embedding predicate were seen in Icelandic. The study corroborates the view that extraction from relative clauses is unacceptable in Icelandic. [ABSTRACT FROM AUTHOR]

Few-shot relation extraction (RE) aims to identity and extract the relation between head and tail entities in a given context by utilizing a few annotated instances. Recent studies have shown that prompt-tuning models can improve the performance of few-shot learning by bridging the gap between pre-training and downstream tasks. The core idea of prompt-tuning is to leverage prompt templates to wrap the original input text into a cloze question and map the output words to corresponding labels via a language verbalizer for predictions. However, designing an appropriate prompt template and language verbalizer for RE task is cumbersome and time-consuming. Furthermore, the rich prior knowledge and semantic information contained in the relations are easily ignored, which can be used to construct prompts. To address these issues, we propose a novel Knowledge-enhanced Meta-Prompt (Know-MP) framework, which can improve meta-learning capabilities by introducing external knowledge to construct prompts. Specifically, we first inject the entity types of head and tail entities to construct prompt templates, thereby encoding the prior knowledge contained in the relations into prompt-tuning. Then, we expand rich label words for each relation type from their relation name to construct a knowledge-enhanced soft verbalizer. Finally, we adopt the meta-learning algorithm based on the attention mechanisms to mitigate the impact of noisy data on few-shot RE to accurately predict the relation of query instances and optimize the parameters of meta-learner. Experiments on FewRel 1.0 and FewRel 2.0, two benchmark datasets of few-shot RE, demonstrate the effectiveness of Know-MP. • Self-supervised PLMs for relation extraction rely on large-scale annotation data. • Prompt-tuning bridges the gap between pre-training and fine-tuning. • We propose a novel knowledge-enhanced meta-prompt method under few-shot setting. • The prior knowledge and semantic information in relations can be fully exploited. • Our method can reduce the impact of noisy data and improve the model's robustness. [ABSTRACT FROM AUTHOR]

Recently, table filling models have achieved promising performance in jointly extracting relation triplets from complex sentences, leveraging their inherent structural advantage of delineating entities and relations as table cells. Nonetheless, these models predominantly concentrate on the cells corresponding to entity pairs within the predicted tables, neglecting the interrelations among other token pairs. This oversight can potentially lead to the exclusion of essential token information. To address these challenges, we introduce the Token Relation-Inspired Network (TR-Net) , a novel framework for the joint extraction of entities and relations. It encompasses a token relation generator that adaptively constructs a token relation table, concentrating on the prominent token cells. Moreover, it also uses a structure-enhanced encoder that integrates the structural and sequential data of sentences via a highway gate mechanism. Our experimental analysis demonstrates that TR-Net delivers considerable enhancements and achieves state-of-the-art performance on four public datasets. • A token relation-inspired model is proposed for joint entity and relation extraction. • A token relation generator adaptively generates a token relation table to guide the table-filling process. • A structure-enhanced entity encoder integrates structural and sequence information. • Experimental results on four public datasets demonstrate the effectiveness of our model. [ABSTRACT FROM AUTHOR]

Document-level Relation Extraction (DocRE) aims to identify semantic relations among multiple entity pairs within a document. Most of the previous DocRE methods take the entire document as input. However, for human annotators, a small subset of sentences in the document, namely the evidence, is sufficient to infer the relation of an entity pair. Additionally, a document usually contains multiple entities, and these entities are scattered throughout various location of the document. Previous models use these entities independently, ignore the global interdependency among relation triples. To handle above issues, we propose a novel framework EAAGRE (Evidence and Axial Attention Guided Relation Extraction). Firstly, we use human-annotated evidence labels to supervise the attention module of DocRE system, making the model pay attention to the evidence sentences rather than others. Secondly, we construct an entity-level relation matrix and use axial attention to capture the global interactions among entity pairs. By doing so, we further extract the relations that require multiple entity pairs for prediction. We conduct various experiments on DocRED and have some improvement compared to baseline models, verifying the effectiveness of our model. • Our model jointly considers the evidence and global interdependency information of relation triples. • Our model directly supervise the top-k attention of tokens for evidence prediction. • Use axial attention to model the global interdependency between relation triples. • Experimental results illustrate the superiority of our EAAGRE model. [ABSTRACT FROM AUTHOR]

