Context Matching Research Articles

Government chatbot: Empowering smart conversations with enhanced contextual understanding and reasoning

Currently, an increasing number of governments have adopted question answering systems (QASs) in public service delivery. As some citizens with limited information literacy often express their questions vaguely when interacting with a chatbot, it is necessary to improve the contextual understanding and reasoning ability of government chatbots (G-chatbots). This goal can be achieved through the optimisation of the matching between question, answer and context. By incorporating the Relational Graph Convolutional Networks (R-GCNs) and fuzzy logic, this study proposes a multi-turn dialogue model that introduces a re-question mechanism and a subgraph matching algorithm. The experiment results show that the model can improve the contextual reasoning ability of G-chatbots by about 10% and generate answers in a more explainable way. This study innovatively integrates a question–answer–context matching approach, re-question mechanism into the MTRF-G-chatbot model, reducing barriers to citizens’ access to government services and enhancing contextual reasoning abilities.

A novel requirement elicitation and evaluation framework for product-service systems based on contextual matching and hybrid decision-making

Most studies concerning requirement elicitation and evaluation for product-service systems (PSSs) disregard the importance of completeness and the preferences of different stakeholders. In order to address these issues, a novel requirement elicitation and evaluation framework for PSSs is proposed. First, a new requirement elicitation method based on contextual matching is introduced to capture the complete PSS requirements. Second, a hybrid decision-making method that considers the preferences of users and enterprise experts, as well as the uncertainty and stochasticity during the decision-making process, is proposed to determine the prioritization of PSS requirements. Finally, an illustrative case study of an automatic guided vehicle PSS (AGV-PSS) is demonstrated to examine and justify the proposed framework and methods.

Variation in encoding context benefits item recognition.

The current study assesses whether varying the encoding context of a repeated event is a potential strategy to improve recognition memory across retrieval contexts. Context variability, also known as encoding variability, has historically been investigated primarily using recall and cued recall tasks, with the consensus being that encoding variability is not necessarily beneficial for episodic retrieval. However, recent studies (see text) suggest that test type may determine the strategy's effectiveness. Aligned with these recent findings, we found consistent benefits to simple item recognition when a word was studied in more variable contexts compared to less variable contexts across four experiments. This main effect of context variability occurred when crossed with a manipulation of repetition spacing and when crossed with a manipulation of encoding-retrieval context match. Variation in encoding contexts beyond the future retrieval context led to better item recognition than repeated study exposures within the future retrieval context. We argue that the current study and other recent findings indicate a need to re-evaluate the historical consensus on encoding variability as a beneficial strategy for learning.

Research on Feature Extraction and Optimization Measures of University English Translation for Machine Learning

Abstract In the background of increasing translation content, it is no longer possible to rely solely on human translation to solve the problem of cross-language communication, and thus machine translation technology has gradually become an important means to solve the language barrier. In this paper, the semantic content features are extracted from university English using a semantic model based on fuzzy semantic mapping relations. Optimize the model feature extraction based on concept set context matching, introduce Super-Concept and Sub-Concept, calculate the concept semantic translation similarity, and add them into the translation decoding to get the translation optimization results. In this way, the university’s English machine translation system is constructed and evaluated. The maximum values of data recall and context matching rate of the system in this paper are 71.1% and 99.6%, respectively, and the BLEU values of this paper’s model are higher than those of the CNN model and the Transformer machine translation model in the tests of different slicing granularity. The system has high data recall, context matching rate, and translation accuracy, which is feasible and practical in university English translation, and lays a research foundation for further optimization of university English machine translation.

Multicontrast MRI Super-Resolution via Transformer-Empowered Multiscale Contextual Matching and Aggregation.

