Wukong

2021 ◽  
Vol 55 (1) ◽  
pp. 77-83
Author(s):  
Rong Chen ◽  
Haibo Chen
Keyword(s):  
Social Networking ◽  
State Of The Art ◽  
Low Latency ◽  
Electronic Payment ◽  
Evolving Graphs ◽  
Graph Data ◽  
Graph Querying ◽  
Locality Preserving ◽  
Urban Monitoring ◽  
System Designs

Querying graph data is becoming increasingly prevalent and important across many application domains, like social networking, urban monitoring, electronic payment, and semantic webs. In the last few years, we have ben working on improving the performance of graph querying by leveraging new hardware features and system designs. Moving towards this goal, we have designed and developed Wukong, a distributed in-memory framework that provides low latency and high throughput for concurrent query processing over large and fast-evolving graph data. This article overviews our architecture and presents four systems that aim to satisfy diverse challenging requirements on graph querying (e. g. high concurrency, evolving graphs, workload heterogencity, and locality preserving). Our systems also significantly outperform state-of-the-art systems in both latency and throughput, usually by orders of magnitude.

scholarly journals Multi-View Attribute Graph Convolution Networks for Clustering

2020 ◽  
Author(s):  
Jiafeng Cheng ◽  
Qianqian Wang ◽  
Zhiqiang Tao ◽  
Deyan Xie ◽  
Quanxue Gao
Keyword(s):  
Neural Networks ◽  
State Of The Art ◽  
Graph Embedding ◽  
Structured Data ◽  
Attention Networks ◽  
Graph Data ◽  
Graph Reconstruction ◽  
Node Attributes ◽  
Graph Neural Networks ◽  
Geometric Relationship

Graph neural networks (GNNs) have made considerable achievements in processing graph-structured data. However, existing methods can not allocate learnable weights to different nodes in the neighborhood and lack of robustness on account of neglecting both node attributes and graph reconstruction. Moreover, most of multi-view GNNs mainly focus on the case of multiple graphs, while designing GNNs for solving graph-structured data of multi-view attributes is still under-explored. In this paper, we propose a novel Multi-View Attribute Graph Convolution Networks (MAGCN) model for the clustering task. MAGCN is designed with two-pathway encoders that map graph embedding features and learn the view-consistency information. Specifically, the first pathway develops multi-view attribute graph attention networks to reduce the noise/redundancy and learn the graph embedding features for each multi-view graph data. The second pathway develops consistent embedding encoders to capture the geometric relationship and probability distribution consistency among different views, which adaptively finds a consistent clustering embedding space for multi-view attributes. Experiments on three benchmark graph datasets show the superiority of our method compared with several state-of-the-art algorithms.

scholarly journals Orthros: A Low-Latency PRF

2021 ◽  
pp. 37-77
Author(s):  
Subhadeep Banik ◽  
Takanori Isobe ◽  
Fukang Liu ◽  
Kazuhiko Minematsu ◽  
Kosei Sakamoto
Keyword(s):  
Hardware Implementation ◽  
Block Cipher ◽  
State Of The Art ◽  
Security Analysis ◽  
Parallel Structure ◽  
Low Latency ◽  
Round Function ◽  
Minimum Latency ◽  
Primary Focus ◽  
Pseudorandom Function

We present Orthros, a 128-bit block pseudorandom function. It is designed with primary focus on latency of fully unrolled circuits. For this purpose, we adopt a parallel structure comprising two keyed permutations. The round function of each permutation is similar to Midori, a low-energy block cipher, however we thoroughly revise it to reduce latency, and introduce different rounds to significantly improve cryptographic strength in a small number of rounds. We provide a comprehensive, dedicated security analysis. For hardware implementation, Orthros achieves the lowest latency among the state-of-the-art low-latency primitives. For example, using the STM 90nm library, Orthros achieves a minimum latency of around 2.4 ns, while other constructions like PRINCE, Midori-128 and QARMA9-128- σ0 achieve 2.56 ns, 4.10 ns, 4.38 ns respectively.

