The Big Data explosion has necessitated the development of search algorithms that scale sub-linearly in time and memory. While compression algorithms and search algorithms do exist independently, few algorithms offer ...
详细信息
ISBN:
(数字)9798350362480
ISBN:
(纸本)9798350362497
The Big Data explosion has necessitated the development of search algorithms that scale sub-linearly in time and memory. While compression algorithms and search algorithms do exist independently, few algorithms offer both, and those which do are domain-specific. We present panCAKES, a novel approach to compressive search, i.e., a way to perform k-NN and ρ-NN search on compressed data while only decompressing a small, relevant, portion of the data. panCAKES assumes the manifold hypothesis and leverages the low-dimensional structure of the data to compress and search it efficiently. panCAKES is generic over any distance function for which the distance between two points is proportional to the memory cost of storing an encoding of one in terms of the other. This property holds for many widely-used distance functions, e.g. string edit distances (Levenshtein, Needleman-Wunsch, etc.) and set dissimilarity measures (Jaccard, Dice, etc.). We benchmark panCAKES on a variety of datasets, including genomic, proteomic, and set data. We compare compression ratios to gzip, and search performance between the compressed and uncompressed versions of the same dataset. panCAKES achieves compression ratios close to those of gzip, while offering sub-linear time performance for k-NN and ρ-NN search. We conclude that panCAKES is an efficient, general-purpose algorithm for exact compressive search on large datasets that obey the manifold hypothesis. We provide an open-source implementation of panCAKES in the Rust programming language.
Deep Learning is changing our world in a way that we never imagined before. In order to solve and reduce the misclassification of galaxy due to lack of abundance of data we focus to classify Galaxies accurately. We st...
详细信息
The deadly COVID-19 coupled with other diseases has proven to be the biggest challenge humans have seen in ages. Healthcare systems, even in the most developed countries, have completely shattered during the peak of w...
详细信息
Static application Security Testing (SAST) tools are an established means of detecting vulnerabilities early in development. Previous studies have reported low detection rates from SAST tools and recommend either comb...
详细信息
The deployment of video streaming in an Internet of Vehicles (IoV) context enhances both driving comfort and road safety. Video transmission provides clear and informative perspectives, offering a wide range of entert...
详细信息
ISBN:
(数字)9798350350265
ISBN:
(纸本)9798350350272
The deployment of video streaming in an Internet of Vehicles (IoV) context enhances both driving comfort and road safety. Video transmission provides clear and informative perspectives, offering a wide range of entertainment and infotainment, including details on restaurants, parking, and tourist attractions. However, maintaining a satisfactory quality of experience (QoE) in an IoV environment, which is characterized by fluctuating bandwidth and changing topology, presents a significant challenge. Furthermore, traffic parameters such as vehicle speed negatively impact video quality. This paper reviews various solutions proposed in the literature to optimize the MPEG-DASH (Dynamic Adaptive Streaming over HTTP) protocol in an IoV environment. The objective is to dynamically adjust video quality based on available resources to enhance QoE. Additionally, this review identifies key areas that researchers should focus on to improve this protocol. We highlight also the important challenges for our future contributions.
The video surveillance system is a key component of the technologies deployed in smart cities. It serves a variety of applications, including public safety, crime prevention, traffic management, and environmental moni...
详细信息
ISBN:
(数字)9798350350265
ISBN:
(纸本)9798350350272
The video surveillance system is a key component of the technologies deployed in smart cities. It serves a variety of applications, including public safety, crime prevention, traffic management, and environmental monitoring. The data captured by these systems includes sensitive information related to privacy, crime and national security, requiring robust protection against data breaches to ensure confidentiality. In this paper, we introduce a video fingerprinting-based method that uses a timestamp and device number, intended to prevent and detect image manipulation or replacement of original images with copies during transmission and of receiving the monitored data. Additionally, we propose the use of a blockchain system with an immutable distributed ledger for traceability and auditing of authentication procedures.
Floods cause immense damage worldwide, resulting in significant financial losses and loss of life. Existing flood detection methods often face limitations, including delays and errors due to manual data interpretation...
详细信息
One significant aspect of smart cities is the emergence of Autonomous Electric Vehicles (AEVs), which are poised to revolutionize transportation systems, offering an intelligent transportation system that surpasses tr...
详细信息
ISBN:
(数字)9798350371628
ISBN:
(纸本)9798350371635
One significant aspect of smart cities is the emergence of Autonomous Electric Vehicles (AEVs), which are poised to revolutionize transportation systems, offering an intelligent transportation system that surpasses traditional modes in terms of reducing carbon emissions, environmental pollution, congestion, transportation costs, and wait times. In this context, ride-hailing systems will play a pivotal role by leveraging shared AEVs to optimize transportation efficiency. This paper presents an in-dept. exploration of an efficient Autonomous Driving System (ARH) that leverages a fleet of autonomous electric vehicles, enabling riders to seamlessly schedule trips while the intelligent system allocates suitable vehicles and integrates multiple ride requests. Our proposed solution aims to minimize arrival times for both AEVs and passengers by accepting the maximum number of feasible requests. To strike a balance between passenger satisfaction and system profitability, this study introduces a novel heuristic approach called Minimization of Passenger and Vehicle Time (MPVT). By adopting this approach, we aim to overcome the challenges associated with complexity and deliver a satisfactory experience for passengers while maximizing profitability for the system operator.
Early and accurate detection of diseases affecting rice plants may have devastating impacts on productivity, and early and accurate detection of these diseases is essential for mitigating their consequences. However, ...
详细信息
Deep neural networks (DNNs) are susceptible to bugs, just like other types of software systems. A significant uptick in using DNN, and its applications in wide-ranging areas, including safety-critical systems, warrant...
详细信息
ISBN:
(纸本)9798350329964
Deep neural networks (DNNs) are susceptible to bugs, just like other types of software systems. A significant uptick in using DNN, and its applications in wide-ranging areas, including safety-critical systems, warrant extensive research on software engineering tools for improving the reliability of DNN-based systems. One such tool that has gained significant attention in the recent years is DNN fault localization. This paper revisits mutation-based fault localization in the context of DNN models and proposes a novel technique, named deepmufl, applicable to a wide range of DNN models. We have implemented deepmufl and have evaluated its effectiveness using 109 bugs obtained from StackOverflow. Our results show that deepmufl detects 53/109 of the bugs by ranking the buggy layer in top-1 position, outperforming state-of-the-art static and dynamic DNN fault localization systems that are also designed to target the class of bugs supported by deepmufl. Moreover, we observed that we can halve the fault localization time for a pre-trained model using mutation selection, yet losing only 7.55% of the bugs localized in top-1 position.
暂无评论