Due to the progress in deep learning technology, techniques that generate spoofed speech have significantly emerged. Such synthetic speech can be exploited for harmful purposes, like impersonation or disseminating false information. Researchers in the area investigate the useful features for spoof detection. This paper extensively investigates three problems in spoof detection in speech, namely, the imbalanced sample per class, which may negatively affect the performance of any detection models, the effect of the feature early and late fusion, and the analysis of unseen attacks on the model. Regarding the imbalanced issue, we have proposed two approaches (a Synthetic Minority Over Sampling Technique (SMOTE)-based and a Bootstrap-based model). We have used the OpenSMILE toolkit, to extract different feature sets, their results and early and late fusion of them have been investigated. The experiments are evaluated using the ASVspoof 2019 datasets which encompass synthetic, voice-conversion, and replayed speech samples. Additionally, Support Vector Machine (SVM) and Deep Neural Network (DNN) have been adopted in the classification. The outcomes from various test scenarios indicated that neither the imbalanced nature of the dataset nor a specific feature or their fusions outperformed the brute force version of the model as the best Equal Error Rate (EER) achieved by the Imbalance model is 6.67 % and 1.80 % for both Logical Access (LA) and Physical Access (PA) respectively. [ABSTRACT FROM AUTHOR]

Relational triple extraction refers to extracting entities and relations from natural texts, which is a crucial task in the construction of knowledge graph. Recently, tagging based methods have received increasing attention because of their simple and effective structural form. Among them, the two-step extraction method is easy to cause the problem of category imbalance. To address this issue, we propose a novel two-step extraction method, which first extracts subjects, generates a fixed-size embedding for each relation, and then regards these relations as known conditions to extract the objects directly with the identified subjects. In order to eliminate the influence of irrelevant relations when predicting objects, we use a relation-special attention mechanism and a gate unit to select appropriate relations. In addition, most current models do not account for two-way interaction between tasks, so we design a feature interactive network to achieve bidirectional interaction between subject and object extraction tasks and enhance their connection. Experimental results on NYT, WebNLG, NYT ⋆ and WebNLG ⋆ datasets show that our model is competitive among joint extraction models. • Propose a novel two-step extraction method to solve the category imbalance. • Design a Feature Interactive Network to achieve two-way interaction between tasks. • Experimental results show that our model outperforms state-of-the-art baselines. • A large of ablation experiments demonstrate each component's rationality. [Display omitted] [ABSTRACT FROM AUTHOR]

Despite recent advancements, existing story generation systems continue to encounter difficulties in effectively incorporating contextual and event features, which greatly influence the quality of generated narratives. To tackle these challenges, we introduce a novel neural generation model, EtriCA, that enhances the relevance and coherence of generated stories by employing a cross-attention mechanism to map context features onto event sequences through residual mapping. This feature capturing mechanism enables our model to exploit logical relationships between events more effectively during the story generation process. To further enhance our proposed model, we employ a post-training framework for knowledge enhancement (KeEtriCA) on a large-scale book corpus. This allows EtriCA to adapt to a wider range of data samples. This results in approximately 5% improvement in automatic metrics and over 10% improvement in human evaluation. We conduct extensive experiments, including comparisons with state-of-the-art (SOTA) baseline models, to evaluate the performance of our framework on story generation. The experimental results, encompassing both automated metrics and human assessments, demonstrate the superiority of our model over existing state-of-the-art baselines. These results underscore the effectiveness of our model in leveraging context and event features to improve the quality of generated narratives. • Novel task composing narratives from an initial context and event sequences. • Two annotated datasets and a method to enhance existing event extraction frameworks. • KeEtriCa leverages context and event sequences with a cross-attention module. • Investigations into how to write a more fluent, relevant, and coherent story. [ABSTRACT FROM AUTHOR]

