Matching Retrieval Research Articles

Does Translation Technology Affect Translators' Performance? A Meta-Analysis

Translation technologies, including computer-assisted translation (CAT) tools, translation memory (TM) systems, and machine translation (MT), are increasingly utilized in professional translation workflows and training. However, the effects of these technologies on translators' performance remain inconclusive. This meta-analysis examines the overall impact of translation technologies on translator performance by synthesizing data from 12 experimental studies published between 2000 and 2023. The study investigates the effectiveness of translation technologies compared to traditional translation methods. The findings reveal a significant positive effect size, indicating that translation technologies have the potential to improve translators’ performance relative to purely human translation. The integration of advanced interactive CAT systems and post-editing MT demonstrates larger advancements compared to basic TM match retrieval. Moreover, experienced professional translators derive greater benefits from incorporating technologies than student translators, highlighting the importance of leveraging automation capabilities alongside human expertise. However, the study identifies significant heterogeneity among the studies, influenced by factors such as translation direction. Translators translating into their native language exhibit greater advancements, emphasizing the advantages of technologies that strengthen fluency in the target language.

Application of fast batch obtaining cited frequency of web of science with VBA

Citation frequency is currently recognized as one of the evaluation indicators in the academic field. Therefore, research related to evaluation requires obtaining the citation frequency of literature, especially the citation frequency of Web of Science. Using programming languages like Java to batch obtain the citation frequency of Web of Science through the Web of Science API interface requires a certain programming foundation and specialized programming tools, which are generally difficult for researchers to master. This article explores a convenient and fast method to batch obtain the citation frequency of Web of Science using VBA. Using Web of Science's literature matching retrieval service, it is possible to obtain the citation frequency of literature in batches. The literature matching retrieval service is implemented in Excel using VBA code. Through testing and practice, it has been proven that batch and rapid retrieval of Web of Science citation frequency can be achieved using VBA in Excel. This article not only provides the entire runnable source code for easy expansion but also gives practical examples for reference.

Early cognitive assessment in premature infants: the discriminatory value of eye-tracking vs. Bayley Scales.

The testing of visuocognitive development in preterm infants shows strong interactions between perinatal characteristics and cognition, learning and overall neurodevelopment evolution. The assessment of anticipatory gaze data of object-location bindings via eye-tracking can predict the neurodevelopment of preterm infants at the age of 3 years; little is known, however, about the early cognitive function and its assessment methods during the first year of life. The current study presents data from a novel assessment tool, a Delayed Match Retrieval (DMR) paradigm via eye-tracking was used to measure visual working memory (VWM) and attention skills. The eye-tracking task that was designed to measure infants' ability to actively localize objects and to make online predictions of object-location bindings. 63 infants participated in the study, 39 preterm infants and 24 healthy full term infants - at a corrected age of 8-9 months for premature infants and similar chronological age for full term infants. Infants were also administered the Bayley Scales of Infant and Toddler Development. The analysis of the Bayley scores showed no significant difference between the two groups while the eye-tracking data showed a significant group effect on all measurements. Moreover, preterm infants' VWM performance was significantly lower than full term's. Birth weight affected the gaze time on all Areas Of Interest (AOIs), overall VWM performance and the scores at the Cognitive Bayley subscale. Furthermore, preterm infants with fetal growth restriction (FGR) showed significant performance effects in the eye-tracking measurements but not on their Bayley scores verifying the high discriminatory value of the eye gaze data. Visual working memory and attention as measured via eye-tracking is a non-intrusive, painless, short duration procedure (approx. 4-min) was found to be a significant tool for identifying prematurity and FGR effects on the development of cognition during the first year of life. Bayley Scales alone may not pick up these deficits. Identifying tools for early neurodevelopmental assessments and cognitive function is important in order to enable earlier support and intervention in the vulnerable group of premature infants, given the associations between foundational executive functional skills and later cognitive and academic ability.

