Script Event Research Articles

Integrating Relational Knowledge With Text Sequences for Script Event Prediction.

Script event prediction aims to infer subsequent events given an incomplete script. It requires a deep understanding of events, and can provide support for a variety of tasks. Existing models rarely consider the relational knowledge between events, they regard scripts as sequences or graphs, which cannot capture the relational information between events and the semantic information of script sequences jointly. To address this issue, we propose a new script form, relational event chain, that combines event chains and relational graphs. We also introduce a new model, relational-transformer, to learn embeddings based on this new script form. In particular, we first extract the relationship between events from an event knowledge graph to formalize scripts as relational event chains, then use the relational-transformer to calculate the likelihood of different candidate events, where the model learns event embeddings that encode both semantic and relational knowledge by combining transformers and graph neural networks (GNNs). Experimental results on both one-step inference and multistep inference tasks show that our model can outperform existing baselines, indicating the validity of encoding relational knowledge into event embeddings. The influence of using different model structures and different types of relational knowledge is analyzed as well.

Integrating rich event-level and schema-level information for script event prediction

Integrating rich event-level and schema-level information for script event prediction

Rich Event Modeling for Script Event Prediction

Script is a kind of structured knowledge extracted from texts, which contains a sequence of events. Based on such knowledge, script event prediction aims to predict the subsequent event. To do so, two aspects should be considered for events, namely, event description (i.e., what the events should contain) and event encoding (i.e., how they should be encoded). Most existing methods describe an event by a verb together with a few core arguments (i.e., subject, object, and indirect object), which are not precise enough. In addition, existing event encoders are limited to a fixed number of arguments, which are not flexible enough to deal with extra information. Thus, in this paper, we propose the Rich Event Prediction (REP) framework for script event prediction. Fundamentally, it is based on the proposed rich event description, which enriches the existing ones with three kinds of important information, namely, the senses of verbs, extra semantic roles, and types of participants. REP contains an event extractor to extract such information from texts. Based on the extracted rich information, a predictor then selects the most probable subsequent event. The core component of the predictor is a transformer-based event encoder that integrates the above information flexibly. Experimental results on the widely used Gigaword Corpus show the effectiveness of the proposed framework.

A Generative Approach for Script Event Prediction via Contrastive Fine-Tuning

Script event prediction aims to predict the subsequent event given the context. This requires the capability to infer the correlations between events. Recent works have attempted to improve event correlation reasoning by using pretrained language models and incorporating external knowledge (e.g., discourse relations). Though promising results have been achieved, some challenges still remain. First, the pretrained language models adopted by current works ignore event-level knowledge, resulting in an inability to capture the correlations between events well. Second, modeling correlations between events with discourse relations is limited because it can only capture explicit correlations between events with discourse markers, and cannot capture many implicit correlations. To this end, we propose a novel generative approach for this task, in which a pretrained language model is fine-tuned with an event-centric pretraining objective and predicts the next event within a generative paradigm. Specifically, we first introduce a novel event-level blank infilling strategy as the learning objective to inject event-level knowledge into the pretrained language model, and then design a likelihood-based contrastive loss for fine-tuning the generative model. Instead of using an additional prediction layer, we perform prediction by using sequence likelihoods generated by the generative model. Our approach models correlations between events in a soft way without any external knowledge. The likelihood-based prediction eliminates the need to use additional networks to make predictions and is somewhat interpretable since it scores each word in the event. Experimental results on the multi-choice narrative cloze (MCNC) task demonstrate that our approach achieves better results than other state-of-the-art baselines. Our code will be available at https://github.com/zhufq00/mcnc.

Script Event Prediction via Multilingual Event Graph Networks

Predicting what happens next in text plays a critical role in building NLP applications. Many methods including count-based and neural-network-based have been proposed to tackle the task called script event prediction: predicting the most suitable subsequent event from a candidate list given a chain of narrative events (context). However, two problems including event ambiguity and evidence bias hinder the performance of these monolingual approaches. The former means that some events in the event chain are ambiguous. The latter means that both the wrong and correct candidate events can obtain sufficient support from the event context. In this article, we propose a novel multilingual approach to address two issues simultaneously. Specifically, to alleviate the event ambiguity problem, we project the monolingual event chains to parallel cross-lingual event chains, which can provide complementary information for monolingual event disambiguation. To deal with the evidence bias problem, we construct two monolingual event graphs and a cross-lingual event aligned graph to fully explore connections between events. What’s more, we design a graph attention mechanism to model the confidence of the complement clues, which controls the information integration from various languages. By modeling the events with graphs instead of pairs or chains, the model can compare the candidate subsequent events simultaneously and choose the more suitable subsequent event as the final answer. Extensive experiments were conducted on the widely used New York Times corpus for script event prediction task and experimental results show that our approach outperforms previous models.