Goal-oriented dialogue systems are becoming pervasive in human lives. To facilitate task completion and human participation in a practical setting, such systems must have extensive technical knowledge and social understanding. Politeness is a socially desirable trait that plays a crucial role in task-oriented conversations for ensuring better user engagement and satisfaction. To this end, we propose a novel task of politeness analysis in goal-oriented dialogues. Politeness analysis consists of two sub-tasks: politeness turn identification and phrase extraction. Politeness turn identification is dependent on textual triggers denoting politeness or impoliteness. In this regard, we propose a Bidirectional Encoder Representations from Transformers-Directional Graph Convolutional Network (BERT-DGCN) based multi-task learning approach that performs turn identification and phrase extraction tasks in a unified framework. Our proposed approach employs BERT for encoding input turns and DGCN for encoding syntactic information, in which dependency among words is incorporated into DGCN to improve its capability to represent input utterances and benefit politeness analysis task accordingly. Our proposed model classifies each turn of a conversation into one of the three pre-defined classes, viz. polite, impolite and neutral, and extracts phrases denoting politeness or impoliteness in that turn simultaneously. As there is no such readily available data, we prepare a conversational dataset, PoDial for mental health counseling and legal aid for crime victims in English for our experiment. Experimental results demonstrate that our proposed approach is effective and achieves 2.04 points improvement on turn identification accuracy and 2.40 points on phrase extraction F1- score on our dataset over baselines. • Aim to incorporate social intelligence in goal-oriented dialogue systems. • Politeness facilitates the development of socially-affective conversational agents. • Propose a multi-task learning framework for politeness analysis task in conversations. • It helps comprehend speakers' social conduct and detect phrases conveying politeness. • Sharing of information between tasks improves the performance of our multi-task model. [ABSTRACT FROM AUTHOR]

In the online processing of long-distance wh-dependencies, native speakers have been found to make use of intermediate syntactic gaps, which has the effect of facilitating dependency resolution. This strategy has also been observed in second language (L2) speakers living in an L2 immersion context, but not in classroom L2 learners. This research note investigates whether there is evidence of use of the intermediate gap among L2 speakers that have received considerable naturalistic exposure to the L2 from a young age, but do not live in a standard immersion context. Two groups of participants, one L1 English and one L1 Afrikaans–L2 English, completed a self-paced reading task involving English sentences containing long-distance wh-dependencies. The data were analysed using Bayesian regression. The results indicate that the intermediate gap facilitated dependency resolution in the L1 English group but not in the L2 group. Increased L2 exposure seemed to increase sensitivity to the intermediate gap among the L2 speakers, but was not associated with faster dependency resolution. The findings suggest that although non-immersive L2 experience affects L2 processing of abstract grammatical cues, it may be less effective than fully immersive experience in engendering nativelike processing of long-distance wh-dependencies. [ABSTRACT FROM AUTHOR]

Pure cotton low-twist single yarn has good softness, bulkiness, and wearing comfort, but its lower strength makes the yarn break easily during the spinning process, which seriously affects the spinning effect and progress. The addition of long-staple cotton helps to increase the average length and uniformity of the raw fiber, thereby improving the breaking load and spinnability of pure cotton low-twist single yarn. In this study, the addition of long-staple cotton, the twist factor, and the twist direction were used as variables to spin 22 kinds of combed 14. 6 tex pure cotton low-twist single yarn with spinnability, and the breaking load, the hairiness, and the unevenness of the yarn were tested. The result shows that the spinnability of pure cotton low-twist single yarn is mainly related to the breaking load, and with the increase of the twist factor, the breaking load of low-twist single yarn shows an obvious upward trend. When the proportion of long-staple cotton is 70%, the breaking load of low-twist single yarn is the largest, and the harmful hairiness index and unevenness of the yarn are significantly improved. [ABSTRACT FROM AUTHOR]