Accelerating spliced alignment of long RNA sequencing reads using parallel maximal exact match retrieval

The genomics landscape has undergone a revolutionary transformation with the emergence of third-generation sequencing technologies. Fueled by the exponential surge in sequencing data, there is an urgent demand for accurate and rapid algorithms to effectively handle this burgeoning influx. Under such circumstances, we developed a parallelized, yet accuracy-lossless algorithm for maximal exact match (MEM) retrieval to strategically address the computational bottleneck of uLTRA, a leading spliced alignment algorithm known for its precision in handling long RNA sequencing (RNA-seq) reads. The design of the algorithm incorporates a multi-threaded strategy, enabling the concurrent processing of multiple reads simultaneously. Additionally, we implemented the serialization of index required for MEM retrieval to facilitate its reuse, resulting in accelerated startup for practical tasks. Extensive experiments demonstrate that our parallel algorithm achieves significant improvements in runtime, speedup, throughput, and memory usage. When applied to the largest human dataset, the algorithm achieves an impressive speedup of 10.78 × , significantly improving throughput on a large scale. Moreover, the integration of the parallel MEM retrieval algorithm into the uLTRA pipeline introduces a dual-layered parallel capability, consistently yielding a speedup of 4.99 × compared to the multi-process and single-threaded execution of uLTRA. The thorough analysis of experimental results underscores the adept utilization of parallel processing capabilities and its advantageous performance in handling large datasets. This study provides a showcase of parallelized strategies for MEM retrieval within the context of spliced alignment algorithm, effectively facilitating the process of RNA-seq data analysis. The code is available at https://github.com/RongxingWong/AcceleratingSplicedAlignment.

Emergency Decision-Making for Middle Route of South-to-North Water Diversion Project Using Case-Based Reasoning and Prospect Theory

The middle route of the South-to-North Water Diversion Project is one of the crucial frameworks of China’s water network and an essential channel for water resource allocation in North China. The safe operation of the project has a huge impact on regional economic development, social stability and other aspects. The objectives of this research are to improve the disposal efficiency of all kinds of accidents during the operation of the Middle Route of the South-to-North Water Diversion Project, reduce people’s property losses and ensure the safety of water supply along the line. This paper will put forward a new emergency decision-making method based on case-based reasoning technology and prospect theory. The method is divided into two parts: (1) Collecting the historical case information and building the case library. The frame representation in the case-based reasoning technology is used to describe the characteristics of historical cases and adopt the two-level method of historical cases fast retrieval and similarity fuzzy matching retrieval to complete the preliminary selection of emergency plans; (2) The decision-making and optimization model of disposal plans based on prospect theory, namely, using the value function and probability weight classification to measure the prospect value of similar schemes and selecting the optimal disposal scheme, in order to improve the science and rationality of the decision-making results. Finally, examples are taken to verify the feasibility and effectiveness of the method.

Investigation on artificial boundary problem in the combined finite-discrete element method (FDEM)

The combined finite-discrete element method (FDEM) is an advanced finite and discrete element coupling method, which can commendably simulate the large deformation and fracture process of materials. The explicit solution method in FDEM is naturally suitable for solving dynamic problems, and the key to dynamic analysis is to establish artificial boundary conditions that can be prepared to simulate infinite domain motion. At present, there is only the viscous boundary condition in the original FDEM, but this is only applicable for solving the internal source wave problems (e.g., blasting, mechanical vibration), and there are no artificial boundary conditions for solving the external source wave problems (e.g., earthquake). Therefore, three artificial boundary conditions have been developed in the present study to enhance the ability of FDEM to conduct dynamic response analysis. Firstly, the basic principle of FDEM and the existing viscous boundary condition are briefly introduced. Then, three newly added boundary conditions are introduced in sequence: (1) viscous-spring boundary condition, which can absorb stress wave energy at the boundary and restore the residual displacement to meet the actual engineering; (2) free-field boundary condition, which is coupled with free-field motion to absorb scattered waves at the lateral boundaries of the model, and the algorithms of mesh matching retrieval and coupling calculation are also introduced; (3) static-dynamic unified boundary condition, which can accurately transform the fixed boundary under the quasi-static analysis into the non-reflective boundary condition under the dynamic analysis. Finally, several classic models are built to verify the feasibility of the newly added boundary conditions. The comparison with the shaking table test also shows that the improved FDEM can be used for seismic response analysis under actual earthquakes.

