Use Of Contextual Information Research Articles

Deep Entity Linking via Eliminating Semantic Ambiguity With BERT

Entity linking refers to the task of aligning mentions of entities in the text to their corresponding entries in a specific knowledge base, which is of great significance for many natural language process applications such as semantic text understanding and knowledge fusion. The pivotal of this problem is how to make effective use of contextual information to disambiguate mentions. Moreover, it has been observed that, in most cases, mention has similar or even identical strings to the entity it refers to. To prevent the model from linking mentions to entities with similar strings rather than the semantically similar ones, in this paper, we introduce the advanced language representation model called BERT (Bidirectional Encoder Representations from Transformers) and design a hard negative samples mining strategy to fine-tune it accordingly. Based on the learned features, we obtain the valid entity through computing the similarity between the textual clues of mentions and the entity candidates in the knowledge base. The proposed hard negative samples mining strategy benefits entity linking from the larger, more expressive pre-trained representations of BERT with limited training time and computing sources. To the best of our knowledge, we are the first to equip entity linking task with the powerful pre-trained general language model by deliberately tackling its potential shortcoming of learning literally, and the experiments on the standard benchmark datasets show that the proposed model yields state-of-the-art results.

Learning Pixel-Level and Instance-Level Context-Aware Features for Pedestrian Detection in Crowds

Pedestrian detection in crowded scenes is an intractable problem in computer vision, in which occlusion often presents a great challenge. In this paper, we propose a novel context-aware feature learning method for detecting pedestrians in crowds, with the purpose of making better use of context information for dealing with occlusion. Unlike most current pedestrian detectors that only extract context information from a single and fixed region, a new pixel-level context embedding module is developed to integrate multi-cue context into a deep CNN feature hierarchy, which enables access to the context of various regions by multi-branch convolution layers with different receptive fields. In addition, to utilize the distinctive visual characteristics formed by pedestrians that appear in groups and occlude each other, we propose a novel instance-level context prediction module which is actually implemented by a two-person detector, to improve the one-person detection performance. Applying with these strategies, we achieve an efficient and lightweight detector that can be trained in an end-to-end fashion. We evaluate the proposed approach on two popular pedestrian detection datasets, i.e., Caltech and CityPersons. The extensive experimental results demonstrate the effectiveness of the proposed method, especially under heavy occlusion cases.

A context-sensitive offloading system using machine-learning classification algorithms for mobile cloud environment

Computational offloading in Mobile Cloud Computing (MCC) has attracted attention due to benefits in energy saving and improved mobile application performance. Nevertheless, this technique underperforms if the offloading decision ignores contextual information. While recent studies have highlighted the use of contextual information to improve the computational offloading decision, there still remain challenges regarding the dynamic nature of the MCC environment. Most solutions design a single reasoner for the offloading decision and do not know how accurate and precise this technique is, so that when applied in real-world environments it can contribute to inaccurate decisions and consequently the low performance of the overall system. Thus, this paper proposes a Context-Sensitive Offloading System (CSOS) that takes advantage of the main machine-learning reasoning techniques and robust profiling system to provide offloading decisions with high levels of accuracy. We first evaluate the main classification algorithms under our database and the results show that JRIP and J48 classifiers achieves 95% accuracy. Secondly, we develop and evaluate our system under controlled and real scenarios, where context information changes from one experiment to another. Under these conditions, CSOS makes correct decisions as well as ensuring performance gains and energy efficiency.

Voice Activity Detection Using an Adaptive Context Attention Model

Voice activity detection (VAD) classifies incoming signal segments into speech or background noise; its performance is crucial in various speech-related applications. Although speech-signal context is a relevant VAD asset, its usefulness varies in unpredictable noise environments. Therefore, its usage should be adaptively adjustable to the noise type. This letter improves the use of context information by using an adaptive context attention model (ACAM) with a novel training strategy for effective attention, which weights the most crucial parts of the context for proper classification. Experiments in real-world scenarios demonstrate that the proposed ACAM-based VAD outperforms the other baseline VAD methods.

Exploiting context for people detection in crowded scenes

The effective use of context information can significantly improve the effect of object detection. This paper proposes a method to exploit the context of a hard example searched by online hard example mining for improving the detection effect for people in crowded scenes. As shown by these experiments, this method can improve the effectiveness of faster R-CNN networks on people detection with a smaller convolutional neural network model. Because smaller convolutional neural network model is used, both the running memory consumption and the computing time can be reduced. Hence, our method can be implemented more easily on embedded devices.