Magnetic resonance imaging (MRI) possesses the unique versatility to acquire images under a diverse array of distinct tissue contrasts, which makes multicontrast super-resolution (SR) techniques possible and needful. Compared with single-contrast MRI SR, multicontrast SR is expected to produce higher quality images by exploiting a variety of complementary information embedded in different imaging contrasts. However, existing approaches still have two shortcomings: 1) most of them are convolution-based methods and, hence, weak in capturing long-range dependencies, which are essential for MR images with complicated anatomical patterns and 2) they ignore to make full use of the multicontrast features at different scales and lack effective modules to match and aggregate these features for faithful SR. To address these issues, we develop a novel multicontrast MRI SR network via transformer-empowered multiscale feature matching and aggregation, dubbed McMRSR ++ . First, we tame transformers to model long-range dependencies in both reference and target images at different scales. Then, a novel multiscale feature matching and aggregation method is proposed to transfer corresponding contexts from reference features at different scales to the target features and interactively aggregate them. Furthermore, a texture-preserving branch and a contrastive constraint are incorporated into our framework for enhancing the textural details in the SR images. Experimental results on both public and clinical in vivo datasets show that McMRSR ++ outperforms state-of-the-art methods under peak signal to noise ratio (PSNR), structure similarity index measure (SSIM), and root mean square error (RMSE) metrics significantly. Visual results demonstrate the superiority of our method in restoring structures, demonstrating its great potential to improve scan efficiency in clinical practice.

FQN Inference in Partial Code by Prompt-tuned Language Model of Code

Partial code usually involves non-fully-qualified type names (non-FQNs) and undeclared receiving objects. Resolving the FQNs of these non-FQN types and undeclared receiving objects (referred to as type inference) is the prerequisite to effective search and reuse of partial code. Existing dictionary-lookup based methods build a symbolic knowledge base of API names and code contexts, which involve significant compilation overhead and are sensitive to unseen API names and code context variations. In this article, we propose using a p rompt-tuned c o de m asked language mod e l (MLM) as a neural knowledge base for type inference, called POME, which is lightweight and has minimal requirements on code compilation. Unlike the existing symbol name and context matching for type inference, POME infers the FQNs syntax and usage knowledge encapsulated in prompt-tuned code MLM through a colze-style fill-in-blank strategy. POME is integrated as a plug-in into web and integrated development environments (IDE) to assist developers in inferring FQNs in the real world. We systematically evaluate POME on a large amount of source code from GitHub and Stack Overflow, and explore its generalization and hybrid capability. The results validate the effectiveness of the POME design and its applicability for partial code type inference, and they can be easily extended to different programming languages (PL). POME can also be used to generate a PL-hybrid type inference model for providing a one-for-all solution. As the first of its kind, our neural type inference method opens the door to many innovative ways of using partial code.

Context-aware Pseudo-true Video Interpolation at 6G Edge

In the 6G network, lots of edge devices facilitate the low-latency transmission of video. However, with limited processing and storage capabilities, the edge devices cannot afford to reconstruct the vast amount of video data. On the condition of edge computing in the 6G network, this article fuses a self-similarity-based context feature into Frame Rate Up-Conversion (FRUC) to generate the pseudo-true video sequences at high frame rate, and its core is the extraction of the context layer for each video frame. First, we extract the patch centered at each pixel and use the self-similarity descriptor to generate the correlation surface. Then, the expectation or skewness of the correlation surface in statistics is computed to represent its context feature. By attaching an expectation or a skewness to each pixel, the context layer is constructed and added to the video frame as a new channel. According to the context layer, we predict the motion vector field of the absent frame by using the bidirectional context match and finally produce the interpolated frame. From the experimental results, it can be seen that by deploying the proposed FRUC algorithm on edge devices, the output pseudo-true video sequences have satisfying objective and subjective qualities.