3D Assembly Model Retrieval Method for a Multiassembly Interface

3D model retrieval is increasingly becoming a hot topic in today’s research. The existing model retrieval methods are limited to the retrieval between similar models, while the matching retrieval for models with assembly relationships is neglected. They cannot realize the prediction of the virtual assembly and make the assemblers to spend extensive time detecting when there are assembly errors in the model. This work proposes a 3D assembly model retrieval method for multiassembly interface, which can retrieve the parts or subassemblies that can be assembled with the input model. In this work, a method for the reuse of the 3D assembly mode is also provided. The method here involves a series of steps, which includes an attribute adjacency graph to express the 3D assembly models, followed by the definition of a conjugated subgraph, and the serialization of the vertices of the graph for the Atlas of the assembly model. Finally, the frequent subgraph mining method is improved, and the 3D assembly model that satisfies the multiassembly interface is extracted. The article claims through the obtained results that this method not only allows the performance of the 3D assembly model retrieval of multiassembly interfaces but also the extraction of the 3D model required by designers, improving, at the same time, the product design efficiency.

A VR-Enabled Chatbot Supporting Design and Manufacturing of Large and Complex Power Transformers

Virtual reality (VR) immersive technology allows users to experience enhanced reality using human–computer interfaces (HCI). Many systems have implemented VR with improved HCI to provide strategic market advantages for industry and engineering applications. An intelligent chatbot is a conversational system capable of natural language communication allowing users to ask questions and receive answers online to enhance customer services. This research develops and implements a system framework for a VR-enabled large industrial power transformer mass-customization chatbot. The research collected 1272 frequently asked questions (FAQs) from a power transformer manufacturers’ knowledge base that is used for question matching and answer retrieval. More than 1.2 million Wikipedia engineering pages were used to train a word-embedding model for natural language understanding of question intent. The complex engineering questions and answers are integrated with an immersive VR computer human interface. The system enables users to ask questions and receive explicit and detailed answers combined with 3D immersive images of industrial sized power transformer assemblies. The user interfaces can be projected into the VR headwear or computer screen and manipulated with a controller. The unique immersive VR consultation chatbot system is to support real-time design consultation for the design and manufacturing of complex power transformers.

Matching Subsequence Music Retrieval in a Software Integration Environment

This paper firstly introduces the basic knowledge of music, proposes the detailed design of a music retrieval system based on the knowledge of music, and analyzes the feature extraction algorithm and matching algorithm by using the features of music. Feature extraction of audio data is the important research of this paper. In this paper, the main melody features, MFCC features, GFCC features, and rhythm features, are extracted from audio data and a feature fusion algorithm is proposed to achieve the fusion of GFCC features and rhythm features to form new features under the processing of principal component analysis (PCA) dimensionality reduction. After learning the main melody features, MFCC features, GFCC features, and rhythm features, based on the property that PCA dimensionality reduction can effectively reduce noise and improve retrieval efficiency, this paper proposes vector fusion by dimensionality reduction of GFCC features and rhythm features. The matching retrieval of audio features is an important task in music retrieval. In this paper, the DTW algorithm is chosen as the main algorithm for retrieving music. The classification retrieval of music is also achieved by the K-nearest neighbor algorithm. In this paper, after implementing the research and improvement of algorithms, these algorithms are integrated into the system to achieve audio preprocessing, feature extraction, feature postprocessing, and matching retrieval. This article uses 100 different kinds of MP3 format music as the music library and randomly selects 4 pieces each time, and it tests the system under different system parameters, recording duration, and environmental noise. Through the research of this paper, the efficiency of music retrieval is improved and theoretical support is provided for the design of music retrieval software integration system.