What happens next? Combining enhanced multilevel script learning and dual fusion strategies for script event prediction

Script event prediction (SEP), aiming at predicting next event from context event sequences (i.e., scripts), has played an important role in many real-world applications such as government decision-making. While most of the existing research only depend on the top-level event prediction, they ignore the influence of other bottom levels or other relationship modeling manners. In this paper, we focus on the problem of SEP via multilevel script learning where the goal of is to explore a multistage, multiprediction and multilevel information fusion model for SEP. This is challenging in (1) simultaneously modeling of the multilevel event relationship semantic information and (2) effectively designing multilevel information fusion strategies. In this paper, we propose a new script event prediction model based on Enhanced Multilevel script learning and Dual Fusion strategies, named EMDF-Net. Specifically, EMDF-Net designs the multilevel (event/chain/segment level) script learning to model both temporal and casual information as well as the rich structural relevance via neural stacking of self-attention mechanism and graph neural networks. Then it proposes dual fusion strategies to fully integrate different-level information by nonlinear feature composition and weighted score fusion. Finally, a deep supervision strategy is utilized to end-to-end train the whole model and provide a good initialization for information fusion. Experimental results on the popular NYT corpus demonstrate the effectiveness and superiority of EMDF-Net.

MERL: Multimodal Event Representation Learning in Heterogeneous Embedding Spaces

Previous work has shown the effectiveness of using event representations for tasks such as script event prediction and stock market prediction. It is however still challenging to learn the subtle semantic differences between events based solely on textual descriptions of events often represented as (subject, predicate, object) triples. As an alternative, images offer a more intuitive way of understanding event semantics. We observe that event described in text and in images show different abstraction levels and therefore should be projected onto heterogeneous embedding spaces, as opposed to what have been done in previous approaches which project signals from different modalities onto a homogeneous space. In this paper, we propose a Multimodal Event Representation Learning framework (MERL) to learn event representations based on both text and image modalities simultaneously. Event textual triples are projected as Gaussian density embeddings by a dual-path Gaussian triple encoder, while event images are projected as point embeddings by a visual event component-aware image encoder. Moreover, a novel score function motivated by statistical hypothesis testing is introduced to coordinate two embedding spaces. Experiments are conducted on various multimodal event-related tasks and results show that MERL outperforms a number of unimodal and multimodal baselines, demonstrating the effectiveness of the proposed framework.

A systematic review of the relationship between perceived life script event age and valence across the life span.

Drawing on life script theory, we examined the relationship between the perceived age and valence of life script events across the life span. A review of the life script literature was conducted based on 14 studies, comprising 28 samples from 12 countries, and encompassing 4,012 participants. A total of 135 life script events between birth and death were identified-roughly four times the number of such events reported in any single sample. As predicted, the relationship between mean perceived life script age and valence for positive life script events during the bump period (i.e., between the ages of 10 and 30 years) was positive in direction within 15 of 18 relevant samples; however, this association was not statistically significant based on the aggregated life script event information. In contrast, consistent with predictions, the age-valence relationship for all life script events across the entire life span was negative in direction in each of the relevant samples and based on the aggregated life script event information. It appears that the life script contains at least 2 developmental stories concerning beliefs about experiences of key events and transitions: Life events get increasingly positive during younger adulthood, but decreasingly positive across the entire life span. (PsycINFO Database Record (c) 2019 APA, all rights reserved).

SAM-Net: Integrating Event-Level and Chain-Level Attentions to Predict What Happens Next

Scripts represent knowledge of event sequences that can help text understanding. Script event prediction requires to measure the relation between an existing chain and the subsequent event. The dominant approaches either focus on the effects of individual events, or the influence of the chain sequence. However, only considering individual events will lose much semantic relations within the event chain, and only considering the sequence of the chain will introduce much noise. With our observations, both the individual events and the event segments within the chain can facilitate the prediction of the subsequent event. This paper develops self attention mechanism to focus on diverse event segments within the chain and the event chain is represented as a set of event segments. We utilize the event-level attention to model the relations between subsequent events and individual events. Then, we propose the chain-level attention to model the relations between subsequent events and event segments within the chain. Finally, we integrate event-level and chain-level attentions to interact with the chain to predict what happens next. Comprehensive experiment results on the widely used New York Times corpus demonstrate that our model achieves better results than other state-of-the-art baselines by adopting the evaluation of Multi-Choice Narrative Cloze task.