Sampling-invariant fully metric learning for few-shot object detection

Few-shot object detection (FSOD) aims to learn models to detect unseen objects with a few annotated exemplars. Despite great success in FSOD, existing metric-based methods heavily rely on class prototypes extracted from limited training data and thus suffer severely from data shift, which can learn inductive bias and produce inconsistent performance across multiple runs. According to statistic law, data shift happens more frequently under low-shot scenarios than that in many-shot scenarios. However, there is insufficient research about utilizing intrinsic and robust properties of limited data for data-stable results instead of blindly achieving high performance on a specific dataset. This inspires us to model intrinsic properties of limited data for building a set of robust class prototypes. To this end, we propose a novel Sampling-Invariant Fully metric learning Network (SIF-Net) that enhances class prototypes in three aspects. Specifically, our model includes three novel modules: (1) a multi-scale context matching (MSCM) module that aggregates more accurate class concepts in a scale-matching manner, (2) a semantic drift correlation (SDC) module that recovers distorted class prototypes through object context, and (3) a fully metric learning (FML) module that encodes both class and spatial priors into class prototypes, which is fundamentally different from previous metric-based approaches. As a result, our SIF-Net is robust to various settings of few-shot data quality. Extensive experiments show that our fully metric-based framework is superior to other metric-based approaches and achieves state-of-the-art performance on PASCAL VOC and MS COCO datasets.

Context-Aware Collaborative Filtering Using Context Similarity: An Empirical Comparison

Recommender systems can assist with decision-making by delivering a list of item recommendations tailored to user preferences. Context-aware recommender systems additionally consider context information and adapt the recommendations to different situations. A process of context matching, therefore, enables the system to utilize rating profiles in the matched contexts to produce context-aware recommendations. However, it suffers from the sparsity problem since users may not rate items in various context situations. One of the major solutions to alleviate the sparsity issue is measuring the similarity of contexts and utilizing rating profiles with similar contexts to build the recommendation model. In this paper, we summarize the context-aware collaborative filtering methods using context similarity, and deliver an empirical comparison based on multiple context-aware data sets.

Retracted] Leveraging Digital Library to Enhance Research and Learning Experience of College Students: An In‐Depth Study

A digital library is a platform that contains collections of books, services, and personnel to support the sharing of knowledge with creation, dissemination, and preservation. In this context, any university library should comprehensively embrace the developmental trend occurring in the library setting, which should be strictly followed by university libraries as a special mission. Digital libraries should also actively promote the updating of the embedded service model and further upgrade the various resources of the university library. Thus, the digital library provides a platform to assist students to develop an inclination towards learning and emotional shaping. Its functional system thus serves both comprehensive and harmonious development of students. Especially, the knowledge service module of the digital library incorporates users’ scientific research context. Most of the existing research studies focus on the individual researcher and neglects the context of the entire research team. Knowledge recommendation for the context of team‐based scientific research activities can better serve more scientific research activities and team cooperation. In this work, we propose a knowledge recommendation algorithm for digital libraries based on team research‐knowledge application in context matching. We leverage the context‐aware learning model to construct the corresponding application context of the digital library knowledge and the context model of team scientific research. Subsequently, we select alternative knowledge and neighbor users as active users and further complete knowledge sorting and recommendation. According to the knowledge recommendation system in a digital library, it is confirmed that the proposed method can effectively deliver the knowledge of researchers in the context of the digital library.

Semantic Context-Aware Image Style Transfer.

To provide semantic image style transfer results which are consistent with human perception, transferring styles of semantic regions of the style image to their corresponding semantic regions of the content image is necessary. However, when the object categories between the content and style images are not the same, it is difficult to match semantic regions between two images for semantic image style transfer. To solve the semantic matching problem and guide the semantic image style transfer based on matched regions, we propose a novel semantic context-aware image style transfer method by performing semantic context matching followed by a hierarchical local-to-global network architecture. The semantic context matching aims to obtain the corresponding regions between the content and style images by using context correlations of different object categories. Based on the matching results, we retrieve semantic context pairs where each pair is composed of two semantically matched regions from the content and style images. To achieve semantic context-aware style transfer, a hierarchical local-to-global network architecture, which contains two sub-networks including the local context network and the global context network, is proposed. The former focuses on style transfer for each semantic context pair from the style image to the content image, and generates a local style transfer image storing the detailed style feature representations for corresponding semantic regions. The latter aims to derive the stylized image by considering the content, the style, and the intermediate local style transfer images, so that inconsistency between different corresponding semantic regions can be addressed and solved. The experimental results show that the stylized results using our method are more consistent with human perception compared with the state-of-the-art methods.