Person Re-Identification Based on Joint Loss and Multiple Attention Mechanism

Person re-identification (ReID) is the use of computer vision and machine learning techniques to determine whether the pedestrians in the two images under different cameras are the same person. It can also be regarded as a matching retrieval task for person targets in different scenes. The research focuses on how to obtain effective person features from images with occlusion, angle change, and target attitude change. Based on the present difficulties and challenges in ReID, the paper proposes a ReID method based on joint loss and multi-attention network. It improves the person re-identification algorithm based on global characteristics, introduces spatial attention and channel attention and constructs joint loss function, improving the characteristic extraction ability of the network and improving the model performance of person re-identification. It analyzes the validity and necessity of each module of the algorithm through the ablation experiment. In addition, it carries out training and evaluation on the two person re-recognition data sets Market1501 and MSMT17, and verifies the advantages of the proposed algorithm in contrast to other advanced algorithms.

Indexing Early Visual Memory Durability in Infancy.

The goal was to examine the scope and development of early visual memory durability. We investigated individual- and age-related differences across three unique tasks in 6- to 12-month-olds (Mage =8.87, N=49) by examining the effect of increased delay on memory performance. Results suggest longer-term memory processes are quantifiable by 8months using a modified Change Detection paradigm and spatial-attention cueing processes are quantifiable by 10months using a modified Delayed Response paradigm, utilizing 500-1,250ms delays. Performance improved from 6 to 12months and longer delays impaired performance. We found no evidence for success on the Delayed Match Retrieval task at any age. These outcomes help inform our understanding of infant visual memory durability and its emergence throughout early development.

An efficient retrieval approach for encrypted speech based on biological hashing and spectral subtraction

In this paper, an efficient retrieval approach for encrypted speech based on biological hashing and spectral subtraction is proposed. The proposed approach improves the impact of noise on the robustness and discrimination of speech hashing scheme, as well as improves the retrieval efficiency, accuracy, and security of search digests, and realizes the authentication of the query result. The speech owner firstly secure the original speech file by encrypting it with two-dimensional Arnold mapping and uploading it to the encrypted speech library on cloud. Then, the pre-processed speech signal is subjected to spectral subtraction and noise reduction, as well as the discrete wavelet transform (DWT) is performed to obtain the wavelet low-frequency coefficient and reconstruct the speech signal, calculating the normalized autocorrelation function to obtain the matrix feature vector, and using the Chebychew mapping algorithm to generate the pseudo-random matrix, and generate the pseudo-random Fourier matrix by fast Fourier transform (FFT). Finally, iterate the matrix feature vector and pseudo-random matrix. After the thresholding, the hash sequence is constructed and uploaded to the system hash index table on cloud. When speech’s user retrieval, the Hamming distance algorithm is used for the matching retrieval operation during the search and integrity authentication of the query result. The experimental results show that the proposed approach effectively reduces the noise of speech, with strong robustness and discrimination, and the retrieval efficiency, accuracy and security have been significantly improved.

Coding of featural information in visual working memory in 2.5-year-old toddlers

The number of objects that infants can remember in visual working memory (VWM) increases rapidly during the first few years of life (Kaldy & Leslie, 2005; Ross-Sheehy, Oakes, & Luck, 2003). However, less is understood about the representational format of VWM: whether storage is determined by fixed-precision memory slots, or the allocation of a limited continuous resource. In the current study, we adapted the Delayed Match Retrieval eye-tracking paradigm (Kaldy, Guillory, & Blaser, 2016), to test 2.5-year-old toddlers’ ability to remember three object-location bindings when the to-be-remembered objects were all unique (Experiment 1) versus when they shared features such as color or shape (Experiment 2). 2.5-year-olds succeeded in Experiment 1, but only performed marginally better than chance in Experiment 2. Interestingly, when incorrect, participants in Experiment 2 were no more likely to select a decoy item that shared a feature with the target item. It seems that the increased similarity of to-be-remembered objects did not impair memory for the objects directly, but instead increased the likelihood of catastrophic forgetting.

