Retrieval Efficiency Research Articles

Analysis of the effect of user-understanding-based retrieval system improvement on the application of intelligent archive management

Abstract As a precipitation and accumulation of history, archives management has gradually tended to be digitalized and informalized with the continuous updating and development of modern technology. In this paper, we first study the retrieval system and focus on the Boolean model, vector model, and probabilistic model in information retrieval technology. The matching relationship between documents and queries is detected from the document set for the user’s query, and a relevance retrieval system based on user understanding is proposed to solve the matching problem. The amount of information that needs to be retrieved is growing exponentially, and how a user perceives the information is crucial to the process. Then, in order to solve the problem of insufficient retrieval efficiency caused by the explosive growth of wisdom files, the retrieval system is creatively optimized on the basis of the ant colony algorithm, which effectively improves the efficiency of wisdom file management. The efficiency of the optimized retrieval system is verified and analyzed in an experimental simulation environment. The findings demonstrate that when the amount of archives rises, the retrieval effectiveness of the improved ant colony algorithm described in this study marginally improves, but in 10~35s. As the inventory of the Smart Archives increases, the content retrieval of the archives will become more and more frequent. This study improves retrieval efficiency and serves as a good demonstration for the construction of archival management information technology.

685 A quality improvement initiative to increase retrieval efficiency and accuracy of cystic fibrosis genetic profiles

685 A quality improvement initiative to increase retrieval efficiency and accuracy of cystic fibrosis genetic profiles

Effective and efficient content-based similarity retrieval of large lung CT images based on WSSLN model.

The in-depth combination and application of AI technology and medical imaging, especially high- definition CT imaging technology, make accurate diagnosis and treatment possible. Retrieving similar CT image(CI)s to an input one from the large-scale CI database of labeled diseases is helpful to realize a precise computer-aided diagnosis. In this paper, we take lung CI as an example and propose progressive content-based similarity retrieval(CBSR) method of the lung CIs based on a Weakly Supervised Similarity Learning Network (WSSLN) model. Two enabling techniques (i.e., the WSSLN model and the distance- based pruning scheme) are proposed to facilitate the CBSR processing of the large lung CIs. The main result of our paper is that, our approach is about 45% more effective than the state-of-the-art methods in terms of the mean average precision(mAP). Moreover, for the retrieval efficiency, the WSSLN-based CBSR method is about 150% more efficient than the sequential scan.

Multi-Scale Feature Fusion Based on PVTv2 for Deep Hash Remote Sensing Image Retrieval

For high-resolution remote sensing image retrieval tasks, single-scale features cannot fully express the complexity of the image information. Due to the large volume of remote sensing images, retrieval requires extensive memory and time. Hence, the problem of how to organically fuse multi-scale features and enhance retrieval efficiency is yet to be resolved. We propose an end-to-end deep hash remote sensing image retrieval model (PVTA_MSF) by fusing multi-scale features based on the Pyramid Vision Transformer network (PVTv2). We construct the multi-scale feature fusion module (MSF) by using a global attention mechanism and a multi-head self-attention mechanism to reduce background interference and enhance the representation capability of image features. Deformable convolution is introduced to address the challenge posed by varying target orientations. Moreover, an intra-class similarity (ICS) loss is proposed to enhance the discriminative capability of the hash feature by minimizing the distance among images of the same category. The experimental results show that, compared with other state-of-the-art methods, the proposed hash feature could yield an excellent representation of remote sensing images and improve remote sensing image retrieval accuracy. The proposed hash feature can gain an increase of 4.2% and 1.6% in terms of mAP on the UC Merced and NWPU-RESISC45 datasets, respectively, in comparison with other methods.

