Baseball fans don’t like lumpy batters: Influence of domain knowledge on the access of subordinate meanings

2018 ◽  
Vol 71 (1) ◽  
pp. 93-102 ◽  
Author(s):  
Jennifer Wiley ◽  
Tim George ◽  
Keith Rayner
Keyword(s):  
Prior Knowledge ◽  
Lexical Access ◽  
Domain Knowledge ◽  
Negative Evidence ◽  
Word Meanings ◽  
Positive Evidence ◽  
Ambiguous Words ◽  
Baseball Fans ◽  
Evidence Condition ◽  
Access Model

Two experiments investigated the effects of domain knowledge on the resolution of ambiguous words with dominant meanings related to baseball. When placed in a sentence context that strongly biased toward the non-baseball meaning (positive evidence), or excluded the baseball meaning (negative evidence), baseball experts had more difficulty than non-experts resolving the ambiguity. Sentence contexts containing positive evidence supported earlier resolution than did the negative evidence condition for both experts and non-experts. These experiments extend prior findings, and can be seen as support for the reordered access model of lexical access, where both prior knowledge and discourse context influence the availability of word meanings.

scholarly journals Defining the Role of Prior Knowledge and Vocabulary in Reading Comprehension: The Retiring of Number 41

1991 ◽  
Vol 23 (4) ◽  
pp. 487-508 ◽  
Author(s):  
Steven A. Stahl ◽  
Victoria Chou Hare ◽  
Richard Sinatra ◽  
James F. Gregory
Keyword(s):  
Prior Knowledge ◽  
Domain Knowledge ◽  
Vocabulary Knowledge ◽  
Word Meanings ◽  
General Domain ◽  
Tenth Graders ◽  
Topic Knowledge ◽  
High Knowledge ◽  
Independent Effects

Although both prior topic knowledge and vocabulary knowledge have been known to affect comprehension in general, less is known about the specifics of the interactions between these factors. Using a magazine article about a ceremony marking the retirement of a baseball player's jersey number, this study examines the effects of knowledge of baseball in general and of the career of Tom Seaver in specific and of knowledge of word meanings in general and of words used in the passage specifically on tenth graders' recall of different aspects of passage content. Vocabulary knowledge tended to affect the number of units recalled overall; prior knowledge influenced which units were recalled. Prior topic knowledge influenced whether subjects produced a gist statement in their recall and how well they recalled numbers relevant to Seaver's career. High knowledge subjects also tended to focus more on information given about his career than low knowledge subjects. Specific and general domain knowledge were so closely related that their effects could not be disentangled. A qualitative analysis of the protocols confirmed the general trends in the quantitative analysis. Results suggest both that domain knowledge and vocabulary have independent effects on comprehension and that these effects are on what is comprehended as well as how much is comprehended.

On-device Prior Knowledge Incorporated Learning for Personalized Atrial Fibrillation Detection

2021 ◽  
Vol 20 (5s) ◽  
pp. 1-25
Author(s):  
Zhenge Jia ◽  
Yiyu Shi ◽  
Samir Saba ◽  
Jingtong Hu
Keyword(s):  
Atrial Fibrillation ◽  
Prior Knowledge ◽  
Domain Knowledge ◽  
Fine Tuning ◽  
Cardiac Monitoring ◽  
Patient Specific ◽  
Detection Accuracy ◽  
Specific Patient ◽  
Deep Model ◽  
Tuning Strategy