Attentional capture by incongruent object/background scenes in patients with Alzheimer disease

AimObject/background association is critical to understand the context of visual scenes but also in daily life tasks like object search. Patients with Alzheimer disease (AD) exhibit impairment in scene processing at different levels: perception, recognition, memory and spatial navigation. We explored whether patients with AD make use of contextual information in congruent and incongruent target/background conditions in three different saccadic choice tasks. ExperimentWe recruited 36 participants (12 young, 12 patients with AD at a moderate stage and 12 age-matched controls). Pairs of scenes (one congruent and one incongruent object/background) were displayed. In a free viewing task we recorded whether the participants spontaneously direct their gaze (eye tracker recordings) toward the congruent or the incongruent scene. In a task referred as “implicit”, the participants had to saccade toward a pre-defined target (animal or piece of furniture) in a scene (congruent or not with the target). In a task, called “explicit”, participants had to saccade towards the congruent scene. ResultsIn contrast with both young and older controls patients with AD showed difficulties to refrain a first saccade toward incongruent scenes in the free viewing and the implicit tasks. They were at chance level in the explicit task. When given time to explore the two scenes, in a manual response condition, they were able to accomplish the implicit and explicit tasks with good accuracy. ConclusionIn contrast to healthy controls patients with AD exhibited a strong, significant, bias towards incongruent object/background scenes, even in the free viewing task, suggesting that unfamiliar or deviant stimuli attract their attention spontaneously. This result is in line with studies showing impairments in filtering out irrelevant distractors in visual search tasks and with studies suggesting inefficient top-down control to select relevant information at early stage of the disease.

Combining Context and Knowledge Representations for Chemical-Disease Relation Extraction.

Automatically extracting the relationships between chemicals and diseases is significantly important to various areas of biomedical research and health care. Biomedical experts have built many large-scale knowledge bases (KBs) to advance the development of biomedical research. KBs contain huge amounts of structured information about entities and relationships, therefore plays a pivotal role in chemical-disease relation (CDR) extraction. However, previous researches pay less attention to the prior knowledge existing in KBs. This paper proposes a neural network-based attention model (NAM) for CDR extraction, which makes full use of context information in documents and prior knowledge in KBs. For a pair of entities in a document, an attention mechanism is employed to select important context words with respect to the relation representations learned from KBs. Experiments on the BioCreative V CDR dataset show that combining context and knowledge representations through the attention mechanism, could significantly improve the CDR extraction performance while achieve comparable results with state-of-the-art systems.

Information Literacy in the Age of Machines that Learn: Desiderata for Machines that Teach

Abstract With the use of machine learning and other advances, modern information search systems make it easy for searchers to access information to meet their most frequent information needs. Building from Kuhlthau’s concepts of exploration and differentiating, this article argues that along with the benefits of greater accessibility, these advances impede the development of information literacy, conceptualized as processes for planning, accessing, judging and communicating information. It is argued that information literacy emerges during interaction with search systems and modern system designs hide or render unworkable the contextual information needed for the judgment processes of information literacy. In response to these concerns, the article contributes desiderata for new designs that facilitate the discovery, navigation and use of context information.

Polarimetric SAR Image Classification Using Geodesic Distances and Composite Kernels

The covariance/coherence matrices are the most common way of representing polarimetric information in the polarimetric synthetic aperture radar (PolSAR) data and have been extensively used in PolSAR classification. Since PolSAR covariance and coherence matrices are Hermitian positive-definite, they form a nonlinear manifold, rather than Euclidean space. Though the geodesic distance measures defined on a manifold are suitable for describing similarities of PolSAR matrix data, the nonlinearity of the manifold often makes the involved optimization problems awkward. To address this problem, we propose to embed the manifold-based PolSAR data into a high (infinite)-dimensional reproducing kernel Hilbert space by Stein kernel and log-Euclidean kernel. Besides, we introduce the composite kernel into the sparse representation classification in order to exploit the spatial context information of PolSAR data. The proposed method is assessed using different PolSAR datasets. Experimental results demonstrate the superior performance compared with the methods without the use of contextual information.