A Clustering Visualization Method for Density Partitioning of Trajectory Big Data Based on Multi-Level Time Encoding

The proliferation of the Internet and the widespread adoption of mobile devices have given rise to an immense volume of real-time trajectory big data. However, a single computer and conventional databases with limited scalability struggle to manage this data effectively. During the process of visual rendering, issues such as page stuttering and subpar visual outcomes often arise. This paper, founded on a distributed architecture, introduces a multi-level time encoding method using “minutes”, “hours”, and “days” as fundamental units, achieving a storage model for trajectory data at multi-scale time. Furthermore, building upon an improved DBSCAN clustering algorithm and integrating it with the K-means clustering algorithm, a novel density-based partitioning clustering algorithm has been introduced, which incorporates road coefficients to circumvent architectural obstacles, successfully resolving page stuttering issues and significantly enhancing the quality of visualization. The results indicate the following: (1) when data is extracted using the units of “minutes”, “hours”, and “days”, the retrieval efficiency of this model is 6.206 times, 12.475 times, and 18.634 times higher, respectively, compared to the retrieval efficiency of the original storage model. As the volume of retrieved data increases, the retrieval efficiency of the proposed storage model becomes increasingly superior to that of the original storage model. Under identical experimental conditions, this model’s retrieval efficiency also outperforms the space–time-coded storage model; (2) Under a consistent rendering level, the clustered trajectory data, when compared to the unclustered raw data, has shown a 40% improvement in the loading speed of generating heat maps. There is an absence of page stuttering. Furthermore, the heat kernel phenomenon in the heat map was also resolved while enhancing the visualization rendering speed.

Hybrid COOT-Reverse Cognitive Fruit Fly Optimization-Based Big Data Services and Virtual Machine Allocation for Cloud Storage System

In recent years, cloud computing technologies have been developed rapidly in this computing world to provide suitable on-demand network access all over the world. A cloud service provider offers numerous types of cloud services to the user. But the most significant issue is how to attain optimal virtual machine (VM) allocation for the user and design an efficient big data storage platform thereby satisfying the requirement of both the cloud service provider and the user. Therefore, this paper presents two novel strategies for optimizing VM resource allocation and cloud storage. An optimized cloud cluster storage service is introduced in this paper using a binarization based on modified fuzzy c-means clustering (BMFCM) algorithm to overcome the negative issues caused by the repetitive nature of the big data traffic. The BMFCM algorithm utilized can be implemented transparently and can also address problems associated with massive data storage. The VM selection is optimized in the proposed work using a hybrid COOT-reverse cognitive fruit fly (RCFF) optimization algorithm. The main aim of this algorithm is to improve the massive big data traffic and storage locality. The CPU utilization, VM power, memory dimension and network bandwidth are taken as the fitness function of the hybrid COOT-RCFF algorithm. When implemented in CloudSim and Hadoop, the proposed methodology offers improvements in terms of completion time, overall energy consumption, makespan, user provider satisfaction and load ratio. The results show that the proposed methodology improves the execution time and data retrieval efficiency by up to 32% and 6.3% more than the existing techniques.

A Fusion Encoder with Multi-Task Guidance for Cross-Modal Text–Image Retrieval in Remote Sensing

In recent years, there has been a growing interest in remote sensing image–text cross-modal retrieval due to the rapid development of space information technology and the significant increase in the volume of remote sensing image data. Remote sensing images have unique characteristics that make the cross-modal retrieval task challenging. Firstly, the semantics of remote sensing images are fine-grained, meaning they can be divided into multiple basic units of semantic expression. Different combinations of basic units of semantic expression can generate diverse text descriptions. Additionally, these images exhibit variations in resolution, color, and perspective. To address these challenges, this paper proposes a multi-task guided fusion encoder (MTGFE) based on the multimodal fusion encoding method, the progressiveness of which has been proved in the cross-modal retrieval of natural images. By jointly training the model with three tasks: image–text matching (ITM), masked language modeling (MLM), and the newly introduced multi-view joint representations contrast (MVJRC), we enhance its capability to capture fine-grained correlations between remote sensing images and texts. Specifically, the MVJRC task is designed to improve the model’s consistency in joint representation expression and fine-grained correlation, particularly for remote sensing images with significant differences in resolution, color, and angle. Furthermore, to address the computational complexity associated with large-scale fusion models and improve retrieval efficiency, this paper proposes a retrieval filtering method, which achieves higher retrieval efficiency while minimizing accuracy loss. Extensive experiments were conducted on four public datasets to evaluate the proposed method, and the results validate its effectiveness.