An eco-friendly and visual pH indicator was developed based on cotton fabrics dyed with natural Lycium ruthenicum extract (LRE), whose color changed from red to purple, blue and green at pH of 2-10. Three methods for dyeing cotton fabrics with LRE were attempted, namely pre-mordanting, simultaneous mordanting, and post- mordanting methods with different dyeing temperatures, dyeing times and dyebath pH values. The cotton fabrics exhibited the highest K/ S value when they were dyed under a dyebath of pH 6 at 20 ° for 90 min in the case of the simultaneous mordanting method. Meanwhile, the dyed cotton fabrics also showed reversible pH-dependent color changes. The developed flexible pH indicator based on renewable natural materials is suitable for multiple applications in environmental monitoring and smart textiles. [ABSTRACT FROM AUTHOR]

Aravind Joshi famously hypothesized that natural language syntax was characterized (in part) by mildly context-sensitive generative power. Subsequent work in mathematical linguistics over the past three decades has revealed surprising convergences among a wide variety of grammatical formalisms, all of which can be said to be mildly context-sensitive. But this convergence is not absolute. Not all mildly context-sensitive formalisms can generate exactly the same stringsets (i.e. they are not all weakly equivalent), and even when two formalisms can both generate a certain stringset, there might be differences in the structural descriptions they use to do so. It has generally been difficult to find cases where such differences in structural descriptions can be pinpointed in a way that allows linguistic considerations to be brought to bear on choices between formalisms, but in this paper we present one such case. The empirical pattern of interest involves wh-movement dependencies in languages that do not enforce the wh-island constraint. This pattern draws attention to two related dimensions of variation among formalisms: whether structures grow monotonically from one end to another, and whether structure-building operations are conditioned by only a finite amount of derivational state. From this perspective, we show that one class of formalisms generates the crucial empirical pattern using structures that align with mainstream syntactic analysis, and another class can only generate that same string pattern in a linguistically unnatural way. This is particularly interesting given that (i) the structurally-inadequate formalisms are strictly more powerful than the structurally-adequate ones from the perspective of weak generative capacity, and (ii) the formalism based on derivational operations that appear on the surface to align most closely with the mechanisms adopted in contemporary work in syntactic theory (merge and move) are the formalisms that fail to align with the analyses proposed in that work when the phenomenon is considered in full generality. [ABSTRACT FROM AUTHOR]

Relational triple extraction is a critical step in knowledge graph construction. Compared to pipeline-based extraction, joint extraction is gaining more attention because it can better utilize entity and relation information without causing error propagation issues. Yet, the challenge with joint extraction lies in handling overlapping triples. Existing approaches adopt sequential steps or multiple modules, which often accumulate errors and interfere with redundant data. In this study, we propose an innovative joint extraction model with cross-attention mechanism and global pointers with context shield window. Specifically, our methodology begins by inputting text data into a pre-trained RoBERTa model to generate word vector representations. Subsequently, these embeddings are passed through a modified cross-attention layer along with entity type embeddings to address missing entity type information. Next, we employ the global pointer to transform the extraction problem into a quintuple extraction problem, which skillfully solves the issue of overlapping triples. It is worth mentioning that we design a context shield window on the global pointer, which facilitates the identification of correct entities within a limited range during the entity extraction process. Finally, the capability of our model against malicious samples is improved by adding adversarial training during the training process. Demonstrating superiority over mainstream models, our approach achieves impressive results on three publicly available datasets. • Solving overlapping triples problems by employing global pointer. • Context shield window is designed to eliminate redundant information. • Merging entity type information with designed entity-type cross attention module. • Adversarial training is added to improve the generalization ability. [ABSTRACT FROM AUTHOR]