Deep Learning Model for Imbalanced Multi-label Surface Defect Classification

2021 ◽  
Author(s):  
Yang Liu ◽  
Yachao Yuan ◽  
Jing Liu
Keyword(s):  
Deep Learning ◽  
Surface Defect ◽  
State Of The Art ◽  
Steel Production ◽  
Learning Model ◽  
The State ◽  
Low Latency ◽  
Defect Classification ◽  
Key Techniques ◽  
Deep Learning Model

Abstract Automatic defect classification is vital to ensure product quality, especially for steel production. In the real world, the amount of collected samples with labels is limited due to high labor costs, and the gathered dataset is usually imbalanced, making accurate steel defect classification very challenging. In this paper, a novel deep learning model for imbalanced multi-label surface defect classification, named ImDeep, is proposed. It can be deployed easily in steel production lines to identify different defect types on the steel's surface. ImDeep incorporates three key techniques, i.e., Imbalanced Sampler, Fussy-FusionNet, and Transfer Learning. It improves the model's classification performance with multi-label and reduces the model's complexity over small datasets with low latency. The performance of different fusion strategies and three key techniques of ImDeep is verified. Simulation results prove that ImDeep accomplishes better performance than the state-of-the-art over the public dataset with varied sizes. Specifically, ImDeep achieves about 97% accuracy of steel surface defect classification over a small imbalanced dataset with a low latency, which improves about 10% compared with that of the state-of-the-art.

Spam 2.0 State of the Art

2012 ◽  
Vol 4 (1) ◽  
pp. 17-36 ◽  
Author(s):  
Pedram Hayati ◽  
Vidyasagar Potdar
Keyword(s):  
Early Detection ◽  
Web 2.0 ◽  
Social Networking ◽  
State Of The Art ◽  
The State ◽  
Online Forums ◽  
Mass Content ◽  
Social Networking Websites ◽  
Comprehensive Survey

Spam 2.0 is defined as the propagation of unsolicited, anonymous, mass content to infiltrate legitimate Web 2.0 applications. A fake eye-catching profile in social networking websites, a promotional review, a response to a thread in online forums with unsolicited content, or a manipulated Wiki page are examples of Spam 2.0. In this paper, the authors provide a comprehensive survey of the state-of-the-art, detection-based, prevention-based and early-detection-based Spam 2.0 filtering methods.

scholarly journals A survey on compact features for visual content analysis

2016 ◽  
Vol 5 ◽  
Author(s):  
Luca Baroffio ◽  
Alessandro E. C. Redondi ◽  
Marco Tagliasacchi ◽  
Stefano Tubaro
Keyword(s):  
Content Analysis ◽  
Large Scale ◽  
Ad Hoc ◽  
State Of The Art ◽  
Large Body ◽  
Visual Features ◽  
Low Latency ◽  
Visual Content ◽  
Feature Encoding ◽  
Memory Footprint

Visual features constitute compact yet effective representations of visual content, and are being exploited in a large number of heterogeneous applications, including augmented reality, image registration, content-based retrieval, and classification. Several visual content analysis applications are distributed over a network and require the transmission of visual data, either in the pixel or in the feature domain, to a central unit that performs the task at hand. Furthermore, large-scale applications need to store a database composed of up to billions of features and perform matching with low latency. In this context, several different implementations of feature extraction algorithms have been proposed over the last few years, with the aim of reducing computational complexity and memory footprint, while maintaining an adequate level of accuracy. Besides extraction, a large body of research addressed the problem of ad-hoc feature encoding methods, and a number of networking and transmission protocols enabling distributed visual content analysis have been proposed. In this survey, we present an overview of state-of-the-art methods for the extraction, encoding, and transmission of compact features for visual content analysis, thoroughly addressing each step of the pipeline and highlighting the peculiarities of the proposed methods.

scholarly journals Nonparametric Bayesian Semi-supervised Word Segmentation

2017 ◽  
Vol 5 ◽  
pp. 179-189 ◽  
Author(s):  
Ryo Fujii ◽  
Ryo Domoto ◽  
Daichi Mochihashi
Keyword(s):  
Social Networking ◽  
Hybrid Model ◽  
Bayesian Methods ◽  
State Of The Art ◽  
Word Segmentation ◽  
Nonparametric Bayesian ◽  
Social Networking Services ◽  
New Words ◽  
Discriminative Model ◽  
Model Structures

This paper presents a novel hybrid generative/discriminative model of word segmentation based on nonparametric Bayesian methods. Unlike ordinary discriminative word segmentation which relies only on labeled data, our semi-supervised model also leverages a huge amounts of unlabeled text to automatically learn new “words”, and further constrains them by using a labeled data to segment non-standard texts such as those found in social networking services. Specifically, our hybrid model combines a discriminative classifier (CRF; Lafferty et al. (2001) and unsupervised word segmentation (NPYLM; Mochihashi et al. (2009)), with a transparent exchange of information between these two model structures within the semi-supervised framework (JESS-CM; Suzuki and Isozaki (2008)). We confirmed that it can appropriately segment non-standard texts like those in Twitter and Weibo and has nearly state-of-the-art accuracy on standard datasets in Japanese, Chinese, and Thai.