Supplemental Material for A Systematic Review of the Relationship Between Perceived Life Script Event Age and Valence Across the Life Span

Supplemental Material for A Systematic Review of the Relationship Between Perceived Life Script Event Age and Valence Across the Life Span

Life Gets Better and Better: Cultural Life Script Theory and Subjective Trajectories for Life Satisfaction

Young adults typically believe that life gets increasingly satisfying over time. We examined the cultural life script as a source of these beliefs. In Study 1 (N = 1244), tabulation of previously published studies indicated that life script events are perceived as becoming increasingly positive over time between the ages of 10 and 30. Further, a specific series of 16 key life script events during this life stage was identified. These results were replicated in Study 2 (N = 100, Mage = 21.14, 51% female) based on young adults‘ perceptions concerning life script events in their personal life stories. Further, the perception that life script events in one's personal life story were becoming increasingly positive over time was linked with more steeply inclining subjective life satisfaction trajectories (i.e. recollected past < current < anticipated future life satisfaction). In Study 3 (N = 261, Mage = 18.5, 93.7% female), manipulating life script event information (number and positivity of events over time) within a personal life story had an additive impact on young adults‘ subjective life satisfaction trajectories. These findings reveal a robust connection between information contained with the cultural life script and the belief that life gets more and more satisfying over time. Copyright © 2016 European Association of Personality Psychology

The neural substrates of script knowledge deficits as revealed by a PET study in Huntington's disease

IntroductionPrevious neuropsychological investigations have suggested that both the prefrontal cortex and the basal ganglia are involved in the management of script event knowledge required in planning behavior. MethodsThis study was designated to map, the correlations between resting-state brain glucose utilization as measured by FDG-PET (positron emission tomography) and scores obtained by means of a series of script generation and script sorting tasks in 8 patients with early Huntington's disease. ResultsThese patients exhibited a selectively greater impairment for the organizational aspects of scripts compared to the semantic aspects of scripts. We showed significant negative correlations between the number of sequencing, boundary, perseverative and intrusion errors and the metabolism of several cortical regions, not only including frontal, but also posterior regions. ConclusionOur findings suggest that, within the fronto-striatal system, the cortical frontal regions are more crucial in script retrieval and script sequencing than the basal ganglia.

Neural Correlates of Script Event Knowledge: a Neuropsychological Study Following Prefrontal Injury

Scripts sequentially link information about daily activities and event knowledge. Patients have difficulty sequencing script events following lesions of the prefrontal cortex while showing intact access to selective aspects of script knowledge. It has been suggested that the sequencing impairment is due to a deficit in an inhibitory gating mechanisms that usually enables selection of an item from competing alternatives. If this is the case, then an inhibitory task should reveal script processing impairments on a script categorization task that is not normally associated with poor performance following prefrontal damage. To test this hypothesis, we administered a simple untimed classification task and a modified Go/NoGo task in which subjects classified events from social and non-social activities (e.g., read the menu, order the food) and related semantic items (e.g., menu, order) in terms of whether they belonged to a target activity. Participants were patients with lesions of the prefrontal cortex and matched controls. The results showed that damage to the right orbitofrontal cortex was associated with social item classification errors in the simple untimed classification task. In addition, the damage to the right prefrontal cortex was associated with increased response times to respond correctly to Go trials in the modified Go/NoGo task. The data demonstrate that damage to the right orbitofrontal cortex results in impairment in the accessibility of script and semantic representations of social activities. This impairment is exacerbated by an inefficient inhibitory gating mechanism.

How Patients with Parkinson's Disease Retrieve and Manage Cognitive Event Knowledge

Several studies have pointed out that basal ganglia are involved in adaptive control of action at both motor and cognitive level. This study aimed to investigate how basal ganglia retrieve and manage script event knowledge required in planning behavior. Script event knowledge was investigated in patients with Parkinson's Disease using three kinds of activity that differed in familiarity. Unlike patients with prefrontal lesions, patients with Parkinson's Disease were able to order events in a typical sequence and obeyed the boundaries and hierarchies between events. In contrast, patients with Parkinson's Disease were impaired in evaluating how important each script event was within the context of goal-oriented planning activity. Our findings indicate that the prefrontal cortex and the basal ganglia are differentially involved in planning. The role of the basal ganglia might consist in providing a feedback about the goodness of each action while building up meaningful sequences of events during learning.

Recalling script-based text: The role of selective processing and schematic cues

Research on the learning of prose material based on scripts has attempted to relate the remembering of script events to how typical the events are of the generic scripts. The novel object effect reported by Bellezza and Bower (1982) shows that a typical script event containing an unexpected object is rated less typical but is recalled better than an event irrelevant to the script. The novel-object effect implies that the relation between script-typicality ratings and remembering may not be a simple one. One explanation of the novel-object effect is that script sentences containing a novel object are processed longer during learning. The second explanation, the inference-contradiction explanation, suggests that the presented word representing a novel object both contradicts and yet becomes associated to information strongly inferred from the script. The results of the experiment reported here support the in ferencecontradiction explanation.