Many languages of the Mayan family restrict the extraction of transitive (ergative) subjects for focus, wh-questions, and relativization (Ā-extraction). We follow Aissen (2017b) in labeling this restriction the ergative extraction constraint (EEC). In this article, we offer a unified account of the EEC within Mayan languages, as well as an analysis of the special construction known as agent focus (AF) used to circumvent it. Specifically, we propose that the EEC has a similar source across the subset of Mayan languages that exhibit it: intervention. The intervention problem is created when an object DP structurally intervenes between the Ā-probe on C0 and the ergative subject. Evidence that intervention by the object is the source of the problem comes from a handful of exceptional contexts that permit transitive subjects to extract in languages which normally ban this extraction, and conversely, a context that exceptionally bans ergative extraction in a language which otherwise allows it. We argue that the problem with A¯ -extracting the ergative subject across the intervening object connects to the requirements of the Ā-probe on C0: the probe on C0 is bundled to search simultaneously for [Ā] and [D] features. This relates the Mayan patterns to recent proposals for extraction patterns in Austronesian languages (e.g. Legate 2014, Aldridge 2017b) and elsewhere (van Urk 2015). Specifically, adapting the proposal of Coon & Keine 2020, we argue that in configurations in which a DP object intervenes between the probe on C0 and an Ā-subject, conflicting requirements on movement lead to a derivational crash. While we propose that the EEC has a uniform source across the family, we argue that AF constructions vary Mayaninternally in how they circumvent the EEC, accounting for the variation in behavior of AF across the family. This article contributes both to our understanding of parametric variation internal to the Mayan family and to the discussion of variation in v-extraction asymmetries and syntactic ergativity crosslinguistically. [ABSTRACT FROM AUTHOR]

The lack of terminology and language resources for the under-resourced South African languages poses a serious problem for effective communication in specialised fields such as law, education, health, agriculture, science, and technology. The ability to use all South African languages in all contemporary fields requires the existence of relevant terminology and resources. The article aims to semi-automatically identify and extract terminology for creating an English, Xitsonga, Siswati, and isiNdebele quadrilingual dictionary. Given parallel texts in the four different languages, we use ParaConc to identify and extract terminology in one language and the corresponding translations in the other languages. In this study, English is used as the source language, while Xitsonga, Siswati, and isiNdebele are the target languages. This process allowed us to identify specific lexical items in the source language (manually) and their translation equivalents in the target languages (automatically). The result was a collection of extracted terminology lists that can be used to compile a specialised quadrilingual dictionary for English and three of the under-resourced languages. We show the usefulness of ParaConc to semi-automatically extract quadrilingual terminology lists, which, by creating quadrilingual dictionaries, will contribute to the development and promotion of multilingualism in South Africa. [ABSTRACT FROM AUTHOR]

Scientific documentation of traditional knowledge on ethnobotanical plants used by the ethnic people of Alipurduar district of West Bengal, India has been done by conducting field surveys during July 2018 to January 2020. With the help of pretested semi-structured questionnaires, knowledgeable ethnic people of the studied areas were interviewed to record their traditional knowledge on plants and their utilization for daily livelihood. The questionnaires covers various aspects (vernacular name(s), parts use, ethno medicinal uses, disease treated, mode of preparation, other uses, among others) of each and every plant species having ethnobotanical values. Standard methods for plant collection and herbarium techniques were followed and identification was done with the help of relevant sources. In the present survey, 84 plant species of 44 families have been documented. Asteraceae dominates (8 species) among the plant families. Based on growth habit, herb (35.71%) dominates among plant types and were mostly collected from their natural habitat (58.33%). A total of 143 ethnobotanical uses were recorded under 5 numbers of used categories, viz. ethnomedicine (EM), edible plant (EP), magico-religious (MR), ethno-veterinary (EV) and narcotic (N), of which all the plants (100.0%) were used under ethno medicinal category. Altogether 77 types of common physical problems, grouped into 22 major problems were found to be treated by the herbal medicine. Leaves were found to be the dominant (40.91%) plant part among others used for the preparation of medicine. In most of the cases medicines were prepared following extraction methodology (41.11%) and were taken orally (62.64%) or topically (36.36%). Collected information clearly suggests that the ethnic people are largely dependent on plant species for their daily needs, especially for primary healthcare. Future research on these medicinal plants can lead to the discovery of unique phytochemical and newer drugs for the treatment of diverse diseases. [ABSTRACT FROM AUTHOR]