Privacy-Preserving Image Retrieval Based on Disordered Local Histograms and Vision Transformer in Cloud Computing

Frequent data breaches in the cloud environment have seriously affected cloud subscribers and providers. Privacy-preserving image retrieval methods can improve the security of cloud image retrieval; however, existing methods have limited accuracy on dynamically updated image databases and mobile lightweight devices. In this study, we propose a privacy-preserving image retrieval method based on disordered local histograms and vision transformer in cloud computing, by designing a multiple encryption method and transformer-based feature model to better mine the local feature value of encrypted images. Specifically, the user performs different value substitution, position substitution, and color substitution on the subblocks of the image to protect the image information. The cloud server extracts the unordered local histogram from the encrypted image and generates retrievable features using transformer. Experiments show that compared with similar CNN schemes, the retrieval accuracy of this method is improved by 8.5%, and the retrieval efficiency is improved by 54.8%.

CTSARF: A Chinese Text Similarity Analysis Model based on Residual Fusion

Semantic textual similarity analysis is a fundamental task in the field of natural language processing. Both representational and interactive models are extensively employed for this purpose. However, representational models often suffer from issues such as semantic defocus, semantic shift, and low accuracy. In contrast, interactive models can accurately capture the semantic focus of a text and extract ample semantic information, but hindered by high time costs and low retrieval efficiency. To address these challenges, this paper introduces a Chinese Text Similarity Analysis model based on Residual Fusion (CTSARF). This model effectively implements the semantic feature fusion of representational and interactive models through a constant mapping method using residual networks. Therefore CTSARF can achieve both high accuracy and retrieval efficiency simultaneously. In this study, we employ the cosine-based loss function CoSENT and the fine-tuned pre-trained model Mengzi to optimize performance in text similarity analysis tasks. Additionally, we utilize a data augmentation method based on text editing to enhance the robustness of CTSARF when evaluating text similarity on unbalanced datasets. We conducted optimization comparison experiments, benchmark model comparison experiments, and ablation experiments of the CTSARF model on four Chinese datasets. The results demonstrate improvements in accuracy, recall, precision, and F1 scores, thereby highlighting the necessity and superiority of the model fusion method based on residual networks.

Deep Learning: A Study of Pattern Recognition for Personalized Clothing

Objectives: This article aims to enhance the efficiency of clothing recognition and retrieval by implementing deep learning algorithms for personalized clothing pattern recognition. Methods: Based on the you only look once version 4 (YOLOv4) algorithm in deep learning, the CSPDarknet-53 in the original algorithm was replaced by GhostNet, and the original Leaky ReLU activation function was replaced by FMish. Then, an improved YOLOv4 algorithm was obtained. Experiments were carried out on the personalized clothing pattern set, the Fashion Mnist dataset, and the DeepFashion dataset to compare and analyze different algorithms. Findings: When replacing CSPDarknet-53 with GhostNet and the Leaky ReLU activation function with FMish, the optimized YOLOv4 algorithm performed significantly better, verifying the reliability of the YOLOv4 improvement. The optimized algorithm achieved an F1 value of 94.22% and a mAP of 95.41% on different datasets, and 39.51% and 49.56% on the DeepFashion dataset, respectively, outperforming other deep learning methods such as the faster-recurrent convolutional neural network. Furthermore, the floating-point operations per second of the optimized YOLOv4 algorithm were 8.72 G, showing a reduction of 49.71% compared to the traditional algorithm. This suggested that it had low complexity and calculation amounts. Novelty: The optimized YOLOv4 algorithm performs excellently in recognizing personalized clothing patterns, which can provide a new and reliable approach for recognition and retrieval in the field of clothing. Doi: 10.28991/HIJ-2023-04-03-03 Full Text: PDF