Atrial Fibrillation (AF), one of the most prevalent arrhythmias, is an irregular heart-rate rhythm causing serious health problems such as stroke and heart failure. Deep learning based methods have been exploited to provide an end-to-end AF detection by automatically extracting features from Electrocardiogram (ECG) signal and achieve state-of-the-art results. However, the pre-trained models cannot adapt to each patient’s rhythm due to the high variability of rhythm characteristics among different patients. Furthermore, the deep models are prone to overfitting when fine-tuned on the limited ECG of the specific patient for personalization. In this work, we propose a prior knowledge incorporated learning method to effectively personalize the model for patient-specific AF detection and alleviate the overfitting problems. To be more specific, a prior-incorporated portion importance mechanism is proposed to enforce the network to learn to focus on the targeted portion of the ECG, following the cardiologists’ domain knowledge in recognizing AF. A prior-incorporated regularization mechanism is further devised to alleviate model overfitting during personalization by regularizing the fine-tuning process with feature priors on typical AF rhythms of the general population. The proposed personalization method embeds the well-defined prior knowledge in diagnosing AF rhythm into the personalization procedure, which improves the personalized deep model and eliminates the workload of manually adjusting parameters in conventional AF detection method. The prior knowledge incorporated personalization is feasibly and semi-automatically conducted on the edge, device of the cardiac monitoring system. We report an average AF detection accuracy of 95.3% of three deep models over patients, surpassing the pre-trained model by a large margin of 11.5% and the fine-tuning strategy by 8.6%.

Semantic ambiguity norms for 530 Spanish words

2016 ◽  
Vol 38 (2) ◽  
pp. 457-475 ◽  
Author(s):  
JUAN HARO ◽  
PILAR FERRÉ ◽  
ROGER BOADA ◽  
JOSEP DEMESTRE
Keyword(s):  
Language Processing ◽  
Ambiguous Word ◽  
Age Of Acquisition ◽  
Emotional Valence ◽  
Objective Measures ◽  
Semantic Ambiguity ◽  
Subjective Ratings ◽  
Subjective Measures ◽  
Word Meanings ◽  
Ambiguous Words

ABSTRACTThis study presents semantic ambiguity norms for 530 Spanish words. Two subjective measures of semantic ambiguity and two subjective measures of relatedness of ambiguous word meanings were collected. In addition, two objective measures of semantic ambiguity were included. Furthermore, subjective ratings were obtained for some relevant lexicosemantic variables, such as concreteness, familiarity, emotional valence, arousal, and age of acquisition. In sum, the database overcomes some of the limitations of the published databases of Spanish ambiguous words; in particular, the scarcity of measures of ambiguity, the lack of relatedness of ambiguous word meanings measures, and the absence of a set of unambiguous words. Thus, it will be very helpful for researchers interested in exploring semantic ambiguity as well as for those using semantic ambiguous words to study language processing in clinical populations.

Chapter 8. A Constraint-Based Approach to Learning and Reasoning

2021 ◽  
Author(s):  
Michelangelo Diligenti ◽  
Francesco Giannini ◽  
Marco Gori ◽  
Marco Maggini ◽  
Giuseppe Marra
Keyword(s):  
Neural Networks ◽  
Prior Knowledge ◽  
Domain Knowledge ◽  
Graphical Model ◽  
Expressive Power ◽  
Training Data ◽  
Markov Logic ◽  
Exact Inference ◽  
Logic Models ◽  
Effectiveness Of Training

Neural-symbolic models bridge the gap between sub-symbolic and symbolic approaches, both of which have significant limitations. Sub-symbolic approaches, like neural networks, require a large amount of labeled data to be successful, whereas symbolic approaches, like logic reasoners, require a small amount of prior domain knowledge but do not easily scale to large collections of data. This chapter presents a general approach to integrate learning and reasoning that is based on the translation of the available prior knowledge into an undirected graphical model. Potentials on the graphical model are designed to accommodate dependencies among random variables by means of a set of trainable functions, like those computed by neural networks. The resulting neural-symbolic framework can effectively leverage the training data, when available, while exploiting high-level logic reasoning in a certain domain of discourse. Although exact inference is intractable within this model, different tractable models can be derived by making different assumptions. In particular, three models are presented in this chapter: Semantic-Based Regularization, Deep Logic Models and Relational Neural Machines. Semantic-Based Regularization is a scalable neural-symbolic model, that does not adapt the parameters of the reasoner, under the assumption that the provided prior knowledge is correct and must be exactly satisfied. Deep Logic Models preserve the scalability of Semantic-Based Regularization, while providing a flexible exploitation of logic knowledge by co-training the parameters of the reasoner during the learning procedure. Finally, Relational Neural Machines provide the fundamental advantages of perfectly replicating the effectiveness of training from supervised data of standard deep architectures, and of preserving the same generality and expressive power of Markov Logic Networks, when considering pure reasoning on symbolic data. The bonding between learning and reasoning is very general as any (deep) learner can be adopted, and any output structure expressed via First-Order Logic can be integrated. However, exact inference within a Relational Neural Machine is still intractable, and different factorizations are discussed to increase the scalability of the approach.