In this paper, we present an advanced domain-specific multi-word terminology extraction method. Our hybrid approach for automatic term identification benefits from both statistical and linguistic approaches. Our main goal is to reduce as much as possible the human effort in term selection tasks as well as to provide a wide-range and representative terminology of a domain. We emphasize in identification of verb or noun phrases multi-word terms, in neologisms and technical jargons. Our architecture applies the term frequency-inverse document frequency (TF-IDF) algorithm to a domain-specific textual corpus in order to measure a unit's importance in it. We also use techniques to filter out nested terms of a candidate term taking into consideration its frequency by itself in the corpus. In addition, the exported terms are filtered out based on a stop-word list and linguistic criteria. To further reduce the size of the candidate terms and achieve accurate and precise terminologies, our method automatically validates them against a general-purpose corpus. Our study based on a small corpus of vibration-based condition monitoring domain shows that most extracted terms have nice correspondence to the domain of condition monitoring concepts and notions. [ABSTRACT FROM AUTHOR]

Copyright of Multi-Knowledge Electronic Comprehensive Journal For Education & Science Publications (MECSJ) is the property of Multi-Knowledge Electronic Comprehensive Journal For Education & Science Publications (MECSJ) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Ambridge argues persuasively for the importance in language learning of a rich database of input exemplars. However, a fuller account must also consider the importance of on-line and developmental competition between rote exemplar-based storage and emergent patterns that can optimize retrieval. [ABSTRACT FROM AUTHOR]

Even though image retrieval is considered as one of the most important research areas in the last two decades, there is still room for improvement since it is still not satisfying for many users. Two of the major problems which need to be improved are the accuracy and the speed of the image retrieval system, in order to achieve user satisfaction and also to make the image retrieval system suitable for all platforms. In this work, the proposed retrieval system uses features with spatial information to analyze the visual content of the image. Then, the feature extraction process is followed by applying the fuzzy c-means (FCM) clustering algorithm to reduce the search space and speed up the retrieval process. The experimental results show that using the spatial features increases the system accuracy and that the clustering algorithm speeds up the image retrieval process. This shows that the proposed system works with texture and non-texture images. [ABSTRACT FROM AUTHOR]

Keyphrase extraction is an important part of natural language processing (NLP) research, although little research is done in the domain of web pages. TheWorldWideWeb contains billions of pages that are potentially interesting for various NLP tasks, yet it remains largely untouched in scientific research. Current research is often only applied to clean corpora such as abstracts and articles from academic journals or sets of scraped texts from a single domain. However, textual data from web pages differ from normal text documents, as it is structured using HTML elements and often consists of many small fragments. These elements are furthermore used in a highly inconsistent manner and are likely to contain noise. We evaluated the keyphrases extracted by several state-of-the-art extraction methods and found that they did not transfer well to web pages. We therefore propose WebEmbedRank, an adaptation of a recently proposed extraction method that can make use of structural information in web pages in a robust manner. We compared this novel method to other baselines and state-of-the-art methods using a manually annotated dataset and found that WebEmbedRank achieved significant improvements over existing extraction methods on web pages. [ABSTRACT FROM AUTHOR]