THE EXTRACTION OF POST-EARTHQUAKE BUILDING DAMAGE INFORMATIOM BASED ON CONVOLUTIONAL NEURAL NETWORK

Abstract. The seismic damage information of buildings extracted from remote sensing (RS) imagery is meaningful for supporting relief and effective reduction of losses caused by earthquake. Both traditional pixel-based and object-oriented methods have some shortcoming in extracting information of object. Pixel-based method can’t make fully use of contextual information of objects. Object-oriented method faces problem that segmentation of image is not ideal, and the choice of feature space is difficult. In this paper, a new stratage is proposed which combines Convolution Neural Network (CNN) with imagery segmentation to extract building damage information from remote sensing imagery. the key idea of this method includes two steps. First to use CNN to predicate the probability of each pixel and then integrate the probability within each segmentation spot. The method is tested through extracting the collapsed building and uncollapsed building from the aerial image which is acquired in Longtoushan Town after Ms 6.5 Ludian County, Yunnan Province earthquake. The results show that the proposed method indicates its effectiveness in extracting damage information of buildings after earthquake.

Neural Cross-Lingual Entity Linking

A major challenge in Entity Linking (EL) is making effective use of contextual information to disambiguate mentions to Wikipedia that might refer to different entities in different contexts. The problem exacerbates with cross-lingual EL which involves linking mentions written in non-English documents to entries in the English Wikipedia: to compare textual clues across languages we need to compute similarity between textual fragments across languages. In this paper, we propose a neural EL model that trains fine-grained similarities and dissimilarities between the query and candidate document from multiple perspectives, combined with convolution and tensor networks. Further, we show that this English-trained system can be applied, in zero-shot learning, to other languages by making surprisingly effective use of multi-lingual embeddings. The proposed system has strong empirical evidence yielding state-of-the-art results in English as well as cross-lingual: Spanish and Chinese TAC 2015 datasets.

Random forest classification based acoustic event detection utilizing contextual-information and bottleneck features

The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique.

Context-aware recommender system based on ontology for recommending tourist destinations at Bandung

Recommender System is software that is able to provide personalized recommendation suits users’ needs. Recommender System has been widely implemented in various domains, including tourism. One approach that can be done for more personalized recommendations is the use of contextual information. This paper proposes a context aware recommender based ontology system in the tourism domain. The system is capable of recommending tourist destinations by using user preferences of the categories of tourism and contextual information such as user locations, weather around tourist destinations and close time of destination. Based on the evaluation, the system has accuracy of of 0.94 (item recommendation precision evaluated by expert) and 0.58 (implicitly from system-end user interaction). Based on the evaluation of user satisfaction, the system provides a satisfaction level of more than 0.7 (scale 0 to 1) for speed factors for providing liked recommendations (PE), informative description of recommendations (INF) and user trust (TR).

Differential patterns of contextual organization of memory in first-episode psychosis

Contextual information is used to support and organize episodic memory. Prior research has reliably shown memory deficits in psychosis; however, little research has characterized how this population uses contextual information during memory recall. We employed an approach founded in a computational framework of free recall to quantify how individuals with first episode of psychosis (FEP, N = 97) and controls (CON, N = 55) use temporal and semantic context to organize memory recall. Free recall was characterized using the Hopkins Verbal Learning Test-Revised (HVLT-R). We compared FEP and CON on three measures of free recall: proportion recalled, temporal clustering, and semantic clustering. Measures of temporal/semantic clustering quantified how individuals use contextual information to organize memory recall. We also assessed to what extent these measures relate to antipsychotic use and differentiated between different types of psychosis. We also explored the relationship between these measures and intelligence. In comparison to CON, FEP had reduced recall and less temporal clustering during free recall (p < 0.05, Bonferroni-corrected), and showed a trend towards greater semantic clustering (p = 0.10, Bonferroni-corrected). Within FEP, antipsychotic use and diagnoses did not differentiate between free recall accuracy or contextual organization of memory. IQ was related to free recall accuracy, but not the use of contextual information during recall in either group (p < 0.05, Bonferroni-corrected). These results show that in addition to deficits in memory recall, FEP differed in how they organize memories compared to CON.