Work Together: Correlation-Identity Reconstruction Hashing for Unsupervised Cross-Modal Retrieval

Unsupervised cross-modal hashing has attracted considerable attention to support large-scale cross-modal retrieval. Although promising progresses have been made so far, existing methods still suffer from limited capability on excavating and preserving the intrinsic multi-modal semantics. In this paper, we propose a Correlation-Identity Reconstruction Hashing (CIRH) method to alleviate this challenging problem. We develop a new unsupervised deep cross-modal hash learning framework to model and preserve the heterogeneous multi-modal correlation semantics into both hash codes and functions, and simultaneously, we involve both the hash codes and functions with the descriptive identity semantics. Specifically, we construct a multi-modal collaborated graph to model the heterogeneous multi-modal correlations, and jointly perform the intra-modal and cross-modal semantic aggregation on homogeneous and heterogeneous graph networks to generate a multi-modal complementary representation with correlation reconstruction. Furthermore, an identity semantic reconstruction process is designed to involve the generated representation with identity semantics by reconstructing the input modality representations. Finally, we propose a correlation-identity consistent hash function learning strategy to transfer the modelled multi-modal semantics into the neural networks of modality-specific deep hash functions. Experiments demonstrate the superior performance of the proposed method on both retrieval accuracy and efficiency. We provide our source codes and experimental datasets at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/XizeWu/CIRH</uri> .

HyNet: A novel hybrid deep learning approach for efficient interior design texture retrieval

Interior designers are suffering from a lack of intelligent design methods. This study aims to enhance the accuracy and efficiency of retrieval textures for interior design, which is a crucial step toward intelligent design. Currently, interior designers rely on repetitive tasks to obtain textures from websites, which is ineffective as a interior design often requires hundreds of textures. To address this issue, this study proposes a hybrid deep learning approach, HyNet, which boosts retrieval efficiency by recommending similar textures instead of blindly searching. Additionally, a new indoor texture dataset is created to support the application of artificial intelligence in this field. The results demonstrate that the proposed method’s ten recommended images achieve a high accuracy rate of 91.41%. This is a significant improvement in efficiency, which can facilitate the design industry’s progression towards intelligence. Overall, this study offers a promising solution to the challenges facing interior designers, and it has the potential to significantly enhance the industry’s productivity and innovation.

A vehicle license plate data access model based on the jump hash consistency algorithm.

The massive amount of vehicle plate data generated by intelligent transportation systems is widely used in the field of urban transportation information system construction and has a high scientific research and application value. The adoption of big data platforms to properly preserve, process, and exploit these valuable data resources has become a hot research area in recent years. To address the problems of implementing complex multi-conditional comprehensive query functions and flexible data applications in the key-value database storage environment of a big data platform, this paper proposes a data access model based on the jump hash consistency algorithm. Algorithms such as data slice storage and multi-threaded sliding window parallel reading are used to realize evenly distributed storage and fast reading of massive time-series data on clustered data nodes. A comparative analysis of data distribution uniformity and retrieval efficiency shows that the model can effectively avoid generating the cluster hotspot problem, support comprehensive analysis queries with various complex conditions, and maintain high query efficiency by precisely positioning the data storage range and utilizing parallel scan reading.