An agricultural audio retrieval method based on inverted index of silence word

With the popularization of Internet and multimedia technology, a large amount of agriculture-related audio information is rapidly accumulated and widely used. The method of keyword matching retrieval for text data content describing audio files can no longer meet the need of agricultural audio files retrieval in terms of retrieval efficiency and content depth. In view of the above problems in agricultural audio retrieval technology, this paper proposes an audio retrieval method based on inverted index of silence word. In this method, a new hybrid retrieval model combining representation retrieval and semantic retrieval is designed and used. The experimental results show that the recall rate and precision rate of this method are above 99.5%, and the average retrieval time is less than 1s. Therefore, this method greatly improves the efficiency and the accuracy of the audio information retrieval, and can be used for rapid retrieval of agricultural audio.

An Encrypted Speech Retrieval Method Based on Deep Perceptual Hashing and CNN-BiLSTM

Since convolutional neural network (CNN) can only extract local features, and long short-term memory (LSTM) neural network model has a large number of learning calculations, a long processing time and an obvious degree of information loss as the length of speech increases. Utilizing the characteristics of autonomous feature extraction in deep learning, CNN and bidirectional long short-term memory (BiLSTM) network are combined to present an encrypted speech retrieval method based on deep perceptual hashing and CNN-BiLSTM. Firstly, the proposed method extracts the Log-Mel Spectrogram/MFCC features of the original speech and enters the CNN and BiLSTM networks in turn for model training. Secondly, we use the trained fusion network model to learn the deep perceptual feature and generate deep perceptual hashing sequences. Finally, the normalized Hamming distance algorithm is used for matching retrieval. In order to protect the speech security in the cloud, a speech encryption algorithm based on a 4D hyperchaotic system is proposed. The experimental results show that the proposed method has good discrimination, robustness, recall and precision compared with the existing methods, and it has good retrieval efficiency and retrieval accuracy for longer speech. Meanwhile, the proposed speech encryption algorithm has a high key space to resist exhaustive attacks.

Neural models for information retrieval without labeled data

Recent developments of machine learning models, and in particular deep neural networks, have yielded significant improvements on several computer vision, natural language processing, and speech recognition tasks. Progress with information retrieval (IR) tasks has been slower, however, due to the lack of large-scale training data as well as neural network models specifically designed for effective information retrieval [9]. In this dissertation, we address these two issues by introducing task-specific neural network architectures for a set of IR tasks and proposing novel unsupervised or weakly supervised solutions for training the models. The proposed learning solutions do not require labeled training data. Instead, in our weak supervision approach, neural models are trained on a large set of noisy and biased training data obtained from external resources, existing models, or heuristics. We first introduce relevance-based embedding models [3] that learn distributed representations for words and queries. We show that the learned representations can be effectively employed for a set of IR tasks, including query expansion, pseudo-relevance feedback, and query classification [1, 2]. We further propose a standalone learning to rank model based on deep neural networks [5, 8]. Our model learns a sparse representation for queries and documents. This enables us to perform efficient retrieval by constructing an inverted index in the learned semantic space. Our model outperforms state-of-the-art retrieval models, while performing as efficiently as term matching retrieval models. We additionally propose a neural network framework for predicting the performance of a retrieval model for a given query [7]. Inspired by existing query performance prediction models, our framework integrates several information sources, such as retrieval score distribution and term distribution in the top retrieved documents. This leads to state-of-the-art results for the performance prediction task on various standard collections. We finally bridge the gap between retrieval and recommendation models, as the two key components in most information systems. Search and recommendation often share the same goal: helping people get the information they need at the right time. Therefore, joint modeling and optimization of search engines and recommender systems could potentially benefit both systems [4]. In more detail, we introduce a retrieval model that is trained using user-item interaction (e.g., recommendation data), with no need to query-document relevance information for training [6]. Our solutions and findings in this dissertation smooth the path towards learning efficient and effective models for various information retrieval and related tasks, especially when large-scale training data is not available.