Comprehension of Japanese Indirect Speech Acts by Chinese Learners: Focusing on the Use of Context Information

Comprehension of Japanese Indirect Speech Acts by Chinese Learners: Focusing on the Use of Context Information

Agent-based simulation with NetLogo to evaluate ambient intelligence scenarios

In this paper, an agent-based simulation is developed in order to evaluate an Ambient Intelligence scenario based on agents. Many AmI applications are implemented through agents but they are not compared with any other existing alternative in order to evaluate the relative benefits of using them. The proposed simulation environment analyses such benefits using two evaluation criteria: First, measuring agent satisfaction of different types of desires along the execution. Second, measuring time savings obtained through a correct use of context information. In this paper, an existing agent architecture, an ontology and a 12-steps protocol to provide AmI services in airports, is evaluated using the NetLogo simulation environment. In our NetLogo model, we are considering scalability issues of this application domain but using FIPA and BDI extensions to be coherent with our previous works and our previous JADE implementation of them. The NetLogo model simulates an airport with agent “passengers” passing through several zones located in a specific order in a map: passport controls, check-in counters of airline companies, boarding gates, and different types of shopping. Although the initial data in each simulation are generated randomly, and the model is just an approximation of real-world airports, the definition of this case of use of AmI through NetLogo agents opens an interesting way to evaluate the benefits of using AmI, which is a significant contribution to the final development of AmI systems.

Chinese Sentence Classification Based on Convolutional Neural Network

Sentence classification is one of the significant issues in Natural Language Processing (NLP). Feature extraction is often regarded as the key point for natural language processing. Traditional ways based on machine learning can not take high level features into consideration, such as Naive Bayesian Model. The neural network for sentence classification can make use of contextual information to achieve greater results in sentence classification tasks. In this paper, we focus on classifying Chinese sentences. And the most important is that we post a novel architecture of Convolutional Neural Network (CNN) to apply on Chinese sentence classification. In particular, most of the previous methods often use softmax classifier for prediction, we embed a linear support vector machine to substitute softmax in the deep neural network model, minimizing a margin-based loss to get a better result. And we use tanh as an activation function, instead of ReLU. The CNN model improve the result of Chinese sentence classification tasks. Experimental results on the Chinese news title database validate the effectiveness of our model.

Petri Net-Based Modelling and Analysis of Context Aware System Considering Context Timeliness

Context-aware applications become very popular due to its providing intelligent and customized services to the front-end users. However, most of them focus on context management flow and don’t consider time constraints of context and business process. To make full use of context information in business environment and provide more accurate services, this paper extends CPN with context and proposes a merging modeling approach considering business process, context and its timeliness. Then to verify whether the time constraint of merged model is satisfied, the available time allowance algorithm is presented. Eventually, a case of intelligent care exemplifies the effectiveness of the model and algorithm.

Use of Contextual Information and Prediction by Struggling Adult Readers: Evidence From Reading Times and Event-Related Potentials

ABSTRACTWe employed self-paced reading and event-related potential measures to investigate how adults of varying literacy levels use sentence context information when reading. Community-dwelling participants read strongly and weakly constraining sentences that ended with expected or unexpected target words. Skilled readers showed N400s that were graded by the cloze probability of the targets, with larger N400s for more unexpected words. Moreover, it took these participants longer to read unexpected targets in strongly than weakly constraining sentences, suggesting a processing cost for revising predictions. Among less skilled readers, a reliable N400 difference was found between expected and unexpected targets only for the strongly constraining sentences. They also took longer when targets were unexpected, regardless of the context. These findings suggest that lower literacy readers could only immediately take advantage of strongly constraining context information to facilitate word processing and that they do not make as much use of predictive processing during comprehension.

Autistic Traits Affect P300 Response to Unexpected Events, regardless of Mental State Inferences.

Limited use of contextual information has been suggested as a way of understanding cognition in people with autism spectrum disorder (ASD). However, it has also been argued that individuals with ASD may have difficulties inferring others' mental states. Here, we examined how individuals with different levels of autistic traits respond to contextual deviations by measuring event-related potentials that reflect context usage. The Autism Spectrum Quotient (AQ) was used to quantify autistic-like traits in 28 university students, and 19 participants were defined as Low or High AQ groups. To additionally examine inferences about mental state, two belief conditions (with or without false belief) were included. Participants read short stories in which the final sentence included either an expected or an unexpected word and rated the word's degree of deviation from expectation. P300 waveform analysis revealed that unexpected words were associated with larger P300 waveforms for the Low AQ group, but smaller P300 responses in the High AQ group. Additionally, AQ social skill subscores were positively correlated with evaluation times in the Unexpected condition, whether a character's belief was false or not. This suggests that autistic traits can affect responses to unexpected events, possibly because of decreased availability of context information.