Query Context Expansion for Open-Domain Question Answering

Humans are accustomed to autonomously associating prior knowledge with the text in a query when answering questions. However, for machines lacking cognition and common sense, a query is merely a combination of some words. Although we can enrich the semantic information of the given query through language representation or query expansion (QE) , the information contained in the query is still insufficient. In this paper, we propose an effective passage retrieval method named query context expansion-based retrieval (QCER) for open-domain question answering (OpenQA) . QCER associates a query with domain information by adding contextual association information based on the pseudo-relevance feedback (PRF) . QCER uses a dense reader to select top-n expansion terms for QE. We implement QCER by appending reader predictions, theoretically present in candidate passages, as contextual information to the initial query to form the new query. QCER with sparse representations (BM25) can improve retrieval efficiency and accelerate query convergence so that the reader can find the desired answer using fewer relevant passages, e.g., 10 passages, as soon as possible. Moreover, QCER can be easily combined with dense passage retrieval (DPR) to achieve even better performance, as sparse and dense representations are often complementary. Remarkably, we demonstrate that QCER achieves state-of-the-art performance in three tasks, passage retrieval, passage reading, and passage reranking, on the Natural Questions (NQ) and TriviaQA (Trivia) datasets under an extractive QA setup.

Cross‐modal person re‐identification based on deep attention hash learning

SummaryPerson re‐identification based on text description is a critical task in modern security systems. However, existing methods primarily focus on performance and overlook the crucial aspect of retrieval efficiency. In this article, we propose a novel two‐stage multimodal re‐discovery algorithm called DCH‐ReID, which leverages attention hashing. First, we introduce a chunked mapping hash learning method that effectively mitigates confusion between hash codes. Second, we propose a hash learning approach based on the channel attention mechanism, assigning higher binary bit weights to important body parts. Finally, to balance retrieval performance and time efficiency, we present a two‐stage retrieval scheme. Through extensive experiments on the CUHK‐PEDES benchmark dataset, we validate that our proposed DCH‐ReID algorithm exhibits superior efficiency and higher accuracy compared to current mainstream text‐based pedestrian re‐identification algorithms.

An efficient scheme for in-orbit remote sensing image data retrieval

The conflict between real-time retrieval of remote sensing image data (RSID) and the limited satellite network transmission resources is becoming more and more prominent. To address it, this paper proposes a novel scheme for RSID retrieval from the remote sensing satellite network, which exploits the spatio-temporal characterization (STC) of sensing images to improve data retrieval efficiency. In particular, based on the spatio-temporal characterization, a data content naming scheme is introduced and an algorithm is accordingly proposed for RSID segmentation in satellite nodes. With the naming scheme, RSID retrieval can be implemented by making use of the Named Data Networking (NDN) architecture. The study aims to investigate the effectiveness of making use of the STC of RSID for the retrieval under the NDN architecture. To the best of our knowledge, this paper is the first to exploit STC to establish connections between user demands, RSID, and data naming for NDN. As a result, the proposed scheme makes the network aware of RSID’s STC and consequently improves the data retrieval performance of the satellite network using NDN. The evaluation results show that the proposed scheme delivers about 65% and 7% higher retrieval hit rates than a retrieval scheme based on the original NDN proposal and a recently proposed scheme — CAP, respectively. In addition, comparing with these two schemes, the proposed one reduces transmission latency by about 20% and 15%, and traffic load by about 48% and 40%, all respectively. These imply that the proposed scheme outperforms in terms of higher data retrieval efficiency and better link utilization.

An efficient long-text semantic retrieval approach via utilizing presentation learning on short-text

Although the short-text retrieval model by BERT achieves significant performance improvement, research on the efficiency and performance of long-text retrieval still faces challenges. Therefore, this study proposes an efficient long-text retrieval model based on BERT (called LTR-BERT). This model achieves speed improvement while retaining most of the long-text retrieval performance. In particular, The LTR-BERT model is trained by using the relevance between short texts. Then, the long text is segmented and stored off-line. In the retrieval stage, only the coding of the query and the matching scores are calculated, which speeds up the retrieval. Moreover, a query expansion strategy is designed to enhance the representation of the original query and reserve the encoding region for the query. It is beneficial for learning missing information in the representation stage. The interaction mechanism without training parameters takes into account the local semantic details and the whole relevance to ensure the accuracy of retrieval and further shorten the response time. Experiments are carried out on MS MARCO Document Ranking dataset, which is specially designed for long-text retrieval. Compared with the interaction-focused semantic matching method by BERT-CLS, the MRR@10 values of the proposed LTR-BERT method are increased by 2.74%. Moreover, the number of documents processed per millisecond increased by 333 times.