scholarly journals k-NDDP: An Efficient Anonymization Model for Social Network Data Release

Electronics ◽  
2021 ◽  
Vol 10 (19) ◽  
pp. 2440
Author(s):  
Shafaq Shakeel ◽  
Adeel Anjum ◽  
Alia Asheralieva ◽  
Masoom Alam
Keyword(s):  
Social Networking ◽  
Social Networking Sites ◽  
Privacy Protection ◽  
Differential Privacy ◽  
State Of The Art ◽  
Internet Technology ◽  
Social Data ◽  
Social Network Data ◽  
The Social ◽  
Vertex Degrees

With the evolution of Internet technology, social networking sites have gained a lot of popularity. People make new friends, share their interests, experiences in life, etc. With these activities on social sites, people generate a vast amount of data that is analyzed by third parties for various purposes. As such, publishing social data without protecting an individual’s private or confidential information can be dangerous. To provide privacy protection, this paper proposes a new degree anonymization approach k-NDDP, which extends the concept of k-anonymity and differential privacy based on Node DP for vertex degrees. In particular, this paper considers identity disclosures on social data. If the adversary efficiently obtains background knowledge about the victim’s degree and neighbor connections, it can re-identify its victim from the social data even if the user’s identity is removed. The contribution of this paper is twofold. First, a simple and, at the same time, effective method k–NDDP is proposed. The method is the extension of k-NMF, i.e., the state-of-the-art method to protect against mutual friend attack, to defend against identity disclosures by adding noise to the social data. Second, the achieved privacy using the concept of differential privacy is evaluated. An extensive empirical study shows that for different values of k, the divergence produced by k-NDDP for CC, BW and APL is not more than 0.8%, also added dummy links are 60% less, as compared to k-NMF approach, thereby it validates that the proposed k-NDDP approach provides strong privacy while maintaining the usefulness of data.

2D2LPI

2013 ◽  
Vol 3 (2) ◽  
pp. 17-31 ◽  
Author(s):  
S Manjunath ◽  
D S Guru ◽  
K B Nagasundara ◽  
M G Suraj
Keyword(s):  
State Of The Art ◽  
Video Summarization ◽  
New Method ◽  
Two Dimensional ◽  
Dimensional Approach ◽  
Person Identification ◽  
Locality Preserving ◽  
Single Instance ◽  
Feature Level Fusion ◽  
Level Fusion

In this paper, a new method of representing images called two directional two dimensional locality preserving indexing called 2D2LPI is presented. It is an extension of the two dimensional locality preserving indexing (2DLPI) method. The authors argue that the recently proposed 2DLPI reduces the dimensions of images in row direction and we propose an alternate way of reducing the dimension in column direction. Later the authors propose a method to reduce the size of an image both in row and column directions. To corroborate the efficacy of the proposed two directional two dimensional approach the authors design a model for person identification based on single instance of finger knuckle print and subsequently the authors propose a feature level fusion of multi-instance finger knuckle print for person identification. Also to study the suitability of the proposed approach on a different domain, a study on video summarization is also presented in this paper. The results of the proposed method are compared with that of the state of the art techniques such as 2D2PCA, 2D2LPP and it is found that the proposed 2D2LPI model is more competitive in terms of accuracy.

Biometrics

2003 ◽  
pp. 111-132
Author(s):  
Stewart T. Fleming
Keyword(s):  
State Of The Art ◽  
Computer Systems ◽  
Interactive Systems ◽  
End User ◽  
User Perceptions ◽  
Current State ◽  
Biometric Systems ◽  
System Designs

This chapter discusses the current state of the art of biometric systems. The use of biometrics is an important new part of the design of secure computer systems. However, many users view such systems with deep suspicion and many designers do not carefully consider the characteristics of biometrics in their system designs. This chapter aims to review the current state of the art in biometrics, to conduct detailed study of the available technologies and systems and to examine end-user perceptions of such systems. A framework is discussed that aims to establish guidelines for the design of interactive systems that include biometrics.

Sign in / Sign up

Export Citation Format

Share Document