Word form "pravda" in periphery of nouns: indexing of first modalation stage

The purpose of the study is to study the degree to which differential features of substantive word forms such as "pravda", "fakt", "chush'", "nonsens" that represent the periphery noun zone on the modalation scale in the typical contexts match, on the one hand, differential features of prototypical representatives of the initial point of transposition, i.e. the nuclear noun, and on the other hand, the prototypical representatives of the endpoint of transposition, i.e. the nuclear denominative modal words. Through the methods of oppositional analysis, indexing, linguistic experiment, distributive and transformational analysis, the author has established and characterized the transposition stages of nouns of the "pravda" type into the category of parenthetic-modal components of the utterance, which explicate the modus subject's assessment of the communicated information in the aspect of persuasiveness. Using the example of the peripheral noun "pravda", the procedure of calculating the degrees of its modalation is shown. Combinatorics and the proportion of features of interacting nouns and parenthetic-modal units are discovered in the structure of the studied word form. Observations and conclusions can be used for a comprehensive description of the mechanism of modalation of words and word forms from different parts of speech.

Zone-Agnostic Greedy Taxi Dispatch Algorithm Based on Contextual Matching Matrix for Efficient Maximization of Revenue and Profit

This paper addresses the taxi fleet dispatch problem, which is critical for many transport service platforms such as Uber, Lyft, and Didi Chuxing. We focus on maximizing the revenue and profit a taxi platform can generate through the dispatch approaches designed with various criteria. We consider determining the proportion of taxi fleets to different destination zones given the expected rewards from the future states following the distribution decisions learned through reinforcement learning (RL) algorithms. We also take into account more straightforward greedy algorithms that look ahead fewer decision time steps in the future. Our dispatch decision algorithms commonly leverage contextual information and heuristics using a data structure called Contextual Matching Matrix (CMM). The key contribution of our paper is the insight into the trade-off between different design criteria. Primarily, through the evaluation with actual taxi operation data offered by Seoul Metropolitan Government, we challenge the natural expectation that the RL-based approaches yield the best result by showing that a lightweight greedy algorithm can have a competitive advantage. Moreover, we break the norm of dissecting the service area into sub-zones and show that matching passengers beyond arbitrary boundaries generates significantly higher operating income and profit.

Variations of impolite responses in YouTube podcast videos

This research looks at how speech partners react to language impoliteness in five YouTube podcast videos. The research data in a descriptive qualitative framework are authentic or natural in the form of spoken conversations between presenters and guest stars that include impoliteness responses. The five titles of Deddy Corbuzier's YouTube podcast videos served as the data source. The research method consists of note-taking and orthographic transcription, both of which employ a free-competence technique. The method of data analysis used was pragmatic and contextual matching. The research's findings reveal three types of responses from speech partners when confronted with language impoliteness. There are several representative variations in each of the response categories. The basis of distinguishing elements is primarily concerned with variations. Offensive countering is a response variation in which the speaker responds with curses as an expression of annoyance, ignores, or dismisses the speech partner. Defensive countering defines one's position in interactions by considering innocence, seeing oneself as distinct from others (more important than others), and demonstrating an action that places oneself in a safe situation. Face attack response attempts to accept impolite acts, confirm the speech partner's accusations, and work out solutions with the speech partner.

Paraphrase type identification for plagiarism detection using contexts and word embeddings

Paraphrase types have been proposed by researchers as the paraphrasing mechanisms underlying acts of plagiarism. Synonymous substitution, word reordering and insertion/deletion have been identified as some of the common paraphrasing strategies used by plagiarists. However, similarity reports generated by most plagiarism detection systems provide a similarity score and produce matching sections of text with their possible sources. In this research we propose methods to identify two important paraphrase types – synonymous substitution and word reordering in paraphrased, plagiarised sentence pairs. We propose a three staged approach that uses context matching and pretrained word embeddings for identifying synonymous substitution and word reordering. Our proposed approach indicates that the use of Smith Waterman Algorithm for Plagiarism Detection and ConceptNet Numberbatch pretrained word embeddings produces the best performance in terms of hbox {F}_1 scores. This research can be used to complement similarity reports generated by currently available plagiarism detection systems by incorporating methods to identify paraphrase types for plagiarism detection.