Improving atmospheric CO2 retrieval based on the collaborative use of Greenhouse gases Monitoring Instrument and Directional Polarimetric Camera sensors on Chinese hyperspectral satellite GF5-02

ABSTRACT The Greenhouse Gas Monitoring Instrument (GMI) onboard the Chinese hyperspectral satellite GF5–02 can provide abundant observations of global atmospheric CO2, which plays an important role in climate research. CO2 retrieval precision is the key to determining the application value of the GMI. To reduce the influence of atmospheric scattering on retrieval, we combined the Directional Polarimetric Camera (DPC) data on the same satellite to improve the anti-interference ability of GMI CO2 retrieval and ensure its retrieval precision. To realize the reliability and feasibility of the collaborative use of the GMI and DPC, this paper designs the pointing registration method of the GMI based on coastline observations, the spatial resolution matching method and the collaborative cloud screening method of the GMI and DPC observations. Combined with the DPC, which supplied the spectral data and aerosol product, the retrieval ability of the coupled bidirectional reflectance distribution function CO2 retrieval (CBCR) method developed for GMI CO2 retrieval was improved, with the retrieval efficiency of CO2 products increasing by 27%, and the CO2 retrieval precision increasing from 3.3 ppm to 2.7 ppm. Moreover, collaborative use not only guaranteed the GMI’s ability to detect global and area CO2 concentration distribution characteristics, such as significant concentration differences between the Northern and Southern Hemispheres in winter and high CO2 concentrations in urban agglomeration areas caused by human activities, but also extended the GMI’s potential for monitoring anomalous events, such as the Tonga volcanic eruption.

A Multi-Level Grid Database for Protecting and Sharing Historical Geographic Urban Data: A Case Study of Shanghai

Historical geographic data play an important supporting role in the study of long-term geographic studies, such as climate change, urban expansion and land-use and land-cover change. These data vary in source, format and accuracy and are widely found in historical documents, old maps, produced vector data, aerial photographs, old photographs, etc. The complex nature of data makes it difficult for researchers to organize, store and manage in a unified manner. Thus, GIS practitioners and social scientists will collectively face the challenge of integrating historical data into spatial databases. Herein, we introduced the concept of a multi-level spatial grid, selecting Shanghai as the study area, to construct the Shanghai historical geographic database and give the conceptual model and processing method. The experiment was performed using the China Historical Geographic Information System (CHGIS), which showed the historical evolution of Shanghai more conveniently. Meanwhile, we simulated one million rows of historical geographic data in Shanghai and compared the retrieval efficiency of the encoding method with the latitude–longitude and geometric object indexing methods, which demonstrated that our method was very effective. This research is important for the construction of a historical urban database, which can better preserve historical resources and promote urban culture with information science and technology.

CMLocate: A cross‐modal automatic visual geo‐localization framework for a natural environment without GNSS information

AbstractIn this paper, a new approach to visual geo‐localization for natural environments is proposed. The digital elevation model (DEM) data in virtual space is rendered and construct a panoramic skyline database is constructed. By combining the skyline database with real‐world image data (used as the “queries” to be localized), visual geo‐localization is treated as a cross‐modal image retrieval problem for panoramic skyline images, creating a unique new visual geo‐localization benchmark for the natural environment. Specifically, the semantic segmentation model named LineNet is proposed, for skyline extractions from query images, which has proven to be robust to a variety of complex natural environments. On the aforementioned benchmarks, the fully automatic method is elaborated for large‐scale cross‐modal localization using panoramic skyline images. Finally, the compound index is delicately designed to reduce the storage space of the positioning global descriptors and improve the retrieval efficiency. Moreover, the proposed method is proven to outperform most state‐of‐the‐art methods.