Charge prediction which aims to determine appropriate charges for criminal cases based on textual fact descriptions, is an important technology in the field of AI&Law. Previous works focus on improving prediction accuracy, ignoring the interpretability, which limits the methods' applicability. In this work, we propose a deep neural framework to extract short but charge-decisive text snippets - rationales - from input fact description, as the interpretation of charge prediction. To solve the scarcity problem ofrationale annotatedcorpus, rationalesare extractedinareinforcement stylewiththe only supervision in the form of charge labels. We further propose a dynamic rationale attention mechanism to better utilize the information in extracted rationales and predict the charges. Experimental results show that besides providing charge prediction interpretation, our approach can also capture subtle details to help charge prediction. [ABSTRACT FROM AUTHOR]

Time is deeply woven into how people perceive, and communicate about the world. Almost unconsciously, we provide our language utterances with temporal cues, like verb tenses, and we can hardly produce sentences without such cues. Extracting temporal cues from text, and constructing a global temporal view about the order of described events is a major challenge of automatic natural language understanding. Temporal reasoning, the process of combining different temporal cues into a coherent temporal view, plays a central role in temporal information extraction. This article presents a comprehensive survey of the research from the past decades on temporal reasoning for automatic temporal information extraction from text, providing a case study on how combining symbolic reasoning with machine learning-based information extraction systems can improve performance. It gives a clear overview of the used methodologies for temporal reasoning, and explains how temporal reasoning can be, and has been successfully integrated into temporal information extraction systems. Based on the distillation of existing work, this survey also suggests currently unexplored research areas. We argue that the level of temporal reasoning that current systems use is still incomplete for the full task of temporal information extraction, and that a deeper understanding of how the various types of temporal information can be integrated into temporal reasoning is required to drive future research in this area. [ABSTRACT FROM AUTHOR]

We introduce "extreme summarization," a new single-document summarization task which aims at creating a short, one-sentence news summary answering the question "What is the article about?". We argue that extreme summarization, by nature, is not amenable to extractive strategies and requires an abstractive modeling approach. In the hope of driving research on this task further: (a) we collect a real-world, large scale dataset by harvesting online articles from the British Broadcasting Corporation (BBC); and (b) propose a novel abstractive model which is conditioned on the article's topics and based entirely on convolutional neural networks. We demonstrate experimentally that this architecture captures long-range dependencies in a document and recognizes pertinent content, outperforming an oracle extractive system and state-of-the-art abstractive approaches when evaluated automatically and by humans on the extreme summarization dataset. [ABSTRACT FROM AUTHOR]

The expedition's complicity in the imperial project of conquest, extraction and settlement has placed it as an object of critique, but largely discredited its significance as a valid research method in the critical social sciences. Yet dismissing the expedition merely as an imperial remnant risks ignoring more nuanced histories that bear no resemblance to myths of conquest and masculine heroics. Instead, this paper considers the expedition as a malleable practice that can be critically appropriated and manipulated in ways that retain and further the critique of violence and knowledge production, while also experimenting with creative alternatives to some of its conventions. [ABSTRACT FROM AUTHOR]

We present algorithms for extracting Hyperedge Replacement Grammar (HRG) rules from a graph along with a vertex order. Our algorithms are based on finding a tree decomposition of smallest width, relative to the vertex order, and then extracting one rule for each node in this structure. The assumption of a fixed order for the vertices of the input graph makes it possible to solve the problem in polynomial time, in contrast to the fact that the problem of finding optimal tree decompositions for a graph is NP-hard. We also present polynomial-time algorithms for parsing based on our HRGs, where the input is a vertex sequence and the output is a graph structure. The intended application of our algorithms is grammar extraction and parsing for semantic representation of natural language. We apply our algorithms to data annotated with Abstract Meaning Representations and report on the characteristics of the resulting grammars. [ABSTRACT FROM AUTHOR]