A general parsing algorithm with context matching for context-sensitive graph grammars

A general parsing algorithm with context matching for context-sensitive graph grammars

Encoding Context Determines Risky Choice

Both memory and choice are influenced by context: Memory is enhanced when encoding and retrieval contexts match, and choice is swayed by available options. Here, we assessed how context influences risky choice in an experience-based task in two main experiments (119 and 98 participants retained, respectively) and two additional experiments reported in the Supplemental Material available online (152 and 106 participants retained, respectively). Within a single session, we created two separate contexts by presenting blocks of trials in distinct backgrounds. Risky choices were context dependent; given the same choice, people chose differently depending on other outcomes experienced in that context. Choices reflected an overweighting of the most extreme outcomes within each local context rather than the global context of all outcomes. When tested in the nontrained context, people chose according to the context at encoding and not retrieval. In subsequent memory tests, people displayed biases specific to distinct contexts: Extreme outcomes from each context were more accessible and judged as more frequent. These results pose a challenge for theories of choice that rely on retrieval as guiding choice.

Multi-View Attention Network for Visual Dialog

Visual dialog is a challenging vision-language task in which a series of questions visually grounded by a given image are answered. To resolve the visual dialog task, a high-level understanding of various multimodal inputs (e.g., question, dialog history, and image) is required. Specifically, it is necessary for an agent to (1) determine the semantic intent of question and (2) align question-relevant textual and visual contents among heterogeneous modality inputs. In this paper, we propose Multi-View Attention Network (MVAN), which leverages multiple views about heterogeneous inputs based on attention mechanisms. MVAN effectively captures the question-relevant information from the dialog history with two complementary modules (i.e., Topic Aggregation and Context Matching), and builds multimodal representations through sequential alignment processes (i.e., Modality Alignment). Experimental results on VisDial v1.0 dataset show the effectiveness of our proposed model, which outperforms previous state-of-the-art methods under both single model and ensemble settings.

A morphing approach for continuous generalization of linear map features.

With the development of web maps, people are no longer satisfied with fixed and limited scale map services but want to obtain personalized and arbitrary scale map data. Continuous map generalization technology can be used to generate arbitrary scale map data. This paper proposes a morphing method for continuously generalizing linear map features using shape context matching and hierarchical interpolation (SCM-HI). More specifically, shape characteristics are quantitatively described by shape context on which shape similarity is measured based on a chi-square method; then, two levels of interpolation, skeleton and detail interpolations, are employed to generate the geometry of intermediate curves. The main contributions of our approach include (1) exploiting both the geometry and spatial structure of a vector curve in shape matching by using shape context, and (2) preserving both the main shape structure as-rigid-as-possible and local geometric details as gradual and smooth as possible for intermediate curves by hierarchical interpolation. Experiments show that our method generates plausible morphing effects and can thus serve as a robust approach for continuous generalization of linear map features.

Strategies and Techniques of Translation in Translating Songs as 21st Century Curriculum

Song is one of English material in 21st century curriculum at senior high school. To produce memorable and beautiful meaning, song lyrics comprise language variation, figurative language, idioms, and pragmatics. However, students will face problems in translating and comprehending the meaning of the song lyric. Therefore, to figure out the message and the intended meaning from the song lyrics, the students should understand techniques and strategies in translation. Regarding with the statements, the aims of this study are to find out kinds of difficulty faced by the students, to analyze kinds of translation technique used by the students and strategies to overcome their difficulties in translating English song. In conducting the research, the writer used descriptive qualitative research based on Creswell theory. The result of the research showed that the students faced difficulties in determining figurative language, idiom, linguistics factors grammatical, referential, connotative, and contextual meaning in song lyric. There are six translation techniques used by the students: literal translation, reduction, modulation, amplification, linguistic compression, and transposition. There are several strategies can be used to help students in translating variation of meaning in English song: shifting shape, shape in meaning, adaptation, contextual matching and matching recorded. Learning techniques of translation and their strategies are useful for receiving better understanding in language enhancement. In addition, to translate the source into target language, students will also have a good comprehension of the song’s content.