scholarly journals A Shift from Phonological Recoding to Direct Access in Reading as a Result of Previous Exposure

1984 ◽  
Vol 16 (2) ◽  
pp. 145-158 ◽  
Author(s):  
Steven G. Zecker ◽  
Mark DuMont
Keyword(s):  
Reaction Time ◽  
Lexical Access ◽  
Direct Access ◽  
Phonological Recoding ◽  
Previous Exposure ◽  
Initial Exposure ◽  
Visual Access ◽  
Access Model

The present study examined the effect of repeated exposures of a visually presented phrase on the mode of lexical access (phonological recoding vs. visual mediation) used. Subjects made meaningfulness decisions about two- and three-word phrases. Following five exposures to each phrase, some of which sounded meaningful but were not (“drops of do”), and others which were neither (“nut and bout”), the significant reaction time advantage on the first exposure for rejecting the latter phrase type was eliminated. Results supported the dual access hypothesis that subjects use phonological recoding upon initial exposure to a phrase, but following repeated exposures are able to use direct visual access. A dual access model compatible with these results is discussed.

Lexical accessing for ambiguous words among advanced Chinese L2 learners

2017 ◽  
Vol 52 (3) ◽  
pp. 209-231
Author(s):  
Helen H. Shen
Keyword(s):  
Lexical Access ◽  
Analysis Method ◽  
Contextual Cues ◽  
Factors Associated ◽  
Word Analysis ◽  
Ambiguous Words ◽  
The Us ◽  
L2 Learners ◽  
Meaning Frequency ◽  
Polysemous Words

Abstract This study investigated factors associated with and strategies used by advanced Chinese L2 learners in accessing the meanings of commonly used polysemous words (lexically ambiguous words) in sentential reading. The participants included 26 learners of Chinese from a Midwest university in the US. The results showed that word frequency, meaning frequency of polysemous words, and learners’ knowledge of polysemous words affected successful lexical access in sentential contexts. Learners mainly used five types of strategies to solve lexical ambiguity problems, of which three were more frequently used: contextual cues, the intra-word analysis method, and the dominant meaning cue. Contextual cues were the most frequently used strategy.

scholarly journals THE PATHOGENESIS OF WAXY DEGENERATION OF STRIATED MUSCLES (ZENKER'S DEGENERATION)

1909 ◽  
Vol 11 (1) ◽  
pp. 1-9 ◽  
Author(s):  
H. Gideon Wells
Keyword(s):  
Lactic Acid ◽  
Striated Muscle ◽  
Negative Evidence ◽  
Striated Muscles ◽  
Positive Evidence ◽  
Striated Muscle Fibers ◽  
Positive Results ◽  
Stimulation Of