Linguistic theory suggests non-canonical sentences subvert the dominant agent-verb-theme order in English via displacement of sentence constituents to argument (NP-movement) or non-argument positions (wh-movement). Both processes have been associated with the left inferior frontal gyrus and posterior superior temporal gyrus, but differences in neural activity and connectivity between movement types have not been investigated. In the current study, functional magnetic resonance imaging data were acquired from 21 adult participants during an auditory sentence-picture verification task using passive and active sentences contrasted to isolate NP-movement, and object- and subject-cleft sentences contrasted to isolate wh-movement. Then, functional magnetic resonance imaging data from regions common to both movement types were entered into a dynamic causal modeling analysis to examine effective connectivity for wh-movement and NP-movement. Results showed greater left inferior frontal gyrus activation for Wh > NP-movement , but no activation for NP > Wh-movement. Both types of movement elicited activity in the opercular part of the left inferior frontal gyrus, left posterior superior temporal gyrus, and left medial superior frontal gyrus. The dynamic causal modeling analyses indicated that neither movement type significantly modulated the connection from the left inferior frontal gyrus to the left posterior superior temporal gyrus, nor vice-versa, suggesting no connectivity differences between wh- and NP-movement. These findings support the idea that increased complexity of wh-structures, compared to sentences with NP-movement, requires greater engagement of cognitive resources via increased neural activity in the left inferior frontal gyrus, but both movement types engage similar neural networks. [ABSTRACT FROM AUTHOR]

Abstract This paper investigates subject extraction from embedded clauses in Jordanian Arabic (JA). Firstly, it shows that JA maintains the division between factive vs. non-factive verbs with respect to (subject) extraction. We argue that non-factive clausal complements are not islands for extraction, and, hence, their subject can undergo movement to the left periphery of the main clause. We use Starke's (2001) proposal of Relativized Minimality to account for the ensuing restriction that bans the object and/or locative adjuncts of the factive clausal complement to appear pre-verbally (while the subject is extracted). We propose that this restriction is due to the richly-featured content of the subject wh-word that acts as a barrier against the movement of elements that fall under its c-command domain. As for factive clausal complements (i.e., embedded under factive verbs), we prove that such complements are strong islands for extraction. We follow Kiparsky and Kiparsky (1970) and related works, that factive complements are, with updated terminology, DPs which are absolute islands in Arabic. With this being the case, this paper challenges several recent approaches to factive complements that have argued either for a reduced left periphery for factive complements (e.g., Haegeman, 2006; de Cuba, 2007) or for the presence of an operator that has the effect to block movement out of these clauses (e.g., Zubizaretta, 2001; Haegeman, 2012). [ABSTRACT FROM AUTHOR]

Background: Extracting relations between important clinical entities is critical but very challenging for natural language processing (NLP) in the medical domain. Researchers have applied deep learning-based approaches to clinical relation extraction; but most of them consider sentence sequence only, without modeling syntactic structures. The aim of this study was to utilize a deep neural network to capture the syntactic features and further improve the performances of relation extraction in clinical notes.Methods: We propose a novel neural approach to model shortest dependency path (SDP) between target entities together with the sentence sequence for clinical relation extraction. Our neural network architecture consists of three modules: (1) sentence sequence representation module using bidirectional long short-term memory network (Bi-LSTM) to capture the features in the sentence sequence; (2) SDP representation module implementing the convolutional neural network (CNN) and Bi-LSTM network to capture the syntactic context for target entities using SDP information; and (3) classification module utilizing a fully-connected layer with Softmax function to classify the relation type between target entities.Results: Using the 2010 i2b2/VA relation extraction dataset, we compared our approach with other baseline methods. Our experimental results show that the proposed approach achieved significant improvements over comparable existing methods, demonstrating the effectiveness of utilizing syntactic structures in deep learning-based relation extraction. The F-measure of our method reaches 74.34% which is 2.5% higher than the method without using syntactic features.Conclusions: We propose a new neural network architecture by modeling SDP along with sentence sequence to extract multi-relations from clinical text. Our experimental results show that the proposed approach significantly improve the performances on clinical notes, demonstrating the effectiveness of syntactic structures in deep learning-based relation extraction. [ABSTRACT FROM AUTHOR]