EL: Local Image Descriptor Based on Extreme Responses to Partial Derivatives of 2D Gaussian Function

We propose a two-part local image descriptor EL (Edges and Lines), based on the strongest image responses to the first- and second-order partial derivatives of the two-dimensional Gaussian function. Using the steering theorems, the proposed method finds the filter orientations giving the strongest image responses. The orientations are quantized, and the magnitudes of the image responses are histogrammed. Iterative adaptive thresholding of histogram values is then applied to normalize the histogram, thereby making the descriptor robust to nonlinear illumination changes. The two-part descriptor is empirically evaluated on the HPatches benchmark for three different tasks, namely, patch verification, image matching, and patch retrieval. The proposed EL descriptor outperforms the traditional descriptors such as SIFT and RootSIFT on all three evaluation tasks and the deep-learning-based descriptors DeepCompare, DeepDesc, and TFeat on the tasks of image matching and patch retrieval.

Two-year-olds succeed at MIT: Multiple identity tracking in 20- and 25-month-old infants

Infants’ ability to remember objects and their locations emerges during the first year of life. However, not much is known about infants’ ability to track objects’ identities in a dynamic environment. Here, we tailored the delayed match retrieval eye-tracking paradigm to study infants’ ability to track two object identities during occlusion—an infant version of multiple identity tracking (MIT). Delayed match retrieval uses virtual “cards” as stimuli that are first shown face up, exposing to-be-remembered information, and then turned face down, occluding it. Here, cards were subject to movement during the face-down occlusion period. We used complex non-nameable objects as card faces to discourage verbal rehearsal. In three experiments (N = 110), we compared infants’ ability to track object identities when two previously exposed cards were static (Experiment 1), were moved into new positions along the same trajectory (Experiment 2), or were moved along different trajectories (Experiment 3) while face down. We found that 20-month-olds could remember two object identities when static; however, it was not until 25 months of age that infants could track when movement was introduced. Our results show that the ability to track multiple identities in visual working memory is present by 25 months.

Focused attention predicts visual working memory performance in 13-month-old infants: A pupillometric study

Attention turns looking, into seeing. Yet, little developmental research has examined the interface of attention and visual working memory (VWM), where what is seen is maintained for use in ongoing visual tasks. Using the task-evoked pupil response – a sensitive, real-time, involuntary measure of focused attention that has been shown to correlate with VWM performance in adults and older children – we examined the relationship between focused attention and VWM in 13-month-olds. We used a Delayed Match Retrieval paradigm, to test infants’ VWM for object-location bindings – what went where – while recording anticipatory gaze responses and pupil dilation. We found that infants with greater focused attention during memory encoding showed significantly better memory performance. As well, trials that ended in a correct response had significantly greater pupil response during memory encoding than incorrect trials. Taken together, this shows that pupillometry can be used as a measure of focused attention in infants, and a means to identify those individuals, or moments, where cognitive effort is maximized.

A retrieval algorithm of encrypted speech based on short-term cross-correlation and perceptual hashing

In order to achieve extraction perceptual features from the encryption speech as a search digest for the content-based encryption speech retrieval, we present a retrieval algorithm of encrypted speech based on short-term cross-correlation and perceptual hashing in this paper. Firstly, the study encrypts the speech file and uploads the encrypted speech data to the encryption speech database in cloud server. Secondly, the sample speech clips are obtained by the cutting operation from the speech file for scrambling encryption. The perceptual hashing sequence of the encrypted speech is constructed by extracting the short-term cross-correlation of the encrypted speech signals as the search digest. These perceptual hashing sequences are uploaded into the hashing index table of cloud server. Finally, the Hamming distance algorithm is used for the matching retrieval operation during the search. The experimental results show that the proposed algorithm of encrypted speech perceptual hashing has a better discrimination, robustness and compactness, and the perceptual hashing sequences can be extracted directly from the encrypted sample speech. Meanwhile, the encryption speech signal has high recall and precision ratios after various content preserving operations. In the whole retrieval process, the downloading and decrypting operations of speech data are not necessary.