In view of theoretical deductions and the positive results obtained in the above experiments, it would seem probable that the production of waxy degeneration depends upon the action of lactic acid which is formed by the living muscle under the stimulation of infecting bacteria or their toxins, the formation of large amounts of lactic acid and its accumulation being perhaps favored by defective circulation through the injured muscle. The hyaline transformation of muscle acted upon by lactic acid is analogous to the swelling of fibrin placed in dilute acids. This view is supported by both negative and positive experimental evidence—the negative evidence being that simple anemic necrosis, aseptic or antiseptic autolysis whether in vivo or in vitro, or the action of bacteria of various sorts on muscle in vitro, are all incapable of causing changes in muscle cells resembling those characteristic of waxy or hyaline degeneration of striated muscle. The positive evidence consists in the demonstration that lactic acid, even in dilutions comparable to the amounts that can be formed in living muscle, can produce a similar or identical waxy transformation of the striated muscle fibers, both in vitro and in vivo; and also the observation that muscles stimulated to exhaustion, under which condition lactic acid is known to accumulate in the muscle, show microscopically changes identical with those of Zenker's waxy degeneration.

Lexical Ambiguity

2018 ◽  
pp. 95-117
Author(s):  
Jennifer Rodd
Keyword(s):  
Brain Imaging ◽  
Language Comprehension ◽  
Lexical Ambiguity ◽  
Brain Regions ◽  
Contextual Cues ◽  
Imaging Studies ◽  
Word Meanings ◽  
Frontal Brain ◽  
Ambiguous Words ◽  
Multiple Meanings

This chapter on lexical ambiguity examines how words with multiple meanings are learned, stored, and processed. Lexical ambiguity is ubiquitous: over 80% of common English words have more than one dictionary entry, with some words having very many different definitions. Being able to learn and process ambiguous words is therefore critical for skilled language comprehension. This chapter reviews experiments that indicate that ambiguous words can be relatively challenging to learn, and that the competition between alternative word meanings can delay processing of these words relative to unambiguous words. However, when ambiguous words occur within sentences readers/listeners can rapidly use contextual cues to select the most likely meaning, and if necessary reinterpret the sentence in the light of subsequent information. The chapter also reviews evidence from brain imaging studies that reveals the network of temporal and frontal brain regions that are known to be important for representing and processing ambiguous words.

Positive evidence for negative evidence

1991 ◽  
Vol 11 (32) ◽  
pp. 219-251 ◽  
Author(s):  
Ernst L. Moerk
Keyword(s):  
Negative Evidence ◽  
Positive Evidence

Positive Evidence and Preemption in the Second Language Classroom

1993 ◽  
Vol 15 (2) ◽  
pp. 181-204 ◽  
Author(s):  
Martha Trahey ◽  
Lydia White
Keyword(s):  
Second Language ◽  
Universal Grammar ◽  
First Language ◽  
Negative Evidence ◽  
Parameter Setting ◽  
L2 Acquisition ◽  
Verb Movement ◽  
Language Programs ◽  
Positive Evidence ◽  
Language Classroom

In this paper we show that supplying positive evidence in the second language (L2) classroom does not necessarily trigger the appropriate L2 value of a parameter of Universal Grammar. The parameter we investigate is the verb movement parameter of Pollock (1989), which accounts for the fact that English and French adverbs differ as to where they occur in relation to the verb: In French the verb raises past the adverb, allowing the order SVAO but not SAV, whereas in English the verb does not raise, allowing SAV but not SVAO. Fifty-four francophone children (aged 11) in intensive English-as-a-second-language programs in Quebec, Canada, were exposed to a 2-week input flood of specially prepared materials containing English adverbs used naturalistically. No form-focused instruction or negative evidence on adverb placement was provided. Subjects were pretested immediately prior to the input flood, posttested immediately afterward, and again 3 weeks later, on four different tasks. On all tasks there is a change between pretest and posttest behavior, namely, a dramatic increase in use of the English SAV order but little or no decline in incorrect usage of SVAO. Results are also compared to groups reported in White (1991a, 1991b); the subjects in the present study differ from both groups in the previous studies. The results of the present study suggest that positive evidence does not serve to preempt the first language parameter setting in this case; acquiring the correct English SAV order did not lead to loss of incorrect SVAO. Implications of this result for theories of preemption and parameter setting in L2 acquisition are discussed.

Sign in / Sign up

Export Citation Format

Share Document